Identification of an ABC Transporter Required for Iron Acquisition and Virulence in Mycobacterium tuberculosis
http://www.100md.com
细菌学杂志 2006年第1期
TB Center, The Public Health Research Institute at the International Center for Public Health, 225 Warren St., Newark, New Jersey 07103
ABSTRACT
Iron availability affects the course of tuberculosis infection, and the ability to acquire this metal is known to be essential for replication of Mycobacterium tuberculosis in human macrophages. M. tuberculosis overcomes iron deficiency by producing siderophores. The relevance of siderophore synthesis for iron acquisition by M. tuberculosis has been demonstrated, but the molecules involved in iron uptake are currently unknown. We have identified two genes (irtA and irtB) encoding an ABC transporter similar to the YbtPQ system involved in iron transport in Yersinia pestis. Inactivation of the irtAB system decreases the ability of M. tuberculosis to survive iron-deficient conditions. IrtA and -B do not participate in siderophore synthesis or secretion but are required for efficient utilization of iron from Fe-carboxymycobactin, as well as replication of M. tuberculosis in human macrophages and in mouse lungs. We postulate that IrtAB is a transporter of Fe-carboxymycobactin. The irtAB genes are located in a chromosomal region previously shown to contain genes regulated by iron and the major iron regulator IdeR. Taken together, our results and previous observations made by other groups regarding two other genes in this region indicate that this gene cluster is dedicated to siderophore synthesis and transport in M. tuberculosis.
INTRODUCTION
As is the case for most living organisms, Mycobacterium tuberculosis, the causative agent of tuberculosis (TB), requires iron as a cofactor for enzymes that are involved in redox reactions and other essential functions. Free iron, however, is not readily available in the mammalian host, since it is mainly kept in solution bound to transferrin, lactoferrin, and ferritin (24). Multiple lines of evidence indicate a critical role for iron acquisition in M. tuberculosis infection. It has been known for long that human serum is tuberculostatic, an effect that can be reversed by the addition of iron (14). More recent evidence obtained from gene expression studies indicates that M. tuberculosis faces iron limitation during growth in human macrophages and lungs (11, 21, 23), and a mutant laboratory strain affected in iron acquisition is attenuated for growth in human macrophages (6). Furthermore, iron availability is known to influence the severity of tuberculosis since abnormally high levels of iron in M. tuberculosis-infected humans and mice are associated with exacerbation of the disease (8). Understanding the process of iron acquisition in this pathogen is therefore highly relevant for the rational design of new ways to control TB.
To overcome iron deficiency, M. tuberculosis synthesizes a cell-associated siderophore (low-molecular-weight Fe+3 chelator) named mycobactin and a secreted one, carboxymycobactin, also known as exomycobactin (18). Although much has been learned about the synthesis and regulation of M. tuberculosis siderophores, the molecules involved in transport of iron into this pathogen remain unknown. In general, bacteria transport Fe(III)-siderophore complexes by a process that involves binding of the complex to specific receptor proteins on the cell surface and active translocation through the plasma membrane by an ABC transporter (3).
To prevent excess intracellular iron that can generate toxic oxygen radicals, expression of genes encoding iron uptake systems is tightly regulated by iron and transcriptional repressors. Our previous studies characterized the iron-responsive changes in gene expression in M. tuberculosis wild type and a mutant of IdeR, the major repressor of iron acquisition genes (20). The gene cluster that includes Rv1344 to Rv1349 was identified in those studies as being repressed by iron and by IdeR. A schematic representation of this cluster including the position of putative IdeR binding sites is shown in Fig. 1. According to the TubercuList web site (genolist.pasteur.fr /TubercuList) Rv1344 encodes a probable acyl-carrier protein and Rv1346 protein is a possible acyl-coenzyme A dehydrogenase (FadE14). Rv1345 and Rv1347 are annotated to encode proteins of unknown function; however, recent studies suggest that the products of these genes might participate in siderophore synthesis (1, 5). The last two genes in this cluster, Rv1348 and Rv1349, encode an ABC transporter (2) highly similar to the YbtPQ system of Yersinia pestis (7). We investigated here the role of this putative ABC transporter in iron acquisition and virulence in M. tuberculosis. Our findings demonstrate that RV1348 and Rv1349 are part of the iron acquisition machinery of M. tuberculosis and are required for maximal survival in iron-deficient conditions in vitro and in vivo in the mouse model of infection.
MATERIALS AND METHODS
Bacteria, plasmids, media, and growth conditions. Escherichia coli JM109 cultures were routinely grown in Luria-Bertani broth or agar medium at 37°C and routinely used in DNA cloning procedures. M. tuberculosis H37Rv was obtained from American Type Culture Collection. The siderophore-deficient mbtB mutant strain (6) was obtained from Clifton E. Barry III at the National Institute of Allergy and Infectious Disease, Rockville, Md. M. tuberculosis strains were maintained in Middlebrook 7H9 broth or on 7H10 agar (Difco), supplemented with 0.2% glycerol, 0.05% Tween 80, and 10% albumin-dextrose-NaCl complex (13). Antibiotics when required were included at the following concentrations: kanamycin (Kan) at 20 μg/ml, streptomycin (Str) at 20 μg/ml, and hygromycin (Hyg) at 150 μg/ml. When indicated, the iron chelator 2'-dipyridyl (DPI) was added at a final concentration of 75 μM.
For M. tuberculosis growth in low iron medium, we used a defined medium (MM) containing 0.5% (wt/vol) asparagine, 0.5% (wt/vol) KH2PO4, 2% glycerol, 0.05% Tween 80, and 10% albumin-dextrose-NaCl complex. The pH was adjusted to 6.8. To lower the trace metal contamination, the medium was treated with Chelex-100 (Bio-Rad Laboratories, Hercules, Calif.) according to the manufacturer's instructions. Chelex was removed by filtration and, before use, the medium was supplemented with 0.5 mg of ZnCl2, 0.1 mg of MnSO4, and 40 mg of MgSO4 liter–1 and the desired concentration of FeCl3.
Plasmid construction and DNA manipulation. For the inactivation of Rv1348 and Rv1349, PCR fragments spanning Rv1348 or Rv1349, respectively, were amplified from M. tuberculosis H37Rv genomic DNA. PCR was carried out by using Pfu Turbo polymerase (Stratagene, La Jolla, Calif.). The oligonucleotides primers (supplied by Integrated DNA Technologies) were as follows: For Rv1348, 5'-AGCGGATGTGGGTTTGGT-3' (forward) and 5'-GCGACAACGGAACAAAAC-3' (reverse); and For Rv1349, 5'-TACGCACGGGACTTCTGG-3' (forward) and 5'- GCCGCTGAGTAGTTGGTT-3' (reverse). PCR products were isolated from agarose gels and cloned into pCR-Blunt TOPO vector (Invitrogen Life Technologies). Constructs were verified as correct by sequencing.
A Hyg resistance cassette was introduced at the unique PmlI site in Rv1348, and the resulting Rv1348::Hyg recombinant fragment was inserted at the NdeI-XbaI of pSM270 (16), a suicide vector that carries sacB and an Str resistance cassette in the plasmid backbone, generating pSM533. A Kan resistance cassette was inserted into the unique PmlI site of Rv1349, and the resulting Rv1349::Kan recombinant fragment was inserted at the NdeI-XbaI of pSM270, generating pSM425.
The complementing plasmid pSM546 was generated by cloning a fragment containing Rv1348-1349 and the promoter region upstream of Rv1348, into vector pYub178 (17), which carries a Kan resistance cassette and the L5 integrase and attachment site (attP). All constructs were verified by sequencing.
Generation of irtAB mutants and complemented strain. M. tuberculosis mutants were generated by using a two-step recombination protocol with a sucrose counter selection. Plasmids PSM533 and PSM425 were electroporated into M. tuberculosis, and recombinants in which the plasmid has integrated by a single crossover were selected by plating on 7H10 plates containing Str and Hyg in the case of PSM533 or Str and Kan in the case of PSM425 transformants. The single crossover at the homologous region was confirmed by Southern blot analysis. A Hygr Strr recombinant from transformation with PSM533 was amplified in the presence of Hyg and plated on 7H10 containing Hyg and 8% sucrose (Suc). Similarly, a Kanr Strr recombinant from transformation with PSM425 was amplified in the presence of Kan and plated on Kan-Suc plates. In each case Hygr Sucr or Kanr Sucr colonies were tested for loss of the plasmid sequences as a result of a second crossover by plating on Str-containing medium. Kanr Sucr Strs and Hygr Sucr Strs colonies from each transformation were analyzed by Southern blot to confirm the allelic exchange. One transformant in which allelic exchange of Rv1349 was confirmed was named ST69, whereas the mutant strain in which allelic replacement of Rv1348 was confirmed was named ST73.
To generate the complemented strain ST96, plasmid pSM546 was electroporated into the mutant strain ST73 and recombinants were selected by plating in 7H10 containing Kan. Integration of pSM546 at the attB site in Kanr colonies was confirmed by Southern blot analysis.
Mycobactin determination. Mycobacterial strains were grown to mid-logarithmic phase in 7H9 medium, and 0.7 ml of culture was spread onto MM agar containing the indicated concentrations of FeCl3. After incubation at 37°C for 10 days, bacteria were scraped from the plate. Subsequently, mycobactin was extracted in ethanol and chloroform and quantified as previously described (20).
Cross-feeding experiments. A logarithmic culture of the mbtB strain grown in 7H10 medium was used to inoculate MM to an optical density (OD) at 540 nm of 0.3. Then, 0.5 ml of this bacterial suspension was mixed with 2.5 ml of MM, and MM supplemented with 2 μM FeCl3 or 2 μM FeCl3 was added to 2.5 ml of the culture filtrate from H37Rv or ST73 cultures grown to the same OD in MM containing no iron. Growth of the mbtB mutant was monitored by measuring the OD.
Carboxymycobactin utilization assay. High-pressure liquid chromatography purified Fe-carboxymycobactin extracted from a chloroform extract of a low-iron culture filtrate of M. tuberculosis Erdman strain (10) was kindly provided by Marcus A. Horwitz at the Department of Medicine, School of Medicine, University of California, Los Angeles. H37Rv, ST73, and ST96 were grown in MM from an OD of 0.05 to 0.6 and then diluted in MM to an OD of 0.1. After 2 days, the cultures were diluted again to an OD of 0.1 in MM containing 5 ng of Fe-carboxymycobactin/ml. The growth of each strain was monitored every day by measuring the OD.
THP-1 infections. M. tuberculosis infections of THP-1-derived macrophages were performed as previously described on (16). Briefly, THP-1 cells were grown in RPMI 1640, supplemented with 0.45% glucose, 0.15% sodium pyruvate, and 4 mM L-glutamine. THP-1 cells were induced to differentiate into macrophages by treatment with 50 nM 12-tetradecanoylphorbol-13-acetate for 24 h. A total of 7.5 x 104 cells per well were incubated for 2 h at 37°C with a bacterial suspension prepared from a logarithmic growing liquid culture of each M. tuberculosis strain at a multiplicity of infection of 1:15 CFU per macrophage. After 2 h the medium was removed, and the cells were washed twice with warm phosphate-buffered saline to remove any residual extracellular bacteria. Next, 100 μl of fresh RPMI was added to each well, and the plate was incubated at 37°C. At the indicated time points, the medium was removed from three wells, the macrophages were lysed with 100 μl of 0.05% sodium dodecyl sulfate, and dilutions of the released intracellular bacteria were plated on 7H10 to determine the CFU.
Mouse aerosol infection. For each strain tested, a 10-ml bacterial suspension of 106 bacilli ml–1 in saline containing 0.04% Tween 80 was used. Aerosols were generated with a Lovelace Nebulizer (In-Tox Products, Albuquerque, NM), and animals were exposed to the aerosol for 30 min. Under these conditions the number of microorganisms detected in the lungs at time zero was ca. 100. At the indicated time points after infection, three mice were sacrificed, and their lungs were removed and homogenized in phosphate-buffered saline-Tween 80. Dilutions of the homogenates were plated on 7H10 agar to determine the CFU.
RESULTS
Sequence analysis. The M. tuberculosis Rv1348 and Rv1349 encode proteins of 859 and 579 amino acids, respectively. There is no intergenic sequence between irtA and irtB, suggesting that they are cotranscribed. A predicted IdeR binding site is located at position –212 upstream of the annotated translational start site for Rv1348 (11). The proteins encoded by Rv1348 and Rv1349 share homology with each other and with members of the ATP binding cassette(ABC) transporter family. Both proteins contain an amino-terminal membrane-spanning domain with six predicted transmembrane helices in Rv1349 and five to seven possible transmembrane helices in Rv1348, fused to a nucleotide-binding domain. Characteristic motifs shared by members of the ABC transporter family (Walker A, ABC signature, Walker B, and Linton and Higgins) have been identified in the carboxy-terminal domain of these proteins (2). Based on these information, we have named these genes and the proteins encoded by them IrtA and IrtB, respectively, for iron-regulated transporters A and B. Since ABC transporters consist of two membrane-spanning domains associated with two cytoplasmic nucleotide binding domains, IrtA and -B are predicted to form a heterodimeric ABC transporter. IrtAB is similar to the Y. pestis YbtPQ transporter, as first noticed by Fetherson et al. (7). Homologs of YbtPQ have also been found in Yersinia enterocolitica (irp6-7) (4) and Corynebacterium diphtheriae (CdtPQ) (15). IrtA shows 46% similarity to YbtP. This similarity is accentuated at the carboxy-terminal end, whereas at the amino-terminal IrtA has an extension of about 292 amino acids not present in YbtP and predicted by computer algorithms (TMHMM and MEMSAT) to be exposed to the outside environment. IrtB shows 46% similarity to YbtQ and 34% identity with the last 578 amino acids of IrtA.
Generation of M. tuberculosis IrtAB mutants. YbtPQ, Irp6-7, and CdtPQ are required for iron transport in Y. pestis, Y. enterocolitica, and C. diphtheriae, respectively (4, 7, 15). The homology between irtA and irtB with these other bacterial iron transport systems prompted us to examine their role in M. tuberculosis iron transport. With that purpose, two mutants strains were created by homologous recombination and allelic replacement: one in which Rv1349 was disrupted by introduction of a Kan resistance cassette (ST69) and another (ST73) in which a Hyg resistance cassette was inserted in Rv1348, as described in Materials and Methods. Since Rv1348 and Rv1349 are organized as an operon with the 5' terminus of Rv1349 overlapping the 3' terminus of Rv1348 by 3 bp, the mutation in ST73 is presumed to be polar, affecting both genes. Gene replacement was confirmed by Southern blot analysis (data not shown). The colony morphology and growth properties of ST69 and ST73 in 7H9 or 7H10 medium were no different from those of the wild-type strain, but exponentially growing cultures of ST69 and ST73 on agar plates exhibited a light orange pigment not observed in the wild-type strain (data not shown).
IrtA and IrtB are required for growth of M. tuberculosis in iron-deficient conditions. The role of IrtAB in survival of M. tuberculosis under iron depletion was evaluated by examining the ability of the mutant strains ST73 and ST69 to grow in 7H9 medium in the presence of the iron chelator DPI. As shown in Fig. 2A to C, inactivation of irtB alone (ST69) or of irtA and irtB (ST73) does not affect growth of M. tuberculosis in iron-sufficient conditions, but under iron-deficient conditions both mutant strains show a growth defect. This defect is more pronounced in the double mutant, suggesting that IrtA alone can partially function. At this point we decided to further analyze the phenotypes of the double-mutant strain ST73. ST73 was complemented with a single copy of irtA and irtB under the control of their own promoter. For this purpose, the integrative plasmid PSM546 was electrophorated into ST73, resulting in strain ST96. Restoring expression of irtA and irtB allows strain ST96 to survive low iron conditions to an extent similar to the wild-type strain, confirming that irtA and irtB are required for normal replication of M. tuberculosis under iron depletion (Fig. 2D). Complementation of ST73 with a plasmid containing only irtA restored growth to the same extent as that shown by the irtB mutant (ST69) (data not shown).
Siderophore production in the irtAB mutant. IrtA and IrtB have fused membrane spanning and ATPase domains, a feature most commonly found in ABC transporters that function as exporters (2). Therefore, we considered the possibility that IrtA and IrtB could be involved in siderophore secretion. A deficiency in siderophore production or secretion would explain the inability of the ST73 mutant to overcome iron deficiency. To test this possibility, we first measured mycobactin production in H37Rv and in ST73 cultured in MM agar containing increasing amounts of FeCl3. ST73 produces comparable amounts of mycobactin as does the wild-type strain in low-iron conditions, indicating that the irtAB mutation does not affect mycobactin synthesis (Fig. 3A). However, mycobactin production is not repressed in the mutant strain as efficiently as in the wild type, and even at the highest concentration tested (50 μM FeCl3) ST73 produces about twice as much mycobactin as the wild-type strain (Fig. 3A). This result suggests that accumulation of intracellular iron levels that signal repression of mycobactin synthesis is less effective in the irtAB mutant strain than in the wild type. Consistent with this interpretation is the observation that irg-1 a previously described iron-repressed gene (19) is expressed at higher levels in the irtAB mutant cultured in high-iron medium (data not shown).
To evaluate production and secretion of carboxymycobactins in ST73, we used a biological assay. The M. tuberculosis strain with a mutation in the mbtB gene does not produce mycobactin or carboxymycobactin and is unable to grow in a low-iron medium (6) (Fig. 3B). However, this strain does grow under those conditions when supplied with an exogenous source of Fe-carboxymycobactin, purified or in the culture filtrate of a carboxymycobactin-producing strain grown in low-iron conditions (unpublished observations). When tested in this assay, the culture filtrate of ST73 was able to support growth of the mbtB strain to the same extent as the culture supernatant obtained from the same number of wild-type bacteria (Fig. 3B). Based on this result and those obtained from mycobactin measurements, we conclude that inactivation of irtA and irtB does not affect siderophore production or secretion in M. tuberculosis.
Fe-carboxymycobactin utilization. In order to test whether IrtAB are involved in iron acquisition from Fe-carboxymycobactin, we examined the effect of the irtAB mutation on the ability of M. tuberculosis to grow in a medium containing Fe-carboxymycobactin as a sole iron source. For this purpose, the mycobacterial strains were pregrown without iron to exhaust intracellular iron reserves and then subcultured in an iron-depleted medium supplemented with Fe-carboxymycobactin. As shown in Fig. 4, the growth of ST73 was significantly limited compared to the wild-type and complemented strains, indicating that the mutation has compromised the ability of this strain to utilize Fe-carboxymycobactin as an iron source.
Effect of the irtAB mutation on replication of M. tuberculosis in macrophages. Macrophages provide an iron-limiting environment for M. tuberculosis (11, 21) and siderophore-mediated iron acquisition is required for efficient multiplication of the bacilli in these cells (6). The effect of inactivation of irtA and irtB on the ability of M. tuberculosis to replicate in macrophages was tested in THP-1 cells. The growth of each strain was determined, as described in Materials and Methods, by the number of CFU obtained at various times after infection. Compared to the wild-type and complemented strains, the irtAB mutant is significantly impeded in the ability to multiply in human macrophages, since on day 7 after infection there is ca. 100 times less mutant than wild-type or complemented mutant bacteria (Fig. 5). This result demonstrates that IrtAB is necessary for normal multiplication of M. tuberculosis in human macrophages.
Effect of irtAB mutation on replication of M. tuberculosis in mice. The effect of the irtAB mutation in virulence of M. tuberculosis was examined in the mouse model. C57B/6 mice were aerosol infected so that ca. 100 CFU of H37Rv, ST73, or ST96 were detected at time zero after infection. At the indicated time points postinfection the mice were sacrificed, and the CFU in the lungs were determined. As shown in Fig. 6 the irtAB mutant is defective in the ability to replicate in mice lungs. This ability was significantly although not completely restored by reintroducing irtAB in the complemented strain.
DISCUSSION
Iron availability during M. tuberculosis infection is a critical factor that influences the course of TB. Altering the ability of this pathogen to acquire iron is likely to profoundly affect the outcome of this infection. In order to develop ways to interfere with iron acquisition of M. tuberculosis, a better understanding of this process is required. In the present study we identified two genes, irtA and irtB, encoding proteins involved in iron acquisition in this pathogenic mycobacterium. The IrtAB system is encoded in a region of the M. tuberculosis chromosome regulated by iron and IdeR (20) that appears to be dedicated to the synthesis and utilization of siderophores. IrtA and IrtB have the motifs typical of ABC transporters, and they closely resemble the Y. pestis iron transporter YbtPQ. IrtA and IrtB are similar to each other, and both have membrane-spanning domains fused to an ATPase domain. This is a feature shared with the YbtPQ system but uncommon among transporters that function as importers which usually have these two domains in different polypeptides (2). ABC transporters are composed of four structural domains two membrane-spanning domains and two cytoplasmic domains containing the ATP binding cassette (2). Therefore, IrtA and IrtB, as is the case with YbtP and YbtQ, are predicted to function as a heterodimer.
Inactivation of irtA and irtB has no effect on growth of M. tuberculosis in high-iron conditions, but it does affect the ability of this bacterium to multiply under iron-deficient conditions (Fig. 2). Inactivation of irtB alone results in a growth defect in low-iron medium that is less severe than the one exhibited by a double irtAB mutant (Fig. 2B and C). This suggests that in the absence of IrtB, IrtA can partially function, forming homodimers or possibly associating with another protein. The limited growth of the irtAB mutant in iron-deficient conditions is probably sustained by additional iron transport systems of lower affinity than the IrtAB system. Similarly, YbtPQ mutants of Y. pestis retain some ability to replicate in low-iron conditions.
The effect of the irtAB mutation on growth under iron deficiency is not due to an effect on siderophore synthesis since this strain continues to produce and secrete siderophores normally (Fig. 3). However, the iron concentration in the medium required to repress mycobactin synthesis and possibly carboxymycobactin production, since they follow a common synthesis pathway, was higher in the irtAB mutant than in the wild-type strain (Fig. 3A). Since there is no evidence suggesting altered IdeR function in the irtAB mutant, we believe that partial derepression of mycobactin synthesis is indicative of lower intracellular iron levels in this strain. Derepression of mycobactin synthesis and formation of Fe-mycobactin complexes on the surface of bacteria cultured in high-iron medium is likely the cause of the orange pigmentation shown by ST69 and ST73. We have observed a similar pigmentation in other iron-deficient strains that also exhibit derepressed mycobactin synthesis (20).
The current model for iron transport in M. tuberculosis suggests a transfer of iron from Fe-carboxymycobactin to mycobactin in the cell surface. Indeed, this transfer of iron can occur (9), but its significance is not clear. The results presented here indicate that mycobactin is not required for uptake of iron from Fe-carboxymycobactin since the mbtB mutant strain, which does not make either of the siderophores, can be fed iron from Fe-carboxymycobactin provided exogenously (Fig. 3B). Thus, an uptake pathway for Fe-carboxymycobactin can function independently of mycobactin. The exact contribution of each siderophore to iron transport in M. tuberculosis awaits the isolation of mutants that are defective in one but not the other siderophore. Inactivation of irtAB renders the mutant unable to efficiently utilize Fe-carboxymycobactin as an iron source, indicating that IrtA and IrtB are required for the uptake of iron from carboxymycobactin (Fig. 4). Based on this result we predict that IrtAB transports Fe-carboxymycobactin complexes into the cytoplasm. However, we should point out that the experiments required to demonstrate Fe-carboxymycobactin internalization in M. tuberculosis have not been conducted. In an early study, incorporation of Fe55 from Fe55-carboxymycobactin complexes into M. bovis was found not to be affected by energy poisons or uncouplers of ATP biosynthesis (22). This observation would apparently argue against the role of an active transporter for Fe-carboxymycobactin in slow-growing mycobacteria. However, in that study the authors did not test the effect of energy poisons on the intracellular iron pool but only on the detected radioactivity associated with whole cells. We believe that under these conditions, adsorption of iron on the cell envelope, including iron transferred from Fe-carboxymycobactin to mycobactin, in a non-energy-requiring process, could mask the effect of energy inhibitors on an active transport process into the cytoplasm. Additional studies to address the energy requirements for transport of iron into M. tuberculosis and the identity of the substrate transported by IrtAB are in progress.
ABC transporters that function as importers generally require a substrate-binding protein (SBP), as well as a translocator in the cytoplasmic membrane. This SBP is located in the periplasm in gram-negative bacteria or is a membrane-bound lipoprotein in gram-positive bacteria (2). In addition, transport in gram-negative bacteria requires an outer membrane receptor that binds the substrate and, using the energy transduced by the tonB system, translocates the complex into the periplasm (3). In gram-positive bacteria the SBP serves also as receptor. No SBP is encoded in the chromosomal region containing irtAB. The same is true for the YbtPQ system of Y. pestis. It is possible that the SBPs used by these systems are encoded elsewhere in the chromosome or that they do not use a typical SBP. Interestingly, a new class of chimeric ABC transporters with fused extracytoplasmic substrate-binding sites has been reported (12). Considering that IrtA has a 292-amino-acid N-terminal domain that is not found in YbtP or in its reported homologs and is predicted to be extracytoplasmic, it is tempting to speculate that it could be a substrate-binding domain. Future studies will address this possibility.
We propose a model of iron transport in which Fe-carboxymycobactin complexes traverse the cell envelope either by diffusion (given its partial hydrophobic character) or, with the aid of a porin, a typical SBP or the amino-terminal end of IrtA and then are translocated into the cytoplasm by IrtAB. Since irtAB mutants still show some growth in low-iron conditions, redundant pathways for Fe-carboxymycobactin utilization must also exist.
The IrtAB system is important for replication not only under iron-deficient conditions in vitro but also in human macrophages and mouse lungs, as shown by attenuation of the mutant strain compared to the wild-type and complemented strains in these infection models (Fig. 5 and 6). The growth defect of the irtAB mutant in mice was not completely complemented, a result that can be due to differences in expression of these genes in the chromosomal context in which the complementing plasmid has integrated. Attenuation of the irtAB mutant in mice shows, for the first time, a direct correlation between the ability to efficiently acquire iron and the capacity to replicate in an in vivo model of tuberculosis infection. The mbtB mutant strain was shown to be attenuated in human macrophages, but its phenotype in mice was not reported.
The work described here documents the first M. tuberculosis iron transporter. Future studies will characterize this transporter at a molecular level to provide information that can be applied for development of antitubercular agents affecting iron uptake.
ACKNOWLEDGMENTS
We are very grateful to Marcus A. Horwitz for providing the purified Fe-carboxymycobactin used in this study and to Clifton E. Barry III and Laura Via for the mbtB mutant strain. We thank Irina Kolesnikova and Jeanie Dubnau for assistance with the mouse infections.
This study was supported by NIH grant AI44856 (I.S.) and awards from the Francis Parker Foundation for Pulmonary Research and the Center for AIDS Research (G.M.R.).
REFERENCES
Babbette, B., D. LaMarca, W. Zhu, J. E. L. Arceneaux, B. R. Byers, and M. D. Lundrigan. 2004. Participation of fad and mbt genes in synthesis of mycobactin in Mycobacterium smegmatis. J. Bacteriol. 186:374-382.
Braibant, M., P. Gilot, and J. Content. 2000. The ATP binding cassette (ABC) transport systems of Mycobacterium tuberculosis. FEBS Microbiol. Rev. 24:449-467.
Braun, V., and H. Killmann. 1999. Bacterial solutions to the iron-supply problem. Trends Biochem. Sci. 24:104-109.
Brem, D., C. Pelludat, C. Rakin, C. A. Jacobi, and J. Heesemann. 2001. Functional analysis of yersiniabactin transport genes of Yersinia enterocolitica. Microbiology 147:1115-1127.
Card, G. L., N. A. Peterson, C. A. Smith, B. Rupp, B. M. Schick, and E. N. Baker. 2005. The crystal structure of Rv1347c, a putative antibiotic resistance protein from Mycobacterium tuberculosis, reveals a GCN5-related fold and suggest an alternative function in siderophore biosynthesis. J. Biol. Chem. 280:13978-13986.
De Voss, J. J., K. Rutter, B. G. Schroeder, H. Su, Y. Zhu, and C. E. Barry III. 2000. The salicylate-derived mycobactin siderophores of Mycobacterium tuberculosis are essential for growth in macrophages. Proc. Natl. Acad. Sci. USA 97:1252-1257.
Fetherson, J. D., V. J. Bertolino, and R. D. Perry. 1999. YbtP and YbtQ two ABC transporters required for iron uptake in Yersinia pestis. Mol. Microbiol. 32:289-299.
Gangaidzo, I. T., M. V. Moyo, E. Mvundura, G. Aggrey, N. L. Murphree, H. Khumalo, T. Saungweme, I. Kasuosve, I. Kasvoesve, Z. A. Gomo, T. Roualt, J. R. Boelaert, and V. R. Gordeuk. 2001. Association of pulmonary tuberculosis with increased dietary iron. J. Infect. Dis. 184:936-939.
Gobin, J., and M. Horwitz. 1996. Exochelins of Mycobacterium tuberculosis remove iron from human iron-binding proteins and donate iron to mycobactins in the M. tuberculosis cell wall. J. Exp. Med. 183:1527-1532.
Gobin, J., C. H. Moore, J. J. R. Reeve, D. K. Wong, B. W. Gibson, and M. A. Horwitz. 1995. Iron acquisition by Mycobacterium tuberculosis: isolation and characterization of a family of iron-binding exochelins. Proc. Natl. Acad. Sci. USA 92:5189-5193.
Gold, B., G. M. Rodriguez, M. P. Marras, M. Pentecost, and I. Smith. 2001. The Mycobacterium tuberculosis IdeR is a dual functional regulator that controls transcription of genes involved in iron acquisition, iron storage, and survival in macrophages. Mol. Microbiol. 42:851-865.
Heide, T. V. D., and B. Poolman. 2002. ABC transporters: one, two, or four extracytoplasmic substrate-binding sites EMBO Rep. 3:938-943.
Jacobs, Jr., W. R., G. V. Kalpana, J. D. Cirillo, L. Pascopella, S. B. Snapper, R. A. Udani, et al. 1991. Genetic systems for mycobacteria. Methods Enzymol. 204:537-555.
Kochan, I. 1971. Mechanisms of tuberculostasis in mammalian serum: role of transferrin in human serum tuberculostasis. J. Infect. Dis. 119:11-18.
Kunkle, C. A., and a. M. P. Schmitt. 2003. Analysis of the Corynebacterium diphtheriae DtxR regulon: identification of a putative siderophore synthesis and transport system that is similar to the Yersinia high-pathogenicity-island-encoded yersiniabactin synthesis and uptake system. J. Bacteriol. 185:6826-6840.
Manganelli, R., M. I. Voskuil, G. K. Schoolnik, and I. Smith. 2001. The Mycobacterium tuberculosis ECF sigma factor E role in global gene expression and survival in macrophages. Mol. Microbiol. 41:423-437.
Pascopella, L., F. M. Collins, J. M. Martin, W. R. Jacobs, Jr., and B. R. Bloom. 1993. Identification of a genomic fragment of Mycobacterium tuberculosis responsible for in vivo growth advantage. Infect. Agents Dis. 2:282-284.
Ratledge, C. 1999. Iron metabolism, p. 260-286. In C. Ratledge and J. Dale (ed.), Mycobacteria: molecular biology and virulence. Blackwell Science Publishers, London, United Kingdom.
Rodriguez, G. M., B. Gold, M. Gomez, O. Dussurget, and I. Smith. 1998. Identification and characterization of two divergently transcribed iron regulated genes in Mycobacterium tuberculosis. Tuberc. Lung Dis. 79:287-298.
Rodriguez, G. M., M. I. Voskuil, B. Gold, G. K. Schoolnik, and I. Smith. 2002. ideR, An essential gene in Mycobacterium tuberculosis: role of IdeR in iron-dependent gene expression, iron metabolism, and oxidative stress response. Infect. Immun. 70:3371-3381.
Schnappinger, D., S. Ehrt, M. I. Voskuil, Y. Liu, J. A. Mangan, I. M. Monahan, G. Dolganov, B. Efron, P. D. Butcher, C. Nathan, and G. K. Schoolnik. 2003. Transcriptional adaptation of Mycobacterium tuberculosis within macrophages: insights into the phagosomal environment. J. Exp. Med. 198:693-704.
Stephenson, M. C., and C. Ratledge. 1980. Specificity of exochelins for iron transport in three species of mycobacteria. J. Gen. Microbiol. 116:521-523.
Timm, J., F. A. Post, L. G. Bekker, G. B. Walther, C. H. Wainwright, R. Manganelli, W. T. Chan, L. Tsenova, B. Gold, I. Smith, G. Kaplan, and J. D. Mckinney. 2003. Differential expression of iron-, carbon-, and oxygen-responsive mycobacterial genes in the lungs of chronicle infected mice and tuberculosis patients. Proc. Natl. Acad. Sci. USA 100:14321-14326.
Weinberg, E. D. 1984. Iron withholding: a defense against infection and neoplasia. Physiol. Rev. 64:65-102.(G. Marcela Rodriguezand I)
ABSTRACT
Iron availability affects the course of tuberculosis infection, and the ability to acquire this metal is known to be essential for replication of Mycobacterium tuberculosis in human macrophages. M. tuberculosis overcomes iron deficiency by producing siderophores. The relevance of siderophore synthesis for iron acquisition by M. tuberculosis has been demonstrated, but the molecules involved in iron uptake are currently unknown. We have identified two genes (irtA and irtB) encoding an ABC transporter similar to the YbtPQ system involved in iron transport in Yersinia pestis. Inactivation of the irtAB system decreases the ability of M. tuberculosis to survive iron-deficient conditions. IrtA and -B do not participate in siderophore synthesis or secretion but are required for efficient utilization of iron from Fe-carboxymycobactin, as well as replication of M. tuberculosis in human macrophages and in mouse lungs. We postulate that IrtAB is a transporter of Fe-carboxymycobactin. The irtAB genes are located in a chromosomal region previously shown to contain genes regulated by iron and the major iron regulator IdeR. Taken together, our results and previous observations made by other groups regarding two other genes in this region indicate that this gene cluster is dedicated to siderophore synthesis and transport in M. tuberculosis.
INTRODUCTION
As is the case for most living organisms, Mycobacterium tuberculosis, the causative agent of tuberculosis (TB), requires iron as a cofactor for enzymes that are involved in redox reactions and other essential functions. Free iron, however, is not readily available in the mammalian host, since it is mainly kept in solution bound to transferrin, lactoferrin, and ferritin (24). Multiple lines of evidence indicate a critical role for iron acquisition in M. tuberculosis infection. It has been known for long that human serum is tuberculostatic, an effect that can be reversed by the addition of iron (14). More recent evidence obtained from gene expression studies indicates that M. tuberculosis faces iron limitation during growth in human macrophages and lungs (11, 21, 23), and a mutant laboratory strain affected in iron acquisition is attenuated for growth in human macrophages (6). Furthermore, iron availability is known to influence the severity of tuberculosis since abnormally high levels of iron in M. tuberculosis-infected humans and mice are associated with exacerbation of the disease (8). Understanding the process of iron acquisition in this pathogen is therefore highly relevant for the rational design of new ways to control TB.
To overcome iron deficiency, M. tuberculosis synthesizes a cell-associated siderophore (low-molecular-weight Fe+3 chelator) named mycobactin and a secreted one, carboxymycobactin, also known as exomycobactin (18). Although much has been learned about the synthesis and regulation of M. tuberculosis siderophores, the molecules involved in transport of iron into this pathogen remain unknown. In general, bacteria transport Fe(III)-siderophore complexes by a process that involves binding of the complex to specific receptor proteins on the cell surface and active translocation through the plasma membrane by an ABC transporter (3).
To prevent excess intracellular iron that can generate toxic oxygen radicals, expression of genes encoding iron uptake systems is tightly regulated by iron and transcriptional repressors. Our previous studies characterized the iron-responsive changes in gene expression in M. tuberculosis wild type and a mutant of IdeR, the major repressor of iron acquisition genes (20). The gene cluster that includes Rv1344 to Rv1349 was identified in those studies as being repressed by iron and by IdeR. A schematic representation of this cluster including the position of putative IdeR binding sites is shown in Fig. 1. According to the TubercuList web site (genolist.pasteur.fr /TubercuList) Rv1344 encodes a probable acyl-carrier protein and Rv1346 protein is a possible acyl-coenzyme A dehydrogenase (FadE14). Rv1345 and Rv1347 are annotated to encode proteins of unknown function; however, recent studies suggest that the products of these genes might participate in siderophore synthesis (1, 5). The last two genes in this cluster, Rv1348 and Rv1349, encode an ABC transporter (2) highly similar to the YbtPQ system of Yersinia pestis (7). We investigated here the role of this putative ABC transporter in iron acquisition and virulence in M. tuberculosis. Our findings demonstrate that RV1348 and Rv1349 are part of the iron acquisition machinery of M. tuberculosis and are required for maximal survival in iron-deficient conditions in vitro and in vivo in the mouse model of infection.
MATERIALS AND METHODS
Bacteria, plasmids, media, and growth conditions. Escherichia coli JM109 cultures were routinely grown in Luria-Bertani broth or agar medium at 37°C and routinely used in DNA cloning procedures. M. tuberculosis H37Rv was obtained from American Type Culture Collection. The siderophore-deficient mbtB mutant strain (6) was obtained from Clifton E. Barry III at the National Institute of Allergy and Infectious Disease, Rockville, Md. M. tuberculosis strains were maintained in Middlebrook 7H9 broth or on 7H10 agar (Difco), supplemented with 0.2% glycerol, 0.05% Tween 80, and 10% albumin-dextrose-NaCl complex (13). Antibiotics when required were included at the following concentrations: kanamycin (Kan) at 20 μg/ml, streptomycin (Str) at 20 μg/ml, and hygromycin (Hyg) at 150 μg/ml. When indicated, the iron chelator 2'-dipyridyl (DPI) was added at a final concentration of 75 μM.
For M. tuberculosis growth in low iron medium, we used a defined medium (MM) containing 0.5% (wt/vol) asparagine, 0.5% (wt/vol) KH2PO4, 2% glycerol, 0.05% Tween 80, and 10% albumin-dextrose-NaCl complex. The pH was adjusted to 6.8. To lower the trace metal contamination, the medium was treated with Chelex-100 (Bio-Rad Laboratories, Hercules, Calif.) according to the manufacturer's instructions. Chelex was removed by filtration and, before use, the medium was supplemented with 0.5 mg of ZnCl2, 0.1 mg of MnSO4, and 40 mg of MgSO4 liter–1 and the desired concentration of FeCl3.
Plasmid construction and DNA manipulation. For the inactivation of Rv1348 and Rv1349, PCR fragments spanning Rv1348 or Rv1349, respectively, were amplified from M. tuberculosis H37Rv genomic DNA. PCR was carried out by using Pfu Turbo polymerase (Stratagene, La Jolla, Calif.). The oligonucleotides primers (supplied by Integrated DNA Technologies) were as follows: For Rv1348, 5'-AGCGGATGTGGGTTTGGT-3' (forward) and 5'-GCGACAACGGAACAAAAC-3' (reverse); and For Rv1349, 5'-TACGCACGGGACTTCTGG-3' (forward) and 5'- GCCGCTGAGTAGTTGGTT-3' (reverse). PCR products were isolated from agarose gels and cloned into pCR-Blunt TOPO vector (Invitrogen Life Technologies). Constructs were verified as correct by sequencing.
A Hyg resistance cassette was introduced at the unique PmlI site in Rv1348, and the resulting Rv1348::Hyg recombinant fragment was inserted at the NdeI-XbaI of pSM270 (16), a suicide vector that carries sacB and an Str resistance cassette in the plasmid backbone, generating pSM533. A Kan resistance cassette was inserted into the unique PmlI site of Rv1349, and the resulting Rv1349::Kan recombinant fragment was inserted at the NdeI-XbaI of pSM270, generating pSM425.
The complementing plasmid pSM546 was generated by cloning a fragment containing Rv1348-1349 and the promoter region upstream of Rv1348, into vector pYub178 (17), which carries a Kan resistance cassette and the L5 integrase and attachment site (attP). All constructs were verified by sequencing.
Generation of irtAB mutants and complemented strain. M. tuberculosis mutants were generated by using a two-step recombination protocol with a sucrose counter selection. Plasmids PSM533 and PSM425 were electroporated into M. tuberculosis, and recombinants in which the plasmid has integrated by a single crossover were selected by plating on 7H10 plates containing Str and Hyg in the case of PSM533 or Str and Kan in the case of PSM425 transformants. The single crossover at the homologous region was confirmed by Southern blot analysis. A Hygr Strr recombinant from transformation with PSM533 was amplified in the presence of Hyg and plated on 7H10 containing Hyg and 8% sucrose (Suc). Similarly, a Kanr Strr recombinant from transformation with PSM425 was amplified in the presence of Kan and plated on Kan-Suc plates. In each case Hygr Sucr or Kanr Sucr colonies were tested for loss of the plasmid sequences as a result of a second crossover by plating on Str-containing medium. Kanr Sucr Strs and Hygr Sucr Strs colonies from each transformation were analyzed by Southern blot to confirm the allelic exchange. One transformant in which allelic exchange of Rv1349 was confirmed was named ST69, whereas the mutant strain in which allelic replacement of Rv1348 was confirmed was named ST73.
To generate the complemented strain ST96, plasmid pSM546 was electroporated into the mutant strain ST73 and recombinants were selected by plating in 7H10 containing Kan. Integration of pSM546 at the attB site in Kanr colonies was confirmed by Southern blot analysis.
Mycobactin determination. Mycobacterial strains were grown to mid-logarithmic phase in 7H9 medium, and 0.7 ml of culture was spread onto MM agar containing the indicated concentrations of FeCl3. After incubation at 37°C for 10 days, bacteria were scraped from the plate. Subsequently, mycobactin was extracted in ethanol and chloroform and quantified as previously described (20).
Cross-feeding experiments. A logarithmic culture of the mbtB strain grown in 7H10 medium was used to inoculate MM to an optical density (OD) at 540 nm of 0.3. Then, 0.5 ml of this bacterial suspension was mixed with 2.5 ml of MM, and MM supplemented with 2 μM FeCl3 or 2 μM FeCl3 was added to 2.5 ml of the culture filtrate from H37Rv or ST73 cultures grown to the same OD in MM containing no iron. Growth of the mbtB mutant was monitored by measuring the OD.
Carboxymycobactin utilization assay. High-pressure liquid chromatography purified Fe-carboxymycobactin extracted from a chloroform extract of a low-iron culture filtrate of M. tuberculosis Erdman strain (10) was kindly provided by Marcus A. Horwitz at the Department of Medicine, School of Medicine, University of California, Los Angeles. H37Rv, ST73, and ST96 were grown in MM from an OD of 0.05 to 0.6 and then diluted in MM to an OD of 0.1. After 2 days, the cultures were diluted again to an OD of 0.1 in MM containing 5 ng of Fe-carboxymycobactin/ml. The growth of each strain was monitored every day by measuring the OD.
THP-1 infections. M. tuberculosis infections of THP-1-derived macrophages were performed as previously described on (16). Briefly, THP-1 cells were grown in RPMI 1640, supplemented with 0.45% glucose, 0.15% sodium pyruvate, and 4 mM L-glutamine. THP-1 cells were induced to differentiate into macrophages by treatment with 50 nM 12-tetradecanoylphorbol-13-acetate for 24 h. A total of 7.5 x 104 cells per well were incubated for 2 h at 37°C with a bacterial suspension prepared from a logarithmic growing liquid culture of each M. tuberculosis strain at a multiplicity of infection of 1:15 CFU per macrophage. After 2 h the medium was removed, and the cells were washed twice with warm phosphate-buffered saline to remove any residual extracellular bacteria. Next, 100 μl of fresh RPMI was added to each well, and the plate was incubated at 37°C. At the indicated time points, the medium was removed from three wells, the macrophages were lysed with 100 μl of 0.05% sodium dodecyl sulfate, and dilutions of the released intracellular bacteria were plated on 7H10 to determine the CFU.
Mouse aerosol infection. For each strain tested, a 10-ml bacterial suspension of 106 bacilli ml–1 in saline containing 0.04% Tween 80 was used. Aerosols were generated with a Lovelace Nebulizer (In-Tox Products, Albuquerque, NM), and animals were exposed to the aerosol for 30 min. Under these conditions the number of microorganisms detected in the lungs at time zero was ca. 100. At the indicated time points after infection, three mice were sacrificed, and their lungs were removed and homogenized in phosphate-buffered saline-Tween 80. Dilutions of the homogenates were plated on 7H10 agar to determine the CFU.
RESULTS
Sequence analysis. The M. tuberculosis Rv1348 and Rv1349 encode proteins of 859 and 579 amino acids, respectively. There is no intergenic sequence between irtA and irtB, suggesting that they are cotranscribed. A predicted IdeR binding site is located at position –212 upstream of the annotated translational start site for Rv1348 (11). The proteins encoded by Rv1348 and Rv1349 share homology with each other and with members of the ATP binding cassette(ABC) transporter family. Both proteins contain an amino-terminal membrane-spanning domain with six predicted transmembrane helices in Rv1349 and five to seven possible transmembrane helices in Rv1348, fused to a nucleotide-binding domain. Characteristic motifs shared by members of the ABC transporter family (Walker A, ABC signature, Walker B, and Linton and Higgins) have been identified in the carboxy-terminal domain of these proteins (2). Based on these information, we have named these genes and the proteins encoded by them IrtA and IrtB, respectively, for iron-regulated transporters A and B. Since ABC transporters consist of two membrane-spanning domains associated with two cytoplasmic nucleotide binding domains, IrtA and -B are predicted to form a heterodimeric ABC transporter. IrtAB is similar to the Y. pestis YbtPQ transporter, as first noticed by Fetherson et al. (7). Homologs of YbtPQ have also been found in Yersinia enterocolitica (irp6-7) (4) and Corynebacterium diphtheriae (CdtPQ) (15). IrtA shows 46% similarity to YbtP. This similarity is accentuated at the carboxy-terminal end, whereas at the amino-terminal IrtA has an extension of about 292 amino acids not present in YbtP and predicted by computer algorithms (TMHMM and MEMSAT) to be exposed to the outside environment. IrtB shows 46% similarity to YbtQ and 34% identity with the last 578 amino acids of IrtA.
Generation of M. tuberculosis IrtAB mutants. YbtPQ, Irp6-7, and CdtPQ are required for iron transport in Y. pestis, Y. enterocolitica, and C. diphtheriae, respectively (4, 7, 15). The homology between irtA and irtB with these other bacterial iron transport systems prompted us to examine their role in M. tuberculosis iron transport. With that purpose, two mutants strains were created by homologous recombination and allelic replacement: one in which Rv1349 was disrupted by introduction of a Kan resistance cassette (ST69) and another (ST73) in which a Hyg resistance cassette was inserted in Rv1348, as described in Materials and Methods. Since Rv1348 and Rv1349 are organized as an operon with the 5' terminus of Rv1349 overlapping the 3' terminus of Rv1348 by 3 bp, the mutation in ST73 is presumed to be polar, affecting both genes. Gene replacement was confirmed by Southern blot analysis (data not shown). The colony morphology and growth properties of ST69 and ST73 in 7H9 or 7H10 medium were no different from those of the wild-type strain, but exponentially growing cultures of ST69 and ST73 on agar plates exhibited a light orange pigment not observed in the wild-type strain (data not shown).
IrtA and IrtB are required for growth of M. tuberculosis in iron-deficient conditions. The role of IrtAB in survival of M. tuberculosis under iron depletion was evaluated by examining the ability of the mutant strains ST73 and ST69 to grow in 7H9 medium in the presence of the iron chelator DPI. As shown in Fig. 2A to C, inactivation of irtB alone (ST69) or of irtA and irtB (ST73) does not affect growth of M. tuberculosis in iron-sufficient conditions, but under iron-deficient conditions both mutant strains show a growth defect. This defect is more pronounced in the double mutant, suggesting that IrtA alone can partially function. At this point we decided to further analyze the phenotypes of the double-mutant strain ST73. ST73 was complemented with a single copy of irtA and irtB under the control of their own promoter. For this purpose, the integrative plasmid PSM546 was electrophorated into ST73, resulting in strain ST96. Restoring expression of irtA and irtB allows strain ST96 to survive low iron conditions to an extent similar to the wild-type strain, confirming that irtA and irtB are required for normal replication of M. tuberculosis under iron depletion (Fig. 2D). Complementation of ST73 with a plasmid containing only irtA restored growth to the same extent as that shown by the irtB mutant (ST69) (data not shown).
Siderophore production in the irtAB mutant. IrtA and IrtB have fused membrane spanning and ATPase domains, a feature most commonly found in ABC transporters that function as exporters (2). Therefore, we considered the possibility that IrtA and IrtB could be involved in siderophore secretion. A deficiency in siderophore production or secretion would explain the inability of the ST73 mutant to overcome iron deficiency. To test this possibility, we first measured mycobactin production in H37Rv and in ST73 cultured in MM agar containing increasing amounts of FeCl3. ST73 produces comparable amounts of mycobactin as does the wild-type strain in low-iron conditions, indicating that the irtAB mutation does not affect mycobactin synthesis (Fig. 3A). However, mycobactin production is not repressed in the mutant strain as efficiently as in the wild type, and even at the highest concentration tested (50 μM FeCl3) ST73 produces about twice as much mycobactin as the wild-type strain (Fig. 3A). This result suggests that accumulation of intracellular iron levels that signal repression of mycobactin synthesis is less effective in the irtAB mutant strain than in the wild type. Consistent with this interpretation is the observation that irg-1 a previously described iron-repressed gene (19) is expressed at higher levels in the irtAB mutant cultured in high-iron medium (data not shown).
To evaluate production and secretion of carboxymycobactins in ST73, we used a biological assay. The M. tuberculosis strain with a mutation in the mbtB gene does not produce mycobactin or carboxymycobactin and is unable to grow in a low-iron medium (6) (Fig. 3B). However, this strain does grow under those conditions when supplied with an exogenous source of Fe-carboxymycobactin, purified or in the culture filtrate of a carboxymycobactin-producing strain grown in low-iron conditions (unpublished observations). When tested in this assay, the culture filtrate of ST73 was able to support growth of the mbtB strain to the same extent as the culture supernatant obtained from the same number of wild-type bacteria (Fig. 3B). Based on this result and those obtained from mycobactin measurements, we conclude that inactivation of irtA and irtB does not affect siderophore production or secretion in M. tuberculosis.
Fe-carboxymycobactin utilization. In order to test whether IrtAB are involved in iron acquisition from Fe-carboxymycobactin, we examined the effect of the irtAB mutation on the ability of M. tuberculosis to grow in a medium containing Fe-carboxymycobactin as a sole iron source. For this purpose, the mycobacterial strains were pregrown without iron to exhaust intracellular iron reserves and then subcultured in an iron-depleted medium supplemented with Fe-carboxymycobactin. As shown in Fig. 4, the growth of ST73 was significantly limited compared to the wild-type and complemented strains, indicating that the mutation has compromised the ability of this strain to utilize Fe-carboxymycobactin as an iron source.
Effect of the irtAB mutation on replication of M. tuberculosis in macrophages. Macrophages provide an iron-limiting environment for M. tuberculosis (11, 21) and siderophore-mediated iron acquisition is required for efficient multiplication of the bacilli in these cells (6). The effect of inactivation of irtA and irtB on the ability of M. tuberculosis to replicate in macrophages was tested in THP-1 cells. The growth of each strain was determined, as described in Materials and Methods, by the number of CFU obtained at various times after infection. Compared to the wild-type and complemented strains, the irtAB mutant is significantly impeded in the ability to multiply in human macrophages, since on day 7 after infection there is ca. 100 times less mutant than wild-type or complemented mutant bacteria (Fig. 5). This result demonstrates that IrtAB is necessary for normal multiplication of M. tuberculosis in human macrophages.
Effect of irtAB mutation on replication of M. tuberculosis in mice. The effect of the irtAB mutation in virulence of M. tuberculosis was examined in the mouse model. C57B/6 mice were aerosol infected so that ca. 100 CFU of H37Rv, ST73, or ST96 were detected at time zero after infection. At the indicated time points postinfection the mice were sacrificed, and the CFU in the lungs were determined. As shown in Fig. 6 the irtAB mutant is defective in the ability to replicate in mice lungs. This ability was significantly although not completely restored by reintroducing irtAB in the complemented strain.
DISCUSSION
Iron availability during M. tuberculosis infection is a critical factor that influences the course of TB. Altering the ability of this pathogen to acquire iron is likely to profoundly affect the outcome of this infection. In order to develop ways to interfere with iron acquisition of M. tuberculosis, a better understanding of this process is required. In the present study we identified two genes, irtA and irtB, encoding proteins involved in iron acquisition in this pathogenic mycobacterium. The IrtAB system is encoded in a region of the M. tuberculosis chromosome regulated by iron and IdeR (20) that appears to be dedicated to the synthesis and utilization of siderophores. IrtA and IrtB have the motifs typical of ABC transporters, and they closely resemble the Y. pestis iron transporter YbtPQ. IrtA and IrtB are similar to each other, and both have membrane-spanning domains fused to an ATPase domain. This is a feature shared with the YbtPQ system but uncommon among transporters that function as importers which usually have these two domains in different polypeptides (2). ABC transporters are composed of four structural domains two membrane-spanning domains and two cytoplasmic domains containing the ATP binding cassette (2). Therefore, IrtA and IrtB, as is the case with YbtP and YbtQ, are predicted to function as a heterodimer.
Inactivation of irtA and irtB has no effect on growth of M. tuberculosis in high-iron conditions, but it does affect the ability of this bacterium to multiply under iron-deficient conditions (Fig. 2). Inactivation of irtB alone results in a growth defect in low-iron medium that is less severe than the one exhibited by a double irtAB mutant (Fig. 2B and C). This suggests that in the absence of IrtB, IrtA can partially function, forming homodimers or possibly associating with another protein. The limited growth of the irtAB mutant in iron-deficient conditions is probably sustained by additional iron transport systems of lower affinity than the IrtAB system. Similarly, YbtPQ mutants of Y. pestis retain some ability to replicate in low-iron conditions.
The effect of the irtAB mutation on growth under iron deficiency is not due to an effect on siderophore synthesis since this strain continues to produce and secrete siderophores normally (Fig. 3). However, the iron concentration in the medium required to repress mycobactin synthesis and possibly carboxymycobactin production, since they follow a common synthesis pathway, was higher in the irtAB mutant than in the wild-type strain (Fig. 3A). Since there is no evidence suggesting altered IdeR function in the irtAB mutant, we believe that partial derepression of mycobactin synthesis is indicative of lower intracellular iron levels in this strain. Derepression of mycobactin synthesis and formation of Fe-mycobactin complexes on the surface of bacteria cultured in high-iron medium is likely the cause of the orange pigmentation shown by ST69 and ST73. We have observed a similar pigmentation in other iron-deficient strains that also exhibit derepressed mycobactin synthesis (20).
The current model for iron transport in M. tuberculosis suggests a transfer of iron from Fe-carboxymycobactin to mycobactin in the cell surface. Indeed, this transfer of iron can occur (9), but its significance is not clear. The results presented here indicate that mycobactin is not required for uptake of iron from Fe-carboxymycobactin since the mbtB mutant strain, which does not make either of the siderophores, can be fed iron from Fe-carboxymycobactin provided exogenously (Fig. 3B). Thus, an uptake pathway for Fe-carboxymycobactin can function independently of mycobactin. The exact contribution of each siderophore to iron transport in M. tuberculosis awaits the isolation of mutants that are defective in one but not the other siderophore. Inactivation of irtAB renders the mutant unable to efficiently utilize Fe-carboxymycobactin as an iron source, indicating that IrtA and IrtB are required for the uptake of iron from carboxymycobactin (Fig. 4). Based on this result we predict that IrtAB transports Fe-carboxymycobactin complexes into the cytoplasm. However, we should point out that the experiments required to demonstrate Fe-carboxymycobactin internalization in M. tuberculosis have not been conducted. In an early study, incorporation of Fe55 from Fe55-carboxymycobactin complexes into M. bovis was found not to be affected by energy poisons or uncouplers of ATP biosynthesis (22). This observation would apparently argue against the role of an active transporter for Fe-carboxymycobactin in slow-growing mycobacteria. However, in that study the authors did not test the effect of energy poisons on the intracellular iron pool but only on the detected radioactivity associated with whole cells. We believe that under these conditions, adsorption of iron on the cell envelope, including iron transferred from Fe-carboxymycobactin to mycobactin, in a non-energy-requiring process, could mask the effect of energy inhibitors on an active transport process into the cytoplasm. Additional studies to address the energy requirements for transport of iron into M. tuberculosis and the identity of the substrate transported by IrtAB are in progress.
ABC transporters that function as importers generally require a substrate-binding protein (SBP), as well as a translocator in the cytoplasmic membrane. This SBP is located in the periplasm in gram-negative bacteria or is a membrane-bound lipoprotein in gram-positive bacteria (2). In addition, transport in gram-negative bacteria requires an outer membrane receptor that binds the substrate and, using the energy transduced by the tonB system, translocates the complex into the periplasm (3). In gram-positive bacteria the SBP serves also as receptor. No SBP is encoded in the chromosomal region containing irtAB. The same is true for the YbtPQ system of Y. pestis. It is possible that the SBPs used by these systems are encoded elsewhere in the chromosome or that they do not use a typical SBP. Interestingly, a new class of chimeric ABC transporters with fused extracytoplasmic substrate-binding sites has been reported (12). Considering that IrtA has a 292-amino-acid N-terminal domain that is not found in YbtP or in its reported homologs and is predicted to be extracytoplasmic, it is tempting to speculate that it could be a substrate-binding domain. Future studies will address this possibility.
We propose a model of iron transport in which Fe-carboxymycobactin complexes traverse the cell envelope either by diffusion (given its partial hydrophobic character) or, with the aid of a porin, a typical SBP or the amino-terminal end of IrtA and then are translocated into the cytoplasm by IrtAB. Since irtAB mutants still show some growth in low-iron conditions, redundant pathways for Fe-carboxymycobactin utilization must also exist.
The IrtAB system is important for replication not only under iron-deficient conditions in vitro but also in human macrophages and mouse lungs, as shown by attenuation of the mutant strain compared to the wild-type and complemented strains in these infection models (Fig. 5 and 6). The growth defect of the irtAB mutant in mice was not completely complemented, a result that can be due to differences in expression of these genes in the chromosomal context in which the complementing plasmid has integrated. Attenuation of the irtAB mutant in mice shows, for the first time, a direct correlation between the ability to efficiently acquire iron and the capacity to replicate in an in vivo model of tuberculosis infection. The mbtB mutant strain was shown to be attenuated in human macrophages, but its phenotype in mice was not reported.
The work described here documents the first M. tuberculosis iron transporter. Future studies will characterize this transporter at a molecular level to provide information that can be applied for development of antitubercular agents affecting iron uptake.
ACKNOWLEDGMENTS
We are very grateful to Marcus A. Horwitz for providing the purified Fe-carboxymycobactin used in this study and to Clifton E. Barry III and Laura Via for the mbtB mutant strain. We thank Irina Kolesnikova and Jeanie Dubnau for assistance with the mouse infections.
This study was supported by NIH grant AI44856 (I.S.) and awards from the Francis Parker Foundation for Pulmonary Research and the Center for AIDS Research (G.M.R.).
REFERENCES
Babbette, B., D. LaMarca, W. Zhu, J. E. L. Arceneaux, B. R. Byers, and M. D. Lundrigan. 2004. Participation of fad and mbt genes in synthesis of mycobactin in Mycobacterium smegmatis. J. Bacteriol. 186:374-382.
Braibant, M., P. Gilot, and J. Content. 2000. The ATP binding cassette (ABC) transport systems of Mycobacterium tuberculosis. FEBS Microbiol. Rev. 24:449-467.
Braun, V., and H. Killmann. 1999. Bacterial solutions to the iron-supply problem. Trends Biochem. Sci. 24:104-109.
Brem, D., C. Pelludat, C. Rakin, C. A. Jacobi, and J. Heesemann. 2001. Functional analysis of yersiniabactin transport genes of Yersinia enterocolitica. Microbiology 147:1115-1127.
Card, G. L., N. A. Peterson, C. A. Smith, B. Rupp, B. M. Schick, and E. N. Baker. 2005. The crystal structure of Rv1347c, a putative antibiotic resistance protein from Mycobacterium tuberculosis, reveals a GCN5-related fold and suggest an alternative function in siderophore biosynthesis. J. Biol. Chem. 280:13978-13986.
De Voss, J. J., K. Rutter, B. G. Schroeder, H. Su, Y. Zhu, and C. E. Barry III. 2000. The salicylate-derived mycobactin siderophores of Mycobacterium tuberculosis are essential for growth in macrophages. Proc. Natl. Acad. Sci. USA 97:1252-1257.
Fetherson, J. D., V. J. Bertolino, and R. D. Perry. 1999. YbtP and YbtQ two ABC transporters required for iron uptake in Yersinia pestis. Mol. Microbiol. 32:289-299.
Gangaidzo, I. T., M. V. Moyo, E. Mvundura, G. Aggrey, N. L. Murphree, H. Khumalo, T. Saungweme, I. Kasuosve, I. Kasvoesve, Z. A. Gomo, T. Roualt, J. R. Boelaert, and V. R. Gordeuk. 2001. Association of pulmonary tuberculosis with increased dietary iron. J. Infect. Dis. 184:936-939.
Gobin, J., and M. Horwitz. 1996. Exochelins of Mycobacterium tuberculosis remove iron from human iron-binding proteins and donate iron to mycobactins in the M. tuberculosis cell wall. J. Exp. Med. 183:1527-1532.
Gobin, J., C. H. Moore, J. J. R. Reeve, D. K. Wong, B. W. Gibson, and M. A. Horwitz. 1995. Iron acquisition by Mycobacterium tuberculosis: isolation and characterization of a family of iron-binding exochelins. Proc. Natl. Acad. Sci. USA 92:5189-5193.
Gold, B., G. M. Rodriguez, M. P. Marras, M. Pentecost, and I. Smith. 2001. The Mycobacterium tuberculosis IdeR is a dual functional regulator that controls transcription of genes involved in iron acquisition, iron storage, and survival in macrophages. Mol. Microbiol. 42:851-865.
Heide, T. V. D., and B. Poolman. 2002. ABC transporters: one, two, or four extracytoplasmic substrate-binding sites EMBO Rep. 3:938-943.
Jacobs, Jr., W. R., G. V. Kalpana, J. D. Cirillo, L. Pascopella, S. B. Snapper, R. A. Udani, et al. 1991. Genetic systems for mycobacteria. Methods Enzymol. 204:537-555.
Kochan, I. 1971. Mechanisms of tuberculostasis in mammalian serum: role of transferrin in human serum tuberculostasis. J. Infect. Dis. 119:11-18.
Kunkle, C. A., and a. M. P. Schmitt. 2003. Analysis of the Corynebacterium diphtheriae DtxR regulon: identification of a putative siderophore synthesis and transport system that is similar to the Yersinia high-pathogenicity-island-encoded yersiniabactin synthesis and uptake system. J. Bacteriol. 185:6826-6840.
Manganelli, R., M. I. Voskuil, G. K. Schoolnik, and I. Smith. 2001. The Mycobacterium tuberculosis ECF sigma factor E role in global gene expression and survival in macrophages. Mol. Microbiol. 41:423-437.
Pascopella, L., F. M. Collins, J. M. Martin, W. R. Jacobs, Jr., and B. R. Bloom. 1993. Identification of a genomic fragment of Mycobacterium tuberculosis responsible for in vivo growth advantage. Infect. Agents Dis. 2:282-284.
Ratledge, C. 1999. Iron metabolism, p. 260-286. In C. Ratledge and J. Dale (ed.), Mycobacteria: molecular biology and virulence. Blackwell Science Publishers, London, United Kingdom.
Rodriguez, G. M., B. Gold, M. Gomez, O. Dussurget, and I. Smith. 1998. Identification and characterization of two divergently transcribed iron regulated genes in Mycobacterium tuberculosis. Tuberc. Lung Dis. 79:287-298.
Rodriguez, G. M., M. I. Voskuil, B. Gold, G. K. Schoolnik, and I. Smith. 2002. ideR, An essential gene in Mycobacterium tuberculosis: role of IdeR in iron-dependent gene expression, iron metabolism, and oxidative stress response. Infect. Immun. 70:3371-3381.
Schnappinger, D., S. Ehrt, M. I. Voskuil, Y. Liu, J. A. Mangan, I. M. Monahan, G. Dolganov, B. Efron, P. D. Butcher, C. Nathan, and G. K. Schoolnik. 2003. Transcriptional adaptation of Mycobacterium tuberculosis within macrophages: insights into the phagosomal environment. J. Exp. Med. 198:693-704.
Stephenson, M. C., and C. Ratledge. 1980. Specificity of exochelins for iron transport in three species of mycobacteria. J. Gen. Microbiol. 116:521-523.
Timm, J., F. A. Post, L. G. Bekker, G. B. Walther, C. H. Wainwright, R. Manganelli, W. T. Chan, L. Tsenova, B. Gold, I. Smith, G. Kaplan, and J. D. Mckinney. 2003. Differential expression of iron-, carbon-, and oxygen-responsive mycobacterial genes in the lungs of chronicle infected mice and tuberculosis patients. Proc. Natl. Acad. Sci. USA 100:14321-14326.
Weinberg, E. D. 1984. Iron withholding: a defense against infection and neoplasia. Physiol. Rev. 64:65-102.(G. Marcela Rodriguezand I)