Bromoquinol (BMQ) was purchased from Alfa Aesar; 6-iodoquinoline (IDQ) and resveratrol (RVT) from TCI Europe; hydrogen peroxide (H₂O₂) from Merck; dimethyl sulfoxide from Fisher Chemical, and the remaining compounds from Sigma Aldrich, namely pentachlorophenol (PCP), triclosan (TCS), salicylate (Sal), benzo[a]pyrene (BaP), congo red, sodium benzoate, menadione sodium bisulfite (MSB), amphotericin B (Amph B), caspofungin (CSP), itraconazole (ITZ), miconazole (Mic) and 2,7-dichlorofluorescin diacetate (DCFH-DA).

Aspergillus fumigatus CEA17 reference strain was propagated at 37°C in solid complete medium [1% D-glucose, 0.2% Peptone, 0.1% Yeast extract, 0.1% Casamino acids, 50 mL of a 20× salt solution, 0.1% trace elements, 0.1% Vitamin solution, 2% agar, pH 6.5 with NaOH]. The composition of the trace elements, vitamins, and nitrate salts has been described previously (Kafer, 1977).

The Minimal Inhibitory Concentration (MIC) for fungal growth inhibition of IDQ, initially acquired from The Pathogen Box (www.pathogenbox.org/), was defined following the standard methodology implemented by the Clinical and Laboratory Standards Institute (CLSI, 2018). The compound’s antifungal activity was analyzed by serial dilutions using MIC assay (0 to 25 μM) in MOPS buffered RPMI 1640 medium (Sigma-Aldrich), pH 7.0 in 96-wells plates. In each well, a total of 1×10⁴ conidia of A. fumigatus wild-type strain was inoculated. Plates were incubated at 37°C without shaking for 48 h. Non-inoculated controls were done in parallel. All experiments were done in triplicate.

The preparation of total RNA samples for RNA-seq Expression Profiling was as follows. Erlenmeyer flasks (125 mL) were used to inoculate 1×10⁷ spores in 30 mL of Vogel’s Minimal Media (Vogel, 1956) and incubated for 16 h at 37°C, 180 rpm. The medium was then exchanged, and 0, 0.5x MIC (=0.35 μM) or 2x MIC (=1.4 μM) of IDQ was added and incubated for 4 h at 37°C, 180 rpm. Six replicates for each condition were prepared. At the end of the incubation period, the cultures were filtered and frozen immediately in liquid nitrogen. Total RNA from six mycelia per condition were extracted using RNeasy Plant Mini Kit (Qiagen), according to the manufacturer’s protocol, a TissueLyser LT (Qiagen) for cell disruption, and approximately 30 mg of poly(vinylpolypyrrolidone) per sample. RNA quality (integrity) was evaluated using a Nucleic Acid QC - Fragment Analyzer.

For single-end RNA sequencing (RNA-seq), libraries were generated using the Smart-Seq2® mRNA assay (Illumina, Inc.) according to the manufacturer’s instructions. Six samples were indexed and sequenced on the Illumina NextSeq550 (20 M reads per sample). Generated FastQ files were analyzed with FastQC, and any low-quality reads were trimmed with Trimmomatic (20). All libraries were aligned to the corresponding model fungus A. fumigatus A1163 genome assembly (ASM15014v1) with gene annotations from Ensembl Fungi v. 45 using HISAT2 v. 2.1.0 (Kim et al., 2015), and only matches with the best score were reported for each read. All RNA-seq experiments were carried out in three biological replicates. Differential expression analysis was performed using DESeq2 v. 1.24.0 (Love et al., 2014). The genes that showed more than log₂ 1-fold expression changes with p-adj value < 0.05 were considered as significantly differentially expressed (IDQ dataset in Supplementary Dataset 1).

The transcriptomics datasets were selected based on the following stringent rules: only studies on aspergilli were considered (i), comprising an exposure period lower than six days (ii) to an organic compound displaying broad environmental, biotechnological or health relevance (iii) with a molar mass below 500 g·mol⁻¹. Consequently, we selected five transcriptome-based datasets: the catabolism of the simple aromatic hydrocarbon salicylate (Sal) in A. nidulans (Martins et al., 2021); the mode of action of the polyphenol resveratrol (RVT) in A. flavus (Wang et al., 2015); the degradation of the polycyclic aromatic hydrocarbon benzo[a]pyrene (BaP) in Aspergillus sp. (Loss et al., 2019); the inhibitory effects of the quinoline bromoquinol (BMQ) (Ben Yaakov et al., 2017) and of 6-iodoquinoline (IDQ) in A. fumigatus (see above). The full gene lists (up-regulated only) used in our analysis are available in Supplementary Dataset 2.

The COCOA strategy uses a set of established bioinformatics tools (Figure 1A) as detailed below. First, we performed the curation and validation of the five selected datasets by reprocessing the raw data to obtain the gene counts using the HISAT2 methodology (Kim et al., 2015) and identifying the differentially expressed genes using the DESeq2 R-based package (Love et al., 2014). Then, a full protein-translated genome orthology transformation to A. nidulans (the “receiver species”) was performed for eight distinct aspergilli available in the FungiDB database: A. aculeatus ATCC16872, A. flavus NRRL3357, A. fumigatus A1163, A. nidulans FGSCA4, A. niger ATCC13496, A. sydowii CBS593.65, A. terreus NIH2624, and A. versicolor CBS583.65 (Stajich et al., 2012). For the orthology analyzes, we used OrthoFinder (Emms and Kelly, 2015, 2017, 2019) because of its proven excellent performance compared to other orthology tools (Emms and Kelly, 2015). The use of these additional genomes than those strictly necessary adds robustness to the construction of gene trees, resulting in better discrimination and stringency levels between orthogroups (Emms and Kelly, 2015, 2017, 2019; Martins et al., 2020). We transposed the list of the up-regulated genes in the five datasets to the corresponding A. nidulans orthologues. After that, the co-expressed genes were identified using the online tool InteractiVenn (Heberle et al., 2015). Finally, we performed a whole-genome protein domain analysis for the protein-translated genome of the “receiver species,” aiming to obtain hints on the putative functions played by the co-expressed genes, necessary when a functional annotation is lacking.

If not mentioned, assays used minimal medium with glucose (10 g·L⁻¹) and nitrate (MMGN), as follows: thiamine (0.01 g·L⁻¹), 5% (v/v) nitrate salts solution [NaNO₃ (120.0 g·L⁻¹), KCl (10.4 g·L⁻¹), MgSO₄·7H2O (10.4 g·L⁻¹) and KH₂PO₄ (30.4 g·L⁻¹)], 0.1% (v/v) trace elements solution [ZnSO₄·7H2O (22.0 g·L⁻¹); H₃BO₃ (11.0 g·L⁻¹); MnCl₂·4H2O (5.0 g·L⁻¹); FeSO₄·7H2O (5.0 g·L⁻¹); CoCl₂·6H₂O (1.7 g·L⁻¹); CuSO₄·5H₂O (1.6 g·L⁻¹); Na₂MoO₄·2H₂O (1.5 g·L⁻¹), and Na₄EDTA (50.0 g·L⁻¹)], and the pH was adjusted to 6.5 with NaOH. Whenever needed, the media was jellified with 1.5% agar (solid cultivation). The MMG is similar to that described above except that sodium nitrate was removed. Most functional assays used MMG supplemented or not with a defined amount of a N source as mentioned in the results section. Moreover, in all comparative assays of mutant versus parental strain, the medium was supplemented with the essential nutrients: uracil (1.12 g·L⁻¹), uridine (1.22 g·L⁻¹), riboflavin (2.5 mg·L⁻¹) and pyridoxine (0.05 mg·L⁻¹).

EC₅₀ levels of each organic compound were assessed on basis of hyphal radial growth rate for the strain A. nidulans FGSC A4 grown in petri dishes (55 mm) containing the MMGN (jellified) and supplemented with either BaP (1.0–3.5 mM); BMQ (0.1–1.0 mM); Sal (75–400 mM); IDQ (0.002–1.0 mM); RVT (0.005–1.5 mM); or sodium benzoate (25–200 mM). Controls without the organic compounds were also made. The assay was carried out by inoculating 2×10⁵ conidia into the center of the plate, and incubate at 30°C for 120 h. Radial growth of mycelia (colony diameter in cm) was measured using a Vernier caliper (error ± 0.05 mm). The plates that did not displayed visible growth were visualized under microscope to confirm growth inhibition. The EC₅₀ values were calculated using the XLSTAT tool (Microsoft Excel) with the Gompertz model.

For the targeted gene expression analysis of the two genes of the AN9181 orthogroup, Aspergillus nidulans (5×10⁵ spores·mL⁻¹) was pre-grown during 24 h in 12-well plates (2 mL of MMGN per well) at 30°C with gentle agitation (100 rpm), in the dark. After the pre-growth phase, the selected organic compounds were added to the cultures at a defined concentration (Time 0). The concentrations used were as follows: 0.2 mM for BaP [similar to the concentration used by Loss et al. (2019)]; 0.5 mM for BMQ (half of the minimal inhibitory concentration described by Ben Yaakov et al. (2017); 0.3 mM for PCP (concentration previously tested before by Varela et al. (2017), and 0.2 mM for TCS to match the range used for the other organic compounds. For the gene expression analysis of AN9181 in the presence of H₂O₂, A1145 strain (1×10⁶ spores·mL⁻¹) was pre-grown during 140 h in 12-well plates (2 mL of MMG per well) at 30°C with gentle agitation (100 rpm), in the dark. After the pre-growth phase, H₂O₂ was added to the cells at a defined concentration (1.0 –4.0 mM) and incubated at 30°C for 30 min. Controls without addition of H₂O₂ were grown as well. Mycelia samples were collected from the 12-well plates, and immediately frozen using liquid nitrogen until further analysis.

Total RNA extraction and cDNA synthesis were performed using the RNeasy Plant Mini Kit (Qiagen) and the iScript cDNA Synthesis Kit (Bio-Rad), and the RT-qPCR performed as previously reported (Martins et al., 2020). The oligonucleotide pairs for specific A. nidulans genes (Supplementary Table 1) were designed using the Primer-Blast web tool (www.ncbi.nlm.nih.gov/tools/primer-blast/), and purchased from STAB Vida Lda. (Portugal). The RT-qPCR analysis was performed in a CFX96 Thermal Cycler (Bio-Rad), using the SsoFast EvaGreen Supermix (Bio-Rad), 250 nM of each oligonucleotide and the cDNA template equivalent to 10 ng of total RNA, at a final volume of 10 μL per well, in at least three biological replicates. The PCR conditions were as follows: enzyme activation at 95°C for 30 s; 40 cycles of denaturation at 95°C for 5 s, and annealing/extension at 60°C for 15 s; and the melting curve obtained from 65 to 95°C, consisting of 0.5°C increments for 5 s. Data were analyzed with the CFX Manager software (Bio-Rad). In more detail, the expression of each gene was taken as the relative expression compared to the time zero (before incubation with the tested compounds) or as the relative expression compared to the no addition of the stressor. The expression of all target genes was normalized by the expression of the histone H2B gene (AN3469) (Supplementary Dataset 3), used as the internal control. Statistical analyzes used the XL-STAT (Addinsoft) software, and multiple Student’s t-tests. Differences in gene expression with a p-value below 0.05 were considered statistically significant.

The gene AN9181 was replaced with Aspergillus fumigatus pyrG gene (pyrG^Afu) in Aspergillus nidulans A1145 and A1149, both auxotrophic strains (pyrG⁻). Deletion cassettes combining the 5′ and 3′-flanking regions of each target gene with pyrG^Afu were obtained by fusion PCR and were used to transform A. nidulans A1145 or A1149 protoplasts and plated onto selective media to generate single-deletion mutant strains (Supplementary Table 2). Isolated transformants were cultivated on the selective media for three generations to assure stable mutations. In greater detail, the deletion cassettes were constructed using a fusion PCR protocol. Six primers (P1-P6) were designed, based on sequences from the Aspergillus Genome Database (www.aspgd.org), and analyzed using the NetPrimer web tool (www.premierbiosoft.com/NetPrimer/AnalyzePrimer.jsp) (listed in Supplementary Table 3). PCR reactions were performed in a T100 Thermal Cycler (Bio-Rad; conditions in Supplementary Table 4). A. fumigatus pyrG was amplified from plasmid pCDS60 (FGSC, Kansas City, MO, United States) using primers CDS164 and CDS165. Flanking fragments upstream and downstream of the gene were amplified with primer pairs P1/P3 and P4/P6, respectively, using genomic DNA from A. nidulans A4 as template. The final cassette was produced by fusing the flanking regions with the A. fumigatus pyrG using nested primers P2 and P5 for the target gene. PCR products were cleaned with NZYGelPure kit (NZYTech). Then, to produce transformable protoplasts, A. nidulans mycelia grown overnight from 10⁸ conidia in 50 mL MMGN with the appropriate nutritional supplements (30°C, 90 rpm), recovered by centrifugation and washed (0.6 M MgSO₄), were digested in 20 mL enzymatic mix (300 mg lysing enzymes from Trichoderma harzianum, 150 μL β-glucuronidase from bovine liver, type B-1, and 150 mg Driselase from Basidiomycetes sp., all from Sigma-Aldrich) in osmotic medium (1.2 M MgSO₄·7H₂O and 10 mM sodium phosphate buffer pH 6.5; final pH adjusted to 5.8 with Na₂HPO₄) for 20 h at 30°C, 90 rpm. The protoplasts suspension was overlaid (2:1) with trapping buffer (0.6 M sorbitol and 100 mM Tris–HCl pH 7.0), and then recovered by centrifugation (1,500 g, 4°C, 15 min, swing-bucket rotor), washed three times with 10 mL of ST10 buffer (1.2 M sorbitol and 10 mM Tris–HCl pH 7.5) and finally resuspended in 1 mL of the same buffer and incubated overnight at 4°C. In the next day, the protoplasts were collected by centrifugation (1,000 g, at 4°C, 2 min), and resuspended in 700 μL cold STC buffer (1.2 M sorbitol, 10 mM Tris–HCl pH 7.5 and 10 mM CaCl₂). The obtained protoplasts (100 μL) were then transformed by mixing with the cleaned fusion PCR product (10 μL), subsequently adding freshly filtered polyethylene glycol (PEG) solution (25% (w/v) in STC buffer; 50 μL) and kept in an ice bath for 25 min. Then, additional PEG solution was added (1 mL), gently mixed using a micropipette and placed at room temperature for 25 min. For the single mutants A1145ΔAN9181 and A1149ΔAN9181, 100 μL of the transformation mix were plated onto a selective medium containing glucose (5.0 g·L⁻¹), yeast extract (5.0 g·L⁻¹), sucrose (342.3 g·L⁻¹), riboflavin (2.5 mg·L⁻¹), pyridoxin (0.05 mg·L⁻¹), 0.1% (v/v) trace elements solution (see above), and 1.5% agar. The selective plates were incubated for 3 to 4 days at 30°C. All transformants were morphologically identical to the parental strain. Two isolated A1145ΔAN9181 transformants (randomly selected) were streaked onto complete medium plates containing glucose (10.0 g·L⁻¹), peptone (2.0 g·L⁻¹), yeast extract (1.0 g·L⁻¹), casein hydrolysate (1.0 g·L⁻¹), 5% (v/v) nitrate salts solution (see above), 0.1% (v/v) trace elements solution, riboflavin (2.5 mg·L⁻¹), pyridoxin (0.05 mg·L⁻¹), pH 6.5, and incubated at 30°C for 4 days. Three generations of each transformant were grown to assure stable mutations and then grown again in MMG without uracil and uridine to confirm the prototrophy to the compounds of the mutant strains (Supplementary Figure 1).

DNA from each transformant was extracted with Quick-DNA™ Fungal/Bacterial Microprep Kit (Zymo Research), and diagnostic PCR was performed with primers P1 and P6 for each gene. Based on amplicon size (3,222 bp for WT and 3,904 bp for A1145ΔAN9181, Supplementary Figure 2A), it was possible to confirm the correct gene replacement of the transformants. To obtain further confirmation, the PCR products were digested for 1 h with the restriction enzymes ApaI or KpnI selected to give differential digestion patterns for mutant and wild-type strains. Supplementary Figure 2B show the products of diagnostic PCR and restriction enzymes digestion for the A1145ΔAN9181 selected strain.

Cell viability was evaluated using the XTT assay in 96-well plates (200 μL per well) using Malt Extract medium or MMG supplemented with 10 mM N source, and an inoculum of 1×10⁶ spores·mL⁻¹, and incubated at 30°C for 24 h (triplicates). After, 10 μL of a solution containing 4.6 mg·mL⁻¹ of XTT and 0.104 mg·mL⁻¹ of menadione was added to each well, incubated for further 2 h, and the absorbance measured (460 nm).

The strains’ spore germination fitness, i.e., number of spores that are able to form a colony, was evaluated by spreading 100 spores onto solid media, then counting the colony forming units (CFUs) daily, during five days (triplicates). The radial growth diameter of each strain (inoculum: 1 μL of a suspension of 2×10⁸ spores·mL⁻¹) in solid MMG, supplemented or not with specific compounds (viz. congo red, H₂O₂, MSB, Amph B, CSP and ITZ) was measured after five days of incubation (30°C, dark, triplicates). The MIC of each antifungal was determined using the micro-broth dilution method (MMG, 96-well plates, 200 μL per well), testing specific concentration ranges for Amph B (250–550 mg·L⁻¹), CSP (60–120 mg·L⁻¹) and ITZ (0.2–0.5 mg·L⁻¹). An inoculum of 1×10⁶ spores·mL⁻¹ was used, and the plates were incubated at 30°C for 48 h. Negative controls were done in parallel. The lowest concentration that showed no growth under microscopic observation was considered the MIC.

Intracellular ROS was quantified using DCFH-DA. Cultures were grown in MMG (140 h, 30°C); then 2.5 μg·mL⁻¹ of DCFH-DA was added and incubated (30°C, 30 min). H₂O₂ was added to the cells at increasing concentrations, from 1 mM to 4 mM, and the incubation step repeated. The cell suspension was disrupted in a TissueLyzer LT (Qiagen) with a metal bead at a maximum speed (3 cycles of 1 min). The fluorescence intensity of the supernatant (recovered by centrifugation: 12000 g, 10 min) was measured using a Tecan Infinite M Nano⁺ Microplate (Männedorf, Switzerland) as follows: excitation length 485/9; emission 528/20; optics, top; read speed, normal; gain, 89; number of flashes, 25; integration time, 40 μs. The fluorescence intensity (per mycelia dry weight) was normalized against the control (no H₂O₂ added).

The orthology analysis (Figure 1A) revealed a total of 11,992 orthogroups among the eight analyzed Aspergillus spp. genomes (Supplementary Dataset 4). Among these, only 258 were species-specific orthogroups, comprising 0.7% of the total number of genes. Figure 1B displays the phylogenetic relations computed by the orthology analysis, and Figures 1C–E provide for each species the total number of genes, the percentage of genes in orthogroups and the number of species-specific orthogroups, respectively. The number of species-specific orthogroups (Figure 1E) is not necessarily correlated with the percentage of genes in orthogroups (Figure 1D) nor the genome size (Figure 1C). For instance, A. versicolor possesses the third larger genome among the eight analyzed Aspergillus spp. yet displays one of the highest percentages of genes present in orthogroups but the lowest number of species-specific orthogroups. On the other hand, A. sydowii, which possesses a genome size comparable to that of A. versicolor and only a slightly lower percentage of genes belonging to orthogroups, is the species presenting the higher number of species-specific orthogroups. This is an indication of a higher occurrence of gene duplication events (Emms and Kelly, 2015) in A. sydowii compared to other aspergilli, a feature also visible in A. aculeatus and A. flavus. Finally, A. fumigatus and A. nidulans present similarly sized genomes, as well as a comparable percentage of genes included in orthogroups and number of species-specific orthogroups.

Upon transposing the differentially expressed genes (up-regulated) from the five transcriptomic datasets analyzed to the corresponding orthologues in the genome of A. nidulans (Supplementary Dataset 2) we analyzed the set of genes present in at least three, four or five transcriptome-based datasets on exposure to selected organic compounds. We observed that, out of the 241 genes present in at least three datasets, several are predicted to be transporters or secondary metabolism-related genes (Supplementary Dataset 5). Twenty-two genes are present in at least four datasets (Table 1), of which a single gene, AN9181, is present in all five datasets (Figure 2A). This gene is part of an orthogroup that contains two genes in A. nidulans, being the other one AN8970.

The AN9181 orthogroup (gene tree displayed in Figure 2B) comprises 25 genes across the eight analyzed aspergilli genomes, namely four in A. aculeatus, four in A. flavus, two in A. fumigatus, two in A. nidulans (as mentioned), two in A. niger, four in A. sydowii, three in A. terreus, and four in A. versicolor. The existence of the AN9181 orthogroup is therefore highly conserved in these aspergilli. The presence of four genes belonging to the AN9181 orthogroup in A. sydowii, A. aculeatus and A. flavus reinforces the indication of higher occurrence of gene duplication events in these species. This is an indication that the AN9181 orthogroup, like many other regulatory genes, is undergoing rapid evolution in Aspergillus sp. as a response to environmental changes and adaptive lifestyles. This plasticity is often associated with the evolution/transition to pathogenic lifestyles in aspergilli (Rokas, 2022). The 25 genes that compose the AN9181 orthogroup are largely uncharacterized with neither relevant information in FungiDB and nor predicted protein interactions. However, the orthologous gene cip1 of Candida albicans is involved in oxidative stress response, being regulated by the transcription factor Cap1p (Wang et al., 2006; Znaidi et al., 2009). This past evidence raises the hypothesis that AN9181 participates in stress responses.

Through a computational protein domain analysis (InterProScan), we observed that the AN9181 encodes a protein containing a phenylcoumaran benzylic ether reductase-like domain (IPR045312) and a Nitrogen metabolite repressor regulator (NmrA)-like domain (IPR008030), while AN8970 encodes a protein containing a NAD(P)-binding domain (IPR016040). A deletion mutant of padA, that encodes a protein also carrying a NmrA-like domain of Dictyostelium discoideum, showed to be more sensitive to ammonia than the wild-type (Núñez-Corcuera et al., 2008). Moreover, disruption of this gene resulted in phenotypic defects in development and growth, namely, the thermosensitive mutant allele padA⁻ showed poor and null growth at permissive and restrictive temperatures, respectively. Deletion of the NmrA gene in A. flavus reduced growth in several nitrogen (N) sources, but increased conidia and sclerotia production (Han et al., 2016). This mutant strain produces less aflatoxin when cultivated in glutamine and alanine supplemented media, and shows reduced virulence and increased sensitivity in response to rapamycin and methyl methanesulfonate, but not in response to the osmotic stressors NaCl and sorbitol (Han et al., 2016).

The genes AN9181 and AN8970 are in separated clusters within the AN9181 orthogroup gene tree (Figure 2B). This observation is consistent with the fact that they possess distinct functional domains (Gabaldón and Koonin, 2013), and therefore, they are not paralogs. Though AN9181 was the initial candidate, we also evaluated the gene expression of AN8970 (same orthogroup) aiming to understand which one would be functionally relevant to regulate stress responses. Therefore, the expression profiles of either gene composing the AN9181 orthogroup (AN9181 and AN8970) were evaluated after 2, 4 and 24 h of exposure to four selected organic compounds in a medium containing a poor N source. The organic compounds comprise two which were found in the selected transcriptome datasets: benzo[a]pyrene (BaP), and bromoquinol (BMQ), and two additional aromatic halogenated based compounds, namely pentachlorophenol (PCP) and triclosan (TCS), classified as persistent organic pollutant and contaminant of emergent concern, respectively (Varela et al., 2017). The last two compounds were tested due to their frequent association to soil and water contamination (Czaplicka, 2004; Morgan et al., 2015), and past studies showing that either compound increased the production of virulent aspergilli conidia within soil colonizing fungi (Martins et al., 2023). The AN9181 revealed lower expression levels compared to AN8970 (Figures 2C,D). However, compared to control conditions, a significant increase in the expression levels of AN9181 was systematically noticed after 4 h of exposure to all the tested organic compounds, as well as in additional time-points, namely after 2 h to PCP, 2 and 24 h to TCS, and 24 h to BaP (Figure 2C). Recently, AN9181 was also found to be up-regulated after fungal growth in a nitrate minimal medium supplemented with either cadmium chloride, congo red or amphotericin B (Antal et al., 2020). In contrast to AN9181, the expression profiles of AN8970 were similar between the control and upon exposure to organic compounds, except for PCP at 24 h and for TCS 4 h (Figure 2D). At the experimental conditions used, the gene expression analysis did not support the idea of a concerted action of the two genes of the AN9181 orthogroup in A. nidulans. Based on these results, we focused the remaining analyzes on the AN9181.

To better understand the functional roles of AN9181, this gene was deleted and functionally analyzed in A. nidulans. The colony morphology of the parental strain and the A1145ΔAN9181 mutant (hereafter referred to as ΔAN9181) on Malt extract agar (MEA) medium were similar (Figures 3A,B). After five days of growth, the colony diameters of either strain in MMG containing a high amount of a non-preferred (i.e., poor) N source were also comparable (i.e., 71 mM sodium nitrate, Supplementary Figure 3). The conidia viability of either strain (i.e., germination fitness) was measured directly by counting the numbers of CFUs. CFUs were similar for both strains when germinated in the rich medium MEA (Figure 3C) and MMG supplemented with either a superior (i.e., rich) N source (10 mM of ammonium sulfate) (Figure 3D) or a non-preferred N source at high concentration (71 mM of sodium nitrate) (Supplementary Figure 4A). However, the ΔAN9181 conidial viability in MMG supplemented with 10 mM sodium nitrate or no added N source were 1.7-fold and 1.5-fold higher, respectively, compared to the parental strain (Figures 3E,F). This result suggests that AN9181 strongly influences the fitness of conidia germinating in medium having low availability of poor nitrogen sources.

The AN9181 encodes for a protein containing a NmrA-like domain. NmrA is a negative transcriptional regulator of several fungi, involved in the post-translational modulation of the GATA transcription factor AreA (Hensel et al., 1998). In A. nidulans, areA regulates the activation of genes involved in the utilization of a broad range of N sources: in the presence of rich N sources, e.g., ammonium and glutamine, NmrA binds to AreA preventing nitrogen catabolic gene expression; contrarily in the presence of nitrate, NmrA and AreA dissociation occurs, hence genes involved in the utilization of alternative nitrogen sources are activated (Han et al., 2016). Based on this, we questioned if AN9181 influences the utilization of different N sources. Specifically, we measured the cellular metabolic activity of both the ΔAN9181 and the parental strain in distinct media. Measurements were similar in both strains grown in media having rich N sources (i.e., MEA and ammonium sulfate) (Figure 3G) and a high availability of sodium nitrate (poor alternative N source) (Supplementary Figure 4B). However, the mutant showed a significant increase in metabolic activity when grown in media with a low availability of a poor N source (10 mM sodium nitrate) or no added N source (Figure 3G). Another mutant - #2ΔAN9181, randomly selected, was used for validation purposes. This mutant showed consistently similar metabolic activity in media having rich N sources and increased metabolic activity in media with low availability of a poor N source or no added N source (Supplementary Figure 5). In addition, the ΔAN9181 and the parental strain were tested in MMG supplemented with 10 mM N of each of the 20 proteinogenic amino acids. Compared to the parental strain, the mutant strain showed higher metabolic activity in tryptophan (aromatic); serine, glycine and methionine (serine family); asparagine (aspartate family); and alanine and valine (pyruvate family) (Figure 3H). Collectively the results suggest that AN9181 influences the utilization of N sources (hence also spore germination fitness) in a nitrogen type specific manner. Overall, the mutant phenotype suggests that AN9181 participates in the regulation of nitrogen catabolism in A. nidulans, resembling NmrA negative regulation of N utilization in many rich N sources. Further assays are however needed to better understand AN9181 regulatory network in the context of nitrogen utilization.

The growth of ΔAN9181 strain in the presence of each organic compound initially covered in the investigated transcriptome datasets was tested. To standardize conditions, all growth assays used MMG supplemented with the compounds under test at their determined EC₅₀ values (Supplementary Table 5). In the presence of BaP, BMQ and IDQ no differences in growth were observed between the parental and the mutant strain. On the contrary, in media supplemented with sodium salicylate or resveratrol, the ΔAN9181 strain grows more than the parental strain (Figure 4A). The cultivation conditions varied from those of the initial studies, including medium composition, time and temperature, as well as the concentration of the aromatic compounds. To test if the latter was influencing the phenotype, we tested the mutant’s growth in media supplemented with increasing concentrations of two selected aromatic compounds. Salicylate is degraded via the catechol branch of the 3-oxoadipate pathway in A. nidulans (Martins et al., 2015). The Ascomycota Phomopsis liquidambari degrades resveratrol into 3,5-dihydroxybenzaldehyde and 4-hydroxybenzaldehyde, which are subsequently oxidized to 3,5-dihydroxybenzoic acid and 4-hydroxybenzoic acid, respectively (Abo-Kadoum et al., 2022). The latter, is an intermediate of the protocatechuate branch of the 3-oxoadipate pathway used by A. nidulans to degrade benzoate. Therefore, it is possible that resveratrol and benzoate pathways are interconnected in A. nidulans, thus the phenotype was tested in increasing concentrations of sodium salicylate and sodium benzoate (the last instead of resveratrol). The results showed that the ΔAN9181 strain growth-phenotype is indeed concentration dependent for sodium salicylate and for sodium benzoate: a clear phenotype is noticed for concentrations >150 mM and > 55 mM, respectively (i.e., concentrations equal or above the EC₅₀ determined for these compounds; see Supplementary Table 5) (Figures 4B,C).

The absence of a phenotype in three (out of five) of the organic compounds herein tested (Figure 4A) is likely related to the fact that the used concentrations were below the threshold to cause major stress effect under the utilized cultivation conditions. Moreover, the response is also influenced by the regulation of nitrogen catabolism as analyzed above. The observation that the growth phenotype of the ΔAN9181 strain is sodium salicylate and sodium benzoate concentration dependent, obvious for concentrations above the corresponding EC₅₀ values, is consistent with the working hypothesis that AN9181 regulates stress responsive metabolism.

Recently, it was reported that AN9181 underwent up-regulation during A. nidulans exposure to congo red or amphotericin B grown in a nitrate minimal medium (Antal et al., 2020). This result expands the regulatory impact of AN9181 under growth in stress conditions. We evaluated if AN9181 deletion influences the susceptibility to congo red and to different antifungal drugs: amphotericin B (targets the cell membrane), caspofungin (targets the cell wall), and itraconazole (inhibits the synthesis of ergosterol, altering the cell membrane permeability). The results showed that ΔAN9181 strain susceptibility to amphotericin B is lower compared to the parental strain (Figure 5A). On the contrary, compared to the parental strain, the susceptibility of the mutant strain was similar and higher to caspofungin and itraconazole, respectively (Figure 5A). Similar results were attained with the A1149ΔAN9181 compared to its corresponding parental strain (Supplementary Figure 6A), notwithstanding that the magnitude of the susceptibility decrease to amphotericin B was more obvious. The determined MICs of each antifungal for the ΔAN9181 mutant strain and its parental strain validate these results (Table 2). Specifically, the MIC of amphotericin B for the ΔAN9181 mutant strain increase 1.4-fold compared to that of the parental strain (Table 2). The determined MICs of itraconazole and caspofungin were similar for the mutant and the parental strains (liquid media), regardless of the mutant’s higher susceptibility to itraconazole (solid media). Differences in the growth phenotypes between liquid and solid media have been reported before (Nichols et al., 2011; Martins et al., 2015). These differences can be related to changes in the drug bioavailability and also carbon availability. We also reassessed the growth phenotype of the strains in the presence of congo red (Figure 5A). A significant decrease in the susceptibility of ΔAN9181 mutant strain compared to the parental strain was noticed. The MIC for congo red in either strain is higher than 256 mg·L⁻¹, regardless that the upper inhibitory limit could not be precisely determined due to the strong red color of the media. All tests were conducted in the same growth media, hence the observed differential responses to the antifungal compounds cannot be simply explained by the regulation of nitrogen utilization by AN9181.

Additionally, the growth phenotype of the strains in the presence of H₂O₂ and in the presence of the superoxide menadione sodium bisulfite was assessed. The susceptibility of the ΔAN9181 mutant strain compared to the parental strain showed a major and minor decrease in the presence of H₂O₂ and MSB, respectively (Figure 5B). Using another mutant – #2ΔAN9181 – it was confirmed that the mutant susceptibility to H₂O₂ was indeed lower than that of the parental strain (Supplementary Figure 6B).

To complement this result, we evaluated the intracellular ROS levels of the parental and the ΔAN9181 mutant strains after exposure to specific compounds. The ROS intracellular levels increased upon exposure of either strain to 3 mM or higher concentrations of H₂O₂ (Figure 5C). ROS levels increased when the parental strain was exposed to either Amph B or RVT at a concentration where its growth was lower than that of the mutant but slight decreased when exposed to either BMQ and ITZ at concentrations where its growth was similar or higher than that of the mutant, respectively (Supplementary Figure 7). Finally, we observed that the expression of AN9181 in the parental strain underwent a 2-fold increase in the presence of H₂O₂ (≥ 1 mM) compared to the negative control; a clear indication that the AN9181 is upregulated under oxidative stress (Figure 5D). Collectively the results suggest that AN9181 participates indeed in stress responses upon exposure to different chemical agents leading to oxidative stress.