The Aspergillus fumigatus Siderophore Biosynthetic Gene sidA ...

INFECTION AND IMMUNITY, Sept. 2005, p. 5493–5503 0019-9567/05/$08.00⫹0 doi:10.1128/IAI.73.9.5493–5503.2005 Copyright © 2005, American Society for Microbiology. All Rights Reserved.

Vol. 73, No. 9

The Aspergillus fumigatus Siderophore Biosynthetic Gene sidA, Encoding L-Ornithine N5-Oxygenase, Is Required for Virulence Anna H. T. Hissen, Adrian N. C. Wan, Mark L. Warwas, Linda J. Pinto, and Margo M. Moore* Department of Biological Sciences, Simon Fraser University, Burnaby, British Columbia V5A 1S6, Canada Received 21 September 2004/Returned for modification 20 October 2004/Accepted 5 May 2005

Aspergillus fumigatus is the leading cause of invasive mold infection and is a serious problem in immunocompromised populations worldwide. We have previously shown that survival of A. fumigatus in serum may be related to secretion of siderophores. In this study, we identified and characterized the sidA gene of A. fumigatus, which encodes L-ornithine N5-oxygenase, the first committed step in hydroxamate siderophore biosynthesis. A. fumigatus sidA codes for a protein of 501 amino acids with significant homology to other fungal L-ornithine N5-oxygenases. A stable ⌬sidA strain was created by deletion of A. fumigatus sidA. This strain was unable to synthesize the siderophores Nⴕ,Nⴖ,Nⵯ-triacetylfusarinine C (TAF) and ferricrocin. Growth of the ⌬sidA strain was the same as that of the wild type in rich media; however, the ⌬sidA strain was unable to grow in low-iron defined media or media containing 10% human serum unless supplemented with TAF or ferricrocin. No significant differences in ferric reduction activities were observed between the parental strain and the ⌬sidA strain, indicating that blocking siderophore secretion did not result in upregulation of this pathway. Unlike the parental strain, the ⌬sidA strain was unable to remove iron from human transferrin. A rescued strain (⌬sidA ⴙ sidA) was constructed; it produced siderophores and had the same growth as the wild type on iron-limited media. Unlike the wild-type and rescued strains, the ⌬sidA strain was avirulent in a mouse model of invasive aspergillosis, indicating that sidA is necessary for A. fumigatus virulence. Aspergillus fumigatus is an opportunistic fungal pathogen and is the leading cause of mold infections worldwide (15). A. fumigatus can cause life-threatening invasive pulmonary aspergillosis and is a particularly serious problem in immunocompromised populations, such as bone marrow and solid organ transplant recipients (51), cancer patients receiving cytotoxic chemotherapy (4), AIDS patients (32), and those with chronic granulomatous disease (45). Antifungal drugs are used in the treatment of invasive aspergillosis, but even with current treatments, mortality rates average 26 to 65% for invasive pulmonary aspergillosis, depending on the severity of the immunosuppression (38). New treatments for invasive aspergillosis are therefore urgently needed. Iron is an essential element for all eukaryotic cells and is required for important cellular functions such as DNA synthesis and repair, respiration, and detoxification of free radicals (9, 10). Iron is abundant in the Earth’s crust but is poorly bioavailable due to its low solubility in aerobic environments at neutral pH. Conversely, excess cellular free iron is damaging because iron can catalyze the formation of deleterious free radicals. Many microbes have adapted to the poor availability of iron by producing siderophores, low-molecular-weight, ferric iron-specific chelators (34). Intracellular siderophores also play a role in iron storage, preventing the formation of damaging free radicals (28). Host animals can limit the growth of pathogenic microor-

ganisms in vivo by significantly reducing free iron levels. Host high-affinity iron-binding molecules, such as transferrin, lactoferrin, heme, and ferritin maintain free iron levels in tissues at concentrations of approximately 10⫺18 M (7), too low to support microbial growth. Transferrin is the predominant ironbinding molecule in plasma and is only about 30% saturated in healthy individuals. In response to infection, nonspecific host defenses decrease the level of transferrin saturation (7), further reducing free iron concentrations. Therefore, strategies for acquisition of iron from host iron-binding compounds are necessary for successful microbial colonization. We have recently demonstrated that A. fumigatus is capable of growth in media containing concentrations of human serum which are inhibitory to the growth of most fungal pathogens (19). Siderophores produced by A. fumigatus are responsible for its ability to access transferrin-bound iron, likely permitting its growth in the presence of serum (21). A. fumigatus has been reported to produce several hydroxamate siderophores, including ferricrocin, ferrichrome, ferrichrome C, and N⬘,N⬙,N⵮-triacetylfusarinine C (TAF) (16). More recent studies have shown that TAF and ferricrocin (Fig. 1) are the siderophores produced by A. fumigatus in the largest quantities (21, 35). TAF and ferricrocin are produced at high concentrations in iron-limited media, including media containing serum (21). Both TAF and ferricrocin have high thermodynamic iron binding constants, with pM values of 31.8 (2) and 26.5 (56), respectively, compared to transferrin, with a pM of 23.6 (20). (pM is the negative log concentration of free iron in a solution containing 1 ␮M Fe3⫹ and 10 ␮M chelator at pH 7.4.) Siderophore production has been demonstrated to contribute to the virulence of several bacterial species, including

* Corresponding author. Mailing address: Department of Biological Sciences, Simon Fraser University, 8888 University Drive, Burnaby, British Columbia V5A 1S6, Canada. Phone: (604) 291-3441. Fax: (604) 291-3496. E-mail: [email protected]. 5493

5494

INFECT. IMMUN.

HISSEN ET AL.

TAF and ferricrocin. We have characterized the ability of the ⌬sidA strain to secrete siderophores, to grow in iron-limited media, and to remove transferrin-bound iron. We also examined ferric reduction activity of A. fumigatus as a possible alternative iron uptake pathway. Finally, we compared the virulence of the ⌬sidA strain to that of the parental strain and a rescued strain in a mouse model of invasive aspergillosis.

MATERIALS AND METHODS

FIG. 1. Chemical structures of the A. fumigatus siderophores N⬘,N⬙,N⵮-triacetylfusarinine C (A) and ferricrocin (B).

Pseudomonas aeruginosa (31), Vibrio species (27, 54), and Burkholderia cepacia (46). Iron is also an important factor in fungal infections. For example, treatment of dialysis patients with the iron-chelating drug desferrioxamine B is a risk factor for zygomycoses, because iron bound to desferrioxamine is efficiently used as an iron source by some species of zygomycetes (1, 11). Some of the genes involved in synthesis of hydroxamate siderophores have been characterized. These include sidA in Aspergillus nidulans (17), dffA in Aspergillus oryzae (57), pvdA in Burkholderia cepacia (46) and Pseudomonas aeruginosa (50), and sid1 in Ustilago maydis (30), all of which encode L-ornithine N5-oxygenases, the first committed step in hydroxamate biosynthesis. Genes coding for L-lysine N6-oxygenases, which are also involved in siderophore biosynthesis, have been characterized in bacteria such as Escherichia coli (12). Several nonribosomal peptide synthetases have also been characterized, including sidC, which catalyzes the final step in ferricrocin synthesis in A. nidulans (17), and sid2, which completes the synthesis of ferrichrome in Ustilago maydis (58). Siderophore biosynthetic pathways are absent in human cells; therefore, these pathways represent potential new targets for antimicrobial chemotherapy. To investigate the importance of siderophore biosynthesis in the virulence of A. fumigatus, we have constructed a ⌬sidA mutant of A. fumigatus by gene deletion. This strain should be unable to produce any hydroxamate siderophores, including

Strains and growth conditions. A. fumigatus (ATCC 13073), originally isolated from a human pulmonary lesion, was obtained from the American Type Culture Collection and maintained on YM slants (0.3% malt extract, 0.3% yeast extract, 0.5% peptone, 0.5% glucose) at 4°C. A. fumigatus 13073 (hygS sidA⫹) is here designated the wild-type (or parental) strain, while 19B4 (hygR sidA⫺) is designated the ⌬sidA strain. R3 is a rescued strain designated the ⌬sidAR strain (hygR sidA⫹). The construction of 19B4 and the rescued strain is described below. 13073, 19B4, and the rescued strain are isogenic except for the disruption/ addition of the sidA gene. A. fumigatus strains were regularly cultured on YM plates or on YM plates containing 200 ␮g/ml hygromycin B (Roche) at 28°C for 6 days until fully conidiated. Conidia were harvested by flooding the culture plate with phosphate-buffered saline (PBS) containing 0.05% Tween 20 and swabbing with a sterile cotton swab. The conidia were vortexed, centrifuged, resuspended in PBS, and filtered through a plug of sterile glass wool to remove hyphae. Concentrations of conidia were determined by counting in a hemacytometer. Growth on different media. A. fumigatus was inoculated into 1-ml volumes of YM medium or 1 ml of modified Grimm-Allen (GA) medium [1 g/liter KHSO4, 3 g/liter K2HPO4, 3 g/liter (NH4)2SO4, 20 g/liter sucrose, 2 mg/liter thiamine, 20 ␮g/liter CuSO4, 1 mg/liter MnSO4, 5.5 mg/liter ZnSO4, 810 mg/liter MgSO4, pH 6.9] (39) in acid-washed test tubes at a concentration of 106 conidia/ml. GA medium was supplemented with 5 ␮M FeCl3 where described. Human serum (male) was obtained from Sigma, stored frozen until use, and added to media at a concentration of 10% (vol/vol). TAF and ferricrocin were purified from A. fumigatus cultures as previously described (21) and treated with 8-hydroxyquinoline to remove all bound iron to yield the desferri- forms. Siderophore solutions were dissolved in 70% ethanol prior to addition to media. Dry weights of A. fumigatus cultures were measured by filtering the entire contents of each tube through preweighed, predried Whatman no. 1 filters and rinsing thoroughly with distilled water. Filters were oven dried and reweighed. Construction of the ⌬sidA strain. Preliminary sequence data for A. fumigatus were obtained from The Wellcome Trust Sanger Institute (www.sanger.ac.uk). sid1 from Ustilago maydis (30) was used as a probe to search the A. fumigatus genome for homologous sequences. Standard molecular techniques were carried out as described by Sambrook et al. (41). Plasmids were propagated in Escherichia coli DH5␣ (Life Technologies). Genomic DNA was extracted from A. fumigatus by standard phenol-chloroform extraction (41) as described by May et al. (29). pID620, composed of pBluescript SK⫹ (Stratagene) containing the hph hygromycin resistance cassette in the EcoRI site (6), was kindly provided by D. W. Holden, Department of Infectious Diseases and Bacteriology, Royal Postgraduate Medical School, Imperial College of London, London, United Kingdom. Custom primers were ordered through Invitrogen. A 1.6-kb DNA fragment containing sidA was PCR amplified from A. fumigatus genomic DNA using primers 5⬘-AAGCTTAAGCTTTTGAACGGAAGTCAG AATCG and 5⬘-TCTAGATCTAGAACAGGTTCCCTCATGTCTGC, which flank the sidA gene and contain restriction sites for HindIII and XbaI, respectively (underlined). This PCR product was digested with HindIII and XbaI and then ligated into HindIII- and XbaI-digested pID620, generating pGAW1. pGAW1 was then digested with SmaI and PstI, excising bases 576 to 1078 of the sidA coding region. The hygromycin resistance cassette (hph) was PCR amplified from pID620 using primers 5⬘-AACGTTAACGTTGTAAAACGACGGCC AGTG and 5⬘-GGAAACAGCTATGACCATG. This PCR product was digested with PstI and ligated to the digested pGAW1, creating the sidA gene replacement vector, pGAW2. The correct disruption of sidA in the transformation vector pGAW2 was confirmed by sequencing the gene deletion construct (University Core DNA and Protein Services, University of Calgary, Calgary, Alberta, Canada). A. fumigatus was transformed using linearized pGAW2 by electroporation, according to the method of Weidner et al. (53). Transformation reaction mixtures were plated on Aspergillus minimal medium (MM) containing 10 g/liter glucose, 0.85 g/liter NaNO3, 0.52 g/liter KCl, 0.52 g/liter MgSO4 · 7H2O, 1.52 g/liter KH2PO4, 40 ␮g/liter Na2B4O7 · 10H2O, 0.4 mg/liter CuSO4 · 5H2O, 1

VOL. 73, 2005

sidA GENE IS REQUIRED FOR A. FUMIGATUS VIRULENCE

mg/liter FePO4 · 4H2O, 0.6 mg/liter MnSO4 · H2O, 0.8 mg/liter Na2MoO4 · 2H2O, 8 mg/liter ZnSO4 · 7H2O, 1 mg/liter nicotinic acid, 2.5 mg/liter riboflavin, 2 mg/liter pantothenic acid, 0.5 mg/liter pyridoxine, 10 ␮g/liter biotin, 0.2 mg/liter p-aminobenzoic acid (PABA), and 10 ␮M TAF, pH 6.5. The plates were incubated at room temperature overnight and then overlaid with 10 ml of MM containing 267 ␮g/ml hygromycin B and 0.7% agar. Plates were then incubated at 37°C for 48 to 72 h until colonies had conidiated. Conidia from putative transformants were screened in 2 ml modified GA medium supplemented with 10 ␮M TAF, incubated at 37°C for 3 days. Cultures which did not produce any orange color upon addition of 200 ␮l of 10 mg/ml FeSO4 were selected. Ten of 140 hygromycin-resistant colonies showed significantly reduced siderophore secretion by this test. These transformants were further screened by PCR for deletion of the sidA gene; all transformants showed deletion of the sidA gene. Gene deletion was confirmed by PCR at the sidA site, using the primers 5⬘-TT GAACGGAAGTCAGAATCG (oof) and 5⬘-ACAGGTTCCCTCATGTCTGC (oor), which flank the sidA gene. Gene deletion was also confirmed by PCR using primers complementary to hph (5⬘-GACATATCCACGCCCTCCTA [hph1] and 5⬘-ACTGTCGGGCGTACACAAAT [hph2]) and to a region external to the sidA gene (5⬘-ACGCCCTCAACTGTATGGAC [be-oor, 1.8 kb upstream of sidA start codon] and 5⬘-TTTCGTGCAAAACAGTGGAG [af-oof, 1.6 kb downstream of sidA stop codon]) (see Fig. 3A). One transformant, 19B4, was selected for further study because it had the lowest levels of measurable siderophore secretion. Southern analysis was carried out on genomic DNA extracted from the wild-type strain and this putative ⌬sidA strain (19B4) by the method of May et al. (29). Genomic DNA was completely digested with EcoRV, PstI, and HindIII, and fragments were separated by electrophoresis on 0.7% agarose and transferred to Hybond N⫹ (Amersham) using standard techniques (41). Probes to the entire sidA gene were constructed using the AlkPhos direct labeling kit with the CDP-star detection reagent (Amersham). Construction of the rescued strain. Genomic 13073 DNA was amplified by PCR with PFU Ultra (Stratagene), using primers corresponding to the wild-type sidA gene (forward, 505 bp upstream of the start codon [GAATTCGAATTCT GTCAAGAGCACCACACCTC] and reverse, 560 bp downstream of the stop codon [GAATTCGAATTCCCATCAGATAACGCGTGAAA]). Both primers contained EcoRI sites to facilitate ligation (underlined). The PCR fragment was cleaned using the QIAquick PCR purification kit (QIAGEN). The plasmid Bluescript SK⫹ was cut with EcoRI (Invitrogen), followed by dephosphorylation with APex heat-labile alkaline phosphatase (Epicenter) and gel purification. The PCR product was ligated to the linear plasmid and the resultant plasmid (pCOMP2) used to transform E. coli DH5␣. Transformants were selected on LB-Amp with X-Gal (5-bromo-4-chloro-3-indolyl-␤-D-galactopyranoside) (1 mg/ plate). Successful transformation was confirmed by restriction mapping and sequencing (data not shown). pCOMP2 was linearized with XbaI and used to transform conidia of A. fumigatus ⌬sidA. The fungi were grown on MM supplemented with TAF, and conidia were swollen in the presence of 10 ␮mol of TAF and ferricrocin. The conidia were electroporated with 0 or 2 ␮g DNA and transformants selected on Grimm-Allen agar containing 150 ␮M dipyridyl. No colonies were observed with the no-DNA controls. In the samples incubated with 2 ␮g DNA, three transformants were obtained, and all three were found to be hygromycin resistant. Successful ectopic integration of sidA was confirmed by PCR using primers homologous to the entire sidA gene (forward, CTCCATAT GGAATCTGTTGAACGGAAG; reverse, CCGAATTCTTATTACAGCATG GCTCGTAGC). PCR also revealed a band that corresponded to the sidA gene disrupted with hph, indicating that the original mutation was present in the rescued strain. One of transformants (R3) was chosen for phenotypic characterization and to confirm the rescue in the mouse model of invasive aspergillosis. Detection of siderophores. Siderophores were purified from wild-type A. fumigatus and the ⌬sidA strain. Conidia (106/ml) were inoculated into 5-ml volumes of YM and incubated at 37°C for 3 days. Cultures were filtered to remove mycelia, and 100 mg/ml FeCl3 was added to the supernatants. Ferrated siderophores were then extracted from the aqueous supernatants with three 1-ml portions of 1:1 phenol-chloroform. Combined phenol-chloroform fractions were washed with 2 ml distilled water and then diluted with 10 ml diethyl ether. The siderophores were extracted from the diethyl ether fraction with two 0.5-ml portions of distilled water. Combined aqueous layers were washed with 0.5 ml diethyl ether and then lyophilized to dryness. Extracts were redissolved in 30 ␮l double-distilled water and analyzed on silica gel thin-layer chromatography sheets, using a running phase of 4:1 dichloromethane-methanol. Purified TAF and ferricrocin were run as standards for identification. Ferric reduction activity. Ferric reductase assays were performed as described by Morrissey et al. (33). A. fumigatus wild-type and ⌬sidA strains were inoculated in GA medium (106 conidia/ml) containing 5 ␮M FeCl3 and incubated at 37°C and 150 rpm for 24 h. Mycelia were filtered through Miracloth (Calbiochem),

5495

washed with distilled water, and transferred to microcentrifuge tubes. Mycelia were incubated with 0.5 ml of 0.2 mM FeCl3 in PBS plus 0.5 ml 2 mM bathophenanthrolinedisulfonic acid (Sigma-Aldrich) in PBS at 37°C and 150 rpm for 1 h, at which time the absorbance of the supernatant was measured at 540 nm. To normalize the absorbance data to biomass, mycelia were then washed and lyophilized and dry weights obtained. Iron removal from holo-transferrin. A. fumigatus (106 conidia/ml) was cultured in 1 ml modified GA medium supplemented with 50 ␮M FeCl3 at 37°C for 24 h. Mycelia were washed three times with PBS and then resuspended in 1 ml minimal essential medium (MEM) containing no phenol red (pH 7.4) (Life Technologies) supplemented with 0.2 mg/ml human holo-transferrin (Sigma). Cultures were incubated at 37°C and 150 rpm, and 100-␮l samples were removed at various intervals. The samples were lyophilized to dryness and redissolved in 10 ␮l water plus 10 ␮l urea-polyacrylamide gel electrophoresis (urea-PAGE) loading buffer (1⫻ Tris-borate-EDTA, 10% glycerol, and 0.2% bromphenol blue). Urea-PAGE was used to determine the proportions of apo-, diferric, and monoferric transferrin in each sample. Urea-PAGE was carried out as described by Wolz et al. (55), using a Protean II xi cell (Bio-Rad). Approximately 10 ␮g transferrin in a 10-␮l volume was loaded onto each lane, and gels were run at 200 V for 18 to 20 h at 4°C. Gels were incubated for 30 min in 0.05% sodium dodecyl sulfate, stained with SYPRO orange (Molecular Probes, Eugene, Oregon), and scanned with a Typhoon 9410 imager (Amersham). Bands were quantified using ImageQuant 5.2 (Molecular Dynamics). Mouse aspergillosis model. Female BALB/c mice weighing from 18 to 22 g were obtained from Charles River Breeders and given 0.5 mg/ml tetracycline in their drinking water throughout the course of the study. Mice were immunosuppressed by subcutaneous injections of 200 mg/kg cortisone acetate (Wiler-PCCA, London, Ontario, Canada) on days ⫺3, 0, 2, and 4. Cortisone acetate was prepared as a 30-mg/ml suspension in sterile saline (Baxter Medical). For these studies, A. fumigatus was cultured on YM agar at 37°C for 4 days to ensure that conidia were fully mature and pigmented. Conidia were harvested as described above and suspended in sterile saline. Mice were randomly assigned to one of four treatment groups: parental strain (n ⫽ 10), ⌬sidA strain (n ⫽ 10), ⌬sidAR strain (n ⫽ 10), and saline (n ⫽ 5). On day 0, mice were anesthetized with isoflurane and 5 ⫻ 106 conidia of either the wild-type, ⌬sidA, or ⌬sidAR strain were instilled intranasally in a 20-␮l volume using a micropipette and a gel loading tip. Control mice were anesthetized and received 20 ␮l saline intranasally. Mice were kept anesthetized until all the liquid was observed to be inhaled. Mice were monitored daily for 14 days to observe any clinical symptoms and were deemed to have reached endpoint if they displayed ruffled fur and one of the following: either (i) labored breathing, hunching, and decreased movement or (ii) disorientation and loss of balance. Mice displaying either set of clinical symptoms were euthanatized. At endpoint or the end of the 14-day study, lungs were fixed by immediately opening the chest of the euthanatized animal, isolating the trachea, and perfusing 10% formalin in PBS into the lung cavities. After 2 minutes, lungs were removed and further fixed at room temperature overnight in PBS containing 10% formalin. Lungs were subsequently paraffin embedded, sectioned, and stained with hematoxylin and eosin. Images were obtained on a Zeiss LSM10 confocal microscope equipped with a QImaging 10-bit camera. Statistics. A. fumigatus growth data were analyzed using analysis of variance with Dunnett’s multiple-comparison test, while mouse survival data were analyzed by log rank analysis. Statistical analyses were carried out using Prism 4.0 software (GraphPad). Nucleotide sequence accession number. The sequence data for A. fumigatus sidA have been submitted to the GenBank database under accession number AY819708.

RESULTS Characterization of sidA. A BLAST search of the A. fumigatus genome revealed a sequence with high homology to published sequences for L-ornithine N5-oxygenases. (The A. fumigatus gene would be termed AfusidA according to the nomenclature recommended by the Aspergillus genome sequencing group [3].) The gene was composed of an open reading frame of 1,564 base pairs. Using GlimmerM from The Institute for Genomic Research (http://www.tigr.org/software /glimmerm), trained for A. fumigatus, an intron of 58 base pairs

5496

HISSEN ET AL.

INFECT. IMMUN.

FIG. 2. Alignment of the A. fumigatus SidA, A. nidulans SidA, and A. oryzae DffA amino acid sequences. The amino acid sequence of A. fumigatus SidA was predicted using GlimmerM from The Institute for Genomic Research, trained for A. fumigatus. Multiple pairwise alignment was performed with ClustalW (8) and the output generated with Boxshade 3.21. Black and gray boxes represent identical and similar residues, respectively.

and an amino acid sequence of 501 amino acids were predicted for the A. fumigatus SidA protein. The coding sequence of A. fumigatus sidA showed very high identity to sidA from A. nidulans (75%) and dff1 from A. oryzae (74%) (Fig. 2). The A. fumigatus sequence contained the three signature sequences typical of amino acid hydroxylase enzymes. The first of these is the conserved putative binding sites for the substrate DXXX(L/F)ATGYXXXXP (47), located at residue 400. Typical of ornithine-binding enzymes, such as pvdA and sid1, the last P of this sequence was not conserved in

sidA and was replaced by H. The flavin adenine dinucleotide binding domain, GXGXXG, was located at residue 45, and the last glycine in this domain of sidA was exchanged for proline, which is typical of siderophore biosynthetic enzymes (47). An NADP binding site, GXGXXG, was observed at residue 254, although, again typical for siderophore biosynthetic genes, the last G in sidA was not conserved (47). Deletion of sidA. To investigate the role of siderophores in iron uptake and virulence of A. fumigatus, the sidA gene was deleted by gene replacement. Southern analysis revealed only

VOL. 73, 2005


one copy of sidA in A. fumigatus genomic DNA (data not shown). sidA was PCR amplified from genomic DNA and cloned into the multiple cloning site of pBluescript SK⫹ to create pGAW1. The hygromycin phosphotransferase (hph) cassette was inserted into the sidA sequence of pGAW1 to create pGAW2, a transformation vector containing a nonfunctional sidA gene and a selectable marker. In pGAW2, bases 576 to 1078 of the sidA coding region were replaced by hph, thereby removing required domains such as the NADP-binding site at residue 254. A. fumigatus strain ATCC 13073 was transformed by electroporation with pGAW2. The resulting hygromycin-resistant strains were screened for absence of siderophore production in low-iron modified GA medium. The GA medium was supplemented with 10 ␮M TAF to support the growth of siderophore secretion mutants. This TAF concentration was low enough to avoid interference with the colorimetric detection of siderophores. Strains which did not produce an orange color upon addition of FeSO4 to 4-day-old culture medium were examined further. The correct disruption and integration of the gene deletion construct were confirmed for one strain, 19B4, by PCR and Southern blot analysis (Fig. 3A and B). sidA is required for growth in low-iron media and serumcontaining media. After 96 h, the ⌬sidA mutant and the parental strain achieved similar biomasses in rich media such as YM (Table 1). However, growth of the ⌬sidA strain was severely restricted in iron-limited defined media such as GA medium. Supplementing GA medium with 5 ␮M FeCl3 partially restored the growth of the ⌬sidA strain, whereas 10 ␮M TAF restored growth of the ⌬sidA strain to wild-type levels (Table 1). Serum is extremely iron limited because it contains partially saturated transferrin, which reduces free iron concentrations to very low levels. Thus, serum is inhibitory to the growth of many microbes, including most fungi. It has previously been demonstrated that siderophores produced by A. fumigatus were able to remove transferrin-bound iron (21). Growth of the siderophore secretion mutant, the ⌬sidA strain, was completely inhibited by 10% human serum but could be restored to wild-type levels by addition of 50 ␮M desferri-TAF or 50 ␮M desferriferricrocin (Table 1). These siderophore concentrations were chosen because they are similar to the concentration of siderophore produced by wild-type A. fumigatus in this medium. The importance of siderophore biosynthesis to growth in iron-limited media was confirmed by the observation that the rescued ⌬sidAR strain grew equally well as the wild type in GA medium or in medium containing serum (Table 1). sidA is required for secretion of TAF and ferricrocin. sidA catalyzes the first committed step in hydroxamate siderophore biosynthesis. A. fumigatus is known to secrete several siderophores of the hydroxamate group, and all should be absent in the ⌬sidA mutant. To determine whether sidA was required for the production of siderophores in A. fumigatus, the ⌬sidA strain and the wild type were cultured in liquid YM medium. YM was used for this experiment because it supports identical growth rates of the wild-type and ⌬sidA strains without inhibiting siderophore secretion by the wild-type strain. Both strains were cultured at 37°C for 3 days, at which time mycelia were removed by filtration, and siderophores were extracted from the culture medium as described in Materials and Methods.

5497

FIG. 3. Restriction map and Southern blot of wild-type and ⌬sidA genomic DNAs. (A) Double-crossover gene disruption, showing binding sites for primers and restriction sites for both A. fumigatus wildtype and ⌬sidA strains. (B) The Southern blot confirms the disruption of sidA. A. fumigatus genomic DNAs from the wild-type and ⌬sidA strains were completely digested with EcoRV, HindIII, or PstI; separated by gel electrophoresis; and transferred to Hybond N⫹. Blots were probed using a full-length sidA probe constructed using the AlkPhos direct labeling kit. Detection was with the CDP-star reagent. I, wild-type genomic DNA; II, ⌬sidA genomic DNA. Numbers on the left represent molecular size markers in kilodaltons.

Solvent extracts were analyzed by thin-layer chromatography using 4:1 dichloromethane-methanol as the mobile phase. TAF, ferricrocin, and two unidentified siderophores were clearly visible in the extract from the wild-type strain; however, neither TAF, ferricrocin, nor the two additional unidentified siderophores could be observed on the thin-layer chromatogram of the ⌬sidA extracts (Fig. 4). Faint yellow spots in the ⌬sidA lane correspond to components extracted from the me-

5498

HISSEN ET AL.

INFECT. IMMUN.

TABLE 1. ⌬sidA can grow in rich medium but not in iron-limited media Strain

Growth (mg [dry wt])a

YM

Wild type ⌬sidA ⌬sidAR

3.9 ⫾ 0.2 5.0 ⫾ 0.6 5.0 ⫾ 0.4

GA


3.9 ⫾ 0.7 0.3 ⫾ 0.5c 7.6 ⫾ 1.6

GA ⫹ FeCl3 (5 ␮M)


6.5 ⫾ 0.4 1.4 ⫾ 1.1c 7.5 ⫾ 0.3

GA ⫹ TAF (ferrated, 10 ␮M)


6.2 ⫾ 0.5 4.4 ⫾ 1.5 6.0 ⫾ 0.1

GA ⫹ FeCl3 ⫹ serumb


9.4 ⫾ 0.7 0.0 ⫾ 0.0c 9.5 ⫾ 0.8

GA ⫹ FeCl3 ⫹ serum ⫹ desferri-TAF (50 ␮M)


10.1 ⫾ 0.4 12.3 ⫾ 1.2 11.3 ⫾ 0.3

GA ⫹ FeCl3 ⫹ serum ⫹ desferriferricrocin (50 ␮M)


9.7 ⫾ 2.0 11.1 ⫾ 1.1 11.0 ⫾ 1.2

Medium

a Data presented are averages ⫾ standard deviations of triplicate values. This experiment was performed three times with similar results. b 10% (vol/vol) human serum. c Significant growth inhibition relative to wild type (P ⬍ 0.03 by analysis of variance with Dunnett’s multiple-comparison test).

dium that were also present in an uninoculated medium blank (data not shown). Thus, sidA is required for production of TAF, ferricrocin, and other, unidentified hydroxamate siderophores. A. fumigatus has ferric reduction activity. Some pathogenic fungi do not produce siderophores and are able to grow in vivo using ferric reductase enzymes to obtain iron. For example, the ferrous iron transport system was found to be required for virulence of Candida albicans (40). In addition, low-molecularweight reductants have also been shown to participate in iron acquisition in Histoplasma capsulatum (49) and Cryptococcus neoformans (36). Both wild-type and ⌬sidA A. fumigatus strains produced measurable ferric reduction activity when grown in GA medium supplemented with 5 ␮M FeCl3 or in YM medium. There was little difference observed in the levels of ferric reduction activity in GA medium supplemented with 5 ␮M FeCl3 compared to YM medium (Table 2), nor was any significant difference observed in the levels of ferric reduction activity expressed by the parental and ⌬sidA strains (Table 2). These results indicate that for A. fumigatus, at least in vitro, ferric reduction activity was not upregulated in the ⌬sidA strain to compensate for the defect in siderophore secretion. A. fumigatus requires siderophores to remove iron from transferrin. The ability of the siderophore secretion mutant, the ⌬sidA strain, to remove iron from human diferric transferrin was assessed in vitro. The ⌬sidA strain and the parental strain were cultured in modified GA medium containing 50

FIG. 4. Thin-layer chromatography of siderophores produced by wild-type and ⌬sidA strains of A. fumigatus. Siderophores were extracted from wild-type and ⌬sidA cultures grown in 5-ml volumes of modified GA medium at 37°C for 3 days as described in Materials and Methods. Extracts were analyzed on silica gel thin-layer chromatography sheets with a mobile phase of 4:1 dichloromethane-methanol, and the locations of authentic TAF and ferricrocin standards run at the same time are noted. Ferricrocin, TAF, and two unidentified siderophores are visible in the wild-type extract (A), while no siderophores are visible in extract from the ⌬sidA strain (B). Faint yellow spots in the ⌬sidA extracts were also present at the same Rf value in the uninoculated control (data not shown).

␮M FeCl3 for 24 h. Both strains produced similar amounts of mycelial biomass after 24 h but had not yet begun to conidiate. Mycelia were washed three times in PBS and transferred to fresh tubes containing MEM plus 0.2 mg/ml human diferric transferrin. This extra incubation step in GA plus FeCl3 was required because ⌬sidA conidia cannot germinate and grow in MEM. The cultures were incubated at 37°C in MEM-diferric transferrin, and samples of supernatant were removed at various intervals. The supernatants were analyzed by urea-PAGE to measure the iron saturation of transferrin. Wild-type cultures converted all the diferric transferrin to monoferric transferrin and apotransferrin within 12 h (Fig. 5). In contrast, the levels of diferric transferrin in the ⌬sidA culture remained identical to those in an uninoculated control over a period of at least 48 h (Fig. 5). These data show that siderophores are required by A. fumigatus for removal of iron from human diferric transferrin. Siderophore biosynthesis contributes to virulence of A. fumigatus. Because the ⌬sidA strain was unable to grow in ironlimited media, including serum, and was unable to remove iron from human diferric transferrin, we compared its virulence to that of the parental and rescued strains in a mouse model of invasive aspergillosis. Mice were immunosuppressed with cor-

TABLE 2. Ferric reduction activity Absorbance at 540 nm/mg [dry wt] (100)a in: Strain

Wild type ⌬sidA

GA ⫹ 5␮M FeCl3

YM

7.6 ⫾ 2.2 4.7 ⫾ 1.7

4.9 ⫾ 0.6 4.0 ⫾ 0.1

a Data are reported as absorbance per milligram [dry weight] of mycelia present in the reductase assay and represent the averages ⫾ standard deviations from three independent experiments.

VOL. 73, 2005


5499

FIG. 5. Iron saturation of human diferric transferrin incubated with wild-type or ⌬sidA A. fumigatus mycelia. A. fumigatus wild-type and ⌬sidA mycelia were incubated in MEM containing 0.2 mg/ml human diferric transferrin at 37°C and 150 rpm. Iron saturation of transferrin was assessed by urea-PAGE. Gels were stained with SYPRO orange and scanned on a Typhoon 9410 imager, and the transferrin bands were quantified using ImageQuant 5.2. Data are normalized and reported as percent total transferrin present in diferric form/percent transferrin in diferric form at time zero. Error bars represent standard deviations of triplicate measurements.

tisone acetate and then challenged intranasally with 5 ⫻ 106 conidia of the ⌬sidA, wild-type, or ⌬sidAR strain. Two distinct sets of clinical symptoms were observed in mice infected with the parental or rescued strain. Some developed clear signs of pulmonary infection, showing labored breathing, hunching, decreased mobility, and ruffled fur. The remaining mice displayed what appeared to be symptoms of sinusitis or central nervous system impairment. These mice had a characteristic head tilt, became disoriented and agitated, and showed a loss of balance. They also displayed ruffled fur, labored breathing, and eventually decreased mobility. Either one of these outcomes was classified as endpoint. One hundred percent of mice receiving the wild-type or ⌬sidAR strain reached endpoint by day 4 postinfection (Fig. 6).

FIG. 6. Survival curve for female BALB/c mice infected with the A. fumigatus wild-type strain, the ⌬sidA strain, or the rescued strain. Mice were immunosuppressed by subcutaneous injection of 200 mg/kg cortisone acetate on days ⫺3, 0, 2, and 4. Mice were infected with 5 ⫻ 106 conidia of either the wild-type, ⌬sidAR, or ⌬sidA strain in 20 ␮l saline on day 0. Control mice were given saline alone. Mice were monitored daily and sacrificed if they displayed symptoms of infection, as described in Results. Survival curves for the wild-type or ⌬sidAR strain versus the ⌬sidA strain are significantly different (P ⬍ 0.0001) by log rank analysis.

In contrast, the ⌬sidA strain was avirulent in this model of invasive pulmonary aspergillosis (Fig. 6). None of the mice infected with ⌬sidA conidia reached endpoint, which was not significantly different from the mortality rate of saline-inoculated, immunocompromised mice (Fig. 6). To determine the extent of fungal growth, lungs were fixed in 10% formalin, embedded in paraffin, sectioned, and stained with hematoxylin and eosin. Lung fixation was performed when mice reached endpoint, or at 14 days postinfection for the remaining mice. Representative sections are shown in Fig. 7. Of the four salinetreated controls whose lungs were examined, all four showed normal lung structure with open airways and no inflammation (Fig. 7A and B). Fungal hyphae were observed in sections from the mice receiving the wild-type strain. Fungal growth was accompanied by an extensive inflammatory infiltrate composed of polymorphonuclear leukocytes and monocytes. Extensive tissue destruction was apparent, including frank necrosis of the airway walls and large blood vessels and complete replacement of alveolar architecture with necrosis and inflammation (Fig. 7C and D). The mice that showed primarily central nervous system symptoms were also found to have fungal colonization of their lungs (Fig. 7C). Mice receiving the rescued strain showed similar fungal growth and tissue destruction (Fig. 7E and F). In the group that received the ⌬sidA mutant strain, sections from seven mice were examined. In the mice receiving the ⌬sidA strain, the lungs of four mice showed evidence of peribronchiolar or perivascular inflammation, whereas the remaining two mice had no evidence of inflammatory response. Only one of the mice exposed to the ⌬sidA strain had evidence of fungi within the lungs at 14 days postinfection. Figure 7G and H show representative sections in which inflammation is evident. In Fig. 7G, inflammatory cells appear to be confined to the lumen of the bronchioles. In Fig. 7H, there are leukocytes present in the airway walls, but in contrast to mice infected with the wild-type strain, little necrosis is evident. One mouse had an extensive inflammatory response consisting of both

5500

HISSEN ET AL.

INFECT. IMMUN.

FIG. 7. Lung tissue sections from cortisone-treated mice exposed to one of the following treatments: saline (A and B), wild-type A. fumigatus conidia (C and D), conidia from the rescued strain (⌬sidAR) (E and F), and ⌬sidA A. fumigatus conidia (G and H). Each panel represents a section from a different animal. (A and B [saline]) Lungs have normal appearance, with clear airways and no inflammatory infiltrate. (C and D [wild type])


VOL. 73, 2005

polymorphonuclear leukocytes and monocytes. Fungal hyphae were evident in the some tissue sections of ⌬sidA-exposed mice; however, the fungal growth was less extensive than for the wild type and remained confined to the bronchiolar lumen (data not shown). DISCUSSION We have characterized the A. fumigatus L-ornithine N5monooxygenase gene, sidA, which encodes the first committed step in hydroxamate siderophore biosynthesis. The A. fumigatus sidA gene showed a high degree of identity (⬃75%) to 5 L-ornithine N -oxygenases from the closely related species A. nidulans and A. oryzae. A. fumigatus sidA contains the NADP-, flavin adenine dinucleotide- and substrate-binding domains (47), which are identical to those of the other characterized Aspergillus L-ornithine N5-monooxygenases. In addition, sidA had a ⬎50% similarity to L-ornithine N5-oxygenases from U. maydis (sid1), as well as to the pvdA genes of Pseudomonas and Burkholderia species. The similarity of sidA to bacterial L-lysine N6-hydroxylases (iucD) of Escherichia coli and Yersinia species was approximately 45%. A ⌬sidA strain of A. fumigatus was constructed by transformation with pGAW2, a sidA gene deletion construct. The ⌬sidA strain did not produce detectable levels of either of the two most common A. fumigatus siderophores, TAF and ferricrocin, nor were any other siderophores detected. To date, all siderophores reported for Aspergillus species are hydroxamate siderophores (22); therefore, it was expected that deletion of sidA would prevent synthesis of all A. fumigatus siderophores. The ⌬sidA strain was unable to grow in low-iron defined medium (GA with no supplements); therefore, siderophore secretion was required for growth under very low iron conditions. This growth inhibition could be overcome by addition of 5 ␮M FeCl3; however, growth of the ⌬sidA strain in GA containing 5 ␮M FeCl3 was less than that of either the parental or ⌬sidAR strain. These results suggest that ferric iron could promote growth. The mechanism by which the A. fumigatus ⌬sidA strain can access ferric iron is not known, but it could involve the reduction of ferric to ferrous iron, followed by uptake by ferrous iron transporters. Many pathogens, both bacterial and fungal, successfully colonize hosts without production of siderophores. Clearly, other iron uptake mechanisms are capable of scavenging iron within the host. Transferrin has low affinity for ferrous iron, and therefore reduction of ferric iron coupled with a ferrous iron transporter presents an alternative method of iron uptake. Ferric reductases have been well characterized in yeast pathogens such as Candida albicans (25) and Cryptococcus neoformans (24). Histoplasma capsulatum produces siderophores and ferric reductases and expresses a cell surface receptor for hemin (18). The ferric reductases of H. capsulatum have been reported to remove iron from both hemin and transferrin (49),

5501

offering an alternative to siderophore production for iron uptake in vivo. A ferrous permease was found to be required for virulence of C. albicans (40). In addition, C. albicans produces the hemolytic molecule mannan (52) and has been demonstrated to access heme-bound iron (42). A. nidulans has been reported to produce an iron-regulated gene, freA, with homology to S. cerevisiae metalloreductases (37). Wild-type and ⌬sidA A. fumigatus strains both displayed ferric reduction activity. We postulated that ferric reduction activity might be upregulated in the ⌬sidA strain to compensate for loss of siderophore-mediated iron uptake. However, no significant differences were observed between levels of ferric reduction activities of the wild-type and ⌬sidA strains in vitro. At low levels of iron (GA medium plus 5 ␮M FeCl3), ⌬sidA strain growth was less than that of the wild type, suggesting that A. fumigatus ferric reductases did not fully compensate for the absence of siderophores under these conditions. The ferric reductase and ferrous transporter molecules involved in iron uptake in yeast have been characterized (44), but it is not known if functionally similar proteins are present in A. fumigatus. Despite the lack of siderophore secretion, growth of the ⌬sidA strain was not inhibited in rich media such as YM, which contains various organic iron sources. These organic iron sources are therefore accessible to the A. fumigatus ⌬sidA strain through alternative iron uptake pathways. The A. oryzae ⌬dffA strain is similar to the ⌬sidA strain of A. fumigatus in that it was also able to grow rich medium without the addition of siderophores (57). The poor growth of the ⌬sidA mutant in iron-limited GA medium was completely rescued by addition of 10 ␮M TAF, indicating that a lack of siderophore secretion was the only cause of growth inhibition in this medium. These results contrast with those for A. nidulans sidA mutants, which were unable to grow in defined medium unless it was supplemented with 10 ␮M TAF or 1.5 mM ferrous iron (18). Thus, unlike A. fumigatus, A. nidulans is unable to access FeCl3 or ferric iron from sources such as citrate without the use of siderophores. These data suggest that the A. nidulans metalloreductase gene, freA, may not promote any significant degree of reductive iron assimilation. More work is necessary to elucidate ferric reductase pathways in both A. fumigatus and A. nidulans, as the two species appear to differ significantly in their non-siderophore-mediated iron uptake pathways. Previous studies in our laboratory have shown that A. fumigatus ATCC 13073 grows very well in the presence of high concentrations (up to 80%) of human serum, conditions which are normally fungistatic due to the low concentration of free iron (19). In contrast, growth of the A. fumigatus ⌬sidA strain was completely inhibited by 10% human serum. Growth in GA medium containing FeCl3 and serum could be restored by the addition of either 50 ␮M desferri-TAF or 50 ␮M desferriferricrocin, indicating that siderophores are required for the growth of A. fumigatus in serum-containing media.

Fungal hyphae (white arrow) within an airway accompanied by a neutrophil and monocyte infiltration; erythrocytes are seen within the airways. (E [rescued strain]) Hyphae are seen within the bronchiole. (F [rescue]) Fungal growth is evident within an airway (arrow), accompanied by necrosis. (G [⌬sidA strain]) Normal lung tissue surrounding a focus of inflammation confined to the lumen of a bronchiole. (H [⌬sidA strain]) No fungi were observed, but foci containing large numbers of macrophages were observed (arrows). Magnifications, ⫻340 (A, B, D, E, G, and H) and ⫻170 (C and F).

5502

HISSEN ET AL.

Wild-type A. fumigatus efficiently removed transferrinbound iron, even when mycelia were separated from diferric transferrin by a dialysis membrane with a 10-kDa-molecularmass cutoff (21). These data suggested that siderophores were capable of removing transferrin-bound iron, but they did not eliminate the possible involvement of other iron uptake mechanisms such as low-molecular-weight ferric reductants. The ⌬sidA strain, which produced no siderophores, did not remove iron from transferrin, demonstrating conclusively that siderophores are required for iron removal from transferrin. Siderophores have been demonstrated to play a role in pathogenesis of many different bacteria. Pyoverdine, a catecholate-hydroxamate siderophore, is required for virulence of Pseudomonas aeruginosa in a burned mouse model of infection (31). In another study, it was found that both pyoverdine and another siderophore, pyochelin, contributed to virulence of P. aeruginosa (48). Similarly, pvdA mutants of the closely related Burkholderia cepacia, which were unable to secrete ornibactins, were less virulent than the parental strain in both chronic and acute models of respiratory infection (46). Aerobactin, a mixed citrate-hydroxamate siderophore, is an important contributor to in vivo extracellular growth of Escherichia coli (13), Vibrio vulnificus strains unable to produce catechol siderophores showed reduced virulence in an infant mouse (27), and anguibactin production has been shown to contribute to virulence of Vibrio anguillarum in juvenile trout (54). For fungi, less is known about the role of siderophores in virulence, because many of the common fungal pathogens do not produce siderophores. However, many fungi are able to use exogenous siderophores, and siderophore transporters have been shown to be required by the yeast Candida albicans for epithelial invasion and penetration (23). ⌬sidA strains of A. fumigatus were avirulent in a mouse model of invasive aspergillosis. Thus, sidA is one of the few identified virulence factors in A. fumigatus. To date, the only other avirulent strains of A. fumigatus are a PABA auxotroph that could not germinate in vivo unless supplied with PABA (5), a pyrG mutant unable to germinate in the absence of uridine and uracil (14), and a lysF mutant unable to synthesize lysine (26). After this paper was submitted, Schrettl et al. (43) published their study of iron assimilation and virulence in Aspergillus fumigatus. Their findings confirm the results of the present study that siderophore biosynthesis represents a true virulence factor in A. fumigatus. Those authors showed that, unlike in A. nidulans, sidA deficiency reduced but did not completely inhibit the growth of A. fumigatus in iron-limited media indicating that siderophore-independent iron assimilation pathways exist in A. fumigatus. They also showed that the ⌬sidA strain was unable to grow on blood agar and that the ⌬sidA strain was avirulent in a mouse model of invasive aspergillosis. In contrast, disruption of the high-affinity iron permease gene, ftrA, did not diminish virulence in the mouse model, suggesting that reductive iron assimilation is not important for the growth of A. fumigatus in vivo. In conclusion, siderophore-mediated iron uptake is required by A. fumigatus for growth in low-iron media, including serumcontaining media, and is the only mechanism by which A. fumigatus can remove iron from transferrin in vitro. Alternative iron uptake pathways are sufficient for growth in rich

INFECT. IMMUN.

media and in defined media supplemented with 5 ␮M FeCl3. Siderophore biosynthesis was required for virulence of A. fumigatus in a mouse model of invasive aspergillosis. Since siderophore biosynthesis pathways are absent in humans, they represent novel targets for antifungal chemotherapy. ACKNOWLEDGMENTS We are grateful to the Natural Sciences and Engineering Research Council of Canada for financial assistance. The Aspergillus fumigatus sequencing consortium was funded by the National Institute of Allergy and Infectious Disease, The Wellcome Trust, and Fondo de Investicagiones Sanitarias. Preliminary A. fumigatus sequence data were obtained from The Wellcome Trust Sanger Institute (www.sanger.ac.uk). We thank Mary Dearden, Loekie van der Wal, and Madeleine Stephens of the Simon Fraser University Animal Care Facility and Clive Roberts of the University of British Columbia for assistance with animal studies. The assistance of V. Pavlova of the University of British Columbia with the lung histology is gratefully acknowledged. REFERENCES 1. Abe, F., H. Inaba, T. Katoh, and M. Hotchi. 1990. Effects of iron and desferrioxamine on Rhizopus infection. Mycopathologia 110:87–91. 2. Adjimani, J. P., and T. Emery. 1987. Iron uptake in Mycelia sterilia EP-76. J. Bacteriol. 169:3664–3668. 3. Anderson, M. J. 2004, posting date. A proposal for the naming of genes in Aspergillus species. [Online.] http://www.aspergillus.man.ac.uk/indexhome .htm?homepagenew/mainindex.htm⬃main. 4. Bodey, G., B. Bueltmann, W. Duguid, D. Gibbs, H. Hanak, M. Hotchi, G. Mall, P. Martino, F. Meunier, S. Milliken, et al. 1992. Fungal infections in cancer patients: an international autopsy survey. Eur. J. Clin. Microbiol. Infect. Dis. 11:99–109. 5. Brown, J. S., A. Aufauvre-Brown, J. Brown, J. M. Jennings, H. Arst, Jr., and D. W. Holden. 2000. Signature-tagged and directed mutagenesis identify PABA synthetase as essential for Aspergillus fumigatus pathogenicity. Mol. Microbiol. 36:1371–1380. 6. Brown, J. S., A. Aufauvre-Brown, and D. W. Holden. 1998. Insertional mutagenesis of Aspergillus fumigatus. Mol. Gen. Genet. 259:327–335. 7. Bullen, J. J. 1981. The significance of iron in infection. Rev. Infect. Dis. 3:1127–1138. 8. Chenna, R., H. Sugawara, T. Koike, R. Lopez, T. J. Gibson, D. G. Higgins, and J. D. Thompson. 2003. Multiple sequence alignment with the Clustal series of programs. Nucleic Acids Res. 31:3497–3500. 9. Crichton, R. R., and R. J. Ward. 1998. Iron homeostasis. Met. Ions Biol. Syst. 35:633–665. 10. Crichton, R. R., and R. J. Ward. 1992. Iron metabolism—new perspectives in view. Biochemistry 31:11255–11264. 11. de Locht, M., J. R. Boelaert, and Y. J. Schneider. 1994. Iron uptake from ferrioxamine and from ferrirhizoferrin by germinating spores of Rhizopus microsporus. Biochem. Pharmacol. 47:1843–1850. 12. de Lorenzo, V., A. Bindereif, B. H. Paw, and J. B. Neilands. 1986. Aerobactin biosynthesis and transport genes of plasmid ColV-K30 in Escherichia coli K-12. J. Bacteriol. 165:570–578. 13. de Lorenzo, V., and J. L. Martinez. 1988. Aerobactin production as a virulence factor: a reevaluation. Eur. J. Clin. Microbiol. Infect. Dis. 7:621–629. 14. D’Enfert, C., M. Diaquin, A. Delit, N. Wuscher, J. P. Debeaupuis, M. Huerre, and J. P. Latge. 1996. Attenuated virulence of uridine-uracil auxotrophs of Aspergillus fumigatus. Infect. Immun. 64:4401–4405. 15. Denning, D. W., M. J. Anderson, G. Turner, J. P. Latge, and J. W. Bennett. 2002. Sequencing the Aspergillus fumigatus genome. Lancet Infect. Dis. 2:251–253. 16. Diekmann, H., and E. Krezdorn. 1975. Metabolic products of microorganisms. 150. Ferricrocin, triacetylfusigen and other sideramines from fungi of the genus Aspergillus, group Fumigatus. Arch. Microbiol. 106:191–194. 17. Eisendle, M., H. Oberegger, I. Zadra, and H. Haas. 2003. The siderophore system is essential for viability of Aspergillus nidulans: functional analysis of two genes encoding L-ornithine N5-monooxygenase (sidA) and a non-ribosomal peptide synthetase (sidC). Mol. Microbiol. 49:359–375. 18. Foster, L. A. 2002. Utilization and cell-surface binding of hemin by Histoplasma capsulatum. Can. J. Microbiol. 48:437–442. 19. Gifford, A. H., J. R. Klippenstein, and M. M. Moore. 2002. Serum stimulates growth of and proteinase secretion by Aspergillus fumigatus. Infect. Immun. 70:19–26. 20. Harris, W. R., C. J. Carrano, S. R. Cooper, S. R. Sofen, A. E. Avdeef, J. V. McArdle, and K. N. Raymond. 1979. Coordination chemistry of microbial iron transport compounds. 19. Stability constants and electrochemical behavior of ferric enterobactin and model complexes. J. Am. Chem. Soc. 101:6097–6104.

VOL. 73, 2005


21. Hissen, A. H., J. M. Chow, L. J. Pinto, and M. M. Moore. 2004. Survival of Aspergillus fumigatus in serum involves removal of iron from transferrin: the role of siderophores. Infect. Immun. 72:1402–1408. 22. Howard, D. H. 1999. Acquisition, transport, and storage of iron by pathogenic fungi. Clin. Microbiol. Rev. 12:394–404. 23. Hu, C. J., C. Bai, X. D. Zheng, Y. M. Wang, and Y. Wang. 2002. Characterization and functional analysis of the siderophore-iron transporter CaArn1p in Candida albicans. J. Biol. Chem. 277:30598–30605. 24. Jacobson, E. S., A. P. Goodner, and K. J. Nyhus. 1998. Ferrous iron uptake in Cryptococcus neoformans. Infect. Immun. 66:4169–4175. 25. Knight, S. A., E. Lesuisse, R. Stearman, R. D. Klausner, and A. Dancis. 2002. Reductive iron uptake by Candida albicans: role of copper, iron and the TUP1 regulator. Microbiology 148:29–40. 26. Liebmann, B., T. W. Muhleisen, M. Muller, M. Hecht, G. Weidner, A. Braun, M. Brock, and A. A. Brakhage. 2004. Deletion of the Aspergillus fumigatus lysine biosynthesis gene lysF encoding homoaconitase leads to attenuated virulence in a low-dose mouse infection model of invasive aspergillosis. Arch. Microbiol. 181:378–383. 27. Litwin, C. M., T. W. Rayback, and J. Skinner. 1996. Role of catechol siderophore synthesis in Vibrio vulnificus virulence. Infect. Immun. 64:2834– 2838. 28. Matzanke, B. F., E. Bill, A. X. Trautwein, and G. Winkelmann. 1987. Role of siderophores in iron storage in spores of Neurospora crassa and Aspergillus ochraceus. J. Bacteriol. 169:5873–5876. 29. May, G. S., J. Gambino, J. A. Weatherbee, and N. R. Morris. 1985. Identification and functional analysis of beta-tubulin genes by site specific integrative transformation in Aspergillus nidulans. J. Cell Biol. 101:712–719. 30. Mei, B., A. D. Budde, and S. A. Leong. 1993. sid1, a gene initiating siderophore biosynthesis in Ustilago maydis: molecular characterization, regulation by iron, and role in phytopathogenicity. Proc. Natl. Acad. Sci. USA 90:903–907. 31. Meyer, J. M., A. Neely, A. Stintzi, C. Georges, and I. A. Holder. 1996. Pyoverdin is essential for virulence of Pseudomonas aeruginosa. Infect. Immun. 64:518–523. 32. Minamoto, G. Y., T. F. Barlam, and N. J. Vander Els. 1992. Invasive aspergillosis in patients with AIDS. Clin. Infect. Dis. 14:66–74. 33. Morrissey, J. A., P. H. Williams, and A. M. Cashmore. 1996. Candida albicans has a cell-associated ferric-reductase activity which is regulated in response to levels of iron and copper. Microbiology 142:485–492. 34. Neilands, J. B. 1995. Siderophores: structure and function of microbial iron transport compounds. J. Biol. Chem. 270:26723–26726. 35. Nilius, A. M., and S. G. Farmer. 1990. Identification of extracellular siderophores of pathogenic strains of Aspergillus fumigatus. J. Med. Vet. Mycol. 28:395–403. 36. Nyhus, K. J., A. T. Wilborn, and E. S. Jacobson. 1997. Ferric iron reduction by Cryptococcus neoformans. Infect. Immun. 65:434–438. 37. Oberegger, H., M. Schoeser, I. Zadra, B. Abt, and H. Haas. 2001. SREA is involved in regulation of siderophore biosynthesis, utilization and uptake in Aspergillus nidulans. Mol. Microbiol. 41:1077–1089. 38. Patterson, T. F., W. R. Kirkpatrick, M. White, J. W. Hiemenz, J. R. Wingard, B. Dupont, M. G. Rinaldi, D. A. Stevens, J. R. Graybill, et al. 2000. Invasive aspergillosis. Disease spectrum, treatment practices, and outcomes. Medicine (Baltimore) 79:250–260. 39. Payne, S. M. 1994. Detection, isolation, and characterization of siderophores. Methods Enzymol. 235:329–344. 40. Ramanan, N., and Y. Wang. 2000. A high-affinity iron permease essential for Candida albicans virulence. Science 288:1062–1064. 41. Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular cloning: a laboratory manual, 2nd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y.

Editor: T. R. Kozel

5503

42. Santos, R., N. Buisson, S. Knight, A. Dancis, J. M. Camadro, and E. Lesuisse. 2003. Haemin uptake and use as an iron source by Candida albicans: role of CaHMX1-encoded haem oxygenase. Microbiology 149:579–588. 43. Schrettl, M., E. Bignell, C. Kragl, C. Joechl, T. Rogers, H. N. Arst, Jr., K. Haynes, and H. Haas. 2004. Siderophore biosynthesis but not reductive iron assimilation is essential for Aspergillus fumigatus virulence. J. Exp. Med. 200:1213–1219. 44. Schroder, I., E. Johnson, and S. de Vries. 2003. Microbial ferric iron reductases. FEMS Microbiol. Rev. 27:427–447. 45. Segal, B. H., E. S. DeCarlo, K. J. Kwon-Chung, H. L. Malech, J. I. Gallin, and S. M. Holland. 1998. Aspergillus nidulans infection in chronic granulomatous disease. Medicine (Baltimore) 77:345–354. 46. Sokol, P. A., P. Darling, D. E. Woods, E. Mahenthiralingam, and C. Kooi. 1999. Role of ornibactin biosynthesis in the virulence of Burkholderia cepacia: characterization of pvdA, the gene encoding L-ornithine N(5)-oxygenase. Infect. Immun. 67:4443–4455. 47. Stehr, M., H. Diekmann, L. Smau, O. Seth, S. Ghisla, M. Singh, and P. Macheroux. 1998. A hydrophobic sequence motif common to N-hydroxylating enzymes. Trends Biochem. Sci. 23:56–57. 48. Takase, H., H. Nitanai, K. Hoshino, and T. Otani. 2000. Impact of siderophore production on Pseudomonas aeruginosa infections in immunosuppressed mice. Infect. Immun. 68:1834–1839. 49. Timmerman, M. M., and J. P. Woods. 2001. Potential role for extracellular glutathione-dependent ferric reductase in utilization of environmental and host ferric compounds by Histoplasma capsulatum. Infect. Immun. 69:7671– 7678. 50. Visca, P., A. Ciervo, and N. Orsi. 1994. Cloning and nucleotide sequence of the pvdA gene encoding the pyoverdin biosynthetic enzyme L-ornithine N5oxygenase in Pseudomonas aeruginosa. J. Bacteriol. 176:1128–1140. 51. Wald, A., W. Leisenring, J. A. van Burik, and R. A. Bowden. 1997. Epidemiology of Aspergillus infections in a large cohort of patients undergoing bone marrow transplantation. J. Infect. Dis. 175:1459–1466. 52. Watanabe, T., M. Takano, M. Murakami, H. Tanaka, A. Matsuhisa, N. Nakao, T. Mikami, M. Suzuki, and T. Matsumoto. 1999. Characterization of a haemolytic factor from Candida albicans. Microbiology 145:689–694. 53. Weidner, G., C. d’Enfert, A. Koch, P. C. Mol, and A. A. Brakhage. 1998. Development of a homologous transformation system for the human pathogenic fungus Aspergillus fumigatus based on the pyrG gene encoding orotidine 5⬘-monophosphate decarboxylase. Curr. Genet. 33:378–385. 54. Wertheimer, A. M., W. Verweij, Q. Chen, L. M. Crosa, M. Nagasawa, M. E. Tolmasky, L. A. Actis, and J. H. Crosa. 1999. Characterization of the angR gene of Vibrio anguillarum: essential role in virulence. Infect. Immun. 67: 6496–6509. 55. Wolz, C., K. Hohloch, A. Ocaktan, K. Poole, R. W. Evans, N. Rochel, A. M. Albrecht-Gary, M. A. Abdallah, and G. Doring. 1994. Iron release from transferrin by pyoverdin and elastase from Pseudomonas aeruginosa. Infect. Immun. 62:4021–4027. 56. Wong, G. B., M. J. Kappel, K. N. Raymond, B. Matzanke, and G. Winkelmann. 1983. Coordination chemistry of microbial iron transport compounds. 24. Characterization of coprogen and ferricrocin, two ferric hydroxamate siderophores. J. Am. Chem. Soc. 105:810–815. 57. Yamada, O., S. N. Nan, T. Akao, M. Tominaga, H. Watanabe, T. Satoh, H. Enei, and O. Akita. 2003. dffA gene from Aspergillus oryzae encodes Lornithine N5-oxygenase and is indispensable for deferriferrichrysin biosynthesis. J. Biosci. Bioeng. 95:82–88. 58. Yuan, W. M., G. D. Gentil, A. D. Budde, and S. A. Leong. 2001. Characterization of the Ustilago maydis sid2 gene, encoding a multidomain peptide synthetase in the ferrichrome biosynthetic gene cluster. J. Bacteriol. 183: 4040–4051.