Although iron and copper are co-ordinately regulated in living cells, the homeostatic effects of each of these metals on the other remain unknown. Here, we show the function of AfMac1, a transcriptional activator of the copper and iron regulons of Aspergillus fumigatus, on the interaction between iron and copper. In addition to the copper-specific AfMac1-binding motif 5′-TGTGCTCA-3′ found in the promoter region of ctrC, the iron-specific AfMac1-binding motif 5′-AT(C/G)NN(A/T)T(A/C)-3′ was identified in the iron regulon but not in the copper regulon by ChIP sequence analysis. Furthermore, mutation of the AfMac1-binding motif of sit1 eliminated AfMac1-mediated sit1 up-regulation. Interestingly, the regulation of gene expression in the iron regulon by AfMac1 was not affected by copper and vice versa. AfMac1 localized to the nucleus under iron- or copper-depleted conditions, and AfMac1 was mostly detected in the cytoplasm under iron- or copper-replete conditions. Taken together, these results suggest that A. fumigatus independently regulates iron and copper homeostasis in a manner that involves AfMac1 and mutual interactions.
Copper and iron are essential cofactors in various enzymatic processes, such as oxidative phosphorylation, nucleotide biosynthesis, and the detoxification of reactive oxidative species, and harmful effects occur when the levels of these metals are limited in cells . However, copper and iron can cause toxic effects through the Fenton reaction when they accumulate to high levels [2–4]. Therefore, living cells regulate the cellular concentrations of copper and iron. Although copper and iron homeostasis has been studied for a long time using various organisms, its detailed mechanism has not yet been determined.
Saccharomyces cerevisiae has been used to study iron and copper metabolism. In S. cerevisiae, reduced transition metals from the environment are transported into cytoplasmic compartments by metal-specific transporters [5–7]. Both reductive and non-reductive iron uptake systems have been identified in S. cerevisiae. In the reductive iron uptake system, reduced ferrous iron is oxidized to ferric iron by Fet3 multicopper oxidase, which is located on the plasma membrane. Four copper ions are post-transcriptionally inserted into this enzyme for activation by Ccc2p at the Golgi apparatus [8,9], and Ftr1 permease then imports the oxidized ferric iron into the cytosol [5,10]. The imported iron is transferred to superoxide dismutase (SOD, specifically Lys7, in this example) , cytochrome c oxidase (Sco1) [12,13], and other proteins. The expression of genes within the iron regulon is regulated by Aft1 and Aft2, iron-sensing transcription factors that are regulated by iron status of the cell [14–16]. Although a reductive iron uptake system exists in filamentous fungi, the non-reductive iron uptake system of fungi affects their virulence. For example, the siderophore biosynthesis and uptake systems are well developed, and they are involved in controlling virulence of the fungal pathogen Aspergillus fumigatus . In A. fumigatus, both reductive and non-reductive iron uptake systems have been identified. However, reductive iron uptake machinery is not required for pathogenic pathway in contrast with Candida albicans and Cryptococcus neoformans [18,19]. In A. fumigatus, FtrA protein is identified as a membrane iron permease, and deletion of FtrA does not affect the growth of A. fumigatus under iron-starved condition . On the other hand, siderophore uptake machinery has been recognized as a virulence factor of A. fumigatus, and MirB and Sit1 are identified as membrane siderophore transporters. MirB and Sit1 transport tri-acetyl fusarinine C and ferrichrome (FC) from environment, respectively, and siderophore uptake activity is up-regulated by iron starvation [21,22]. Under iron-starved condition, siderophore uptake is activated by up-regulating expression of the genes encode siderophore transporters. Furthermore, siderophore biosynthesis is activated under same condition . The genes involved in siderophore biosynthesis are an important virulence factor in A. fumigatus . And expression of the genes involved in high-affinity iron uptake system such as reductive and siderophore uptake is up-regulated by HapX, which is an iron-responsive transcription activator . HapX maintains iron homeostasis in A. fumigatus by activating iron uptake system but inhibiting iron consumption under iron-depleted condition and is a virulence factor of A. fumigatus . In fact, C. albicans and Cr. neoformans also use siderophores as iron sources, and several membranous siderophore transporters have been identified [18,19]. However, the reductive iron uptake system affects virulence in C. albicans and Cr. neoformans, and siderophore transporters are not involved in their pathogenicity. Thus, with respect to their iron uptake systems, these species have developed differently from A. fumigatus.
In the case of copper, the reduced metal is transported into the cytosol matrix by the Ctr1/Ctr3 high-affinity copper uptake system of S. cerevisiae [6,24]. Copper is targeted to organelles or various proteins by various copper chaperones. Ccc2, a P-type ATPase transporter, incorporates four copper molecules into a post-Golgi compartment of the secretory pathway for Fet3 multicopper oxidase [8,9,25]. Atx1, a cytosolic copper-binding protein, delivers copper to Ccc2 . Cox17 is a metallochaperone that is involved in the transport of copper into mitochondria, and Ccs transports copper to Cu/Zn SOD [11,26]. When copper is limited in cells, copper regulons are activated by the metallotranscription factor ScMac1 [27–29]. Like the Aft1 protein, yeast ScMac1 is a copper-dependent transcriptional activator. Copper has been shown to play an important role as a virulence factor in C. albicans and Cr. neoformans [30,31]. The ScMac1 homologs CaMac1 and Cuf1 of C. albicans and Cr. neoformans, respectively, have been identified. CaMac1 regulates the expression of CaCtr1, which is a high-affinity copper transporter, and the working mechanism of CaMac1 is very similar to that of yeast ScMac1. Cuf1 also functions as a copper-dependent transcription factor, regulating the gene expression of CTR4 at the transcriptional level and maintaining copper homeostasis. Quite recently, the function of copper in the virulence of A. fumigatus has been reported. AfMac1, which is a copper-dependent transcription factor, regulating the gene expression of ctrC at the transcriptional level and maintaining copper homeostasis, has been identified by several groups [32–34], and the role of copper in pathogenesis of A. fumigatus has been reported . These reports indicate that copper homeostasis is one of the virulence factor of A. fumigatus.
Both iron metabolism and copper metabolism are strictly regulated in microbes and mammals, and some features show a close interaction between copper and iron metabolism, even though they, at first, appear to be independent. Copper and iron donate and accept an electron freely, which indicates that these metals have similar chemical and physical properties and that excess concentrations of these elements can cause harmful effects . These properties of copper and iron allow them to act as cofactors in various enzymatic reactions; therefore, metal homeostasis is tightly regulated . Additionally, in S. cerevisiae, Fet3 multicopper oxidase, an iron permease, requires four copper ions as cofactors. When copper is limited, intracellular iron availability is also limited . The mammalian homolog of Fet3, ceruloplasmin, also requires copper for its activity . Furthermore, the divalent metal-ion transporter 1 (DMT1) functions in the uptake of both copper and iron , and this feature implies that iron and copper interact with each other in mammalian cells. Surface ferrireductase also plays an important role in copper and iron uptake. After copper and iron are reduced by Cu/Fe reductases, the reduced transition metals are imported to the cytosol . These transition metals, which are frequently involved in electron transfer reactions for energy production, are commonly used by electron transfer proteins because these metals may have two oxidation/reduction potentials depending on the environment.
Investigators have long recognized the importance of copper and iron metabolism and have devoted considerable effort to determining the connection between copper and iron homeostasis [40–42]. However, direct proof of an interaction between iron and copper has not yet been provided. Here, to further investigate the relationship between copper and iron homeostasis, we cloned and characterized Afmac1, a copper-/iron-sensing transcription factor of A. fumigatus, and found that AfMac1 co-ordinates the control of copper and iron homeostasis.
Materials and methods
Strains, media, and culture conditions
The A. fumigatus strain A1160 [akuB (KU80)-delta pyrG1] and congenic mutants Δafmac1 [akuB (KU80)-delta pyrG1, Δafmac1::pyrG], Δctrc [akuB (KU80)-delta pyrG1, Δctrc::pyrG], and Δsit1 [akuB (KU80)-delta pyrG1, Δsit1::pyrG] were used in the present study. The S. cerevisiae strain BY4741 (MATa, his3Δ1, leu2Δ0, met15Δ0, ura3Δ0) and congenic Δscmac1 (MATa, his3Δ1, leu2Δ0, met15Δ0, ura3Δ0, Δscmac1::KanMX4) were used in the present study. A. fumigatus strains were grown at 37°C in Aspergillus minimal medium (AMM) and complete medium (CM), according to the procedure described by Pontecorvo et al. , with 1% glucose as a carbon source. For copper-omitted or iron-omitted Hunter's trace element solution, the indicated concentrations of copper (copper sulfate) and iron (ammonium ferrous sulfate) added to AMM. The E. coli strain XL-10G was used to construct the plasmids employed in the present study. Fungal plate assays were performed on AMM plates for 3 days. To activate the gene driven by the xylose-inducible promoter , minimal medium was supplemented with 0.5% xylose as a carbon source.
Plasmid construction and site-directed mutagenesis
A. fumigatus AfMac1 was cloned from total RNA extracted from A. fumigatus FGSC A1160 using the AfMac1-F/AfMac1-R primer pair; the resulting PCR product was then cloned into the pGEM T-easy vector. To characterize the Afmac1 gene in the A1160 host strain, Afmac1 and its derivatives driven by a xylose-inducible promoter originating from Penicillium chrysogenum were constructed in the pPTRII fungal vector (Takara, cat # 3622). To investigate the localization of AfMac1 in A. fumigatus, the Apa1 enzyme site for tagging the GFP gene to the Afmac1 gene was introduced into the C-terminal coding region, generating the Afmac1-GFP plasmid on the pPTRII vector; to generate the ApaI enzyme site, the Afmac1-apa/Afmac1-kpn primer pair was used. To assess whether AfMac1 binds to the promoter region of the sit1 gene, the genomic portion of sit1, including the 5′-upstream and 3′-downstream regions, was cloned using the sit1PPF/R primer set, and mutation of the putative AfMac1-binding motif was conducted using the sit1PP-Mup/dw primer set. To generate the deletion fungal strain of Afmac1, plasmids were constructed using the pSL1190 vector containing the pyrG blaster cassette. The following primer pairs were used to generate the Afmac1 deletion cassette: for the 5′-flanking region, Afmac1-5URF/Afmac1-5URR; and for the 3′-flanking region, Afmac1-3UTRF/Afmac1-3URR. The Afmac1 deletion cassette was completed by inserting the 5′-flanking and 3′-flanking regions on the corresponding sides of the pyrG blaster cassette. To construct the plasmid extracting the AfMac1-GFP protein, Afmac1-GFP was inserted into the pGEX4T-1 vector. The primers used in this work are listed in Supplementary Table S1.
Northern and southern blot analyses
Total RNA was extracted from various fungal strains grown at 37°C for 24 h in AMM and separated on a 1% formaldehyde agarose gel. Probes were prepared by PCR amplification of the internal regions of the open reading frames of the genes of interest, as follows: for sreA, sreA-norF/R; for ctrC, ctrCnorF/R; for hapX, hapX-norF/R; for mirD, mirD-norF/R; for mirB, mirB-norF/R; for sit1, sit1-norF/R; for sidA, sidA-norF/R; for ftrA, ftrA-norF/R; for fetC, fetC-norF/R; for Afmac1, Afmac1-F/Afmac1-621; for FET3, the FET3-5/3 primer set; and for CTR1, the CTR1-5/3 primer set. For southern blot analysis, genomic DNA (gDNA) was extracted using the CTAB method from fungal transformants that had been grown overnight in CM. The extracted gDNA was separated by electrophoresis at 100 V after enzymatic digestion using BamH1 to probe Afmac1 using the 3′-flanking region.
To generate the Afmac1 deletion mutant strains, a PCR product amplified by Afmac1-5URF/Afmac1-3URR primers was transformed into an uracil auxotroph of A. fumigatus A1160 strain. Briefly, the A1160 strain was grown in CM containing uridine and uracil for 24 h. Protoplasts were prepared via incubation with a glucan lytic enzyme at 30°C for 2 h after washing with osmotic buffer (0.6 M KCl and 10 mM NaCl). The protoplasts were collected using Miracloth and washed with cold osmotic buffer. Next, the washed protoplasts were resuspended in STC buffer [1.2 M sorbitol, 10 mM Tris–HCl (pH 7.5), and 10 mM CaCl2]. The protoplasts were incubated with DNA fragments in 25% PEG 6000 solution in STC buffer for 20 min on ice, and a 25% PEG 6000 solution and KC buffer (0.6 M KCl and 50 mM CaCl2) were then added. The protoplasts containing the DNA fragment were poured into a Petri dish with 1% soft agar without uracil, and the plates were incubated at 37°C for 3 days. Transformants were screened by PCR analysis. Homologous genomic integration was confirmed by southern blot analysis.
To visualize the location of AfMac1 in the mycelium, fungal cells harboring a GFP-tagged Afmac1 gene in the genomic DNA were grown on a cover slide in 1 ml of AMM containing 0.5% (w/v) xylose and the indicated concentrations of BCS (Bathocuproinedisulfonic acid), BPS (Bathophenanthrolinedisulfonic acid), iron, and/or copper for 2 days. After the mycelia were removed from the surface of the medium, the cover slide containing the fungal cells was fixed with 1% formaldehyde for 30 min and incubated with 0.2 M glycine for an additional 5 min. The fixed cells were then washed twice with PBS, and the slides were overlaid with mounting solution containing 50 ng/ml DAPI. The cells were viewed and imaged using a fluorescence microscope (AXIO imager A1/M1, Carl Zeiss, Germany).
Electrophoretic mobility shift assay
Recombinant AfMac1-GFP protein was purified from a 1-l culture of E. coli strain JM109 carrying plasmid pGEX4T_AfMac1-GFP. Purified AfMac1-GFP protein was used for the electrophoretic mobility shift assay (EMSA) experiment. The EMSA probes for the Afmac1 and hapX genes were amplified from the genomic DNA of A. fumigatus using the primer pairs AfMac1EMSA-F/R and HapXEMSA_F/R, respectively. The EMSA oligonucleotides that were used to identify the binding motif of AfMac1 were annealed using the following primers: for the mirB oligo, mirBup/dw; for the mirB mutant oligo M2, mirB-M2up/dw; for the mirB mutant oligo M3, mirB-M3up/dw; and for the mirB mutant oligo M8, mirB-M8up/dw. The EMSA experiment was performed as follows. Binding reactions were conducted in a 10-µl volume containing 2 µg of AfMac1-GFP and 20 000 cpm of the labeled probe for 30 min at RT; the reactions contained binding buffer [10 mM Tri–HCl (pH 8.0), 50 mM KCl, 5 mM MgCl2, 1 mM DTT, 1% (w/v) glycerol, and 0.5 µg poly(dIdC)]. Next, 10 µM iron or copper was added to the binding reactions to determine whether iron or copper was required for the transcriptional activity of AfMac1. The DNA-binding specificity was assessed by the addition of a 10-, 50-, and 100-fold molar excess of unlabeled homologous competitor DNA to the binding reaction. Following incubation, the DNA–protein complexes were electrophoretically resolved from unbound DNA probes on an 8% (w/v) non-denaturing polyacrylamide gel at 100 V using 0.5× Tris borate/EDTA buffer. The gels were dried, and the DNA–protein complexes were visualized by autoradiography.
Iron uptake assay
To investigate the iron uptake ability of Afmac1 deletion mutant strains, the fungal cells were cultured at 37°C for 2 days in AMM. The mycelia of wild-type (WT) and the Afmac1 deletion strain were harvested by filtration through Miracloth and washed with sterile water. The washed mycelia were further washed with citrate buffer, and the fungal cells were transferred to a spin column after complete removal of the buffer from the mycelium. Citrate buffer containing 1 µM 55Fe was added to the column containing the fungal cells, and the cells were incubated at 37°C for 1 h. The activated cells were washed three times with distilled water, and the columns were dried at 60°C for 30 min. The dried mycelia were weighed on an electronic scale and immersed in scintillation cocktail solution. Radioactivity was determined using a liquid scintillation counter. Iron uptake assay was performed as triplicates.
Microarray platform, labeling, and hybridization
Total RNA from WT and Δafmac1 with and without BCS treatment was extracted using TRIzol. The control and test RNAs were isolated from the WT and Δafmac1 strains, respectively. For the control and test RNAs, the synthesis of target cRNA probes and hybridization was performed using Agilent's two-color Low Input Quick Amp WT Labeling Kit (Agilent Technology, U.S.A.) according to the manufacturer's instructions. Briefly, 0.2 mg of total RNA was mixed with WT primer mix and incubated at 65°C for 10 min. cDNA master mix (5× first-strand buffer, 0.1 M DTT, 10 mM dNTP mix, RNase-Out, and MMLV-RT) was prepared and added to the reaction mixture. The samples were incubated at 40°C for 2 h, after which RT and dsDNA synthesis was terminated by incubation at 70°C for 15 min. The transcription master mix was prepared according to the manufacturer's protocol (4× transcription buffer, 0.1 M DTT, NTP mix, 50% PEG, RNase-Out, inorganic pyrophosphatase, T7-RNA polymerase, and cyanine 3/5-CTP). Transcription of dsDNA was performed by adding the transcription master mix to the dsDNA reaction samples and incubating the reaction at 40°C for 2 h. Amplified and labeled cRNA was purified on an RNase mini-column (Qiagen) according to the manufacturer's protocol. The labeled cRNA target was quantified using an ND-1000 spectrophotometer (NanoDrop Technologies, Inc., Wilmington, DE, U.S.A.). After evaluating the labeling efficiency, cyanine-3-labeled and cyanine-5-labeled cRNA targets were combined and fragmented by the addition of 10× blocking agent and 25× fragmentation buffer and incubated at 60°C for 30 min. The fragmented cRNA was resuspended in 2× hybridization buffer and directly pipetted onto an assembled A. fumigatus Af293 30 K microarray (MYcroarray.com, U.S.A.). The arrays were hybridized at 57°C for 17 h using an Agilent hybridization oven (Agilent Technology, U.S.A.). The hybridized microarrays were washed according to the manufacturer's washing protocol (Agilent Technology, U.S.A.), and the results were visualized using an Axon GenePix 4000B Scanner (Axon Instruments, CA). The data were quantified using GenePix Pro 6.0 (Axon Instruments). The average fluorescence intensity of each spot was calculated, and the local background was subtracted. LOESS normalization was performed using GenoWiz 4.0 (Ocimum Biosolutions, India).
ChIP sequencing and multiple EM for motif elicitation analysis
For the ChIP assay, A. fumigatus cells harboring the AfMac1-GFP plasmid were grown in AMM containing 200 ng/ml pyrithiamine and then cross-linked with 1% formaldehyde and lysed in SDS lysis buffer. The lysed samples were sonicated at high power for 30 min (cycles of 30-s pulses with a 30-s pause) using a Bioruptor. After the sample was precleared with salmon sperm DNA/protein A-agarose, GFP antibody was added to the protein extract to immunoprecipitate AfMac1/DNA complexes. The resulting immune complexes were extensively washed, and the DNA was recovered using a PCR cleanup kit. ChIP sequencing was conducted using the eluted DNA. The library construction was performed using the NEBNext® UltraTM DNA Library Prep Kit for Illumina (New England Biolabs, U.K.) according to the manufacturer's instructions. Briefly, the ChIPped DNA was ligated with adaptors. After purification, PCR was performed with adaptor-ligated DNA and index primers for the multiplexing sequencing. The library was purified using magnetic beads to remove all reaction components. The size of the library was assessed using an Agilent 2100 Bioanalyzer (Agilent Technologies, Amstelveen, The Netherlands). High-throughput sequencing was performed as paired-end 100 sequencing using HiSeq 2000 (Illumina, Inc., U.S.A.). The reads were trimmed and aligned using Bowtie2 version 2.1.0. Bowtie2 indices were generated either from the genome assembly sequence or from representative transcript sequences to align the genome and transcriptome. We used Model-based Analysis of ChIP-Seq (MACS) to identify peaks in the alignment file. Briefly, total reads of input and GFP by HiSeq2000 were 43 Mreads and 40 Mreads, respectively, and 428 input peaks and 2586 GFP peaks were identified. The intensity of input was minimum — 1.27 and maximum — 3.13, and the intensity of GFP was minimum — 3.17 and maximum — 58.96. In total, 409 genes and 2076 genes were identified from input and GFP, respectively, from ChIP-seq analysis. To identify the putative AfMac1-binding motif based on the ChIP-seq results, the upstream regions of ftrA, sreA, mirB, mirC, mirD, sidA, sidC, and sidD were subjected to multiple EM for motif elicitation (MEME) analysis.
Inductively coupled plasma/optical emission spectrometry
All plastic containers and conical tubes were washed in 6 N HCl and thoroughly rinsed with MilliQ water to remove trace amounts of copper prior to use. The WT and Δafmac1 fungal strains were used to inoculate liquid CM cultures, which were grown at 180 rpm and 37°C for 24 h prior to harvesting with Miracloth. The harvested cells were washed for 20 min with MilliQ water. The mycelia were dried in a drying oven at 70°C for 24 h and weighed. The dried mycelia were digested overnight at 70°C in a typical mixture of 30% H2O2 and 60% HNO3 (1 : 3, v/v) (12 ml of mixture per gram of mycelium wet weight). The residue was diluted with MilliQ water (1 : 1, v/v) and filtered through a 0.2-µm filter. The final solution was prepared by adding 200 ml of sample to 3 ml of MilliQ water, followed by ICP-OES (Inductively coupled plasma/optical emission spectrometry; Agilent 730 Series, U.S.A.) measurement. All calibration solutions were prepared from a 1000 µg/ml ICP-grade Cu standard solution in 4% HNO3 (PlasmaCAL, ICP-AES/MS standard, SCP Science, Baie D'Urfé, QC, Canada).
Afmac1 regulates iron metabolism in
We previously reported that AfMac1 functions as a copper-dependent transcriptional activator. To further characterize the function of AfMac1, we investigated the global gene expression profiles of WT and Afmac1 deletion mutants of A. fumigatus by cDNA microarray analysis. The expression of genes encoding putative copper transporters was down-regulated following Afmac1 deletion (Figure 1A). Surprisingly, we found that the genes involved in iron metabolism were also down-regulated following Afmac1 deletion. The putative siderophore transporters mirB, mirD, mirC, and sit1 [36,37] and the genes encoding the reductive iron transporters ftrA and fetC  were also down-regulated. These results indicate that AfMac1 also regulates expression of the iron regulon and that it is involved in iron metabolism. These results also support a close association between copper and iron metabolic pathways of the cell.
AfMac1 regulates the expression of genes within the iron regulon.
The putative iron-binding motifs XCSCX and XCLCX were identified in the N-terminal region of AfMac1 , supporting the likely involvement of AfMac1 in iron metabolism (Figure 1B). The iron content of the cells decreased dramatically following the deletion of AfMac1 (Supplementary Figure S1). Free iron levels, ferrichrome (FC), and ferrioxamine B (FOB) uptake activity also decreased dramatically following Afmac1 deletion (Figure 1C). However, free iron uptake activity remained unchanged following the deletion of ScMAC1, and FC and FOB uptake activity increased even when ScMAC1 was deleted in S. cerevisiae. These results indicate that the role of AfMac1 in the regulation of copper and iron metabolism differs from that of ScMac1.
We also examined whether the product of the cloned Afmac1 gene functioned properly. The introduction of Afmac1 restored the iron uptake activity of the Afmac1 deletion mutant, indicating that the cloned Afmac1 gene product functioned properly and that Afmac1 is involved in iron metabolism (Supplementary Figure S2). To investigate the effect of AfMac1 on cell growth, we performed a plate assay with the Afmac1 deletion mutant (Figure 2A). We added iron and copper to cultures of WT and Afmac1 deletion mutants to investigate the effects of copper and iron on growth of the Afmac1 deletion mutant. Under low iron conditions, cell growth was very slow, and exogenous iron partially supported cell growth. Under low copper and low iron conditions, cell growth was restored by exogenous copper, but the cells formed colonies that were pale in color. However, under copper- and iron-replete conditions, the cells displayed the same color and growth rate as WT cells. These results indicate that the Afmac1 deletion mutant requires both copper and iron for normal growth. To confirm the microarray data, northern blot analysis was performed. The genes encoding the putative siderophore transporters sit1 and mirB, the putative reductive iron transporter fetC, the siderophore biosynthesis gene sidA, and the iron-responsive regulator hapX were down-regulated following Afmac1 deletion (Figure 2B). hapX has been reported to be a major transcriptional activator under iron-depleted conditions, and we found that the expression of hapX was also regulated by AfMac1. To confirm the relationship between Afmac1 and hapX, hapX was cloned and introduced into the Afmac1 deletion strain. The expression of fetC, sit1, mirB, and sidA was down-regulated by Afmac1 deletion (Figure 2C). However, when hapX was introduced into the Afmac1 deletion mutant, expression of the iron regulon was up-regulated. This result indicates that AfMac1 binds upstream of hapX in the iron metabolism pathway. Furthermore, we performed an EMSA against hapX (Figure 3A). The binding activity of AfMac1 to the 5′-promoter region of hapX decreased in the presence of the unlabeled 5′-promoter region of hapX and iron (Figure 3B).
Exogenous copper and iron supplementation suppresses the growth defect of the
Afmac1 deletion mutant.
AfMac1 regulates the gene expression of
hapX by binding to the promoter region of hapX.
Afmac1 binds to the AfMac1-binding motif in the promoter region of the iron regulon
Despite the apparent co-ordinate regulation of the copper and iron regulons, no AfMac1-binding motifs (5′-TGTGCTCA-3′) were found in the promoter region of the iron regulon ; therefore, we searched for AfMac1-binding motifs using ChIP-seq analysis. Using MEME of the peaks from the majority of the iron regulon, the AfMac1-bound DNA motif was identified as a 10-bp region, and AfMac1 binding to the AfMac1-binding motif was confirmed by EMSA using mutants with changes in the conserved sequences (Figure 4A). The AfMac1-binding motif was identified as 5′-(A/G)AT(C/G)(A/G)(G/A)(A/T)T(A/C)(A/T)-3′ in the 5′-promoter region of the iron regulon. Furthermore, the mutation of the third thymine to cytosine resulted in a failure to compete with AfMac1 binding to the 5′-promoter region of sit1 in the EMSA assay (Figure 4B). However, the same motif was not found in the genes involved in copper metabolism in A. fumigatus.
AfMac1 binds to the AfMac1 conserved binding motif 5′
To confirm the binding of AfMac1 to the AfMac1-binding motif of sit1, the core region of the AfMac1-binding motif of the 5′-promoter region of sit1 was mutated from 5′-ATC-3′ to 5′-TAG-3′ and introduced into the sit1 deletion mutant of A. fumigatus. The expression of sit1 was not up-regulated in response to the mutation, even when iron was depleted (Figure 4C). These results suggest that AfMac1 binds specifically to the 5′-(A/G)AT(C/G)(A/G)(G/A)(A/T)T(A/C)(A/T)-3′ motif in the promoter region of the iron regulon.
Afmac1 expression itself is not regulated by iron or copper, and its effects on the expression of the iron and copper regulons are independently regulated by iron and copper, respectively
To further identify the mechanism of AfMac1 function, we examined the expression of Afmac1 under iron- and copper-limited conditions. The expression of Afmac1 was extremely low and difficult to detect (Figure 5A). Furthermore, we investigated the effect of iron and copper on the expression levels of ctrC and sit1, respectively. To investigate whether cross-regulation by iron and copper occurred, we exposed the cells to copper- and iron-depleted conditions (Figure 5B,C). The expressions of sit1 and ctrC were up-regulated by iron and copper depletion, respectively. And, the mutation of the AfMac1- or Mac1-binding motifs failed to up-regulate sit1 or ctrC by iron- and copper-depletion, respectively (Figure 5B,C). However, depletion of iron and copper failed to up-regulate ctrC and sit1, respectively. Furthermore, the exogenous addition of high concentrations of iron and copper failed to inhibit the expression of ctrC and sit1, respectively (Figure 5D), and depletion of copper failed to activate iron uptake activity (Supplementary Figure S3). These results suggest that the expression of the iron and copper regulons by AfMac1 is regulated independently by iron and copper. We next investigated how AfMac1 maintains iron and copper homeostasis. Although the depletion of both iron and copper failed to up-regulate AfMac1 (Figure 5A), AfMac1 regulated the expression of the iron or copper regulon by monitoring the availability of iron or copper, respectively, and by regulating the gene expression of specific target genes.
Afmac1 expression is not regulated by iron or copper, and its effects on the expression of the iron and copper regulons are independently regulated by iron and copper.
The cellular localization of AfMac1 was then investigated. Following the depletion of copper, most of the AfMac1 was found in the nucleus (Figure 6). However, AfMac1 was detected in whole cells treated with 10 µM copper. We also treated cells with iron to test the effect of iron on the localization of AfMac1. Interestingly, most of the AfMac1 was found in the nucleus following iron depletion. However, a small amount of AfMac1 was also found in the nucleus when cells were treated with 10 µM iron, unlike the effect of copper. These results indicate that the intracellular localization of AfMac1 is regulated by copper and that iron utilization and AfMac1 localization may depend on the presence of both iron and copper. Furthermore, the results suggest a dual role for AfMac1 in which copper metabolism and iron metabolism are regulated independently. Interestingly, we found that the addition of copper failed to affect the cellular localization of AfMac1 in the absence of iron and vice versa (Figures 5A and 6).
AfMac1 localized to the nucleus under iron- or copper-depleted conditions.
To date, several lines of evidence have indicated a close relationship between iron and copper metabolism. First, ceruloplasmin and Fet3 from humans and S. cerevisiae, respectively, have been well studied ; copper is required for their ferroxidase activity, which, in turn, is required for iron transport. In the liver, copper deficiency causes iron accumulation and vice versa . In the intestine, copper-deficient animals showed impaired activity of iron efflux from enterocytes, and this was shown to be related to hephaestin activity, another ferroxidase . Second, the divalent metal transporter DMT1 takes up both iron and copper, and there is competition between iron and copper for uptake . High concentrations of copper can decrease DMT1 protein and mRNA levels . Third, the ferrireductase CYBRD1 reduces ferric iron to ferrous iron, and DMT1 takes up ferrous iron. The expression of CYBRD1 is up-regulated by iron depletion and is involved in iron homeostasis . Interestingly, CYBRD1 can reduce copper, implying the existence of an interaction between iron and copper. In S. cerevisiae, the ferrireductase Fre1 is also a membrane reductase and is involved in reductive iron uptake . Fre1 is regulated by both iron and copper, similar to the example of CYBRD1. Although these reports suggest the possibility of a link between iron and copper metabolism, no direct regulatory mechanism that simultaneously regulates iron and copper had been identified.
Here, we identified the mechanism by which iron and copper separately regulate AfMac1 activity during cellular copper and iron metabolism. The function of AfMac1 in copper uptake and conidia pigment of A. fumigatus was recently reported [32–34]. Our group also described the detailed working mechanism of AfMac1 action at the copper regulon at the transcriptional level . In the present report, we identified a novel function of AfMac1 in copper and iron metabolism. As shown in the Results section, copper and iron inhibited AfMac1 binding to the 5′-promoter regions of ctrC and hapX, respectively, as determined by EMSA. However, iron and copper deficiencies failed to activate gene expression of the copper and iron regulons, respectively. The depletion of iron resulted in sit1, but not in ctrC, up-regulation. Additionally, ctrC, but not sit1, up-regulation was detected when copper was depleted. Furthermore, high concentrations of copper or iron failed to inhibit the expression of the iron or copper regulons, respectively, when the other metal was limited, and a high concentration of copper or iron failed to inhibit the expression of the iron or copper regulons, respectively, when the other metal was depleted. These regulatory mechanisms were mediated by AfMac1, showing that AfMac1 functions as a coordinator of iron and copper metabolism.
Based on these results, we constructed a hypothetical working model of the effect of AfMac1 on expression of the iron and copper regulons. The binding of iron and copper affects the transcriptional activity of AfMac1 and results in differential binding affinities of AfMac1 for iron and copper; these affinities require the iron and copper-dependent binding motifs of AfMac1 and independently regulate gene expression of the iron and copper regulons (Figure 7). When iron and copper are abundant, both ions bind to AfMac1 and reduce its activity at the iron and copper regulons. These results suggest that AfMac1 co-ordinates iron and copper metabolism. To date, AfMac1 homologs have not been identified in higher organisms. However, the function of AfMac1 indicates that living organisms have required iron and copper throughout evolution, that mechanisms for the control of their intracellular levels have evolved together and that their regulatory pathways have differentiated to permit subtle regulation for eukaryotes. Our results provide clues into the interaction between iron metabolism and copper metabolism in humans and can explain the occurrence of chlorosis in women. Women who worked in copper factories were not found to suffer from chlorosis .
Postulated working model of AfMac1 dependence on copper and iron in
We next addressed the possible reasons for the evolution of AfMac1 as a bi-functional transcription factor of copper and iron metabolism in A. fumigatus, a situation that is distinct from that found in other organisms. Using a genome database of fungal species, we found that the iron-regulatory and copper-regulatory domains of AfMac1 are conserved in a single transcription factor in many genera of filamentous fungi, including Aspergillus, Penicillium, and Fusarium, but not in yeast genera, such as Saccharomyces, Candida, and Cryptococcus. Filamentous fungi reside in oligotrophic environments; the dual functions of AfMac1 could help maintain iron and copper homeostasis under conditions of low metal concentrations in a nutrient-deficient environment and improve the efficiency of iron and copper metabolism compared with regulation by two different proteins. For example, during fungal infection, the host sequesters iron in the infected region as a means of competing with fungal pathogens , and siderophore secretion by fungi is a mechanism by which the fungal pathogen combats iron sequestration by its host . AfMac1 regulates the expression of genes that encode proteins involved in siderophore biosynthesis and secretion. Copper has also been recognized as a virulence factor in fungal infection. AfMac1 regulates both iron and copper homeostasis in host cells; under these conditions, fungal pathogens can survive more efficiently than when iron and copper are regulated separately by different proteins. Under these conditions, iron sequestration by the host does not restrict copper utilization by A. fumigatus, and the pathogen survives longer in the host despite the fact that iron is restricted. Our data suggest that iron and copper metabolism have a close relationship and that they are closely associated within cells. Our data also provide new insights into the interaction of iron and copper in fungi and suggest the possibility of the existence of similar regulatory mechanisms in higher organisms.
Aspergillus minimal medium
divalent metal-ion transporter 1
electrophoretic mobility shift assay
multiple EM for motif elicitation
- STC buffer
sorbitol, Tris–HCl, and CaCl2
All authors participated in the experiments and data collection. C.-W.Y. designed all experiments and wrote the manuscript. Y.-S.P. performed most of experiments. H.S. and S.K. revised the manuscript. All authors approved the final version of manuscript.
This work was carried out with the support of ‘Cooperative Research Program for Agriculture Science and Technology Development (Project No: PJ01368101)’, Rural Development Administration, Republic of Korea.
The Authors declare that there are no competing interests associated with the manuscript.