Devices & Hardware

The Aspergillus nidulans GATA Factor SREA Is Involved in Regulation of Siderophore Biosynthesis and Control of Iron Uptake

The Aspergillus nidulans GATA Factor SREA Is Involved in Regulation of Siderophore Biosynthesis and Control of Iron Uptake
of 8
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
  The  Aspergillus nidulans  GATA Factor SREA Is Involved inRegulation of Siderophore Biosynthesis and Control of Iron Uptake* (Received for publication, October 15, 1998, and in revised form, December 1, 1998) Hubertus Haas‡, Ivo Zadra, Georg Sto¨ffler, and Klaus Angermayr  From the Department of Microbiologie Medical School, University of Innsbruck, Fritz-Pregl Str. 3, A-6020 Innsbruck, Austria  A gene encoding a new GATA factor from  Aspergillusnidulans ,  sreA , was isolated and characterized. SREA displays homology to two fungal regulators of sid-erophore biosynthesis: about 30% overall identity toSRE from  Neurospora crassa  and about 50% identity toURBS1 from Ustilagomaydis over a stretch of 200 aminoacid residues containing two GATA-type zinc finger mo-tifs and a cysteine-rich region. This putative DNA bind-ingdomain,expressedasafusionproteinin  Escherichiacoli , specifically binds to GATA sequence motifs. Dele-tion of   sreA  results in derepression of   L -ornithine-  N  5 -oxygenase activity and consequently in derepression of the biosynthesis of the hydroxamate siderophore  N  ,  N   ,  N   -triacetyl fusarinine under sufficient iron sup-ply in  A. nidulans . Transcription of   sreA  is confined tohighironconditions,underscoringthefunctionofSREA as a repressor of siderophore biosynthesis under suffi-cient iron supply. Nevertheless, overexpression of   sreA does not result in repression of siderophore synthesisunder low iron conditions, suggesting additional mech-anisms involved in this regulatory circuit. Consistentwith increased sensitivity to the iron-activated antibiot-ics phleomycin and streptonigrin, the  sreA  deletion mu-tant displays increased accumulation of   59 Fe. These re-sults demonstrate that SREA plays a central role in ironuptake in addition to siderophore biosynthesis.  All microorganisms, with the exception of certain lactobacillithat utilize manganese and cobalt as biocatalysts in place of iron, require iron for their growth (1). The electron transferability of the iron atom, modified by diverse ligand environ-ments, makes it essential for redox reactions ranging fromrespiration to ribonucleotide synthesis. Although iron is thefourthmostabundantelementintheearthcrust,theamountof naturally utilizable iron is very limited for organisms as mostof this metal exists in extremely insoluble complexes of ferrichydroxide in nature. On the other hand, excess of free iron inthecellisdetrimental,becauseironcancatalyzetheproductionof cell-damaging hydroxyl radicals in the presence of oxygen.Therefore, the concentration of iron in biological fluids istightly controlled. Because most species lack an excretory routefor iron, control is primarily accomplished by regulating therate of iron uptake. Thereby, diverse species, from  Escherichiacoli  to human, have developed various and most often multipleiron transport mechanisms.The genus  Aspergillus  is one of the most ubiquitous micro-organisms worldwide and various  Aspergillus  species are fac-ultative pathogens involved in human infections capable of causing severe diseases like allergic bronchopulmonary as-pergillosis, aspergilloma, and invasive pulmonary aspergillo-sis. In particular, invasive pulmonary aspergillosis representsa life-threatening disease of increasing incidence in immuno-compromised patients, which has been attributed to increasing iatrogenic immunosuppression and successful life-supportmeasures for neutropenic patients. Because iron is tightly se-questered in mammalian hosts by high affinity iron-binding proteins, microbes require efficient iron-scavenging systems tosurvive and proliferate within the host. Under conditions of iron starvation, most fungi synthesize and excrete low molec-ular weight, iron-specific chelators called siderophores, whichhave therefore often been proposed as virulence factors. Sid-erophore biosynthesis and its impact on bacterial pathogenicityis well established (1). In contrast, the knowledge of the bio-chemical basis for siderophore biosynthesis and its regulationon the molecular level is rather fragmentary in eukaryoticorganism. This may be in part because of the fact that theleading model organism for molecular genetic analysis,  Saccha-romyces cerevisiae , lacks the ability to synthesize siderophores,although it can utilize siderophores produced by other species(2). In the basidiomycete  Ustilago maydis , a gene encoding atranscriptional repressor of siderophore biosynthesis belonging to the GATA protein family of transcription factors has beencharacterized (3). Subsequent genetic and molecular studieshave shown that URBS1 acts as a transcriptional repressor bydirect binding to GATA motifs in the promoter of at least  sid1 ,encoding   L -ornithine  N  5 -oxygenase, the first committed step inthe biosynthesis of the two  Ustilago  siderophores ferrichromeand ferrichrome A (4, 5).In a general search for GATA factor-encoding genes, we haverecently isolated a gene from the ascomycete  Penicillium chry-sogenum  displaying significant homology to  Ustilago urbs1 ,and the disruption of the corresponding gene from  Neurosporacrassa  proved that this gene encodes a repressor of siderophorebiosynthesis (6, 7). In this study, we report the cloning andcharacterization of   sreA , a GATA factor-encoding gene respon-sible for repression of siderophore biosynthesis in  Aspergillusnidulans  and show its involvement in the control of iron up-take. This study represents a first important step in the inves-tigation of the regulation of siderophore biosynthesis in  As- pergilli  and provides the basis for evaluation of the impact of iron metabolism on pathogenicity in these fungal species. EXPERIMENTAL PROCEDURES  Strains, Vectors, Growth Media, and General Molecular Tech-niques— The vectors and plasmids were propagated in  E. coli  DH5  supplied from Life Technologies, Inc. Generally,  A. nidulans  strain A4(Glasgow wild type) provided from the Fungal Genetic Stock Center,Kansas City, KS, was used. For fungal transformation,  A. nidulans * This work was supported in part by Austrian Science FoundationGrant FWF-P13202-MOB (to H. H.). The costs of publication of thisarticle were defrayed in part by the payment of page charges. Thisarticle must therefore be hereby marked “ advertisement ” in accordancewith 18 U.S.C. Section 1734 solely to indicate this fact. The nucleotide sequence(s) reported in this paper has been submittedtotheGenBank TM /EBIDataBankwithaccessionnumber(s) AF095898. ‡ To whom correspondence should be addressed. Tel.: 43-512-507-3608; Fax: 43-512-507-2866; E-mail: T HE  J OURNAL OF  B IOLOGICAL  C HEMISTRY   Vol. 274, No. 8, Issue of February 19, pp. 4613–4619, 1999 © 1999 by The American Society for Biochemistry and Molecular Biology, Inc.  Printed in U.S.A. This paper is available on line at  4613   b  y g u e  s  t   onM a  y1 2  ,2  0 1  6 h  t   t   p :  /   /   w w w . j   b  c  . or  g /  D o wnl   o a  d  e  d f  r  om  strain WG355 ( biA1 bgaO argB2 ), kindly provided by Dr. A. Brakhage,was used (8). SRKO1 and SROE3, generated in this work (see below),represent  A. nidulans  strains with a deletion of the  sreA  gene and astrain overexpressing   sreA , respectively. Generally,  A. nidulans  wasgrown in minimal medium according to Pontecorvo  et al.  (9) containing 1% glucose as the carbon source and 30 m M  ammonium tartrate as thenitrogen source. High iron medium contained 10   M  FeSO 4 , and forpreparation of low iron medium, addition of FeSO 4  was omitted. If required, biotin (20   g/liter) or  L -arginine (200 mg/liter) was added tothe media. YAG (2% glucose, 0.5% yeast extract, 2% agar, trace ele-ments) was used as complete medium.Standard molecular techniques were performed as described by Sam-brook  et al.  (11). Fungal DNA was isolated according to Yelton  et al. (12); for RNA isolation, RNAzol TM (Biotex Laboratories, Inc) was used.  Preparation of Maltose-binding Protein-SREA Fusion Proteins— Forexpression of the N-terminal SREA zinc finger (NZF), a fragment en-coding the SREA amino acids 80–175 was amplified from cDNA em-ploying the primers 5  -TTTTGAATTCGAGACTCCAATGAACG (OSR1)and 5  -TTTTCTGCAGTTAACCTTCGCTTCCAGTA (OSR2), carrying add-on restriction enzyme cleavage sites for  Eco RI and  Pst I. Subse-quently, the amplified product was cleaved with  Eco RI and  Pst I andligated into the respective restriction sites of pMal-cRI (New EnglandBiolabs). After verification of proper integration of the relevant frag-ment into the plasmid by sequencing, the resulting plasmid was trans-formed into  E. coli  DH5  . Expression and purification of the maltose-binding protein-NZF fusion protein were carried out according to themanufacturer’s recommendations. For expression of the C-terminalSREA zinc finger (CZF, amino acid residues 229–315) and the peptidecontaining both zinc fingers (NCZF, amino acid residues 80–315), thesame strategy was employed. For amplification of the CZF-encoding fragment primers, 5  -TTTTGAATTCCCGTCTCCAGAGGCTGA (OSR5)and 5  -TTTTCTGCAGTTAATGGGTAGCAGCAGTC (OSR6) were used,and the NCZF-encoding fragment was amplified employing the primersOSR1 and OSR6.  Northern Analysis— Generally, 15   g of total RNA was electrophore-sed on 1.2% agarose-2.2  M  formaldehyde gels and blotted onto HybondN membranes (Amersham Pharmacia Biotech). Hybridization probeswere generated by PCR 1 using oligonucleotides 5  -CTCGCCCCCATAC-TAAA and 5  -CTTGCTATCATTCTTGC for the  sreA  probe A, 5  -TTTC-GAGTCGCTAGGCT and 5  -TCCGTCCTCTCCCCTTT for the  sreA probe B, 5  -TTCGCTCCGTACTCAAG and 5  -GAGTAGCGACAGCA- ATG for the  argB  probe C, 5  -ATCGCCAGAAGCATCGT and 5  -ACT-GAAAAGGTTATCGCT for the  sreA  probe D (see Fig. 4), and 5  -CGG-TGATGAGGCACAGT and 5  -CGGACGTCGACATCACA for  actA (GenBank TM accession number U61733).  Electrophoretic Mobility Shift Assays of DNA-Protein Interactions— The 130-bp fragment (F130) containing two GATA core elements de-scribedinHaas  etal. (10)waslabeledbyend-fillingof5  overhangswithKlenow DNA polymerase (Boehringer Mannheim) and [  32 P]dATP (11).For assessing binding specificity, each oligonucleotide described in Fig.3 was annealed with its complementary oligonucleotide by incubationin 50 m M  Tris-HCl, pH 7.5, 100 m M  NaCl for 5 min at 90 °C andsubsequent cooling to 4 °C. Radioactive end-labeling was performed byfillinginwithKlenowfragmentofDNApolymeraseIusing[  - 32 P]dCTP(11). Electrophoretic mobility shift assays were carried out as describedfor NREB (10). Gene Disruption and Overexpression of sreA— For disruption of   sreA, the 2.4-kbp  Nco I-  Nsi I fragment of the  sreA  containing the 4.6-kbp  Eco RI fragment was replaced by the 2.6-kbp  Pst I-  Bam HI  argB -encod-ing fragment from pILJ16 (13) subsequent to blunt-ending of the  Nco Iand  Bam HI restriction sites by filling in using Klenow fragment of   E.coli  polymerase I. pILJ16 was provided by Dr. M. X. Caddick (Univer-sity of Liverpool, UK). The resulting plasmid was digested with  Eco RI,and the 4.8-kbp fragment was gel-purified before transformation of   A.nidulans  strain WG355.For overexpression of   sreA  in  A. nidulans,  a translational fusion of the  sreA -coding region with the  gpdA  promoter was constructed. There-fore, a 1.8-kbp  sreA  fragment was amplified from genomic DNA em-ploying primers 5  -TTCTTTGGATCCACTTCTCTTGTCATTGA and 5  -TTCTTTCCATGGTAGCGTCCATCCCGCAC carrying add-on restric-tion enzyme cleavage sites for  Nco I and  Bam HI, as primers. Subse-quently, the amplified product was cleaved with  Nco I and  Bam HI andligated into the respective restriction sites of pAN52–1 (Ref. 14; Gen-Bank TM accession number Z32697). pAN52–1 was provided by Dr. P.Punt (TNO, Nutrition and Food Research Institute, The Netherlands).The junctions as well as the coding regions of the fused product were verified by sequencing. The resulting plasmid was co-transformed into  A. nidulans  WG355, with the  argB  carrying vector pILJ16.Transformation of   A. nidulans  was carried out according to Tilburn  et al.  (15). Screening of positive clones was performed by PCR fromfungal colonies. For screening of   sreA  disruption strains, the primersOsre, 5  -CGCTAATCCCGCCATCG, and Oarg, 5  -TTCCTCTGCT-GCGTCCG, were employed (see Fig. 4). To obtain homokaryotic trans-formants, colonies from single homokaryotic spores were picked, andgenomic integration was analyzed by Southern blot analysis.  Identification and Quantification of Siderophores— The siderophoreconcentration was determined by means of the ferric perchlorate assay,the chrome azurol S liquid assay, and quantitative HPLC analysis (16,17). For reversed phase HPLC analysis of siderophores according toKonetschny-Rapp  et al.  (17), siderophores were isolated according to theprocedure described for isolation of ferrichrome from  Ustilago sphaero- gena  (16). Isolated ferric siderophores were photometrically quantifiedat 435 nm. The siderophores neocoprogen I, neocoprogen II, coprogen,and triacetylfusarinin C used as standards for HPLC analysis were agift of Dr. G. Winkelmann (Universita¨t Tu¨bingen, Germany).  L -Ornithine-N  5 -oxygenase Enzyme Assay— L -ornithine-  N  5 -oxygenaseenzyme activity was determined as already described (7). Quantification of Resistance/Sensitivity to Drugs— Quantification of resistance/sensitivity of   A. nidulans  to phleomycin and streptonigrinwas performed as described by de Souza  et al.  (18). 10 9 conidia of therespective  Aspergillus  strain were spread onto a YAG plate and incu-bated for 12 h at 37 °C to produce a nonsporulating mycelial mat.Mycelial “plugs” were cut out from the mats using the wide end of asterile Pasteur pipette and transferred in three repetitions onto platescontaining various drugs or chelators mentioned below and incubatedfor 72 h at 37 °C. The diameters of the resulting colonies were meas-ured, and the radial growth was compared with the average diameter of colonies on zero-dose plates expressed in %. Subsequently, the growthrate of the respective strain was then correlated to that of the  A.nidulans  wild type. This technique avoids problems of delayed germi-nation and ensures measurement of true exponential-phase growthrate. Drugs were used in YAG with the following concentration: phleo-mycin (Sigma), 50   g/ml; streptonigrin (Sigma), 10   g/ml; benomyl(Sigma), 1   g/ml; EDTA and bathophenanthrolinedisulfonic acid, 0.1m M  each. In the case of low iron conditions, minimal medium lacking iron was used, and growth was compared with that on minimal mediumcontaining 10   M  FeSO 4 .  Iron Uptake— For studying iron uptake, plates containing 50 nCi of  59 FeCl 3  (Amersham Pharmacia Biotech)/ml of medium were overlaidwith a dialysis membrane (Serva, 67 mm). Subsequently, mycelialplugs of the different  Aspergillus  strains, prepared as described in Quantification of Resistance/Sensitivity to Drugs  (18), were transferredon top of these dialysis membranes. After incubation for 72 h at 37 °C,the dialysis membranes carrying the fungal colonies were removed fromthe plates, and the  59 Fe content of the colonies was determined using aPhosphor Storage Imaging System, model Storm 840 (MolecularDynamics). RESULTS AND DISCUSSION  Isolation and Characterization of sreA— The various mem-bers of the GATA protein family are characterized by a highdegree of similarity within their DNA binding domain. Toisolate the  sreP  homologous gene from  A. nidulans,  a PCR-aided strategy similar to the one used for the cloning of theGATA factor-encoding genes  sreP  and  nreB  from  P. chrysoge-num  was employed (6, 10). Using the degenerated primers5  -ACNCCNYTNTGGMG and 5  –APNCCPCANGCPTTPCA (Y     T or C; M    A or C; N    any nucleotide; P    A or G)derived from the amino acid sequences Thr-Pro-Leu-Trp-Arg and Cys-Asn-Ala-Cys-Gly-Leu, conserved in most fungal GATA factors, a 56-bp fragment was amplified from genomic  Asper- gillus  DNA. The obtained fragments were subcloned into plas-mid vectors and sequenced whereby the deduced amino acidsequence of one fragment displayed only a single amino acidexchange compared with the corresponding region of SREP.The encoding fragment was radiolabeled and utilized to probethe  Aspergillus  cosmid library constructed in pWE15 provided 1 The abbreviations used are: PCR, polymerase chain reaction; bp,base pair(s); kbp, kilobase pair(s); HPLC, high performance liquid chro-matography; NZF, N-terminal SREA zinc finger; CZF, C-terminalSREA zinc finger; NCZF, both zinc fingers.  Regulation of Iron Homeostasis in Aspergillus 4614   b  y g u e  s  t   onM a  y1 2  ,2  0 1  6 h  t   t   p :  /   /   w w w . j   b  c  . or  g /  D o wnl   o a  d  e  d f  r  om  by the Fungal Genetic Stock Center (19). Five hybridizing clones, W04B12, W09A09, W09E09, W10A09, and W10E09,were detected. According to the information made available bythe Fungal Genetic Stock Center, the inserts of all five cosmidssrcinate from chromosome VIII, localizing   sreA  to chromosome VIII as well. Subsequently, a 4.6-kbp  Eco RI fragment carrying  sreA  was subcloned from cosmid W09E09 into pBluescriptKSand sequenced in its entirety. Additionally, 0.4 kbp of the3  -downstream region of   sreA  have been sequenced directlyfrom the cosmid W09E09. The transcription start points andthe polyadenylation site of   sreA  were mapped by 5  - and 3  -rapid amplification of cDNA ends protocols according to Fro-hman  et al.  (20). Two major transcription start points werelocalized 1073 bp and 211 bp upstream of the putative trans-lation start codon. The polyadenylation site was mapped 435 bpdownstream of the translation stop codon. Two introns weremapped by amplification of the coding region from mRNA byreverse-transcribed PCR.Comparison of the genomic and cDNA sequences revealed anopen reading frame of 1647 bp interrupted by two introns, 55and 60 bp in length. The 5   and 3   borders of the introns as wellas the putative lariat formation sites perfectly match the con-sensus sequences for fungal introns (21). The deduced SREA protein displays a calculated mass of 58.8 kDa. Searches inseveral data bases using the BLAST alignment computer pro-gram confirmed that the cloned gene encodes a member of theGATA protein family (38). In contrast to all other identifiedfungal GATA factors, SREA as well as SREP from  P. chryso- genum , SRE from  N. crassa , URBS1 from  U. maydis , andGAF2p from  Schizosaccharomyces pombe  contain two zinc fin-gers (3, 6, 7, 22). Additionally, the sequence homology in thesefive proteins is not limited to the two zinc finger regions butalso extends into the intervening region, where a conserved27-amino acid residue sequence with four cysteines is present(Fig. 1). Moreover, SREA, SREP, and SRE share a commonhighly conserved C terminus predicted to form a coiled coilstructure by computer analysis using the ExPASy tools soft-ware package, indicative of a putative protein-protein interac-tion domain (23). SREA displays the highest overall identity to  Penicillium  SREP and  Neurospora  SRE with 61% and 35%,respectively.The positions of the two introns are perfectly conserved in  A.nidulans sreA ,  P. chrysogenum sreP , and  N. crassa sre , dating the intron srcin before the evolutionary split of the ancestorsof these three ascomycetes (6, 7). In contrast,  urbs1  from  U.maydis  and  GAF2p  from  S. pombe  do not contain any introns(3, 22).  Expression of sreA— Northern blot analysis indicates thatexpression of   sreA  is sensitive to the iron content of the medium(Fig. 2). The steady state  sreA  mRNA levels are significantlyelevated in mycelia grown in high iron medium or transferredfrom low iron medium into high iron medium. In contrast, inmycelia grown in low iron medium or transferred from highiron medium into low iron medium, no  sreA  transcripts can bedetected. In this respect, transcription of   sreA  differs signifi-cantly from that of   Ustilago urbs1  and  Neurospora sre,  where it F IG . 1.  Alignment of   A. nidulans  SREA,  P. chrysogenum  SREP,  N. crassa  SRE,  U. maydis  URBS1, and  S. pombe  GAF2p.  Amino acidresidues identical in at least three of the compared proteins are  boxed  in  gray . The GATA-type zinc fingers ( GTZ ), the cysteine-rich region ( CRR ),and the conserved C terminus ( CT)  are  boxed  in  black.  Regulation of Iron Homeostasis in Aspergillus  4615   b  y g u e  s  t   onM a  y1 2  ,2  0 1  6 h  t   t   p :  /   /   w w w . j   b  c  . or  g /  D o wnl   o a  d  e  d f  r  om  was found to be constitutive (3, 7). Analogous to the situation in  P. chrysogenum ,  sreA  is expressed via two transcripts, 3.2 kband 2.4 kb in length. Northern blot analysis employing twodifferent hybridization probes, probe A placed in the regionbetween the two major transcription start points and probe Bsrcinating from the coding region, proved that the two tran-scripts are indeed because of different initiations of transcrip-tion found by the 5  -rapid amplification of cDNA ends proce-dure (Fig. 2). sreA Encodes a GATA-binding Protein— GATA factors aredefined as GATA-binding proteins that recognize the core motif GATA. To investigate whether  sreA  in fact encodes a DNA-binding protein, three SREA peptides containing the NZF, theCZF, and both zinc fingers (NCZF), respectively, were ex-pressed as fusion products with the maltose-binding protein in  E. coli  and purified by affinity chromatography. In  A. nidulans ,an SREA-regulated gene has not been isolated so far; therefore,no putative target binding sequence is available. Recently itwas shown that NIT-2, WC-1, WC-2, and NGF1, four GATA-factors of   N. crassa  that are involved in diverse regulatorycircuits, bind to the same nucleotide sequence containing GATA motifs, thus displaying little preference for flanking regions (24). Therefore, a fragment containing two GATA mo-tifs, which was already shown to bind the  Penicillium  GATA factor NREB, was employed to test the DNA binding potentialof the SREA peptides (10). As shown in Fig. 3, both zinc fingersare required for specific DNA binding, at least to the sequenceprovided, because only NCZF was able to shift the 130-bp DNA fragment (F130). In contrast to the SREA homologous proteins,all other fungal GATA factors identified so far contain only asinglezincfinger,whichprovedtobesufficientforspecificDNA binding. On the other hand, the requirement of two GATA-typezinc fingers for recognition of certain DNA sequences has beendemonstrated for vertebrate GATA factors: in murine andchicken GATA-1, the C-terminal finger is sufficient for specificDNA binding and certain functions  in vivo , but the N-terminalfinger can help to achieve binding specificity and stability,whereby binding to palindromic GATA sites is dependent onthe presence of both finger domains (25–28). Interestingly, amutation of the conserved amino acid residue Arg-350 in theN-terminal  URBS1  zinc finger does not affect regulation of siderophore biosynthesis in  U. maydis , whereby the corre-sponding mutation in the C-terminal zinc finger abolishes thefunction of URBS1 (4). It cannot be ruled out that the C-terminal finger of SREA might recognize specific DNA se-quences on its own, but for recognition of the provided se-quence, obviously both fingers are necessary.To investigate the involvement of the two GATA motifs pres-ent in F130, 48 bp of double-stranded oligonucleotides contain-ing both GATA motifs as well as mutations of one or both siteswere employed in mobility shift analysis. As shown in Fig. 3,the mutation of either single GATA site has no apparent effecton the  in vitro  binding of NCZF, whereas the mutation of bothGATA sites eliminates binding. These results prove that  sreA indeed encodes a GATA-binding protein. In comparison,  Usti-lago  URBS1 also specifically binds to a single GATA motif inthe  sid1  promoter  in vitro , but for  in vivo  function, two clus-tered sites are necessary (5). sreA Encodes a Repressor of Siderophore Biosynthesis— Dis-ruption of   sreA  was achieved using the 4.6-kbp  Eco RI  sreA genomic fragment in which the promoter region and the firsttwo exons of   sreA  were replaced by the  argB -encoding genefrom  A. nidulans  (Fig. 4). The resulting 4.8-kbp  Eco RI frag-ment was used to transform an  argB  strain of   A. nidulans (WG355).  Aspergillus argB  /sreA  strains were isolated andcharacterized by PCR (using primers Osre and Oarg, see Fig.4). Southern blot analysis proved that the  sreA  disruptionstrain, called SRKO1, does not contain the first two  sreA  exons(probe B) and that these exons are replaced by the  argB  gene,resulting in a 4.8-kbp  Eco RI fragment instead of the srcinal4.6-kbp fragment (probe C), also confirmed by hybridizationwith probe D, srcinating from the third  sreA  exon (Fig. 4). F IG . 2.  Northern analysis of   sreA  expression in  A. nidulans wild type, SRKO1, and SROE3.  A. nidulans  strains were grown for36 h in high iron minimal medium (Fe  ) or low iron minimal medium(Fe  ). Subsequently, mycelial pads were washed and transferred fromFe  into Fe  media (Fe -S ) or from Fe  into Fe  media (Fe  S ), respec-tively, and grown for another hour. Total RNA was isolated from my-celia of the different growth conditions, separated on an agarose/form-aldehyde gel, blotted onto membrane filter, and hybridized with theradiolabeled  sreA probe A  or  B. Probe A  corresponds to the regionbetween the two transcription start sites, and  probe B  srcinates fromthe coding region (see Fig. 5). As the control for loading and quality of RNA, blots were hybridized with the    –actin-encoding gene of   A. nidu-lans  (39).F IG . 3.  Electrophoretic gel mobility shift analysis of SREA andtarget DNA.  A , sequences of the 48-bp oligonucleotides used in gelmobility shift analysis.  O1  contains both wild-type GATA motifs pres-ent in  F130, O2  contains only the 5  -GATA motif,  O3  contains only the3  -site, and  O4  lacks any GATA sequence.  B,  gel shift analysis employ-ing   32 P-labeled  F130 , the 130-bp fragment of the  nreB  5  -upstreamregion, or radiolabeled  O1-O4. NCZF   represents a maltose-binding fusion protein containing both the N- and C-terminal zinc fingers of SREA.  NZF   and  CZF   contain only the N-terminal or the C-terminalzinc finger, respectively. The amount of fusion protein used was 0.2   g.  Regulation of Iron Homeostasis in Aspergillus 4616   b  y g u e  s  t   onM a  y1 2  ,2  0 1  6 h  t   t   p :  /   /   w w w . j   b  c  . or  g /  D o wnl   o a  d  e  d f  r  om  Moreover, in SRKO1, no  sreA  transcripts could be detected(Fig. 2).To check if   sreA  indeed encodes a regulator of siderophorebiosynthesis,  Aspergillus  wild-type and SRKO1 strain weregrown in high iron and low iron medium, respectively, andsiderophore production was analyzed and quantified by re- versed phase HPLC (Fig. 5). The amount of siderophores pro-duced was confirmed employing the perchlorate assay and thechrome azurol S liquid assay (data not shown). As described byCharlang   et al.  (29),  A. nidulans  wild type was found to excreteexclusively the siderophore type  N  ,  N   ,  N   -triacetyl fusarinineC in significant amounts, which holds for most other  Aspergilli but only under iron starvation conditions (30). In contrast,SRKO1 produced siderophores irrespective of the iron contentof the growth medium, whereby again only  N  ,  N   ,  N   -triacetylfusarinine C was found in the culture broth. However, sid-erophore production was not completely derepressed in SRKO1under high iron conditions; it reached only about 10–20% theamount produced under iron starvation. These data indicatethat in addition of lifting of repression by SREA, other factorsare presumably involved in the expression of siderophore bio-synthetic genes that are not active under sufficient iron supplyas proposed for  N. crassa  (7). L -ornithine-  N  5 -oxygenase represents the first committedstep in the biosynthesis of most hydroxamate type siderophores(31). In  Aspergillus  wild type, only 10% of this enzyme activitycan be detected under growth with high iron supply relative togrowth under iron limiting conditions. In contrast, SRKO1contains only about 70%  L -ornithine-  N  5 -oxygenase activitycompared with wild type but similar enzyme activity in highand low iron conditions (data not shown), indicating that theexpression of the encoding gene is under control of SREA,comparable with the situation in  U. maydis  (31).Because transcription of   sreA  is confined to high iron condi-tions, it was tempting to speculate that constitutive expressionof   sreA  might block siderophore biosynthesis under low ironconditions. Therefore, the  sreA -coding region was placed undercontrol of the promoter of the glycerine aldehyde 3-phosphatedehydrogenase (  gpdA )-encoding gene using the  A. nidulans expression vector pAN52–1 and co-transformed with a plasmidcarrying the  argB  gene into WG355.  Aspergillus  strains carry-ing the  sreA  expression cassette were screened by Southernblot analysis after selection for arginine prototrophy. InSROE3, an  Aspergillus  strain containing a single copy of the sreA  expression cassette (data not shown), siderophore produc-tion was unaffected under iron starvation despite high levelconstitutive  sreA  transcription, as proved by Northern analysis(Fig. 2). Important to note, the  sreA  mRNA is significantlyshorter in SROE3, because in the  sreA  expression vector, thelong   sreA  5  -untranslated region is replaced by the small one of the  gpdA  gene. These data indicate that posttranscriptionalregulation is involved in the activation of repressor function of SREA as proposed for the respective corresponding gene prod-ucts in  N. crassa  and  U. maydis.  In the latter two organisms, F IG . 4.  Deletion of the  sreA  gene in  A.nidulans .  A  and  B , restriction maps of the 4.6- and 4.8-kbp  Eco RI fragments con-taining the wild-type sreA sequence (  A )and the construct used for  sreA  deletion(  B ), respectively.  Open boxes  represent sreA  exons, and the  gray box  marks the argB  gene.  Osre  and  Oarg  symbolizeprimers used for PCR screening of   sreA deletion mutants.  Hatched boxes  repre-sent the hybridization probes used inNorthern and Southern analysis.  tsp  and  pas  mark the transcription start pointsand the polyadenylation site, respec-tively.  C , Southern analysis of the  sreA deletion strain SRKO1 in comparison tothe wild type.F IG . 5.  Reversed phase HPLC analysis of siderophore produc-tionof   A.nidulans wildtype,SRKO1,andSROE3. The  Aspergillus strains were grown for 48 h in high iron minimal medium (Fe  ) or lowiron minimal medium (Fe  ). Subsequently, siderophores were isolatedfrom 50 ml of growth medium as described under “Experimental Pro-cedures” and analyzed. As siderophore standards, neocoprogen I (  NI  ),neocoprogen II (  NII  ), coprogen ( C ), and  N  ,  N   ,  N   -triacetylfusarinin C( TFC ) were used. The absorption at 380 nm is denoted by milliabsorp-tion units ( mAU  ).  Regulation of Iron Homeostasis in Aspergillus  4617   b  y g u e  s  t   onM a  y1 2  ,2  0 1  6 h  t   t   p :  /   /   w w w . j   b  c  . or  g /  D o wnl   o a  d  e  d f  r  om
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!