Aspergillus fumigatus ffmA Encodes a C2H2-Containing Transcriptional Regulator That Modulates Azole Resistance and Is Required for Normal Growth

ABSTRACT The production of a collection of deletion mutant strains corresponding to a large number of transcription factors from the filamentous fungal pathogen Aspergillus fumigatus has permitted rapid identification of transcriptional regulators involved in a range of different processes. Here, we characterize a gene designated ffmA (favors fermentative metabolism) as a C2H2-containing transcription factor that is required for azole drug resistance and normal growth. Loss of ffmA caused cells to exhibit significant defects in growth, either under untreated or azole-challenged conditions. Loss of FfmA caused a reduction in expression of the AbcG1 ATP-binding cassette transporter, previously shown to contribute to azole resistance. Strikingly, overproduction of the AtrR transcription factor gene restored a wild-type growth phenotype to an ffmAΔ strain. Overexpression of AtrR also suppressed the defect in AbcG1 expression caused by loss of FfmA. Replacement of the ffmA promoter with a doxycycline-repressible promoter restored nearly normal growth in the absence of doxycycline. Finally, chromatin immunoprecipitation experiments indicated that FfmA bound to its own promoter as well as to the abcG1 promoter. These data imply that FfmA and AtrR interact both with respect to abcG1 expression and also more broadly to regulate hyphal growth. IMPORTANCE Infections associated with azole-resistant forms of the primary human pathogen Aspergillus fumigatus are associated with poor outcomes in patient populations. This makes analysis of the mechanisms underlying azole resistance of A. fumigatus a high priority. In this work, we describe characterization of a gene designated ffmA that encodes a sequence-specific transcriptional regulator. We identified ffmA in a screen of a collection of gene deletion mutant strains made in A. fumigatus. Loss of ffmA caused sensitivity to azole drugs and also a large reduction in normal growth. We found that overproduction of the AtrR transcription factor could restore growth to ffmA null cells. We provide evidence that FfmA can recognize promoters of genes involved in azole resistance as well as the ffmA promoter itself. Our data indicate that FfmA and AtrR interact to support azole resistance and normal growth.

consists of a compound mutation corresponding to a duplication of a 34-bp region in the cyp51A promoter along with a single-amino-acid substitution in the encoded enzyme (4). This allele is referred to as TR34 (34-bp promoter duplication) L98H cyp51A. Together, these two alterations cause a dramatic increase in voriconazole MIC. Infections associated with this cyp51A allele are associated with a significantly worse outcome in particular populations (5) and make understanding regulation of cyp51A in particular and azole resistance in general of high priority in this pathogen.
While mutations associated with the cyp51A gene are found with high frequency in azole-resistant A. fumigatus isolates, more recent studies have demonstrated that not all azole resistance in this organism can be explained by mutations at cyp51A (reviewed in references 6 and 7). Mutations in the gene encoding HMG coenzyme A (CoA) reductase have been described that lead to strong decreases in voriconazole susceptibility with a wildtype cyp51A locus present (8,9). Additionally, an alteration in the hapE gene encoding a subunit of the CCAAT-binding complex also produced a strain with an elevated voriconazole MIC (10). Loss of normal hapE function leads to a large increase in cyp51A expression with associated elevation of azole resistance (11). Coupled with the well-described increase in cyp51A expression caused by the presence of the TR34 promoter (11,12), these data illustrate the critical contribution of transcriptional regulation to azole resistance.
An additional non-cyp51A mode of azole resistance is illustrated by the overproduction of the ATP-binding cassette (ABC) transporter AbcG1 (also known as Cdr1B/abcC). Clinical isolates have been described in which elevated transcription of the gene encoding this membrane protein leads to enhanced azole resistance (13). The presence of the abcG1 gene is required for the normal increase in voriconazole resistance seen in strains containing the TR34 L98H form of cyp51A, demonstrating the role of this ABC transporter even in the presence of a strongly resistant allele of cyp51A (14). We and others have found a transcription factor designated AtrR (ABC transporter regulator) that is required for high-level transcription of both abcG1 and cyp51A (15,16). Transcriptional control of cyp51A is a key contributor to azole resistance, but much remains to be uncovered concerning the range of transcriptional inputs to drug resistance in A. fumigatus.
We have previously described a collection of deletion mutants in genes that encode transcription factors (17). The first large-scale screening of this mutant collection employed itraconazole and led to the identification of the negative transcriptional regulator nctA-B. This mutant library also confirmed that deletion mutants in either the sterol gene regulator srbA (18) or atrR caused large increases in itraconazole susceptibility. We counterscreened these itraconazole-susceptible mutant strains for effects on expression of the AbcG1 ABC transporter using an antibody directed against this membrane protein (14). A poorly characterized gene called rfeC (recently renamed ffmA) (19) was identified that caused a large decrease in AbcG1 expression. The ffmA gene encodes a protein containing a C 2 H 2 -containing sequence-specific DNA-binding domain (20). We also found that ffmA null mutants have a large defect in normal growth along with the predicted azole hypersensitivity. Interestingly, overexpression of the AtrR transcription factor was able to suppress both the growth defect of an ffmA null mutant and restore AbcG1 expression. These data indicate the presence of a genetic interaction between AtrR and FfmA.

RESULTS
Screening the transcription factor deletion mutant library for genes that affect azole resistance. Using a previously described transcription factor deletion mutant library (17), we identified additional mutant strains that exhibited increased susceptibility to itraconazole. These individual strains were grown from each library isolate, whole-cell protein extracts were prepared, and these were analyzed by Western blotting using an anti-AbcG1 antiserum (21). The results of this Western blot comparison are shown in Fig. 1.
We found clear and reproducible reduction in the level of AbcG1 in strains lacking either the atrR or AFUB_026340 (Afu2g10550) gene. The latter gene has recently been characterized as a locus enhancing expression of genes involved in fermentation (favors fermentative metabolism, ffmA [19]), and we will use this designation here. The ffmA gene has also been characterized as rfeC (regulator of FLO11) on the basis of activating the FLO11 promoter when expressed as a cDNA clone in Saccharomyces cerevisiae (22). We selected ffmA for further study to examine its role as a regulator of abcG1 expression.
Phenotypes caused by loss of the ffmA gene. To confirm the effects of the ffmA gene on phenotypes, we prepared a new ffmAD allele in a different genetic background (AfS35). We also used the original deletion mutant library strain for comparison. Spores were prepared from the two wild-type strains and their isogenic ffmAD derivatives. A radial growth assay was performed for these 4 different strains ( Fig. 2A).
Irrespective of the genetic background employed, the clearest effect caused by loss of ffmA is a profound growth defect. We estimate the growth rate of the ffmAD strain to be roughly 25% that of the wild-type cells. This was confirmed by quantitation of a radial growth assay (see Fig. S1 in the supplemental material). This defect in growth rate complicated the analysis of azole susceptibility in the ffmAD strain. We also observed the accumulation of an orange/brown pigmentation upon loss of ffmA in both strain backgrounds. To compare the voriconazole susceptibility phenotypes of these strains, we used a zone-of-inhibition assay and placed a filter disk containing this azole drug in the center of the plate of spores from each wild-type and isogenic ffmAD strain. These plates were incubated at 37°C and then photographed (Fig. 2B).
Loss of ffmA from the A1160 wild-type strain caused a reproducible increased susceptibility to voriconazole, but this was masked by the extreme growth defect seen in the AfS35 genetic background. The ffmAD derivative of AfS35 appeared to be more seriously growth compromised than its A1160 ffmAD counterpart. The indicated strains were grown overnight and whole-cell protein extracts prepared. These extracts were resolved on SDS-PAGE and then analyzed by Western blotting using an anti-AbcG1 polyclonal antiserum (21). To ensure equivalent loading of protein extracts, the membrane was stained with Ponceau S dye after transfer.
Since loss of the ffmA gene caused such a pronounced defect in growth, we examined the effect of placing ffmA under the control of a strong promoter. We hypothesized that overproduction of FfmA caused by replacement of the ffmA promoter with the corresponding region from the highly expressed hspA (AFUA_1G07440, Hsp70 family chaperone) gene might affect voriconazole resistance. Using CRISPR/cas9-mediated recombination, the hspA promoter was inserted in place of the native ffmA promoter region. Appropriate transformants were recovered and analyzed for several ffmA-dependent phenotypes.
The presence of the hspA-ffmA fusion gene reduced radial growth to ;35% of normal at 37°C (Fig. 3A), similar to that seen for the ffmAD strain. This growth defect likely contributed to the lack of any effect seen on voriconazole susceptibility.
To confirm that use of the hspA promoter to drive ffmA expression led to overproduction of FfmA, we assessed protein levels of FfmA using Western blotting. To The ffmAD strain present in the transcription factor deletion mutant collection (A1160 background) and an ffmAD strain constructed in the AfS35 genetic background were grown with their isogenic wild-type counterparts and spores isolated. Equal numbers of spores (;100) were placed on minimal medium and allowed to grow at 37°C. Note the orange color of the mycelial side of the ffmAD strains. (B) Voriconazole susceptibility of an ffmAD strain. An equivalent number of spores from the strains indicated were spread on minimal medium, and then a filter disk containing the indicated concentration of voriconazole was placed in the center of each plate. Plates were allowed to develop at 37°C for either 72 or 96 h. Diameters around each disk are listed. accomplish this, a rabbit polyclonal antiserum directed against bacterially expressed FfmA was prepared. Isogenic wild-type and hspA-ffmA strains were grown with or without voriconazole treatment, and levels of FfmA were analyzed by Western blotting.
As shown on the left side of Fig. 3B, treatment of wild-type cells with voriconazole led to a 2-fold increase in FfmA levels compared to untreated cells. The presence of the hspA-ffmA allele caused a 2-fold increase in FfmA levels compared to the wild-type strain, consistent with a higher transcriptional level supported by the hspA promoter as we have seen before (23). Interestingly, exposure of hspA-ffmA cells to voriconazole caused a further increase in FfmA induction to roughly 5-fold above the non-azoletreated wild-type levels. The hspA promoter was inserted into the ffmA locus in place of the wild-type ffmA version. Equal numbers of spores from isogenic wild-type and hspA-ffmA strains were placed on minimal medium alone or containing 0.05 mg/liter voriconazole. Plates were incubated at 37°C for 3 days and then photographed. (B) Whole-cell protein extracts were prepared from isogenic strains containing either the wild-type (wt) or hspA-driven (hspA-ffmA) ffmA gene. These strains were grown either in the presence or absence of 0.05 mg/liter voriconazole. All strains were grown at 37°C for 16 h, except the hspA-ffmA strain in the presence of voriconazole, which was grown for 24 h due to its slow growth under these conditions (indicated by the asterisk). Equal amounts of extracts were analyzed by Western blotting using the indicated rabbit antisera. The membranes were stained by Ponceau S dye to ensure equal loading and transfer. Western blot analysis using the anti-FfmA antiserum is shown on the left, while blotting with the anti-AtrR and anti-AbcG1 antisera is shown on the right. The numbers above each immunoreactive band refer to expression levels normalized to the wild type in the absence of drug.
Our finding that loss of ffmA led to a decrease in AbcG1 expression ( Fig. 1) prompted us to examine the effect of hspA-ffmA on expression of both AbcG1 and the AtrR transcription factor that is key to the regulation of this ABC transporter-encoding gene (15,16). The same protein extracts analyzed in Fig. 3B were evaluated for the level of AbcG1 and AtrR in Fig. 3C using appropriate antisera. Expression of AtrR was increased in the hspA-ffmA strain compared to the wild type by approximately 50%, with a further increase to nearly 2-fold when the hspA-ffmA cells were challenged with voriconazole. Expression of the ABC transporter AbcG1 was seen to increase by 3-fold in hspA-ffmA cells in the presence of voriconazole compared to equivalently treated wild-type cells. These data further support the view that AbcG1 expression responds to changes in ffmA expression and suggest that the atrR gene is another target of FfmA.
Overproduction of AtrR can suppress ffmA-dependent phenotypes. Our finding of this potential link between AtrR and FfmA led us to examine the interaction between the two genes encoding these transcriptional regulators. To test the epistasis of atrR and ffmA, we used a hypermorphic form of atrR that we characterized earlier: the hspA-atrR fusion gene. This allele of atrR was found to cause overproduction of abcG1 along with other AtrR-regulated target genes (16). We introduced the ffmAD null allele into a strain containing the hspA-atrR fusion at the normal atrR locus. Transformants were recovered and confirmed by PCR analyses.
The most obvious phenotype seen was the striking suppression of the growth defect of an ffmAD strain (Fig. 4A). While the radial growth assay demonstrated that the hspA-atrR allele restored wild-type levels of growth to the ffmAD strains, production of orange pigment was retained, indicating that this ffmAD phenotype was not suppressed. This indicates that while there is important overlap between FfmA-and AtrR-dependent phenotypes, this overlap is not universal between these factors.
To examine the ability of the hspA-atrR fusion to suppress the defect in AbcG1 expression caused by the ffmAD lesion, we assessed levels of AbcG1 protein in these strains by Western blotting (Fig. 4B). Loss of ffmA decreased AbcG1 expression in an otherwise wild-type background, while the atrRD null allele reduced AbcG1 expression even further and there was no detectable immunoreactivity in an abcG1D strain. Importantly, the hspA-atrR allele could restore AbcG1 expression to ;80% of wild-type levels in an ffmAD null strain. Overexpression of AtrR from the hspA promoter was sufficient to restore near-normal growth and parental levels of AbcG1 expression to a strain lacking ffmAD.
Doxycycline-dependent expression of ffmA. Both loss of ffmA and overproduction of FfmA from the hspA promoter caused a serious defect in normal growth along with reduction of both AbcG1 expression and voriconazole resistance. To develop a strain that could be grown and tested in a controllable manner, we constructed a doxycycline-repressible form of ffmA. This was accomplished by integrating a doxycyclinerepressible (dox-off; DO) promoter into the ffmA locus to replace the wild-type promoter region. This DO promoter was integrated either immediately upstream of the wild-type ffmA (DO-ffmA) or with a single Flag epitope fused to the N terminus of FfmA (DO-Flag-ffmA). To confirm that the presence of the DO promoter did not cause a severe growth defect in the absence of doxycycline, appropriate transformants were placed on medium containing or lacking this compound and allowed to grow at 37°C.
The presence of either DO allele of ffmA led to nearly normal growth in the absence of doxycycline (Fig. 5A). The growth rate of the DO allele-containing strains was estimated to be 80% of the wild-type strain in the absence of doxycycline. The addition of 25 mg/liter doxycycline caused a reduced growth phenotype to 25% of the wild-type level, resembling that of the ffmAD or hspA-ffmA strain analyzed above.
We next used the doxycycline-repressible nature of the DO-ffmA and DO-Flag-ffmA genes to determine the effect of acutely repressing FfmA production on AbcG1. Both DO promoter-driven forms of ffmA were grown for 8 h and then either left untreated or had doxycycline added to the culture. After 12 additional hours of growth, wholecell protein extracts were prepared and analyzed by Western blotting using anti-FfmA or anti-AbcG1 antisera (Fig. 5C).
Both DO-driven forms of ffmA produced more FfmA protein than was observed in wild-type cells. FfmA levels were elevated to 3.6-fold higher than wild-type levels in the DO-ffmA fusion, while the DO-Flag-ffmA fusion gene protein produced ;50% more FfmA protein than wild-type cells. Surprisingly, AbcG1 expression was lowered in the presence of both DO-ffmA alleles when doxycycline was absent. The DO-ffmA-containing strain produced only 20% of normal AbcG1 protein, while the DO-Flag-ffmA fusion led to production of 70% of wild-type AbcG1 levels. Both DO fusion genes were strongly repressed by the addition of doxycycline to the medium, with both FfmA and AbcG1 levels nearly undetectable. These data support the interpretation that loss of AbcG1 expression and the growth defect caused by loss of ffmA represent an authentic consequence of the absence of FfmA and are not an indirect effect caused by response to the growth defect elicited by FfmA loss.
FfmA binds to the ffmA and abcG1 promoters in vivo. The presence of a C 2 H 2 DNA-binding domain in the FfmA protein sequence as well as its effect on abcG1 and atrR expression is consistent with this protein playing a role as a transcription factor. To determine if FfmA is directly associating with target gene promoters, we carried out single-gene chromatin immunoprecipitation (ChIP) experiments using either the DO-Flag-ffmA fusion gene or wild-type cells. We used anti-Flag antibody to enable ChIP in the DO-Flag-FfmA strain and anti-FfmA in wild-type cells. Fixed chromatin samples To test the relationship between atrR and ffmA, the ffmAD cassette was introduced into an hspA-atrR fusion-containing strain. This double mutant strain was grown along with isogenic wild-type and ffmAD strains, spotted onto minimal medium, and allowed to grow at 37°C. Plates were photographed after 3 days. (B) Whole-cell protein extracts were made from the strains described above and analyzed by Western blotting using the anti-AbcG1 antiserum. The top panel shows Ponceau S staining of the membrane prior to blotting. from these two strains were sheared, immunoprecipitated with the appropriate antibody, and then analyzed by quantitative PCR (qPCR).
ChIP using anti-Flag antibodies in samples from the DO-Flag-ffmA-containing strain showed strong enrichment of the Flag-FfmA protein on both the abcG1 and ffmA Plates were developed at 37°C as before. (C) The strains used for panels A and B were grown in minimal medium with no doxycycline overnight and then diluted into fresh medium lacking (2) or containing (1) the indicated concentration of doxycycline. These cultures were allowed to grow for an additional 12 h and whole-cell protein extracts prepared. Extracts were prepared after normalizing by mycelial dry weight of each culture. Equivalent amounts of each protein extract were analyzed for total loading and levels of FfmA, AbcG1, and tubulin using appropriate antibodies. The numbers below each specific immunoreactive band indicate expression level normalized to the wild-type strain. promoters (Fig. 6A). Note that in the case of this strain, the DO promoter was inserted between the normal ffmA promoter and the ffmA coding region, leading to retention of the wild-type ffmA promoter, although it is no longer controlling expression of ffmA. We also examined ChIP of FfmA in unaltered AfS35 cells using anti-FfmA antibody. ChIP of FfmA in this strain showed similar enrichment of FfmA to both the fully native ffmA and abcG1 promoters. In the case of the Flag-FfmA and wild-type FfmA proteins, only marginal enrichment was seen at the atrR promoter, although it was greater than that seen on the actin promoter (act1) used as a presumptive negative control. These data are consistent with FfmA directly controlling gene expression of ffmA and abcG1 by direct promoter binding and potentially indirectly controlling atrR expression.

DISCUSSION
Our interest in ffmA emerged from the effect of this gene on itraconazole resistance as shown in the early analysis of a collection of transcription factor gene deletion mutants (17). More detailed analysis of the phenotype of a ffmAD strain clearly indicated that loss of this gene had a modest azole drug phenotype compared to its . Strains were grown in minimal medium and fixed chromatin prepared. After shearing, FfmAbound DNA was recovered with anti-FLAG immunoprecipitation (top) or anti-FfmA immunoprecipitation (bottom). Immunoprecipitated DNA was analyzed using primer sets specific for the indicated 4 promoters. A control immunoprecipitation was performed with omission of the primary antibody as a control. The fold enrichment refers to the ratio of immunopurified DNA recovered in the presence of primary antibody compared to the absence. The numbers above each bar correspond to the fold enrichment from a representative trial. Note that act1 immunopurification is used as a negative-control promoter.
pronounced defect in growth. Other screenings of this transcription factor deletion mutant library led to the detection of ffmA as a determinant required for resistance to cell wall stress agents but also showed a very similar general growth defect (19). Based on both this and the previous study of the phenotype of the ffmAD strain, it seems that the primary phenotype caused by loss of ffmA is a severe growth deficit. This is in contrast to deletion of genes such as srbA (18) or atrR (15) that cause striking sensitivity to azole drugs but grow relatively well in drug-free media. FfmA is necessary for normal expression of abcG1, but our data indicate a wider role in cell growth beyond azole resistance.
Based on the presence of a C 2 H 2 zinc finger-containing DNA binding domain in its protein sequence, we believe that FfmA is a sequence-specific transcription factor. Our ChIP experiments support this view, as FfmA is found to enrich on the abcG1 promoter, which would be consistent with FfmA acting to positively regulate expression of this gene. Additionally, we found that this same abcG1 fragment was bound by AtrR (16), suggesting that FfmA and AtrR bind the same or adjacent regions of the abcG1 promoter. We also found that FfmA showed enrichment on its own promoter, consistent with its expression being autoregulated.
The shared role for AtrR and FfmA in control of abcG1 expression led us to examine the interaction between these two genes. Overproduction of AtrR using the hspA promoter was well tolerated by the cell and led to both increased azole resistance (16) and strong suppression of the growth phenotype caused by loss of ffmA. Owing to the difficulty in growing ffmAD strains, we disrupted ffmA in the hspA-atrR background. These transformants were readily obtained and grew normally. Furthermore, the contribution of FfmA to AbcG1 expression was strongly suppressed by the increased levels of AtrR. This genetic interaction led us first to suspect that FfmA acts upstream of AtrR in terms of function, but ChIP experiments on the atrR promoter have not yet revealed strong enrichment for FfmA, although we have not scanned the entire promoter. Another possibility is that both AtrR and FfmA stimulate expression of abcG1 transcription, and the increased levels of AtrR are adequate to reverse the transcriptional defect caused by loss of ffmA.
The precise role of FfmA in the cell remains to be determined. Clearly, A. fumigatus is exquisitely sensitive to modifications of FfmA expression. As we show here, either loss of ffmA or its overproduction from the hspA promoter caused a strong defect in normal growth. Driving either a Flag-tagged or untagged FfmA from the dox-off promoter did allow growth that was similar to that of the wild type but remained only 80% of normal. This sensitivity to variation in expression level has been noted before in the analysis of human disease genes (24), in which genes linked to disease are rarely seen to be associated with copy number variation owing to their critical importance. These genes are often associated with developmental processes, as disturbances in the expression profiles of genes of this type lead to loss of viability. In the case of ffmA dosage changes, cells are still viable but clearly suffer significant impacts on growth. Since this ffmAD-mediated growth defect can be robustly suppressed with atrR overproduction, we suggest that the downstream functions of these transcription factors overlap. As discussed above, this is certainly true for control of abcG1 expression, but as abcG1D strains have no detectable growth defect in the absence of azole drugs, other target genes must be shared that impact growth under unstressed conditions. Identification of these genes will be an important future goal.

MATERIALS AND METHODS
A. fumigatus strains, growth conditions, and transformation. The lab strains that were used in this study are listed in Table 1. For initial shortlisting of strains that were defective for abcG1 expression, the itraconazole-susceptible members of the transcription factor deletion mutant were used (17). A. fumigatus strains were typically grown at 37°C in rich medium (Sabouraud dextrose; 0.5% tryptone, 0.5% peptone, 2% dextrose [pH 5.6 6 0.2]). Selection of transformants was performed using minimal medium (MM; 1% glucose, nitrate salts, trace elements, 2% agar [pH 6.5]). Trace elements, vitamins, and nitrate salts are as described in the appendix of reference 25, supplemented with 1% sorbitol and either 50 mg/ liter phleomycin (after adjusting the media to pH 7) or 200 mg/liter hygromycin gold (both InvivoGen).
For solid medium, 1.5% agar was added. Doxycycline promoter shut-off experiments were performed by adding 25 mg/liter doxycycline (BD Biosciences). Wild-type and doxycycline-repressible ffmA-containing strains were grown overnight in the absence of doxycycline. Doxycycline repression was initiated and incubation continued for 12 h prior to sample preparation and analysis.
Transformation and generation of ffmA mutants was done using in vitro-assembled cas9-guide RNA ribonucleoproteins coupled with 50-bp microhomology repair templates (26). For generation of ffmA deletion mutants, 2 CRISPR RNAs (59-ATCTAGGATCCATCATGAAG, corresponding to the 59 end of the gene, and 59-ATAGTCAATGGCTCAAGGGA, corresponding to the 39 end of the gene) were used to replace ffmA with the hygromycin resistance marker cassette amplified from the plasmid pSP62 (16) using ultramergrade oligonucleotides from IDT harboring 50-bp homology to the upstream and downstream junctions of the ffmA gene. The doxycycline-repressible (DO) promoter (with or without a Flag tag) marked with the phleomycin resistance cassette was inserted upstream of the ffmA gene using the single CRISPR RNA 59-ATCTAGGATCCATCATGAAG. For generating the DO-ffmA allele, a microhomology repair template containing the resistance marker cassette and DO promoter were PCR amplified from plasmid pSP114. The plasmid pSP114 was generated from pSK606-ptrA (provided by J. Fortwendel), from which the ptrA marker was replaced by a phleomycin resistance gene (Ble) by cloning the latter from pCH008-PhleoR (also from J. Fortwendel) at the KpnI and PstI sites. To generate the Ble-DO-Flag cassette, complementary oligonucleotides corresponding to a 1Â Flag tag was annealed and ligated to the ends of PmeI-linearized pSP114. This plasmid was named pSP115 and was used as the template to generate microhomology repair constructs to generate the Ble-DO-Flag-ffmA allele. Transformants were genotypically confirmed by diagnostic PCR of the novel upstream and downstream junction formed upon targeted integration as well as by PCR amplification to confirm lack of the DNA binding domain of ffmA in the case of ffmA deletion mutants. At least 3 independent, properly targeted transformants were phenotyped for all ffmA mutants, of which a representative strain is depicted in the data presented.
Radial growth/drug disc diffusion assay. Fresh spores of A. fumigatus were suspended in 1Â phosphate-buffered saline (PBS) supplemented with 0.01% Tween 20 (1Â PBST). The spore suspension was counted using a hemocytometer to determine the spore concentration. Spores were then appropriately diluted in 1Â PBST. For the drug diffusion assay, 1 Â 10 6 spores were mixed with 10 ml soft agar (0.7%) and poured over 15 ml of regular agar containing (1.5%) minimal medium. A paper disk was placed on the center of the plate, and 10 ml 1 mg/liter voriconazole was spotted onto the sterile filter paper. For the radial growth assay, ;100 spores (in 4 ml) were spotted on minimal medium with or without the drug. The plates were incubated at 37°C and inspected for growth every 12 h.
Generation of an FfmA antibody. A region of 502 bp (corresponding to amino acids 195 to 340 based on homology to the DNA binding domain of ScAdr1) from ffmA was PCR amplified and cloned in frame as an NdeI/HindIII fragment upstream of the C-terminal 6ÂHis tag in pET29a1 (EMD Millipore, Inc.) to form plasmid pSP113 and was transformed into the Escherichia coli strain BL21(DE3). Two liters of transformed bacteria was grown to log phase and induced with isopropyl-b-D-thiogalactopyranoside for 90 min. Cell lysates were prepared using a French press and subsequent protein purification accomplished using Talon metal affinity resin (TaKaRa Bio USA, Inc.) as described by the manufacturer. Protein fractions were analyzed by staining them with Coomassie blue and by Western blotting using His-specific antibodies. The purified proteins were then lyophilized and sent to Pacific Immunology (Ramona, CA) for injection into rabbits to generate polyclonal antibodies against FfmA. Antiserum generated from these rabbits was received and tested for immunoreactivity against A. fumigatus cell lysates. The antiserum was then purified using an AminoLink Plus coupling resin (Thermo Scientific, Inc.) according to the manufacturer's instructions, and the affinity-purified antiserum was used to detect the FfmA protein from A. fumigatus cell lysates.
Western blotting. Western blotting was performed as described in reference 14. The FfmA polyclonal antibody used here has been detailed in the reference mentioned above and was used at a 1:500 dilution. The anti-Flag M2 monoclonal antibody (F1804) was procured from Sigma and used at a 1:2,000 dilution. AbcG1 polyclonal antibody (21) was used at a dilution of 1:500.
Chromatin immunoprecipitation. Chromatin immunoprecipitation was done as described in reference 16, with the following modifications: 30 ml was reserved as an input control (IC) fraction for reverse cross-linking to verify sonication and control for ChIP and qPCR. The sheared chromatin was incubated with either anti-Flag M2 monoclonal antibody (F1804; Sigma) at 1:250 dilution or with anti-FfmA polyclonal antibody (described above) at a dilution of 1:50 overnight (16 h) on a nutator at 4°C. This sample was further incubated with 50 ml of washed Dynabeads (Life Technologies) conjugated to either protein G (when using anti-Flag) or protein A (when using anti-FfmA) for another 8 h. Real-time PCR of ChIP DNA was also performed as described in reference 15, with the following modifications: 0.5 ml of ChIP or input (diluted 30-fold to bring it to 1%) DNA was used in a 20-ml total volume reaction mix using SYBR green master mix (Bio-Rad) and 0.4 mM each primer. Fold enrichment was calculated to determine the enrichment of the promoter region for each gene. The oligonucleotide primer pairs used to check promoter enrichment consist of ffmA-ChIP-F (59-GCTGAAATATGGATGCCTCTC), ffmA-ChIP-R (59-ACCTTCT GTACTTCGTGGTAAC), atrR-ChIP-F (59-ACGGGATCCGTTTTGATACTC), atrR-ChIP-R (59-CTGAACGAAGA GTCCGTCTC), abcG1-ChIP-F (59-CGCTAATCATGAATCATCCCAC), abcG1-ChIP-R (59-TCTCTTTTCTTGGAC CCGAC), actA-ChIP-F (59-GCCACCTAAGCGTTACCACT), and actA-ChIP-R (59-GCCGCTTCGTATAGGAGACC). The positions of each amplicon produced for these promoters are the following: ffmA, 22338 to 21973; atrR, 21035 to 2700; abcG1, 2766 to 2445; and actA, 2916 to 2580. Note the numbering is relative to the ATG for each gene.

SUPPLEMENTAL MATERIAL
Supplemental material is available online only. FIG S1, PDF file, 0.4 MB.