Glucose minimal medium (MMG) and MMG with 0.1% yeast extract (MMY) with appropriate supplements were used for general culture of A. fumigatus strains (Käfer, 1977). For pyrimidine and arginine auxotrophic mutant strain (AF293.6) (Xue et al., 2004), MMY was supplemented with 5 mM uridine, 10 mM uracil (for pyrG1), and 0.1% arginine (for argB1). For liquid submerged culture and phenotypic analyses on air-exposed culture were performed as described previously (Jung et al., 2016). To examine secondary metabolite production, spores of relevant strains were inoculated 50 ml of liquid MMY and incubated at 250 rpm at 37°C for 4 days.

The oligonucleotides used in this study are listed in Supplementary Table 1. The rgsC gene was deleted in A. fumigatus AF293.6 (pyrG1 argB1) strain (Xue et al., 2004). The deletion construct generated employing double-joint PCR (DJ-PCR) (Yu et al., 2004) containing the Aspergillus nidulans selective marker (AnargB⁺) with the 5′ and 3′ franking regions of the rgsC gene was introduced into the recipient strain AF293.6 (Szewczyk et al., 2006). The selective marker was amplified from A. nidulans FGSC4 genomic DNA with the primer pair oligo 214/oligo 215. The rgsC null mutant was isolated and confirmed by PCR, followed by restriction enzyme digestion (Supplementary Figure 1; Yu et al., 2004). To complement rgsC null mutant, a single joint PCR (SJ-PCR) method was used (Yu et al., 2004). The ORF of rgsC gene with a promoter and terminator was amplified with primer pairs where the 3′ reverse primer carries overlapping sequences with the ptrA gene's 5′ end. Amplification of the ptrA gene was carried out with primer pairs where the 5′ forward primer carries overlapping sequences with rgsC gene's 3′ end. The final amplicon was amplified with the nested primer pair oligo 781/oligo 731 and introduced into a ΔrgsC strain.

To isolate genomic DNA from A. fumigatus, about 10⁶ conidia were inoculated in 2 ml of liquid MMY, and stationary cultured at 37°C for 24 h. The mycelial mat was collected and squeeze-dried, and genomic DNA was isolated as described (Yu et al., 2004). The deletion mutant was confirmed by PCR amplification of the coding region of the gene followed by restriction enzyme digestion of the PCR amplicon. Total RNA isolation was carried out as previously described (Han et al., 2004a; Mah and Yu, 2006). Quantitative RT-PCR (qRT-PCR) assays were performed according to the manufacturer's instruction (Qiagen, USA) using a Rotor-Gene Q (Qiagen, USA). Each run was assayed in triplicate in a total volume of 20 μl containing the RNA template, One Step RT-PCR SYBR Mix (Doctor Protein, Korea), reverse transcriptase, and 10 pmole of each primer (Supplementary Table 1). Reverse transcription was performed at 42°C for 30 min. PCR conditions were 95°C/5 min for one cycle, followed by 95 and 55°C/30 s for 40 cycles. Amplification of one single specific target DNA was checked by melting curve analysis (+0.5°C ramping for 10 s, from 55 to 95°C). The expression ratios were normalized to EF1α expression and calculated according to the ΔΔCt method (Livak and Schmittgen, 2001).

Germination rates were measured as previously described with a slight modification (Ni et al., 2005). To examine germination levels, conidia of WT and mutant were inoculated in 5 ml of liquid MMY, or liquid medium lacking a carbon source, and incubated at 37°C. Levels of germination were examined every 2 h after inoculation under a microscope. Various media were used to assess the roles of RgsC in stress responses. For oxidative stress test, hydrogen peroxide (5 mM), menadione (100 μM), and paraquat (100 μM) were added to the YG media after autoclaving. To assess cell wall stress, Congo red (100 μg/ml), calcofluor white (50 μg/ml), caspofungin (0.1 μg/ml) were added to the YG media after autoclaving. The production of gliotoxin (GT) was determined as described previously (Bok and Keller, 2004). The chloroform extracts were air-dried and resuspended in 100 ml of methanol. Ten micro liter aliquots of each sample were applied to a thin-layer chromatography (TLC) silica plate (Kiesel gel 60, E. Merck). The TLC plate was developed with toluene:ethyl acetae:formic acid (5:4:1, v/v/v) and GT standard was purchased from Sigma (USA). To test alternative nitrogen sources, MMG with nitrogen free salts was used as the base medium. Three grams per liter peptone and yeast extract, 6.0 g/l NaNO₃, or 8.13 g/l proline was added, and these media were compared to MMG (NH₄Cl).

For catalase and superoxide dismutase (SOD) activity assays, protein was extracted as previous method (Jung et al., 2016). Catalase activity on gels was detected by ferricyanide-negative stain (Wayne and Diaz, 1986) and SOD activity was visualized by inhibition of the reduction of nitro blue tetrazolium (NBT, Sigma) according to the method of Beauchamp and Fridovich (1971).

The insect survival assay was performed as previously described with some modifications (Fuchs et al., 2010; Jung et al., 2016). Briefly, sixth instar Galleria mellonella were infected by injecting the fresh conidia (1 × 10⁵) into the last left pro-leg and incubated at 37°C in the dark for the duration of the experiment. Larvae were checked daily for survival and Kaplan-Meier survival curves were analyzed using the Log-Rank (Mantel-Cox) test for significance (p < 0.01).

The synthesis of target cDNA probes and hybridization were performed using Agilent's Low Input Quick Amp WT Labeling Kit (Agilent Technology, USA) according to the manufacturer's instructions. Briefly, 100 ng total RNA was mixed with WT primer mix and incubated at 65°C for 10 min. cDNA master mix (5 × First strand buffer, 0.1 M DTT, 10 mM dNTP mix, RNase-Out, and MMLV-RT) was prepared and added to the reaction mixture. The samples were incubated at 40°C for 2 h, and then the RT and dsDNA synthesis reactions were terminated by incubating at 70°C for 10 min. The transcription master mix was prepared as directed by the manufacturer's protocol (4 × Transcription buffer, 0.1 M DTT, NTP mix, 50% PEG, RNase-Out, inorganic pyrophosphatase, T7-RNA polymerase, and Cyanine 3/5-CTP). Transcription of dsDNA was performed by adding the transcription master mix to the dsDNA reaction samples and incubating at 40°C for 2 h. Amplified and labeled cRNA was purified and labeled cRNA target was quantified. After checking labeling efficiency, each of cyanine 3-labeled and cyanine 5-labeled cRNA target were mixed, and fragmentation of cRNA was performed by adding 10 × blocking agent and 25 × fragmentation buffer and incubating at 60°C for 30 min. The fragmented cRNA was resuspended with 2 × hybridization buffer and directly pipetted onto assembled MYcroarray.com (A. fumigatus AF293) 30 K Microarray. The arrays hybridized at 57°C for 17 h using an Agilent Hybridization oven (Agilent Technology, USA). The hybridized microarrays were washed as per the manufacturer's washing protocol (Agilent Technology, USA). Hybridization images were analyzed by an Agilent DNA microarray Scanner (Agilent Technology, USA), and the data quantification was performed using Agilent Feature Extraction software 10.7 (Agilent Technology, USA). The average fluorescence intensity for each spot was calculated and local background was subtracted using Gene Pix Pro 6.0 (Axon Instruments, USA). Loess normalization and selection of fold-changed genes were performed using GenoWiz 4.0 (Ocimum biosolutions, India). The data is available in the Gene Expression Omnibus (GEO) at NCBI (the accession number is GSE83200).

The ORF of rgsC of A. fumigatus (AFUA_1G09040) consists of 3,718 bp nucleotides with 1 intron, predicted to encode a 1,216 aa length protein. As shown in Figure 1A, the domain structure of RgsC contains a transmembrane (31–53 aa), PXA (101–290 aa, E-value; 3.34e-27), RGS (419–556 aa, E-value; 8.34e-16), 4 low complexity, and 1 PX domain (863–976 aa, E-value; 2.78e-24). With these protein sequences, we further identified additional RgsC-like proteins in other fungi and carried phylogenetic analyses (Figure 1B). As presented, the A. fumigatus RgsC is closely related to that of Aspergillus fischeri, Aspergillus clavatus, Aspergillus nomius, Aspergillus flavus, Aspergillus niger, A. nidulans, and Penicillium spp., but phylogenetically distinct from RgsC of dimorphic fungi and dermatophytic fungi (Figure 1B). To characterize the rgsC gene, levels of rgsC mRNA at different time points in the life cycle were examined, and found to be low during the early vegetative growth and increased at the later phase of vegetative growth (Figure 1C).

To characterize functions of rgsC, we generated the ΔrgsC mutant by replacing its ORF with the A. nidulans argB+ marker and we also generated complemented strains (C′) via re-introducing the wild type (WT) allele of rgsC to a deletion strain. Multiple ΔrgsC and C′ strains displaying identical phenotypes were isolated and further examined. When inoculated on solid medium, the rgsC deletion mutant formed a very distinct colony. The color of colony was very faint except a center region and the reverse side of colony was also light compared to WT and C' strains (Figure 2A). Moreover, whereas the colony edge of WT and C′ strains showed abundant conidiophores, the ΔrgsC mutant exhibited a very few number of conidiophores (Figure 2A, right panels). Conidia per plate further demonstrated that asexual spore production in the ΔrgsC mutant (1.8 × 10⁸ conidia/plate) was significantly decreased (p < 0.05) to a level that was only about 70% of WT and C′ strains (Figure 2B). Another noticeable change was that, the deletion of rgsC resulted in a significant reduction (about 85% of WT) in radial colony growth (Figure 2C). Further examination of mRNA levels of key asexual developmental regulators, abaA, brlA, vosA, and wetA in WT and ΔrgsC strains revealed a significantly reduced accumulation of these key developmental activators by the absence of rgsC (Figure 2D). As shown in Figure 2D, accumulation of abaA and wetA mRNAs increased from 12 h, peaked at 36 h, and decreased after 48 h post developmental induction in WT. Accumulation of brlA mRNA increased from 6 h, peaked at 12 h, and decreased after 24 h post developmental induction. However, the deletion of rgsC resulted in significantly low levels (p < 0.05) of these mRNAs at almost all times tested (Figure 2D). These results suggest that RgsC is necessary for proper growth and development in A. fumigatus.

G protein signaling plays a positive role in spore germination (Fillinger et al., 2002; Lafon et al., 2005). If RgsC attenuates a G-protein signaling pathway activating germination, the absence of RgsC may result in elevated spore germination. To test this, we first inoculated conidia of WT, ΔrgsC mutant, and C′ strains in liquid MMY and analyzed the kinetics of germ tube emergence. As shown in Figure 3A, WT and C′ strains exhibited about 30% conidial germination at 8 h and near 100% germination at 14 h in liquid submerged culture. On the other hand, the ΔrgsC strain showed about 40% conidial germination at 8 h and near 100% germination at 12 h in liquid medium. To test further, we examined germination rates in the absence of external carbon source by inoculating conidia of WT, ΔrgsC, and C′ strains in liquid MMY (without glucose). After 16 h inoculation, whereas only about 10% of WT and C′ conidia showed germling formation, 40% of the ΔrgsC conidia germinated (Figure 3A), suggesting that RgsC may negatively regulate conidial germination potentially sensing the external carbon source. To investigate whether RgsC mediates sensing of carbon sources, germination of the ΔrgsC mutant conidia in comparison to that of WT and C′ strain conidia were monitored in the presence of various carbon sources. The germination rate was significantly elevated in the rgsC null mutant in all but glucose medium (Figure 3B), suggesting that RgsC is necessary for the proper control of spore germination in response to varying carbon sources.

To evaluate functions of RgsC in oxidative stress response, we incubated WT, ΔrgsC, and C′ strains in the presence of H₂O₂ and the reactive oxygen species (ROS) generating compounds menadione (MD) and paraquat (PQ). As shown in Figure 4A, while the ΔrgsC mutant was hypersensitive to MD, it exhibited a slightly reduced tolerance to H₂O₂ and PQ. To further investigate the role RgsC, we analyzed activities of the ROS detoxifying enzymes catalase and SOD. Activities of both conidia-specific (CatA) and mycelia-specific (Cat1) catalases were decreased about 5 to 10-fold in the ΔrgsC mutant compared to those of WT and C′ strains (Figure 4B). In A. fumigatus, four genes encoding SODs have been identified and SOD1 and SOD2 were shown to play a major role to detoxify intracellular superoxide anions (Lambou et al., 2010). As catalases activities, activities of SOD1 and SOD2 in the ΔrgsC mutant was only 60 and 20% of WT strain, respectively, suggesting that the reduced tolerance of the ΔrgsC mutant to oxidative stresses could be due to low detoxifying enzymes activities. We then examined whether the absence of rgsC affected mRNA levels of catalases and SODs. We found that mRNA levels of catA, sod1, and sod2 were significantly decreased (p < 0.05) in the ΔrgsC mutant (Figure 4C). Taken together, these results suggest that RgsC is needed for protection against external oxidative stresses.

To examine whether RgsC mediate cell wall stress response, the mutant were exposed to a variety of cell wall damaging compounds including Congo red (100 μg/ml), calcofluor white (50 μg/ml), and caspofungin (0.1 μg/ml). Growth of the ΔrgsC mutant was slightly inhibited by the tested compounds compared to that of WT and C′ strain (Figure 5A), suggesting cell wall biosynthesis and/or integrity may be affected by the deletion of rgsC. We further analyzed mRNA levels of the MADS-box transcription factor rlmA and APSES transcription factor swi4 and swi6 which regulate genes involved in cell wall integrity and biogenesis (Kim et al., 2010; Rocha et al., 2016). In the ΔrgsC mutant it appears that levels of all tested genes' mRNA decreased after 48 h compared to WT (Figure 5B). These results indicate that RgsC is associated with proper cell wall biogenesis and cell wall integrity.

In A. fumigatus, the gliotoxin (GT) production is partially regulated by the asexual developmental activator BrlA (Shin et al., 2015). The deletion of RGSs including FlbA, GprK, and Rax1 resulted in lowered brlA expression and GT production (Mah and Yu, 2006; Jung et al., 2016; Igbalajobi et al., 2017). As the deletion of rgsC resulted in significantly defective conidiation and brlA mRNA levels (Figure 2), we examined levels of GT by TLC. As shown in Figure 6A, the ΔrgsC mutant produced undetectable levels of GT. Then we analyzed mRNA levels of several key GT biosynthetic genes by qRT-PCR using total RNA of WT, mutant, and C' strains. The mRNA levels of the gliM, gliT, and gliZ genes were significantly lower (p < 0.05) in the ΔrgsC mutant than in WT and C′ strains (Figure 6B). We next examined the effect of RgsC on virulence using the G. mellonella larvae survival test. Conidia of WT, ΔrgsC, and C′ strains were inoculated in G. mellonella larvae, and the larvae survival rates were recorded as a function of time. The virulence of ΔrgsC strains in wax moth was significantly reduced compared to WT and C′ strains (Figure 6C). The Log-Rank test revealed that the survival curves of WT and ΔrgsC were significantly different (p < 0.002). These results indicate an important role of RgsC in proper production of GT and the virulence of the fungus.

To obtain a more comprehensive insight into the RgsC mediated processes in A. fumigatus, we performed microarray analysis using ΔrgsC and WT cells collected at 12 h post asexual-developmental induction. Two biological replicates showed a high level of correlation (r = 0.885, Figure 7A). As shown in Figure 7B, the hierarchical clustering heat map based on transcriptome analysis showed that a majority of genes are down-regulated in ΔrgsC strain compared to WT. Of the 8,608 probes, 384 genes (4.5%) showed at least 1.5-fold (p < 0.05) differentially expressed, in which 82 genes (1.0%) were up-regulated and 302 genes (3.5%) were down-regulated (Supplementary Table 2). Table 1 lists the genes with increase in expression at least 2.0-fold (p < 0.01) following the deletion of rgsC. The highest up-regulated gene was predicted to encode a conserved hypothetical protein (AFUA_8G06430), with salicylate hydroxylase (AFUA_2G00770) identified as the up-regulated known gene with the maximum fold change in mutant relative to WT. The existence of transcripts corresponding to the conserved hypothetical proteins were first confirmed by qRT-PCR on the same RNA used for the microarray library construction (data not shown). Most of the down-regulated genes were related to nitrogen transport (Table 2), including small oligopeptide transporter (AFUA_2G15240), high affinity nitrate transporter NrtB (AFUA_1G17470), MFS peptide transporter (AFUA_1G12240), ammonium transporter MeaA (AFUA_2G05880), and nitrate transporter CrnA (AFUA_1G12850). These findings led us to test a role for the RgsC in nitrogen source sensing. The ΔrgsC mutant was grown on a variety of nitrogen sources such as, NaNO₃, NH₄Cl, peptone, proline, and yeast extract. As shown in Figure 8, while there were no differences in growth on organic nitrogen sources, growth of the ΔrgsC mutant was significantly (p < 0.01) restricted with NaNO₃, NH₄Cl, and proline as a nitrogen source, suggesting that RgsC may play a role in inorganic and simple nitrogen sensing.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. The reviewer JM and handling Editor declared their shared affiliation.