The Putative C2H2 Transcription Factor MtfA Is a Novel Regulator of Secondary Metabolism and Morphogenesis in Aspergillus nidulans

Secondary metabolism in the model fungus Aspergillus nidulans is controlled by the conserved global regulator VeA, which also governs morphological differentiation. Among the secondary metabolites regulated by VeA is the mycotoxin sterigmatocystin (ST). The presence of VeA is necessary for the biosynthesis of this carcinogenic compound. We identified a revertant mutant able to synthesize ST intermediates in the absence of VeA. The point mutation occurred at the coding region of a gene encoding a novel putative C2H2 zinc finger domain transcription factor that we denominated mtfA. The A. nidulans mtfA gene product localizes at nuclei independently of the illumination regime. Deletion of the mtfA gene restores mycotoxin biosynthesis in the absence of veA, but drastically reduced mycotoxin production when mtfA gene expression was altered, by deletion or overexpression, in A. nidulans strains with a veA wild-type allele. Our study revealed that mtfA regulates ST production by affecting the expression of the specific ST gene cluster activator aflR. Importantly, mtfA is also a regulator of other secondary metabolism gene clusters, such as genes responsible for the synthesis of terrequinone and penicillin. As in the case of ST, deletion or overexpression of mtfA was also detrimental for the expression of terrequinone genes. Deletion of mtfA also decreased the expression of the genes in the penicillin gene cluster, reducing penicillin production. However, in this case, over-expression of mtfA enhanced the transcription of penicillin genes, increasing penicillin production more than 5 fold with respect to the control. Importantly, in addition to its effect on secondary metabolism, mtfA also affects asexual and sexual development in A. nidulans. Deletion of mtfA results in a reduction of conidiation and sexual stage. We found mtfA putative orthologs conserved in other fungal species.


Introduction
Fungal species produce numerous secondary metabolites [1,2,3], including compounds with detrimental effects, such as mycotoxins [4], capable of causing disease and death in humans and other animals [4,5]. Aspergillus nidulans, a model filamentous fungus studied for more than fifty years, produces the mycotoxin sterigmatocystin (ST). This mycotoxin, ST, and the well-known carcinogenic compounds called aflatoxins (AF), produced by related species such as A. flavus, A. parasiticus, and A. nomius [6], are both synthesized through a conserved metabolic pathway [7,8,9] where ST is the penultimate precursor. The genes responsible for ST/AF production are clustered. Within these clusters, the regulatory gene aflR encodes a transcription factor that acts as a specific cluster activator [10,11,12].
The range of secondary metabolites produced by A. nidulans also includes bioactive compounds with demonstrated beneficial effects and applications for medical treatments, including antibiotics, such as the beta-lactam penicillin (PN) [13,14], or anti-tumoral metabolites such as terrequinone [15,16], with potential direct application in the medical field. In both cases the genes involved in the synthesis of these compounds are also found clustered [16,17].
In fungi, secondary metabolism is often found to be governed by genetic mechanisms that also control asexual and sexual development [18]. One of these principal common regulatory links is the global regulatory gene veA, first described to be a developmental regulator in A. nidulans [19,20]. In 2003 we described for the first time the connection between veA and the synthesis of diverse fungal secondary metabolites, including ST [21]. Absence of the veA gene in A. nidulans prevents aflR expression and concomitant ST biosynthesis. A similar effect was also observed in Aspergillus flavus and Aspergillus parasiticus veA deletion mutants, that lost the capacity to produce AFs [22,23,24]. Furthermore, veA also regulates the biosynthesis of other mycotoxins, for example cyclopiazonic acid and aflatrem in Aspergillus flavus [22]. veA is extensively conserved in Ascomycetes [25] and its global regulatory effect on mycotoxin biosynthesis was also observed in other fungal genera, for example, on the synthesis of trichothecenes in F. graminerum [26], and on fumonisins and fusarins in Fusarium spp, including F. verticillioides and F. fujikuroi [25,27]; all these mycotoxins can cause severe impacts on the health of humans and other vertebrates [4]. Interestingly, VeA also regulates the production of other secondary metabolites with beneficial properties, for instance PN in A. nidulans and P. chrysogenum [21,28] as well as cephalosporin C in Acremonium chrysogenum [29].
VeA has also been found to affect fungal infection of plants and animals. For example, a decrease in virulence was observed in deletion veA mutants of A. flavus when infecting plant tissue [24]. This effect was also observed in mycotoxigenic Fusarium species, such as F. verticillioides [25], F. graminearum [26] and F. fujikuroi [27]. In the case of animal infections, deletion of the veA homolog in Histoplasma capsulatum also leads to a reduction in virulence in a murine model [30], although in Aspergillus fumigatus veA is dispensable for virulence in the neutropenic mouse infection model [31].
Most of the studies to elucidate the veA regulatory mechanism of action have been carried out using the model fungus A. nidulans. It is known that the KapA a-importin transports the VeA protein to the nucleus, and that this transport is promoted by darkness [32,33]. In the nucleus, VeA interacts with light-responsive proteins that also modulate mycotoxin production and fungal development, such as the red phytochrome-like protein FphA, which interacts with the blue sensing proteins LreA-LreB [34,35]. VeA also sustains other nuclear protein interactions with VelB and LaeA [36,37]. LaeA, a chromatin modifying protein, is also required for the synthesis of ST and other secondary metabolites [38,39]. Absence of VelB, another protein of the velvet family [37], decreases and delays ST biosynthesis, indicating a positive role in ST biosynthesis [36].
To identify novel veA-dependent genetic elements involved in the regulation of ST biosynthesis in the model system A. nidulans, we performed a mutagenesis in a deletion veA strain to generate revertant mutants that regained the capacity to produce toxin [40]. Several revertant mutants (RM) were obtained. In the present study we characterized one of these selected revertants, RM7. This revertant mutant presented a point mutation in a gene that we denominated mtfA (master transcription factor A) encoding a novel putative C 2 H 2 zinc finger domain type transcription factor. We show that the mtfA effect on ST production is veAdependent. Additionally, mtfA regulates the expression of other secondary metabolite gene clusters, such as those of terrequinone and PN. Furthermore, mtfA is also important for normal sexual and asexual development in A. nidulans.

Fungal Strains and Growth Conditions
Fungal strains used in this study are listed in Table 1. Media used include glucose minimal media (GMM) [41], YGT (0.5% yeast extract, 2% dextrose, trace elements prepared as described [41], and oat meal media (OMM) (1% oat meal). Supplements for auxotrophic markers were added as required [41]. Glucose was substituted with threonine (100 mM) in threonine minimal medium (TMM) for induction of alcA promoter. Solid medium was prepared by adding 10 g/liter agar. Strains were stored as 30% glycerol stocks at 280uC.

Identification of the Revertant Mutation in RM7
To find the mutation in RM7, A. nidulans genomic library pRG3-AMA1-NOT1 was utilized to transform the RM7-R2 (DstcE, X 2 ) strain. Plasmid DNA was rescued from fungal transformants presenting wild-type phenotype. Both end regions of the DNA inserts in the isolated plasmids were sequenced and the complete insert sequences were found in the A. nidulans genome database (http://www.aspgd.org) by BLAST analysis. The exact location of the mutation in RM7 was identified by sequencing of the PCR product amplified from the same locus in RM7.

Sequence Search and Alignment
The deduced protein sequence of MtfA (AN8741.2) was compared against databases from different fungal genera, using the BLAST (blastp) tool provided by National Center for Biotechnology Information (NCBI), (http://www.ncbi.nlm.nih. gov/). The gene entry with the highest percentage of identity and the lowest e-value for each of the species was selected (Table  S1). Pairwise sequence alignment of the proteins was performed using the EMBOSS Needle tool (http://www.ebi.ac.uk/Tools/ psa/emboss_needle/) from EMBL-EBI (European Molecular Biology Laboratory's European Bioinformatics Institute). Percentage of similarities and percentage of identities were tabulated for each of the alignments (Table S1). Multiple sequence alignment was performed with MtfA (A. nidulans) and orthologs found across various fungal species using MAFFT version 6.0 (http://mafft. cbrc.jp/alignment/server/index.html), followed by shading using the BoxShade tool version 3.21. for presentation (http://www.ch. embnet.org/software/BOX_form.html).

Phylogenetic Analysis
Phylogenetic analysis was performed for the following 20 species: A. oryzae, A. flavus, A. kawachii, A. niger, A. terreus, N. fischeri, A. fumigatus, A. clavatus, A. nidulans, P. chrysogenum, P. marneffei, A. capsulatus, U. reesii, C. immitis, F. oxysporum, M. oryzae, N. tetrasperma, N. crassa, C. globosum and B. fuckeliana. Orthologs of MtfA (A.nidulans) were identified in the above described genomes by searching against each other using the BLAST (blastp) tool from NCBI. Multiple sequence alignment was performed using MUSCLE v3.8.31 [43]. The alignment was used to build a Hidden Markov model (HMM), followed by realignment of sequences against the generated HMM, using the hmmbuild and hmmalign tools in HMMER v3.0b2 (http://hmmer.org/). A maximum likelihood phylogeny reconstruction method implemented in the software PhyML v3.0 [44,45], whose workflow is available at iPlant collaborative TM (http://www.iplantcollaborative.org/) was used for tree construction with default settings. The resulting tree was viewed using FigTree v1.4.0 (http://tree.bio.ed.ac.uk/software/figtree/). Midpoint rooting [46] of the tree was chosen in order to minimize the large distances from the root to any leaf. The numbers on the branches indicate the approximate likelihood branch support values in percentages [47].

Generation of the mtfA Deletion Strain
The entire mtfA coding region was replaced in RDAE206 and RJMP1.49 strains ( Table 1). The DNA cassette used to delete mftA by gene replacement with the pyrG marker was obtained from FGSC (http://www.fgsc.net). Polyethylene glycol-mediated transformation of RDAE206 and RJMP1.49 protoplasts was carried out as described previously [48]. Transformants were selected on appropriate selection medium without uridine or uracil and confirmed by Southern blot analysis as previously described [49]. The deletion strains were designated as TDAEDmtfA and TRVDmtfA respectively.
A complementation strain was also obtained by transforming DmtfA (TRVDmtfA) strain with the mtfA wild-type allele. The complementation vector was generated as follows: A DNA fragment contained the entire mtfA coding region and 59 and 39 UTRs was first amplified with primers RM7com1 and RM7com2 (Table 2) from FGSC4 A. nidulans genomic DNA. Then the PCR product was digested with SacII and KpnI and cloned into pSM3 vector, containing the pyroA transformation marker, previously digested with the same enzymes, resulting in the plasmid pSM3-rm7com. This vector was transformed into DmtfA protoplasts and the transformants were selected on appropriate selection medium without pyridoxine. Complementation was confirmed by PCR and Southern blot analysis. The complemented strain was designated as DmtfA-com. Strains that were isogenic with respect to the auxotrophic markers were generated and used in this study.

Generation of mtfA Over-expression Strain
To generate the mtfA over-expression strain, the entire mtfA coding region was first amplified using the RM7-OE1 and RM7-OE2 primers ( Table 2). The PCR product was then digested with KpnI and PacI and ligated into pmacro plasmid, containing the A. nidulans alcA promoter, trpC terminator and pyrG marker, resulting in the plasmid pMacroMtfAOE. The pMacroMtfAOE vector was transformed into RJMP1.49 and transformants were selected on appropriate selection medium without uridine and uracil, and confirmed by PCR using RM7-OE1 and RM7-OE2 primers   (Table 2) and Southern blot analysis (data not shown). The resulting selected transformant was denominated TRV60.

Toxin Analysis
Culture plates containing 25 mL of solid GMM or OMM with appropriate supplements were top-agar inoculated with approximately 5610 6 spores/mL. The cultures were incubated in the dark. Three cores (16 mm diameter) from each replicate plate were collected and extracted with chloroform. Alternatively, strains were grown in GMM liquid shaken cultures (10 6 spores/ mL) and incubated at 37uC. Twenty-four h and 48 h old culture supernatants were analyzed for ST and mycelia were collected for RNA analysis. For analysis of mycotoxin production in overexpression mtfA and control cultures, GMM was inoculated with conidia (10 6 conidia/mL) from the mtfA over-expression strain (TRV60) or its isogenic control (TRV50.1), and incubated for 16 h at 37uC and 250 rpm. At that time, equal amounts of mycelia were then spread onto inducing medium TMM. Culture supernatants were collected for toxin analysis and mycelia were collected for RNA analysis 24 and 48 hrs after shift. Culture supernatants were also extracted with chloroform.
The chloroform extracts were dried overnight and then resuspended in 200 ml of chloroform. Samples were fractionated by silica gel thin-layer chromatography (TLC) using benzene and glacial acetic acid [95:5 (v/v)] as solvent system for ST analysis and chloroform:acetone:n-hexane (85:15:20) for NOR analysis. Aluminum chloride (15% in ethanol) was sprayed on the plates, that were subsequently baked for 10 min at 80uC. Both ST and NOR bands present on TLC plates were visualized under UV light (375-nm).

Penicillin Analysis
The PN bioassay analysis was carried out as previously described [50] with some modifications, using Bacillus calidolactis C953 as testing organism. Briefly, strains were inoculated with approximately 10 6 spores mL 21 in 20 mL of seed culture medium, and incubated at 26uC for 24 h at 250 rpm. Mycelia were collected with Miracloth (Calbiochem, USA) and transferred to PN-inducing medium containing lactose, 40 g/L; corn steep liquid (50%), 40 g/L; KH2PO4, 7 g/L; and phenoxyacetic acid, 0.5 g/ L; pH was adjusted to 6.0 with 10 M KOH. Cyclopentanone (10 mM) was added to induce expression of the alcA promoter when the mtfA over-expression strain and its isogenic control were used. Twenty mL of PN-inducing medium was inoculated with 1 mL of mycelia suspension (containing equal amounts of mycelium), and mycelial samples were collected at 24 h and 48 h of incubation for RNA analysis. After 96 h, the culture supernatants were collected for PN analysis. The experiment was carried out with three replicates. Three hundred mL of Tryptone-Soy Agar was inoculated with 20 mL of B. calidolactis C953 culture and plated on three 150-mm-diameter Petri dishes. Supernatant aliquots of each culture supernatant were then added to 7-mmdiameter wells. Bacteria were cultured at 55uC for 16 h and inhibition halos were visualized and measured. To confirm that the observed antibacterial activity was due to the presence of PN and not to the presence of other fungal compounds in the supernatant, controls containing commercial penicillinase from Bacillus cereus (Sigma, MO, USA) were also used. A standard curve using various concentrations of PN G (Sigma, MO, USA) was utilized to determine PN concentration in each sample.

Study of MtfA Subcellular Localization
Aspergillus nidulans RJMP1.49 strain (Table 1) was transformed with mtfA::gfp::pyrG A.fum as described previously [48]. Primers used in the generation of the fusion PCR product utilized for transformation are listed in Table 2. Plasmid p1439 [32] was used as template for the PCR amplification of the intermediate fragment. Correct integration was confirmed by PCR and Southern blot analysis (data not shown). Conidia from the selected transformant (i.e. TSD12.1, Table 1) were inoculated as described previously [32]. Briefly, conidia were allowed to germinate on coverslip submerged in Watch minimal medium [51] in light or dark. After 16 h samples were washed in 16PBS and stained with DAPI (10 ng/mL) in 50% glycerol and 0.1% Triton X-100. Samples were observed with a Nikon Eclipse E-600 equipped with Nomarski optics and fluorochromes for GFP and UV using a 1006objective. Micrographs were taken using Hamamatsu ORCA-ER high sensitivity monochrome CCD camera using Microsuite 5 imaging software. The exposure time for DIC, DAPI and GFP was 50 ms, 200 ms and 1 s respectively.

Morphological Studies
Plates containing 25 mL of solid GMM with the appropriate supplements were top-agar inoculated with 5 mL of top agar containing 10 6 spores/mL of A. nidulans strains TRV50.2 control, DmtfA or DmtfA-com ( Table 1). The cultures were incubated in dark or in light at 37uC. Cores were removed from each culture and homogenized in water. Conidia and Hülle cells were counted using a hemacytometer. Identical cores were taken to visualize cleistothecia under a dissecting microscope. To improve visualization of fruiting bodies, the cores were sprayed with 70% ethanol to remove conidiophores.
For radial growth analysis, each strain was point inoculated and incubated under light or dark conditions at 37uC for 6 days, when colony diameter was measured. Experiments were performed with three replicates.

Gene Expression Analysis
Total RNA was extracted from lyophilized mycelia using RNeasy Mini Kit (Qiagen) or Trizol (Invitrogen), following the manufacturer's instructions. Gene expression levels were evaluated by Northern blots or quantitative reverse transcription-PCR (qRT-PCR) analysis. The templates used for making probes for Northern blots were obtained as follows: ipnA, a 1.1-kb HindIII-EcoRI fragment of pUCHH(458) [52]; aflR, stcU, aatA, acvA, dtiA, and dtiB probe templates were amplified by PCR from A. nidulans genomic DNA with primers indicated in Table 2.
For qRT-PCR, 2 mg of total RNA was treated with RQ1 RNase-Free DNase (Promega). cDNA was synthesized with Moloney murine leukaemia virus (MMLV) reverse transcriptase (Promega). qRT-PCR was performed with the Applied Biosystems 7000 Real-Time PCR System using SYBR green dye for fluorescence detection. The primer pairs used for qRT-PCR are listed in Table 1. The expression data for each gene was normalized to the A. nidulans actin gene expression and the relative expression levels were calculated using the 2 2DCT method.

Locus AN8741.2, Mutated in RM7, Encodes a Putative C 2 H 2 Type Transcription Factor
In our previous study, we generated seven revertant mutants (RMs) capable of restoring normal levels in the production of the orange ST intermediate norsonolinic acid (NOR) in a DstcE strain lacking the veA gene (RDAE206) [40]. Classical genetics analysis revealed that these RMs belong to different linkage groups (data not shown). In the current study we identify the mutated gene in RM7 that restores toxin production in a deletion veA genetic background (Figure 1). The mutation in RM7 was recessive (data not shown) and the specific affected locus was found by complementation of RM7-R2 with an A. nidulans genomic library (pRG3-AMA1-NOT1, [53]).Several positive transformants showing wild-type phenotype were obtained. Sequencing of the rescued plasmids from these fungal transformants and comparison of these sequences with the A. nidulans genomic database (http://www. aspgd.org) by BLAST analysis indicated that they contained the same genomic insert including two ORFs, one of them encoding a putative C 2 H 2 finger domain protein, and another encoding an unknown hypothetical protein (Figure 2). In order to determine where the mutation was located in RM7, the corresponding genomic DNA fragment was PCR-amplified. Sequencing of this PCR product revealed that the mutation occurred in a gene encoding the novel putative C 2 H 2 transcription factor, that we designated mtfA (master transcription factor A). The mutation was a G-T transversion at nucleotide +3 of the mtfA coding region, changing the start codon from ATG to ATT (Figure 2A).

MtfA Orthologs are Present in Other Fungal Species
The deduced amino acid sequence of A. nidulans MtfA revealed significant identity with ortholog proteins from other Aspergillus spp., such as A. clavatus (64% identity), A. terreus (61%), A. flavus (61)%, or A. fumigatus (59%). Further analysis of other fungal genomic databases indicated that MtfA is also conserved in other fungal genera in Ascomycetes (Table S1, Figure S1 and S2). The C 2 H 2 DNA binding domain is highly conserved among these putative orthologs. A MtfA ortholog was not found in the strictyeast fungus Saccharomyces cerevisiae. Similarly, MtfA putative orthologs were not found in plants or animals. Orthologs from other fungal genera are listed in Table S1. An extensive alignment and phylogenetic tree is shown in Figure S1 and S2. MtfA orthologs were particularly conserved among Aspergillus spp. The MtfA tree topology was consistent with established fungal taxonomy. MtfA presents similarity to other A. nidulans C 2 H 2 DNA binding domain proteins (Table S2), showing the highest similarity with FlbC (25.3% identity in the full protein comparison and 29% identity when comparing the DNA binding domains).

mtfA Regulates Mycotoxin Biosynthesis
To confirm that NOR production in RM7 (DveA, X-) was indeed due to a loss-of-function mutation in mtfA, and to assess the effect of this mutation on ST production in a strain with a wildtype veA allele (veA+), we performed a complete deletion of mtfA in RDAE206 (DveA) and RJMP1.49 (veA+), obtaining TDAEDmtfA and TRVDmtfA strains, respectively ( Figure S3). Deletion of mtfA in these strains was confirmed by Southern blot analysis, using the 59 UTR as probe template P1 ( Figure S3B). This probe revealed a 7.1 kb PstI fragment in the wild-type control and a 6.3 kb PstI fragment in the deletion mutants as expected. Also, hybridization with the transformation marker gene used for gene replacement, AfpyrG (specific probe template P2), revealed 6.3 kb and 2.2 kb PstI fragments in mtfA deletion mutants, while these bands were absent in the wild-type control ( Figure S3B), as predicted.
To elucidate the role of mtfA in mycotoxin biosynthesis in a strain with a veA wild-type genetic background (veA+) we analyzed ST production in the TRVDpmtfA strain and compared it with that of the isogenic wild-type control strain and the complementation strain. Interestingly, our results indicated that TRVpDmtfA mutant did not produce ST after 48 h of incubation under both light and dark conditions in the veA wild-type background, whereas the wild type and complementation strain produced clearly detectable levels of ST ( Figure 3A). At 72 h only very low levels of ST were detected in the TRVDmtfAp culture under these experimental conditions ( Figure 3A). In addition, the TLC analysis indicated that deletion of mtfA also resulted in a delay in the synthesis of two additional unknown compounds in cultures growing in the dark ( Figure 3A).

mtfA Controls aflR Expression and Activation of the ST Gene Cluster
Expression of the specific ST regulatory gene aflR, and expression of stcU, gene encoding a ketoreductase that is used as indicator for cluster activation [21,54], were analyzed in liquid shaken cultures of wild type, deletion mtfA and complementation strain at 24 h and 48 h after spore inoculation. Neither aflR nor stcU were expressed in the mtfA deletion mutant, while transcripts for both genes accumulated at the 48 h time point analyzed ( Figure 3B). The presence of these transcripts coincided with the presence of ST in the control cultures. Mycotoxin was not detected in the mtfA deletion cultures under the experimental conditions assayed ( Figure 3C). Analysis of later time points also showed a notable reduction of ST production as well as a reduction in aflR expression in the DmtfA strain with respect to the controls ( Figure  S4), Over-expression of mtfA (alcA(p)::mftA, veA+) also prevented the transcription of aflR and stcU as well as ST production under conditions that allowed the control strains to activate the transcription of ST genes and mycotoxin production ( Figure 4).

Deletion of mtfA does not Recover Mycotoxin Biosynthesis in a Deletion laeA Genetic Background
Since VeA and LaeA proteins can interact in the nucleus and are, at least in part, functionally dependent, we examined whether loss of mtfA results in rescue of ST production in a DlaeA strain. For this purpose, double DmtfADlaeA mutants were generated in veA1 and veA+ genetic backgrounds by meiotic recombination from crosses between RJW34-1 (pyrG89; wA3; DstcE::argB; DlaeA::methG; trpC801; veA1) and TRVDmtfA (Table 1). Our TLC analysis showed that deletion of mtfA did not recover ST biosynthesis in the strains with laeA deletion ( Figure S5).

mtfA Positively Regulates PN Biosynthesis by Controlling the Expression of the PN Gene Cluster
Results from our chemical analysis indicated that mtfA also affects the synthesis of other metabolites ( Figure 5). Based on this finding, we also examined whether mtfA controls PN biosynthesis. We evaluated the production of this antibiotic in TRVpDmtfA and  compared it with PN levels in the isogenic wild-type control and complementation strain. We used a strain of B. calidolactis as testing organism. Deletion of mtfA decreases penicillin production approximately 7-fold with respect to the wild type ( Figure 5A), indicating that mftA is necessary for wild-type levels of penicillin biosynthesis. Our gene expression analysis revealed that acvA, ipnA and aatA, genes in the PN gene cluster [17], are down-regulated in the mftA deletion mutant ( Figure 5B-C), particularly at the 24 h time point (24 h after mycelium is transferred to PN induction medium).
Over-expression of mtfA clearly increases production of PN (approximately 5-fold) with respect to the PN production levels obtained in the wild-type strain ( Figure 6A). Expression of acvA, ipnA and aatA, was greater in the mtfA over-expression strain than in the control strain ( Figure 6B-C). The experiment was repeated several times with similar results.

mtfA Regulates the Expression of Terrequinone Genes
We also tested whether mtfA controls the expression of genes involved in terrequinone biosynthesis, a compound known for its anti-tumoral properties [15]. Specifically we examined the expression of tdiA and tdiB [16,55]. At 24 h and 48 h of incubation, expression of tdiA and tdiB was detected in the wildtype control and complementation strains, while transcripts of these genes were absent in the mtfA deletion mutant ( Figure 7A). Similarly to the case of ST production, over-expression of mtfA negatively affected the expression of tdiA and tdiB ( Figure 7B); Although transcripts were detected for both genes in the mtfA overexpression strain, tdiA expression levels were drastically reduced compared with the control at both 24 and 48 h after induction, and tdiB expression was only detected at 24 h in the overexpression mtfA at very low levels, while it was clearly detectable in the control strain at both time points analyzed ( Figure 7B).

MtfA Subcellular Localization
We further studied the function of the A. nidulans mtfA gene product by examining its subcellular localization in both light and dark conditions. Because the predicted MtfA has a C 2 H 2 DNA binding domain we predicted that it could be found in nuclei. We generated a strain containing MtfA fused to GFP. Our observations using fluorescence microscopy indicated that indeed MftA localizes in nuclei, as revealed when compared with DAPI staining. Nuclear localization of MtfA was independent of the presence or absence of light ( Figure 8).

mtfA Regulates Asexual and Sexual Development in A. nidulans
Deletion of mtfA results in slightly smaller colonies than the wildtype (Figure 9), indicating that mtfA positively influences fungal growth in both light and dark conditions. The mtfA deletion colonies presented a brownish pigmentation which is absent in the control strain. mtfA was expressed at similar levels under conditions promoting either asexual or sexual development, increasing transcript accumulation over time ( Figure S6), Conidiophore formation and conidial production was drastically reduced in the mtfA deletion strains with respect to the wild type ( Figure 10). This effect was observed in both light and dark cultures. The differences in conidiation levels were more pronounced in the light, a condition that promotes asexual development in A. nidulans [56]. In addition, the conidiophores produced by the DmtfA strain presented fewer metula and phialides than the control strains ( Figure S7A). The reduction in conidiation observed in DmtfA coincided with alterations in the expression of brlA ( Figure 10C), a key transcription factor in the initiation of conidiophore formation [57]. Reduction in brlA expression was observed after 48 h of incubation in the light, condition that promotes conidiophore formation. In the dark brlA levels in the wild type were low, as expected. However, expression of this gene in the mtfA mutant was abnormally high in the dark, a condition that represses conidiation [56]. The increase of brlA expression in DmtfA in the dark not only did not result in hyperconidiation, but the conidial production was as low as that observed in DmtfA growing in the light.
Sexual development is also influenced by mtfA. Absence of mtfA in A. nidulans results in a more than 2-fold reduction in Hülle cells, nursing cells participating in the formation of cleistothecia (fruiting bodies) ( Figure 10D) [56]. Cleistothecial production was delayed and decreased in this mutant ( Figures 10A, 10E and S7B-C). The cleistothecia present in DmtfA were of reduced size ( Figure 10A). Expression of nsdD and steA, encoding transcription factors necessary for the activation of sexual development in A. nidulans [58,59] did not significantly change in the absence of mtfA under the experimental conditions assayed (data not shown). Complementation of the deletion mutant with the mtfA wild-type allele restored wild-type morphogenesis.

Discussion
This study revealed and characterized a new putative C 2 H 2 transcription factor, MtfA. This protein, located in the cell nuclei, acts as master regulator in the production of several important secondary metabolites. In addition to this role, MtfA also affects asexual and sexual development in A. nidulans. MtfA presents two C 2 H 2 zinc finger DNA-binding domains at the C-terminal region. In A. nidulans these C 2 H 2 zinc finger domains have been found previously in other regulatory proteins, such as BrlA [57], SteA [60], PacC [52], SltA [61], CrzA [62], CreA [63] and FlbC [64]. Of the A. nidulans C 2 H 2 zinc finger DNA-binding domain transcription factors examined, MtfA showed the highest similarity to FlbC with 25.3% identity. Our in silico analysis revealed that MtfA orthologs are present in many filamentous fungi, and they are not found in S. cerevisiae or in higher eukaryotes.
Our study indicated that A. nidulans MtfA controls the expression of aflR, a gene encoding another transcription factor specifically necessary for the activation of the ST gene cluster [10,11,12], and therefore, affecting the production of the ST toxin. We observed that either absence of mtfA or forced over-expression of mtfA results in a reduction of aflR transcription and decrease in ST biosynthesis, suggesting that only wild-type levels of mtfA gene product, in a balanced stoichiometry with other present factors, is conducive to normal ST levels. This delicate balance among regulatory factors has been previously observed with other regulators. For instance, in case of the global regulator VeA, where both deletion or over-expression of the gene encoding this protein were detrimental to the biosynthesis of the antibiotic PN [21,65]. In addition, our study indicated that the mtfA role in regulating ST production is veA-dependent, which could, at least in part, explain the existence of these biological thresholds for proper function in the case of MtfA abundance in the cell.
VeA has been shown to be functionally associated with LaeA, a chromatin remodeling putative methyltransferase [39,66], that forms part of the velvet complex in the nucleus [36,37]; however, our study showed that deletion of mtfA did not suffice to rescue toxin production in strains where laeA is absent, in both veA1 and veA+ genetic backgrounds. This indicates that both laeA and mtfA are necessary for normal ST production in either veA+ or veA1 background, Similar results were also observed in the case of the A. nidulans rtfA deletion mutant [40], suggesting that although VeA and LaeA are partially functionally connected, they also present differences in their regulatory output. It is possible that mtfA function could be associated with other components of the velvet complex. Future studies in our laboratory will provide further insight on mtfA mechanism of action and its possible connections with other known genetic regulatory networks.
Interestingly, mtfA showed a broad effect influencing the expression of other secondary metabolism gene clusters. Our results revealed that mtfA affects the expression of genes in the terrequinone gene cluster. In this case also both deletion or overexpression of mtfA lead to a reduction in the expression of tdiA, and tdiB, that encode asterriquinone synthetase and a protein with homology to fungal indole prenyltransferases, respectively, involved in the biosynthesis of terrequinone A, a known anti-tumoral bisindole alkaloid with applications in the medical field [16,55].
In addition, our study showed that the master transcription factor, mtfA, also controls the expression of genes in the PN gene cluster and consequently the amounts of PN produced. While deletion of mtfA resulted in a reduction of PN production, differently from the case of ST and terrequinone A, mtfA overexpression enhances expression of PN genes, acvA, ipnA and aatA, and increases PN production with respect to the wild-type levels. This is relevant since PN and PN-derivatives have an important and well established commercial value, since these compounds are amongst the most important small molecules in clinical use for the last 70 years [67,68,69].
A relationship between the genetic regulation of fungal secondary metabolism and development has been previously observed [18,35,37,40]. Our studies in A. nidulans indicated that mtfA not only regulates different aspects of secondary metabolism, but it also affects asexual and sexual development. The mtfA deletion strain presented a reduction in the number of conidiophores formed with respect to the wild type, leading to a notable decrease in conidial production in both light and dark cultures, and therefore reducing the potential of fungal dissemination. Analysis of the expression levels of brlA, an indispensable gene in the control of conidiophore formation [57,70], indicated that brlA expression is dependent on mtfA. Furthermore, A. nidulans sexual development was also influenced by mtfA. The formation of Hülle cell was decreased in the mtfA deletion strain. This is relevant since Hülle cells are thick-walled cells that nourish cleistothecial primordia as they mature [19]. The decrease in Hülle cells observed at this early stage of sexual development could contribute to the delay and reduction in the number of mature cleistothecia formed in the mtfA deletion strain.
In conclusion, we found a novel master transcription factor, MtfA, that controls the activation of several secondary metabolism gene clusters and regulates asexual and sexual morphological development in A. nidulans. Our study indicated that MtfA is conserved in numerous filamentous fungi, particularly among Ascomycetes, many of them species of importance in industrial application, in agriculture or in the medical field. mtfA homologs were not found in plant or animal genomes, suggesting that mtfA or its gene product could have great potential to be used as a genetic target to reduce the detrimental effects of fungi, for instance production of mycotoxin, while enhancing those traits that are beneficial, such as increase in the production of antibiotics and other medical drugs of fungal origin.  Table  S1 in the supplemental material. (TIF) Figure S3 Targeted mtfA deletion. A) Diagram showing PstI sites (P) in the wild-type mtfA locus, and the same locus after gene replacement of mtfA by the A. fumigatus pyrG gene (AfpyrG), used as selection marker for fungal transformation. Recombination events between the the flanking regions are indicated with crosses (X). Primers used for the construction of the deletion cassette are indicated by small arrows as described by FGSC. Fragments used as probe templates for Southern blot analyses are also shown. B) Southern blot analyses. The DmtfA deletion construct was transformed in RDAE206 and RJMP 1.49 strains (Table 1). PstI