| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Previous Article | Next Article 
Eukaryotic Cell, August 2007, p. 1339-1353, Vol. 6, No. 8
1535-9778/07/$08.00+0 doi:10.1128/EC.00111-07
Copyright © 2007, American Society for Microbiology. All Rights Reserved.
,
Stuart B. Krasnoff,2
Donna M. Gibson,2 and
B. Gillian Turgeon1*
Department of Plant Pathology and Plant-Microbe Biology, 334 Plant Science Building, Cornell University, Ithaca, New York 14853,1 Plant Protection Research Unit USDA-ARS, Tower Road, Ithaca, New York 148532
Received 8 April 2007/ Accepted 18 June 2007
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
nps2) was defective in ascus/ascospore development in homozygous nps2 crosses. Homozygous crosses of the remaining 11 nps strains showed wild-type (WT) fertility. Phylogenetic, expression, and biochemical analyses demonstrated that the NRPS encoded by NPS2 is responsible for the biosynthesis of ferricrocin, the intracellular siderophore of C. heterostrophus. Functional conservation of NPS2 in both heterothallic C. heterostrophus and the unrelated homothallic ascomycete Gibberella zeae was demonstrated. G. zeae nps2 strains are concomitantly defective in intracellular siderophore (ferricrocin) biosynthesis and sexual development. Exogenous application of iron partially restored fertility to C. heterostrophus and G. zeae nps2 strains, demonstrating that abnormal sexual development of nps2 strains is at least partly due to their iron deficiency. Exogenous application of the natural siderophore ferricrocin to C. heterostrophus and G. zeae nps2 strains restored WT fertility. NPS1, a G. zeae NPS gene that groups phylogenetically with NPS2, does not play a role in sexual development. Overall, these data demonstrate that iron and intracellular siderophores are essential for successful sexual development of the heterothallic ascomycete C. heterostrophus and the homothallic ascomycete G. zeae. |
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Our objective is to determine the natural biological functions of NRPS metabolites in the organisms that produce them; for this, we focus on the genetically tractable heterothallic ascomycete pathogen of maize Cochliobolus heterostrophus. Twelve genes that encode NRPSs (NPS1 to NPS12) have been identified in the C. heterostrophus genome (18, 23). Initially, we focused on the role of NRPS metabolites in virulence to the plant host, since the best-known NRPS metabolites in phytopathogenic ascomycetes are phytotoxins, such as AM-toxin, made by the apple pathotype of Alternaria alternata, and HC-toxin, made by race 1 of Cochliobolus carbonum, both crucial for the pathogenicity of the producing organisms to their hosts. Characterization of C. heterostrophus mutant strains carrying a single-gene deletion of each of the 12 NPS genes, with respect to their virulence to maize, revealed that only NPS6 is required for this activity (18, 23). Further characterization revealed that nps6 strains were sensitive also to oxidative (18) and low-iron (23) stress, phenotypes that, in combination, led to the recognition that the product of the NPS6 protein was a siderophore. Mass spectrometry (MS) and high-performance liquid chromatography (HPLC) analyses identified the siderophore products of the NPS6 protein as coprogen and derivatives, extracellular siderophores, demonstrating that extracellular siderophores play a role in fungal virulence to plant hosts (23).
Although iron is an essential nutrient for virtually all organisms, including fungi, bioavailable forms are scarce in aerobic environments because of oxidation and formation of minimally soluble iron hydroxides (19). When starved for iron, many fungi and bacteria secrete siderophores to acquire iron extracellularly. The strong iron-chelating activity of these low-molecular-weight organic compounds can solubilize iron effectively from the environment (25, 35). Indeed, as noted above, preventing extracellular siderophore biosynthesis by deletion of NPS6 in C. heterostrophus leads to hypersensitivity to iron depletion, which highlights the major contribution of siderophores to iron gathering in this species (23). Similarly, nps6 strains of two other filamentous ascomycetes, a second cereal pathogen, Gibberella zeae/Fusarium graminearum, and the dicot pathogen Alternaria brassicicola, lacking the ability to biosynthesize extracellular siderophores, are hypersensitive to iron depletion and reduced in virulence to their hosts, indicating that siderophore-mediated iron metabolism is generally important among filamentous ascomycete phytopathogens (23). Except for the polycarboxylates identified in zygomycetes (31), all fungal siderophores are products of NRPSs.
While characterizing the role of C. heterostrophus NPS6 in extracellular siderophore biosynthesis, we identified ferricrocin as the intracellular siderophore of this species and determined also that the NRPS encoded by NPS6 is not involved in its biosynthesis (i.e., the nps6 strain produces wild-type [WT] levels of ferricrocin) (23). Intracellular siderophores, in contrast to extracellular ones, have been proposed to play a role in iron storage in mycelia and spores (36) and have been recognized as asexual spore germination factors of Neurospora crassa, Penicillium chrysogenum, and Aspergillus nidulans (9, 15). In these species, treatment of asexual spores with solutions with low water activity (high salinity) led to loss of intracellular siderophores stored in the spores and to inhibition or delay of spore germination. In addition, A. nidulans intracellular siderophores are essential for efficient asexual sporulation (12). In contrast, deletion of C. heterostrophus NPS6, which is responsible for extracellular siderophore production, leads to a reduction of asexual sporulation, as well as to a reduction of virulence to the host (23). Recently, intracellular siderophores have been reported also to play a role in the self-compatible sexual development of A. nidulans (13). In addition to siderophores, other small molecules, such as the oxylipin psi factors, have been demonstrated to alter the ratio of asexual to sexual sporulation in A. nidulans (5, 6, 33) and a small 
-butyrolactone-containing molecule, butyrolactone I, has been reported as an inducer of asexual sporulation in A. nidulans (27). This close relationship between secondary-metabolite or small-molecule production and fungal development has been documented comprehensively for Aspergillus spp. (reviewed in references 7 and 37). It has been demonstrated that sporulation and sterigmatocystin-aflatoxin production are regulated by a common G-protein/cyclic-AMP/protein kinase A signaling pathway (7).
The primary objective of the present study was to determine if one or more C. heterostrophus NRPS metabolites function in sexual development, by mating assay screening of our collection of mutant strains, each with 1 of the 12 NPS genes deleted. We found that NPS2 is involved in ascospore development. nps2 strains fail to form asci and ascospores in crosses in which the nps2 mutation is homozygous; otherwise, sexual development is normal up to this stage. Through biochemical analyses of WT and nps2 strains, we identified the product of the NRPS2 enzyme as the intracellular siderophore ferricrocin, thus connecting the production of this small molecule to a fundamental fungal development process. Furthermore, we demonstrated functional conservation of NPS2 in ferricrocin biosynthesis and in sexual development in the heterothallic ascomycete C. heterostrophus and in the unrelated homothallic ascomycete G. zeae. We propose that iron and intracellular siderophores are essential for successful sexual development in both self-incompatible and self-compatible ascomycetes.
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
|
DNA manipulations. Fungal genomic DNA was prepared as described previously (11, 23). Unless otherwise mentioned, all PCRs were performed with PCR master mix (Promega) by following the manufacturer's recommendations.
Deletion of the complete set of C. heterostrophus NPS genes. For targeted gene deletion, split marker constructs were prepared (see Table S1 in the supplemental material) as previously described (8, 23). In previous work on NPS genes (18), partial gene deletions were generated for 11 of the NPS genes. NPS12 was not included. For the present study, the open reading frame (ORF) of each of genes NPS1 to -3 and NPS5 to -12 was completely deleted and 94% of that of NPS4 was deleted (see Table S1 in the supplemental material) by replacement with the hygB cassette from pUCATPH (20), which confers resistance to hygromycin B. Transformation of C. heterostrophus was carried out as described previously (23).
Deletion of G. zeae NPS1 and NPS2 genes.
We deleted 94% of the G. zeae NPS1 and NPS2 ORFs (see Table S1 in the supplemental material) and replaced the deleted portions with the hygB gene cassette from pUCATPH. G. zeae nps1 nps2 strains were constructed from strain Fgnps2-6-1 (nps2 hygBR) (Table 1), which carries a deletion of NPS2. The NPS1 ORF was replaced with the nptII gene (23), which confers resistance to G418. Transformation of G. zeae was carried out as described by Oide et al. (23).
Complementation. Reintroduction of the WT C. heterostrophus NPS2 ORF was carried out with strain Chnps2-3p (Table 1), which carries a deletion of the first thiolation domain of the ChNPS2 ORF, as described by Lee et al. (18). The construct for reintroduction of the C. heterostrophus NPS2 ORF was prepared by PCR with genomic DNA and Platinum Taq High Fidelity (Invitrogen) by following the manufacturer's directions. A 2,241-bp DNA fragment including the deleted region (342 bp) and the surrounding flanking sequences (829-bp 5'flank and 1,070-bp 3'flank) was amplified. Strain Chnps2-3p was transformed with approximately 5 µg of the PCR product and 10 µg of plasmid pII99 (23) carrying the nptII gene. Initial screening of transformants was carried out by overlaying a 1% agar solution containing 600 µg/ml G418 on the transformation plates. Secondary screening of transformants was carried out on CM without salts (23) containing 400 µg G418/ml. Transformants resistant to G418 were selected and screened for sensitivity to hygromycin B. Replacement of the hygB gene in G418R hygBS transformants, at the Nps2 locus, with the NPS2 fragment was confirmed by PCR as described previously for NPS6 (23).
Crosses. For C. heterostrophus, crosses were set up as described previously (17), except that minimal medium (MM) without glucose (MMNOS) was used instead of Sach's medium (21). MMNOS is easier to prepare than Sach's medium, and to date, no differences in the fertility of any cross tested have been observed when pseudothecia from each medium type were compared.
For G. zeae, crosses were set up as described previously (10) except that carrot medium was replaced with carrot juice (CJ) medium. In our experiments, WT G. zeae showed better fertility on CJ medium than on medium made with fresh carrots. CJ medium is 50 ml of 100% carrot juice (LakeWood, Miami, FL), 3.0 g of CaCO3, and 950 ml of H2O per liter of medium. For mating, growing mycelial tips were placed in the centers of CJ agar plates and the plates were sealed with Parafilm and incubated under continuous black light at 24°C for 7 days. Approximately 1 ml of sterile 2.5% Tween 60 solution was applied to the plate, and mycelia growing on the plate were knocked down with a rubber policeman, which induced sexual development. Excess Tween 60 solution was discarded and the unsealed plates were incubated under the original conditions for an additional 7 days.
Screening of all C. heterostrophus nps strains for a role in sexual development. (i) Crosses of C. heterostrophus nps strains to the WT.
The complete set of 12 nps strains was screened for a role in sexual development. For all experiments, crosses between C. heterostrophus strains C4 (MAT1-2) and CB7 (MAT1-1 alb1) were set up as WT controls. The fertility of nps strains in heterozygous crosses was evaluated by crossing each of the 12 original nps strain (all MAT1-2, pigmented) to tester strain CB7 (MAT1-1 alb1). At least two independent transformants were examined for each of the 12 NPS genes deleted, and each cross combination was duplicated. The ability of strains to act both male and female (as hermaphrodites) was tracked by crossing pigmented nps strains to albino WT testers of the opposite mating type. If a nps strain retains WT hermaphroditic ability, both pigmented and albino pseudothecia should form, as pseudothecium color is determined by the female partner. Thus, the ability of the pigmented nps strains to act female was assessed by monitoring for the presence or absence of fertile pigmented pseudothecia, while the ability of each nps strain to act male was assessed by monitoring for the presence or absence of fertile albino pseudothecia.
The number of pseudothecia per square centimeter of corn leaf was recorded for each cross. For the initial screening, at least 10 pigmented and 10 albino pseudothecia were opened and the number of asci in each pseudothecium was recorded.
(ii) C. heterostrophus nps x nps strain crosses.
The fertility of C. heterostrophus nps strains in homozygous crosses was evaluated in essentially the same way as described for heterozygous crosses. C. heterostrophus nps MAT1-1 strains were obtained as follows. hygBR progeny were recovered from a cross between the original nps MAT1-2 transformants and CB7 (MAT1-1 alb1), and their mating type was determined by crossing them to MAT1-2 (strain C4) and MAT1-1 (strain CB7) testers. For each homozygous nps cross, at least two different combinations of independent nps strains were set up (e.g., for Chnps2 x Chnps2, 1449-T1-1 x 1449-T1-5 and 1451-T1-2 x 1451-T1-3 were examined; see Table S2 in the supplemental material and Table 1). Each cross was duplicated.
Evaluation of the fertility of C. heterostrophus nps2 strains and G. zeae nps1, nps2, and nps1 nps2 strains. (i) C. heterostrophus.
After the initial analyses, the fertility of nps2 strains was examined more extensively. The fertility of nps2 strains in a heterozygous cross was examined also in a MAT1-1 background in order to exclude the possibility that the observed abnormal sexual development was connected in some way to the mating type. Note that only nps2 strains with MAT1-2 backgrounds were examined in the initial characterization of fertility in a heterozygous cross.
Ten replicates of each WT x WT, WT x nps2 MAT1-1, WT x nps2 MAT1-2, and nps2 MAT1-1 x nps2 MAT1-2 cross were set up, and the number and color of pseudothecia per square centimeter of leaf were recorded for each cross. The data were analyzed by analysis of variance (ANOVA). More than 30 pseudothecia were opened for each type of cross, and the number of asci in each pseudothecium was recorded. The data were statistically evaluated by ANOVA (note that no asci are formed in nps2 x nps2; hence, the data were excluded from the statistical test).
(ii) G. zeae.
For homothallic G. zeae, each mutant strain was selfed and its fertility was evaluated by counting the perithecia on each plate and determining the number of asci per perithecium. A selfed WT strain was set up for all experiments as a control. For evaluation of perithecium formation, 10 plates were set up for each genotypic class (i.e., nps1, nps2, nps1 nps2, and WT) and the number of perithecia formed on each plate was recorded. Data were analyzed by ANOVA. At least 30 perithecia were opened for each genotypic class, and the number of asci was recorded for each perithecium. Data on WT and Gznps1 strains were analyzed by ANOVA.
Evaluation of fertility when iron is supplied. For C. heterostrophus, MMNOS with or without Fe(III)EDTA was prepared. Before autoclaving, stock solutions of Fe(III)EDTA (10 mM in water) were added to MMNOS such that a final concentration of 250 or 500 µM was achieved. Fertility was evaluated as described above.
For G. zeae, CJ medium with or without Fe(III)EDTA or ferric citrate was prepared (23). Before autoclaving, stock solutions of Fe(III)EDTA (10 mM) or ferric citrate (10 mM) were added to CJ medium such that a final concentration of 125 or 250 µM was achieved in both cases. Crosses were set up as described above. Induction of sexual development was carried out with 2.5% Tween 60 with or without 125 µM Fe(III)EDTA or ferric citrate. Fertility was evaluated as described above.
Evaluation of fertility when ferricrocin is supplied. For C. heterostrophus, MMNOS with or without 200 µM ferricrocin was prepared. Ferricrocin (a generous gift from H. Haas, Innsbruck Medical University, Innsbruck, Austria) was applied to the medium after autoclaving. Crosses were set up by the standard method, and fertility was evaluated as described above.
For G. zeae, CJ medium with or without 100 µM ferricrocin was prepared. Crosses were set up as usual, except that induction of sexual development was carried out with 2.5% Tween 60 solutions with or without 100 µM ferricrocin. Fertility was evaluated as described above.
Expression analyses. Expression of C. heterostrophus NPS2 and G. zeae NPS1 and NPS2 under iron-depleted conditions was examined as described previously (23). Briefly, total RNA was extracted from 48-h cultures grown in 30 ml MM or in MM without iron. Reverse transcription-PCR was carried out with SuperScript III (Invitrogen) by following the manufacturer's recommendations. cDNA samples were normalized on the basis of the expression of the gene that encodes C. heterostrophus glyceraldehyde phosphate dehydrogenase (GPD1) or G. zeae actin (ACT1). The primers used for this study are available on request.
Sensitivity to iron depletion.
Sensitivity to iron depletion was examined as described previously (23), with the iron chelators 2,2'-dipyridyl (2DP) and bathophenathroline disulfonic acid (BPS). The MIC of the membrane-permeable iron chelator 2DP for each strain was determined by growing each strain on MM with twofold increasing concentrations of 2DP. Sensitivity to the membrane-impermeable iron chelator BPS was evaluated by comparing colony diameters of WT and nps2 strains (5 days old) on MM with or without 50 µM BPS.
Isolation, identification, and quantitative analysis of siderophores.
Fungal cultures for HPLC analyses were prepared as described previously (23). Extraction of extracellular siderophores from culture broth was carried out as described by Oide et al. (23). For isolation of intracellular siderophores, freeze-dried mycelia were ground in approximately 40 ml of 15 mM potassium phosphate buffer (pH 7.5) with a pestle and a spatula and the suspension was subjected to sonication for 10 min. The insoluble fraction was removed by centrifugation, and the supernatant was collected. The insoluble fraction was resuspended in fresh potassium phosphate buffer (
40 ml) and subjected to sonication for 10 min. After centrifugation to remove the insoluble fraction, the soluble fractions (about 80 ml) were combined and subjected to siderophore extraction on XAD-16 resin as described previously (23).
For quantitative analysis of siderophore production, crude siderophore fractions were treated to solid-phase extraction as described previously (23), reconstituted in 200 µl of a 50:50 mixture of methanol and HPLC mobile components (50:50 acetonitrile-15 mM ammonium acetate, pH 4.5), filtered through a 0.4-µm-pore-size polyvinylidene difluoride membrane, and then analyzed by HPLC with an injection volume of 10 µl for broth samples and 20 µl for mycelial samples. Analytical HPLC was conducted on a Polaris C18 5-µm column (250 by 4.6 mm) with a binary mobile phase consisting of acetonitrile (A) and 15 mM ammonium acetate buffer (pH 4.5) (B) with gradient elution at a flow rate of 1.0 ml/min (2.5% A for 1 min, a linear ramp up to 60% A in 13 min, 60% A for 2 min, and a linear ramp down to 2.5% A in 1 min). Detection was by absorption at 215 and 435 nm. Quantitative estimates of siderophore production were made by HPLC after determination of standard curves. Limits of detection were established conventionally at a signal-to-noise ratio of 3.
Coprogen, neocoprogen I, and neocoprogen II, used as reference standards, were purified as described previously (23). Ferricrocin and triacetylfusarinine C (TAFC), used as reference standards, were purified as described by Oberegger et al. (22). Electrospray mass spectra were obtained as described previously (23).
|
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
nps strains (carrying individual deletions of NPS1 to NPS12) was crossed to an albino WT strain of the opposite mating type, each resulting strain was fully fertile (data not shown). Both pigmented and albino pseudothecia formed, as in WT crosses, indicating that each nps strain could act either male or female (compare Fig. 1A and B to D and E, data shown for the nps2 strain). No clear difference was observed in the average number of asci per pseudothecium of nps crosses, compared to that of WT crosses (compare Fig. 1C with F). No significant reduction in the number of pseudothecia formed or in the number of asci per pseudothecium was observed, whether a nps strain acted male (as evidenced by the presence and fertility of albino pseudothecia) or female (as evidenced by the presence and fertility of pigmented pseudothecia) (Fig. 1N, left three bars).
|
nps2 strains are defective in sexual development.
When each of the C. heterostrophus nps strains with one of the 12 NPS genes deleted was homozygous in a cross, one type of cross (nps2 x nps2) was sterile. When any of the other 11 C. heterostrophus nps strains participated in a homozygous cross, the resulting strains were as fertile as in crosses to the WT (data not shown).
In the homozygous crosses with nps2 strains, WT numbers of both pigmented and albino pseudothecia were produced (Fig. 1J, K, and M, right bar); however, these were barren (Fig. 1L and N, right bar). No mature asci or ascospores developed in these pseudothecia, even when cross plates were incubated for 30 days after cross setup. WT crosses are generally mature by 21 days. Furthermore, results were the same when the deletion of the NPS2 gene was in a MAT1-1 background (compare Fig. 1G, H, and I to D, E, and F; middle bars in panel M; data not shown for ascus/ascospore development).
Reintroduction of the C. heterostrophus WT NPS2 ORF into the nps2 strain restored WT fertility capability to the nps2 strain (data not shown).
C. heterostrophus NPS2 encodes an NRPS responsible for ferricrocin biosynthesis.
Previous phylogenetic analyses indicated that C. heterostrophus NPS2 is closely related to NPS genes involved in ferrichrome-type siderophore biosynthesis in other fungal species (i.e., sidC of A. nidulans or sid2 of U. maydis) (18), suggesting that NPS2 also plays a role in ferrichrome-type siderophore biosynthesis in C. heterostrophus. In contrast to C. heterostrophus nps6 strains, which are defective in extracellular siderophore biosynthesis (23), C. heterostrophus nps2 strains do not show any increase in sensitivity to iron depletion on media including the iron chelator 2DP or BPS, compared to WT strains (Fig. 2A and B). Nevertheless, expression of ChNPS2 was very weakly induced under iron-depleted conditions (Fig. 2C), which suggested to us that C. heterostrophus NPS2 may have a role in iron metabolism. HPLC analyses of C. heterostrophus WT and nps2 strains revealed that the nps2 strain lacks the ability to biosynthesize ferricrocin, the intracellular siderophore of C. heterostrophus (Fig. 3, peak 2). The C. heterostrophus nps2 strain, however, still produces extracellular siderophores (broth extract, data not shown) and their presence is detectable in mycelial extracts (Fig. 3, peaks 1, 3, and 4; see Fig. 5 in reference 23), demonstrating that the NRPS encoded by C. heterostrophus NPS2 plays a role in intracellular siderophore biosynthesis but not in extracellular siderophore production.
|
|
|
NPS2 function is conserved between heterothallic C. heterostrophus and homothallic G. zeae.
To examine the functional conservation of NPS2, strains carrying a deletion of G. zeae NPS2 were characterized with respect to sexual development in this unrelated homothallic ascomycete. Like C. heterostrophus NPS2, G. zeae NPS2 was weakly up-regulated under iron-depleted conditions (Fig. 2C) and G. zeae nps2 strains showed WT tolerance to iron depletion (data not shown). As observed for nps2 strains of C. heterostrophus, G. zeae nps2 strains are defective in sexual development, in this case, in self-compatible sexual development. G. zeae nps2 strains form perithecia (Fig. 4D) as well as the selfed WT does (Fig. 4A), and there is no significant difference in the number of perithecia formed (Fig. 4C, left two bars). However, no mature asci or ascospores are found in these perithecia (Fig. 4E and F, middle bar), as for nps2 strains of C. heterostrophus.
|
nps2 strain revealed that it completely lacks ferricrocin in its cells, demonstrating that the NRPS encoded by G. zeae NPS2 is responsible for the biosynthesis of ferricrocin (Fig. 5, compare the top panel [arrow] to the panel second from the top). Biosynthesis of TAFC, the extracellular siderophore of G. zeae, was observed in the broth extract of the nps2 strain (data not shown), indicating that the NRPS encoded by G. zeae NPS2 does not play a role in TAFC biosynthesis. Quantitative analyses of TAFC biosynthesis indicated that the amount of TAFC produced was increased in nps2 strains (Fig. 6A and B, second bar from the right) compared to that in WT strains (Fig. 6A and B, left bar).
|
nps2 strains, G. zeae nps1 strains are as tolerant to iron depletion as the WT is (data not shown). In contrast to selfed strains of G. zeae nps2, however, selfed strains of G. zeae nps1 show WT fertility, producing perithecia (compare Fig. 4A and G) with mature asci and ascospores (compare Fig. 4B and H). No clear reduction in the number of perithecia (see Fig. S1A in the supplemental material) or in the number of asci per perithecium (see Fig. S1B in the supplemental material) was observed, compared to the WT.
Biosynthesis of both intracellular (arrow, Fig. 5, second row from the bottom) and extracellular (not shown) siderophores was observed in G. zeae nps1 strains. Quantitative HPLC analyses of WT and nps1 strains of G. zeae, however, indicated that the amount of both extracellular (Fig. 6A and B, second bar from the left) and intracellular (Fig. 6C and D, right bar) siderophore made by the nps1 strain was reduced compared to that of the WT strain (Fig. 6A to D, left bar), suggesting that the NRPS encoded by G. zeae NPS1 is somehow involved in siderophore biosynthesis.
The sexual development phenotype of strains carrying deletions of both G. zeae NPS1 and NPS2 is identical to that of G. zeae nps2 strains.
Double deletion of G. zeae NPS1 and NPS2 does not affect tolerance to iron depletion (data not shown). The nps1 nps2 strain shows a defect in sexual development similar to that observed for nps2 strains; perithecia form as in the WT and nps2 strains (Fig. 4I and C, right bar; compare with panels A, D, and C, left two bars), but there are no mature asci or ascospores (Fig. 4J and F, right bar [absence of]; compare with panels E and F, middle bar [absence of]).
As expected, G. zeae nps1 nps2 strains completely lacked the ability to biosynthesize ferricrocin (Fig. 5, bottom row), while they were still able to biosynthesize extracellular siderophores (not shown). Quantitative HPLC analyses showed that the amount of extracellular (Fig. 6A and B, right bar) siderophore made by the nps1 nps2 strain was increased compared to that made by the WT or nps1 strain (Fig. 6A and B, compare the right bar to the first and second bars from the left). No significant difference in extracellular siderophore biosynthesis was observed between the nps1 nps2 and nps2 strains (Fig. 6A and B, compare the right bar to the second bar from the right).
Exogenous application of iron partially restores fertility to C. heterostrophus nps2 strains.
The connection between the abnormal sexual development of C. heterostrophus nps2 strains and their lack of intracellular siderophore biosynthesis was further examined by characterizing the sexual development of nps2 strains supplied with iron during mating. Application of 250 µM Fe(III)EDTA partially restored fertility to C. heterostrophus nps2 strains (Fig. 7, compare panel B with panel D). C. heterostrophus nps2 strains form pseudothecia with some mature asci and ascospores, although the number of asci per pseudothecium is less than that of a WT cross (Fig. 7E; also, compare panel D with panels A and C). No significant difference in viability was observed between the ascospores recovered from these crosses and those from WT crosses (data not shown). Note that a reduction in the number of asci per pseudothecium was observed in WT crosses when iron was applied, indicating a deleterious effect of iron overload on the sexual reproduction of C. heterostrophus WT strains.
Exogenous application of iron partially or completely restores fertility to G. zeae nps2 and nps1 nps2 strains.
Similarly, an exogenous supply of iron restores fertility to nps2 and nps1 nps2 strains of G. zeae (Fig. 8; see Fig. S2 in the supplemental material). On a plate supplied with 250 µM Fe(III)EDTA, both the G. zeae nps2 and nps1 nps2 strains formed perithecia with mature asci and ascospores (see Fig. S2E and F in the supplemental material; compare to Fig. 8B and C, respectively), although the average number of asci per perithecium was less than that of a WT strain (Fig. 8). Application of 250 µM ferric citrate, however, restored WT levels of fertility to G. zeae nps2 and nps1 nps2 strains (see Fig. S2H and I in the supplemental material; compare to Fig. 8B and C, respectively). No statistically significant difference (P > 0.05) was observed in the number of asci per perithecium among the WT, nps2, and nps1 nps2 strains when ferric citrate was applied (Fig. 8). Thus, the G. zeae nps2 and nps1 nps2 strains show similar responses to an exogenous supply of iron, indicating that there is no significant difference between their defects in iron metabolism during sexual development.
|
nps2 strains demonstrate that the abnormal sexual development seen is at least partly due to their deficiency in nutrient iron resulting from loss of intracellular siderophores.
Exogenous application of ferricrocin restores WT fertility to nps2 strains.
Finally, we asked if application of ferricrocin, the intracellular siderophore of C. heterostrophus and G. zeae, is able to restore WT fertility to the nps2 strains. When ferricrocin was supplied, both the C. heterostrophus (Fig. 9) and G. zeae nps2 (Fig. 10; see Fig. S3 in the supplemental material) strains were as fertile as WT strains were. No significant differences in the number of asci per C. heterostrophus pseudothecium (Fig. 9E) or G. zeae perithecium (Fig. 10) were observed between the nps2 and WT strains, demonstrating that the observed defect in the sexual development of nps2 strains is due to loss of the ability to biosynthesize ferricrocin.
|
|
|
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
nps deletion strains, we have determined that a natural biological function of the intracellular siderophore ferricrocin, a metabolite biosynthesized by the NPS2 enzyme of the organisms that produce it, e.g., C. heterostrophus and G. zeae, is to participate in a fundamental fungal developmental process, i.e., production of sexual spores in meiosis. Thus, the observation that there is a close relationship between secondary-metabolite or small-molecule production and fungal development (7) is additionally documented. Furthermore, this observation provides support for the argument that products of multifunctional enzymes (e.g., NRPSs, best known as producers of secondary metabolites) can play fundamental roles in fungal cells themselves. The well-studied effects of these products on other organisms, either beneficial (e.g., as antibiotics) or harmful (e.g., as mycotoxins), may not address the primary "reason" why fungi and bacteria biosynthesize so-called secondary metabolites. The primary activities of these metabolites, one supposes, are those that benefit the fungal cell itself and not those that benefit pharmaceutical concerns.
Ferricrocin, the intracellular siderophore of C. heterostrophus, plays a role in sexual development.
In this study, we characterized our collection of 12 nps strains of C. heterostrophus with respect to a possible role in sexual development and found that the nps2 strains fail to develop asci in homozygous nps2 crosses. Phylogenetic (18) and expression analyses of NPS2 indicated that the NRPS encoded by NPS2 is involved in siderophore biosynthesis. MS and HPLC analyses demonstrated that the product of the NRPS encoded by NPS2 is ferricrocin, the intracellular siderophore of C. heterostrophus (see Fig. 3 and 5 in reference 23). Exogenous application of iron restored the ability of C. heterostrophus nps2 strains to develop asci and ascospores, albeit at a reduced level. Exogenous application of ferricrocin itself fully restored WT fertility to the nps2 strains. To our knowledge, this is the first demonstration that iron and siderophores play a role in the sexual development of a heterothallic ascomycete fungus.
Nutrient iron stored by intracellular siderophores is essential for ascus and ascospore development in C. heterostrophus.
Intracellular siderophores have been proposed to play a role in iron storage in fungi (36). In this study, exogenous application of iron in the form of Fe(III)EDTA partially restored the ability of C. heterostrophus nps2 strains to develop asci and ascospores, indicating that their defect in sexual development is at least partly due to loss of the ability to store iron and thus is due to starvation for nutrient iron. Whether a nps2 strain acts male or female makes no difference in crosses between WT and nps2 strains; these crosses are fully fertile, demonstrating that the stored iron supply in one of the two mating partners is sufficient for successful sexual development and that the mating type has no connection to the sexual developmental phenotype of nps2 strains. In WT x nps2 crosses, iron bound to intracellular siderophores is accessible to the nps2 mating partner presumably only after cell-cell fusion with a WT mating partner; thus, nutrient iron stored by intracellular siderophores must be relatively unimportant for development prior to cell-cell fusion.
Furthermore, C. heterostrophus nps2 strains form WT pseudothecia even in homozygous nps2 crosses, demonstrating that absence of iron storage capability by siderophores does not affect pseudothecium development. The block in the sexual development of C. heterostrophus nps2 strains appears to be before initiation of ascus formation, although rare immature ascus-like structures can be found in pseudothecia of homozygous nps2 crosses (not shown). In summary, nutrient iron stored by intracellular siderophores is essential for ascus/ascospore development in C. heterostrophus but dispensable for early sexual development, including pseudothecium formation.
Extracellular and intracellular siderophores play fundamentally different roles in fungal developmental biology.
Inhibition of extracellular siderophore biosynthesis through deletion of NPS6 leads to a reduction of pigmentation and asexual sporulation in C. heterostrophus (23). By contrast, nps2 strains show WT pigmentation and WT levels of asexual sporulation (data not shown), indicating that a major source of iron during asexual development of C. heterostrophus is iron influx through extracellular siderophores, rather than iron stored by intracellular siderophores. Extracellular siderophores are required for full virulence of C. heterostrophus to its host, while intracellular siderophores are dispensable for fungal infection, as evidenced by the WT virulence of the nps2 strain to the host. Thus, iron acquired by extracellular siderophores is a major source of iron also in planta. Extracellular siderophores are apparently dispensable for sexual development, since nps6 strains show WT fertility both in heterozygous and in homozygous nps6 crosses, indicating that iron stored in intracellular siderophores is more important for iron metabolism of C. heterostrophus during sexual development. On the basis of the data presented here and our previous work on extracellular siderophores of C. heterostrophus (23), we propose that intra- and extracellular siderophores play fundamentally different roles in C. heterostrophus.
Intracellular siderophores, reactive oxygen species (ROS), and fungal sexual development.
In addition to their role in iron storage, fungal intracellular siderophores are proposed to play a role also in sequestering iron to prevent Haber-Weiss/Fenton reactions, which generate highly cytotoxic hydroxyl radicals (22). The Haber-Weiss/Fenton reaction, summarized as O2·– + H2O2 
O2 + OH· + OH–, is catalyzed by intracellular free iron (14). To avoid this cytotoxic reaction, many organisms sequester intracellular iron in iron-binding proteins, such as transferrin and ferritin (16, 24). The antioxidant property of ferritins, through sequestration of iron, has been well characterized in bacteria, plants, and animals (3, 30). Ferritins, if present in fungi, are largely uncharacterized. A ferritin-like protein in a fungus was reported for the first time only recently (29). In fungal cells, the cytoprotective role that ferritin plays in other organisms (against hydroxy radicals produced by Haber-Weiss/Fenton reactions) is proposed for intracellular siderophores (22).
In A. nidulans, deletion of sidC, which encodes the NRPS responsible for intracellular siderophore biosynthesis, leads to an increase in intracellular free-iron levels and to hypersensitivity of mycelia to the redox-cycling agent paraquat (12). Asexual spores of an A. nidulans sidC strain show hypersensitivity to ROS more clearly than mycelia (13). In our experiments, no significant difference in sensitivity to ROS was observed between WT and nps2 strains of C. heterostrophus during vegetative growth (see Fig. S4 in the supplemental material). Likewise, a mutant strain of Schizosaccharomyces pombe which lacks the ability to biosynthesize intracellular siderophores is as tolerant to paraquat as is the WT (28). The role of intracellular siderophores as an antioxidant may be more important in fungal spores than in mycelia, at least in some fungi.
Sensitivity to iron overload stress was also examined in C. heterostrophus, but no clear difference was observed between the WT and nps2 strains (data not shown). Nevertheless, exogenous application of iron led to reduced fertility in a homozygous WT cross of C. heterostrophus, suggesting that excess iron does have adverse effects on the sexual development of this fungus. It is noteworthy that exogenous application of iron bound to ferricrocin did not cause a reduction of fertility in a homozygous WT cross of C. heterostrophus, indirectly supporting the hypothesis of the cytoprotective role of intracellular siderophores. Exogenous application of iron only partially restored fertility to C. heterostrophus nps2 strains in a homozygous nps2 cross, while application of ferricrocin fully restored fertility. A simple interpretation of these results is that iron was not as efficiently taken up or distributed by fungal cells as was ferricrocin. Alternatively, the results could imply that there is an as-yet-unknown function of intracellular siderophores in fungal sexual development, in addition to their role as a source of nutrient iron. Recent studies imply a role for ROS in fungal development (reviewed in reference 1). For example, expression of cpeA, which encodes a catalase-peroxidase in A. nidulans, is induced during sexual development (26). Perhaps fungal intracellular siderophores play an antioxidant role in sexual development, in addition to their role as a source of nutrient iron.
The roles of NPS2 in ferricrocin biosynthesis and in sexual development are conserved between the heterothallic ascomycete C. heterostrophus and the homothallic ascomycete G. zeae. To date, all of the known fungal siderophores are of the hydroxamate type (34), except for polycarboxylate siderophores produced by zygomycetes (31). Hydroxamate siderophores are largely divided into three groups, the ferrichrome type, the coprogen type, and the fusarinine type (25). Our previous study identified the gene (NPS6) that encodes an NRPS responsible for the biosynthesis of coprogen and its derivatives in C. heterostrophus (23). NPS6 is one of only a few NPS genes conserved among diverse species of fungi (18). We demonstrated that the NRPSs encoded by NPS6 orthologs are responsible for the biosynthesis of coprogen- and fusarinine-type extracellular siderophores in A. brassicicola and G. zeae, respectively, and proposed the functional conservation of NPS6 in coprogen- or fusarinine-type siderophore biosynthesis among filamentous ascomycetes (23). In the present study, we revealed that the NRPS encoded by C. heterostrophus NPS2 is responsible for biosynthesis of ferricrocin, a ferrichrome-type siderophore. NPS2 is also conserved among diverse species of fungi, including ascomycetes and basidiomycetes (18). The present study and an independent study by Tobiasen et al. (32) identified ferricrocin as the product of the NRPS encoded by G. zeae NPS2. sidC and sid2, to which NPS2 genes show strong similarity, encode the NRPSs responsible for the biosynthesis of ferricrocin and ferrichrome in A. nidulans and U. maydis, respectively (12, 38). Schwecke et al. (28) demonstrated that the NRPS encoded by the NPS2 ortholog of S. pombe (sib1) is responsible for ferrichrome biosynthesis. On the basis of in silico analyses, they also predicted that the NRPSs encoded by NPS2 homologs are likely to produce similar products in different species of fungi. Taking the findings together, we propose that the function of the NRPS encoded by NPS2 in ferrichrome-type siderophore biosynthesis is conserved among diverse species of fungi.
Our study on NPS6 demonstrated that the role of extracellular siderophores in fungal virulence to plant hosts is conserved among filamentous ascomycetes (23). Here, we demonstrated that the role of intracellular siderophores in sexual development is conserved in the heterothallic ascomycete C. heterostrophus and the homothallic ascomycete G. zeae. Deletion of the NPS2 ortholog in G. zeae resulted in concomitant defects in intracellular siderophore biosynthesis and ascus/ascospore development, as observed in C. heterostrophus nps2 strains. Application of ferricrocin fully restored WT fertility to the G. zeae nps2 strain. Recently, Eisendle et al. (13) reported that a sidC strain is impaired in sexual development in the homothallic ascomycete A. nidulans. Application of ferricrocin restored the ability of the sidC strain to undergo WT sexual development. Whether or not intracellular siderophores play a role in sexual development also in other species, such as the basidiomycete U. maydis and the fission yeast S. pombe, is the subject of future studies.
GzNPS1 has a potential role in the siderophore biosynthesis of G. zeae.
Like G. zeae NPS2, G. zeae NPS1 shows high similarity to A. nidulans sidC and U. maydis sid2 and groups with these genes and C. heterostrophus NPS2 in phylogenetic analyses, suggesting a role in ferrichrome-type siderophore biosynthesis. Tobiasen et al. (32) proposed also that G. zeae NPS1 is likely to encode the NRPS involved in siderophore biosynthesis, although no functional data were provided. Indeed, we found that G. zeae NPS1 is strongly up-regulated under iron-depleted conditions, further indicating its involvement in iron metabolism. In this study, however, we failed to obtain clear evidence that the NRPS encoded by G. zeae NPS1 has a role in siderophore biosynthesis. Deletion of NPS6 in G. zeae abolished extracellular siderophore biosynthesis (23), and deletion of G. zeae NPS2 led to complete loss of intracellular siderophore biosynthesis, demonstrating that the NRPSs encoded by NPS6 and NPS2 fully account for siderophore biosynthesis in G. zeae. Consistently, HPLC analyses of G. zeae nps1 strains confirmed the biosynthesis of ferricrocin and TAFC, the major intracellular and extracellular siderophores of G. zeae. Furthermore, any characteristic phenotype (sensitivity to low-iron and oxidative stress) associated with iron was not observed in G. zeae nps1 strains and the phenoty
