RgsA Attenuates the PKA Signaling, Stress Response, and Virulence in the Human Opportunistic Pathogen Aspergillus fumigatus

The regulator of G-protein signaling (RGS) proteins play an important role in upstream control of heterotrimeric G-protein signaling pathways. In the genome of the human opportunistic pathogenic fungus Aspergillus fumigatus, six RGS protein-encoding genes are present. To characterize the rgsA gene predicted to encode a protein with an RGS domain, we generated an rgsA null mutant and observed the phenotypes of the mutant. The deletion (Δ) of rgsA resulted in increased radial growth and enhanced asexual sporulation in both solid and liquid culture conditions. Accordingly, transcripts levels of the key asexual developmental regulators abaA, brlA, and wetA are elevated in the ΔrgsA mutant. Moreover, ΔrgsA resulted in elevated spore germination rates in the absence of a carbon source. The activity of cAMP-dependent protein kinase A (PKA) and mRNA levels of genes encoding PKA signaling elements are elevated by ΔrgsA. In addition, mRNA levels of genes associated with stress-response signaling increased with the lack of rgsA, and the ΔrgsA spores showed enhanced tolerance against oxidative stressors. Comparative transcriptomic analyses revealed that the ΔrgsA mutant showed higher mRNA levels of gliotoxin (GT) biosynthetic genes. Accordingly, the rgsA null mutant exhibited increased production of GT and elevated virulence in the mouse. Conversely, the majority of genes encoding glucan degrading enzymes were down-regulated by ΔrgsA, and endoglucanase activities were reduced. In summary, RgsA plays multiple roles, governing growth, development, stress responses, virulence, and external polymer degradation—likely by attenuating PKA signaling.


Introduction
Aspergillus fumigatus is a widespread saprophytic fungus in nature, typically found in soil and decaying vegetation [1,2]. This ubiquitous fungus is the most prevalent airborne fungal pathogen, causing a multitude of diseases in humans, such as allergic bronchopulmonary aspergillosis, aspergilloma, and invasive aspergillosis [3]. Most importantly, as an opportunistic human pathogen, A. fumigatus can cause a serious invasive pulmonary aspergillosis leading to an approximately 90%

RgsA Attenuates Hyphal Growth and Asexual Development
To characterize the function of rgsA in asexual development, we used the DJ-PCR method to generate the ∆rgsA mutant by replacing the ORF with the A. nidulans pyrG + marker [21]. As shown in Figure 1A-D, the radial growth of the ∆rgsA mutant was significantly higher than that of WT and complemented (C ) strains. Conidia per growth area further indicated that conidia production in the ∆rgsA mutant (1.47 × 10 8 conidia/cm 2 ) was increased (p < 0.01) to about 160% of the WT strain ( Figure 1D). Even in the stationary culture, asexual sporulation of the mutant was rapid and increased in comparison with that of WT and C strains ( Figure 1B). We then performed quantitative RT-PCR (qRT-PCR) for examination of the mRNA levels of the key asexual developmental regulators brlA, abaA, and wetA in WT and ∆rgsA mutant strains. As shown in Figure 1E, mRNA levels were significantly higher in the ∆rgsA mutant than that of WT, especially at 24 h post developmental induction. These data suggest that RgsA is needed for adequate control of growth and development in A. fumigatus.  (B) Growth of three strains on liquid MMY stationary culture at indicated time. (C) Colony diameters of WT (black bar), ∆rgsA (white bar), and C (gray bar) strains. (D) Conidia numbers produced by each strain per growth area. (E) mRNA levels of the key asexual developmental regulators in the ∆rgsA mutant relative to WT at 24 (white bar) and 48 h (gray bar) determined by qRT-PCR. Fungal cultures were grown in liquid MMY and mRNA levels were normalized using the ef1α gene. Data are presented as the means ± standard deviations from three independent experiments. ANOVA test: ** p < 0.01.

RgsA Down-Regulates a cAMP-Dependent Protein Kinase A Signaling Pathway
A previous study demonstrated that the absence of rgsA resulted in elevated conidial germination in A. nidulans, with or without an external carbon source [20]. To investigate a potential role of RgsA in controlling spore germination, we monitored the kinetics of germ tube emergence in the ∆rgsA mutant in comparison to that of WT every 2 h after inoculation. As shown in Figure 2A, conidia of WT and ∆rgsA strains began to germinate after 6 h of incubation and showed no significant differences in the presence of glucose. To test further, conidia of WT and ∆rgsA strains were inoculated in a no carbon source medium and observed for germination. Both strains started form germ tubes, about 2% and 3%, respectively at 6 h. At 12 h of incubation, while only about 10% of WT conidia were germinated, 35% of the ∆rgsA conidia germinated (Figure 2A), indicating that RgsA may negatively regulate conidial germination, potentially sensing the external carbon source. To investigate the relationship between RgsA and a cAMP-PKA signaling pathway, we assessed PKA's activity using a peptide substrate, kemptide. While all the strains we tested exhibited similar levels of PKA activity in the presence of cAMP, the ∆rgsA mutant showed PKA activity even in the absence of cAMP ( Figure 2B). To investigate this further, we analyzed mRNA levels of PKA pathway related genes. As shown in Figure 2C, acyA, and pkaC1 mRNA levels were significantly higher in the ∆rgsA mutant than in WT strain. These results indicate that RgsA is required for properly controlling the expressions of acyA and pkaC1, and may negatively regulate a cAMP-PKA signaling pathway.

RgsA Down-Regulates a cAMP-Dependent Protein Kinase A Signaling Pathway
A previous study demonstrated that the absence of rgsA resulted in elevated conidial germination in A. nidulans, with or without an external carbon source [20]. To investigate a potential role of RgsA in controlling spore germination, we monitored the kinetics of germ tube emergence in the ΔrgsA mutant in comparison to that of WT every 2 h after inoculation. As shown in Figure 2A, conidia of WT and ΔrgsA strains began to germinate after 6 h of incubation and showed no significant differences in the presence of glucose. To test further, conidia of WT and ΔrgsA strains were inoculated in a no carbon source medium and observed for germination. Both strains started form germ tubes, about 2% and 3%, respectively at 6 h. At 12 h of incubation, while only about 10% of WT conidia were germinated, 35% of the ΔrgsA conidia germinated (Figure 2A), indicating that RgsA may negatively regulate conidial germination, potentially sensing the external carbon source. To investigate the relationship between RgsA and a cAMP-PKA signaling pathway, we assessed PKA's activity using a peptide substrate, kemptide. While all the strains we tested exhibited similar levels of PKA activity in the presence of cAMP, the ΔrgsA mutant showed PKA activity even in the absence of cAMP ( Figure 2B). To investigate this further, we analyzed mRNA levels of PKA pathway related genes. As shown in Figure 2C, acyA, and pkaC1 mRNA levels were significantly higher in the ΔrgsA mutant than in WT strain. These results indicate that RgsA is required for properly controlling the expressions of acyA and pkaC1, and may negatively regulate a cAMP-PKA signaling pathway. , ∆rgsA, and complemented (C ) strains, as monitored by gel electrophoresis. A phosphorylated substrate migrates toward the + anode. Each strain was grown in MMG for 3 days at 37 • C and mycelial extract was analyzed. (C) Levels of acyA, pkaC1, and pksP mRNA in three strains analyzed by qRT-PCR. * p < 0.05.

Differential Roles of RgsA in Oxidative Stress Response
To test the potential role of RgsA in the oxidative stress response, we incubated WT, ∆rgsA, and C strains on solid MMY containing paraquat (PQ) or H 2 O 2 . As shown in Figure 3A, the ∆rgsA mutant exhibited a significantly increased tolerance to PQ and H 2 O 2 ( Figure 3B). To further dissect RgsA's function, we analyzed activities of catalase and SOD. In A. fumigatus, superoxide dismutases (SODs) encode four genes and Sod1p and Sod2p are concerned with the detoxification of the intracellular superoxide radical [22]. Sod1p is a cytoplasmic, cyanide-sensitive Cu/ZnSOD and Sod2p is a mitochondrial, cyanide-insensitive MnSOD [22]. The activity of Sod1p was slightly reduced in the ∆rgsA mutant, but Sod2p's activity was higher than that of WT strain ( Figure 3C), suggesting that the increased resistance of the ∆rgsA mutant to PQ might be associated with higher Sod2p activities. While the activity of conidia-specific catalase (CatAp) was increased about 1.7-fold, mycelia-specific catalase (Cat1p) was decreased about 1.5-fold in the ∆rgsA mutant compared to the WT and C strains ( Figure 3D). In addition, mRNA levels of genes encoding stress-activated signaling pathway elements were increased about 1.2 to 2.6-fold by ∆rgsA ( Figure 3E). Taken together, these results suggest that RgsA is needed for proper regulation of responses to external oxidative stresses. , ΔrgsA, and complemented (C′) strains, as monitored by gel electrophoresis. A phosphorylated substrate migrates toward the + anode. Each strain was grown in MMG for 3 days at 37 °C and mycelial extract was analyzed. (C) Levels of acyA, pkaC1, and pksP mRNA in three strains analyzed by qRT-PCR. * p < 0.05.

Differential Roles of RgsA in Oxidative Stress Response
To test the potential role of RgsA in the oxidative stress response, we incubated WT, ΔrgsA, and C′ strains on solid MMY containing paraquat (PQ) or H2O2. As shown in Figure 3A, the ΔrgsA mutant exhibited a significantly increased tolerance to PQ and H2O2 ( Figure 3B). To further dissect RgsA's function, we analyzed activities of catalase and SOD. In A. fumigatus, superoxide dismutases (SODs) encode four genes and Sod1p and Sod2p are concerned with the detoxification of the intracellular superoxide radical [22]. Sod1p is a cytoplasmic, cyanide-sensitive Cu/ZnSOD and Sod2p is a mitochondrial, cyanide-insensitive MnSOD [22]. The activity of Sod1p was slightly reduced in the ΔrgsA mutant, but Sod2p's activity was higher than that of WT strain ( Figure 3C), suggesting that the increased resistance of the ΔrgsA mutant to PQ might be associated with higher Sod2p activities. While the activity of conidia-specific catalase (CatAp) was increased about 1.7-fold, mycelia-specific catalase (Cat1p) was decreased about 1.5-fold in the ΔrgsA mutant compared to the WT and C′ strains ( Figure 3D). In addition, mRNA levels of genes encoding stress-activated signaling pathway elements were increased about 1.2 to 2.6-fold by ΔrgsA ( Figure 3E). Taken together, these results suggest that RgsA is needed for proper regulation of responses to external oxidative stresses.

RgsA Down-regulates GT Production and Virulence
Gliotoxin is an important determinant of the virulence of A. fumigatus [23]. Among the putative GT biosynthetic 13-gene cluster in the genome of A. fumigatus [24,25], gliA, gliC, gliF, gliG, gliI, gliJ, gliM, gliN, gliP, and gliT have been identified as the up-regulated genes with the maximum fold changes in the mutant relative to WT. These findings led us to test a role for the RgsA in GT production per se. We examined levels of GT in WT and ∆rgsA strains by TLC, and found that the ∆rgsA mutant produced a significantly higher amount of GT than WT and C strains ( Figure 5A). Next, we tested the effects of RgsA on virulence using the outbred CD1 non-neutropenic mice that were generated by subcutaneous injection of cortisone acetate (10 mg/mouse). Conidia (1 × 10 7 ) of WT, ∆rgsA, and C strains were inoculated to the mouse, and the survival was recorded every 12 h. The most rapid lethality appeared with the ∆rgsA mutant and this strain was significantly more virulent than the other strains (p = 0.0006). As shown in Figure 5B, mice inoculated with conidia of either WT or C strains began to die at day 2.5 post-inoculation, with the survival rate continuing to decrease over the course of the experiment. Five days after infection, about 50% of mice in both strains died. However, the mortality rates caused by ∆rgsA strain were higher than those of WT and C strains: about 90% of mice had died in 4.5 days and all mice had died by 7.5 days after infection. Furthermore, the loss of rgsA significantly enhanced the pulmonary fungal burden of mice ( Figure 5C). To better understand the fate of A. fumigatus inoculated into mouse, the lung tissues of infected mice were fixed with formalin 72 h post-infection and sectioned for histopathology. Figure 5D shows Hematoxylin and Eosin (H&E) and periodic acid-Schiff (PAS) stained sections of infected mice with three strains. Histopathological examination of lung tissues from WT, ∆rgsA, and C strains of infected mice demonstrated that different mortality rates among the groups were associated with different pulmonary fungal burdens. While WT-strain infected mice revealed marked increases of neutrophil recruitment within the bronchi and the peribronchial region; mild necrosis; and the infiltration of numerous red blood cells into the parenchyma, ∆rgsA strain infected mice showed bronchial wall disruption, severe necrosis, scattered regional inflammation, and mild hemorrhaging in lung parenchyma ( Figure 5D). Next, to investigate whether more severe lung damage observed in mice infected with ∆rgsA could be attributed to the extent of fungal invasion, lung sections from three different groups were stained with PAS. WT-strain infected mice revealed the presence of the most conidia within bronchi and the peribronchial region, and rare hyphae growth was detectable. In contrast, lung sections from ∆rgsA-strain infected mice displayed numerous spore germinations and aggressive invasion of hyphae into parenchyma ( Figure 5D). Of note, C strain infected mice showed similar phenotypes to those of WT-strain infected mice, suggesting that RgsA is a negative regulator of fungal virulence.

RgsA Plays a Positive Role in Endoglucanase Activity
In RNA-seq analysis, we found that genes encoding numerous carbohydrate-metabolizing enzymes were significantly down-regulated (Log 2 FC = -2.07 to -3.96) in the ∆rgsA strain compared to WT (Supplementary Table S3; Figure 6A). Furthermore, most of those genes are predicted to encode cellulolytic enzymes, including cellobiohydrolase (AFUA_6G07070, AFUA_6G11610, and AFUA_3G01910), endo-1,4-β-glucanase (AFUA_3G03870 and AFUA_1G05290), β-glucosidase (AFUA_1G14710 and AFUA_1G05770), and cellobiose dehydrogenase (AFUA_2G17620). To test whether RgsA affects endoglucanase activity, we determined the enzyme activity by culturing three strains in liquid MMY containing 0.1% β-glucan. As shown in Figure 6B, both β-1,4 carboxymethyl cellulose (CMC) and β-1,3 (laminarin) endoglucanase activities were significantly reduced irrespective of location, except for the intracellular β-1,4-endoglucanase activity. Those results suggest that RgsA plays a positive role in the degradation of glucan and may play an important roles in saprophytic nutrition. suggest that RgsA plays a positive role in the degradation of glucan and may play an important roles in saprophytic nutrition.

Discussion
G-protein signaling is conserved in all eukaryotes, playing a key role in sensing and transmitting signals into cells to lead to appropriate responses [26]. RGS proteins constitute a diverse and multifunctional family of proteins that play an important role in controlling heterotrimeric G protein signaling [9,27]. In A. nidulans, the homologue of rgsA was identified as a specific RGS protein that negatively regulates GanB function. This RGS-Gα pair plays an important role in upstream regulation of asexual sporulation, germination, and stress responses to environmental changes [20]. In this study, the deletion of rgsA resulted in increased radial growth and conidiation and elevated mRNA levels of central asexual development regulators abaA, brlA, and wetA compared to the WT strain ( Figure 1). Based on all that, we can propose that RgsA is crucial for the proper control of growth and asexual development in A. fumigatus.
Previous studies described that A. nidulans RgsA negatively regulates GanB (G protein α subunit) and that the RgsA-GanB pair governs upstream regulation of spore germination and carbon source sensing, in part via the cAMP-PKA pathway [11,20,28]. It has been shown that activation of cAMP-PKA pathway in A. nidulans can induce germination by various carbon sources. In A. fumigatus, the ΔrgsA mutant and WT began to germinate after 4 h of incubation and the ΔrgsA mutant showed higher germination rates than those of WT at 8 h. After 12 h of incubation, both the ΔrgsA mutant and WT strains exhibited nearly 100% germination. In addition, we found that the germination rates of the ΔrgsA mutant conidia were significantly higher than those of WT strain in the absence of a carbon source. Furthermore, the ΔrgsA strain showed higher PKA activity in the absence of cAMP and mRNA levels of PKA signaling components. Collectively, these results imply that RgsA serves as an important negative, likely upstream regulator of the cAMP-PKA signaling pathway (Figure 2).

Discussion
G-protein signaling is conserved in all eukaryotes, playing a key role in sensing and transmitting signals into cells to lead to appropriate responses [26]. RGS proteins constitute a diverse and multifunctional family of proteins that play an important role in controlling heterotrimeric G protein signaling [9,27]. In A. nidulans, the homologue of rgsA was identified as a specific RGS protein that negatively regulates GanB function. This RGS-Gα pair plays an important role in upstream regulation of asexual sporulation, germination, and stress responses to environmental changes [20]. In this study, the deletion of rgsA resulted in increased radial growth and conidiation and elevated mRNA levels of central asexual development regulators abaA, brlA, and wetA compared to the WT strain ( Figure 1). Based on all that, we can propose that RgsA is crucial for the proper control of growth and asexual development in A. fumigatus.
Previous studies described that A. nidulans RgsA negatively regulates GanB (G protein α subunit) and that the RgsA-GanB pair governs upstream regulation of spore germination and carbon source sensing, in part via the cAMP-PKA pathway [11,20,28]. It has been shown that activation of cAMP-PKA pathway in A. nidulans can induce germination by various carbon sources. In A. fumigatus, the ∆rgsA mutant and WT began to germinate after 4 h of incubation and the ∆rgsA mutant showed higher germination rates than those of WT at 8 h. After 12 h of incubation, both the ∆rgsA mutant and WT strains exhibited nearly 100% germination. In addition, we found that the germination rates of the ∆rgsA mutant conidia were significantly higher than those of WT strain in the absence of a carbon source. Furthermore, the ∆rgsA strain showed higher PKA activity in the absence of cAMP and mRNA levels of PKA signaling components. Collectively, these results imply that RgsA serves as an important negative, likely upstream regulator of the cAMP-PKA signaling pathway (Figure 2).
The ability to sense external stress and respond appropriately is vital for the survival of fungal cells in stress conditions, and is a key factor for A. fumigatus to establish successful invasive aspergillosis.
A. fumigatus is exposed to variety of stressors in both the environment and the host. During infection, A. fumigatus encounters various microenvironments, which force the fungus to grow under different stress conditions during pathogenesis [29]. These stress conditions include temperature, oxidative stress, pH changes, and the low availability of macro and micro-nutrients [1,3,[30][31][32][33]. Our results show that RgsA is involved in the response to oxidative stress, since the rgsA deletion mutant was highly resistant to PQ and H 2 O 2 (Figure 3).
The human pathogen A. fumigatus produced secondary metabolites such as GT, fumagillin, fumitremorgin, gibberellin, and helvolic acid a [34]. GT is a major and the most potent toxin of this fungus, and a member of the epipolythiodioxopiperazine class of toxins. Transcriptome analysis revealed that 440 genes were significant differentially expressed by the loss of rgsA (at least 2.0-fold, p-value < 0.05) with 229 up-regulated and 211 downregulated genes ( Figure 4). Interestingly, most of the up-regulated genes in the ∆rgsA mutant were predicted to encode genes in the GT biosynthesis gene cluster (Supplementary Table S2). GliP is the first step in biosynthesis of gliotoxin (GT) and is essential for GT production and the absence of this gene resulted in GT deficient strains [23,[35][36][37]. GliA is a member of the major facilitator superfamily of transporters. The GT oxidoreductase GliT functions in the protection of the fungus against external GT and is indispensable for GT biosynthesis [24,38]. The gliI gene is predicted to encode aminotransferase and represents the first gene of a functionally proven C-S lyase involved in a secondary metabolite pathway [39]. Glutathione S-transferase GliG is not involved in the self-protection of A. fumigatus against exogenous gliotoxin [40]. The gliM and gliN genes are predicted to encode an O-methyltransferase and methyltransferase respectively [37]. GliJ, a putative dipeptidase, shows higher expression in biofilms and gliC is the predicted cytochrome P450 monooxygenase and acts upstream of gliG in the gliotoxin biosynthesis pathway [39]. The ∆rgsA mutant produced a significantly higher amount of GT than WT and C strains, and caused higher the mortality rates in mice. The lung sections of the ∆rgsA-strain infected mice displayed numerous spore germinations and the aggressive invasion of hyphae into parenchyma ( Figure 5), suggesting that RgsA down-regulates GT production and fungal virulence.
A. fumigatus has a saprophytic lifestyle in decaying organic and plant material, and this fungus has a numerous genes that encode a wide range of hydrolases with the capacity to degrade the major plant cell wall polymers [41]. The majority of down-regulated genes, by the loss of rgsA, are carbohydrate-metabolism-related genes (Supplementary Table S3; Figure 6A). The cellobiohydrolase celD involved in cellulose breakdown and its predicted gene pair with cbhB (AFUA_6G11610) are similar to cbhA and cbhB in A. nidulans (AN5176 and AN0494), which were induced by cellulose and repressed by glucose [42]. The endoglucanase activity of the mutant was lower than WT and C strains in β-glucan containing medium ( Figure 6B), indicating that RgsA is involved in the degradation of β-glucan and may play an important role in saprophytic nutrition.
In summary, we propose a genetic model depicting the regulatory roles of RgsA in A. fumigatus (Figure 7). In the absence of rgsA, the cAMP-PKA and stress-activated signaling pathway may be activated, which might lead to increased GT production, virulence, and tolerance to environmental stresses, whereas polysaccharide breakdown activity may be inhibited by the loss of rgsA. Further studies are needed to identify the targeted G protein α subunit(s) of RgsA and the potential downstream signaling pathways governing the stress responses and virulence of A. fumigatus.

Figure 7.
A model depicting the predicted roles of RgsA. RgsA negatively regulates the cAMP-PKA signaling pathway and stress-activated signaling pathway. This is predicted to be accomplished by RgsA's role in attenuating the target heterotrimeric G proteins. The lack of RgsA leads to the accelerated and prolonged activation of the target G proteins; thus, enhanced activation of the cAMP-PKA signaling pathway and stress-activated signaling pathway, which in turn leads to increased production of GT, enhanced virulence, and an elevated oxidative stress response. In contrast, polysaccharide breakdown activity may be inhibited by the lack of rgsA.

Fungal Strains and Culture Conditions
A. fumigatus AF293.1 (AfpyrG1) was used to generate the ΔrgsA mutant and AF293 was used as a wild type (WT). Fungal strains were grown on glucose minimal medium (MMG) or MMG with 0.1% yeast extract (MMY) with appropriate supplements, as described previously [43]. For liquidsubmerged culture, about 5 × 10 5 conidia/mL were inoculated into liquid MMY and incubated at 37 °C, 250 rpm. For phenotypic analyses of respective strains on air-exposed culture, conidia (1 × 10 5 ) of relevant strains were spotted on solid medium and incubated at 37 °C for 4 days.

Generation of the rgsA Null Mutant
The double-joint PCR method was used to generate the rgsA deletion mutant [21] and the oligonucleotides used in this study are listed in Supplementary Table S1. Briefly, using WT (AF293) genomic DNA as a template, 5′ and 3′-flaking regions of the rgsA ORF were amplified with the primer pairs of Oligo926:Oligo927 and Oligo928:Oligo929, respectively. The selective marker A. nidulans pyrG was amplified with Oligo697 and 698 using A. nidulans FGSC A4's genomic DNA as a template. The three amplicons were fused and the second round of PCR was performed. Using the second round PCR product as a template, the deletion construct was amplified with nested primer pairs Oligo930 and Oligo931, and the resulting PCR amplicon was introduced into AF293.1 via transformation. The rgsA deletion mutant colonies were isolated and confirmed by PCR followed by Figure 7. A model depicting the predicted roles of RgsA. RgsA negatively regulates the cAMP-PKA signaling pathway and stress-activated signaling pathway. This is predicted to be accomplished by RgsA's role in attenuating the target heterotrimeric G proteins. The lack of RgsA leads to the accelerated and prolonged activation of the target G proteins; thus, enhanced activation of the cAMP-PKA signaling pathway and stress-activated signaling pathway, which in turn leads to increased production of GT, enhanced virulence, and an elevated oxidative stress response. In contrast, polysaccharide breakdown activity may be inhibited by the lack of rgsA.

Fungal Strains and Culture Conditions
A. fumigatus AF293.1 (AfpyrG1) was used to generate the ∆rgsA mutant and AF293 was used as a wild type (WT). Fungal strains were grown on glucose minimal medium (MMG) or MMG with 0.1% yeast extract (MMY) with appropriate supplements, as described previously [43]. For liquid-submerged culture, about 5 × 10 5 conidia/mL were inoculated into liquid MMY and incubated at 37 • C, 250 rpm. For phenotypic analyses of respective strains on air-exposed culture, conidia (1 × 10 5 ) of relevant strains were spotted on solid medium and incubated at 37 • C for 4 days.

Generation of the rgsA Null Mutant
The double-joint PCR method was used to generate the rgsA deletion mutant [21] and the oligonucleotides used in this study are listed in Supplementary Table S1. Briefly, using WT (AF293) genomic DNA as a template, 5 and 3 -flaking regions of the rgsA ORF were amplified with the primer pairs of Oligo926:Oligo927 and Oligo928:Oligo929, respectively. The selective marker A. nidulans pyrG was amplified with Oligo697 and 698 using A. nidulans FGSC A4's genomic DNA as a template. The three amplicons were fused and the second round of PCR was performed. Using the second round PCR product as a template, the deletion construct was amplified with nested primer pairs Oligo930 and Oligo931, and the resulting PCR amplicon was introduced into AF293.1 via transformation. The rgsA deletion mutant colonies were isolated and confirmed by PCR followed by SmaI digestion (Supplementary Figure S1). To complement the rgsA null mutant, the single joint PCR (SJ-PCR) method was used [21]. The ORF of rgsA with a promoter and a terminator was amplified with primer pairs; the 3 reverse primer carried overlapping sequences with the ptrA gene's 5 end. Amplification of the ptrA gene was carried out with primer pairs; the 5 forward primer carried overlapping sequences with ptrA gene's 3 end. The final amplicon was amplified with the nested primer pair oligo 930/oligo 931 and introduced into a ∆rgsA strain (Supplementary Table S1).

Nucleic Acid Isolation and Manipulation
Total RNA was isolated using Trizol reagent (Invitrogen, USA). RNA quality was assessed by Agilent 2100 bioanalyzer using the RNA 6000 Nano Chip (Agilent Technologies, Amstelveen, The Netherlands), and RNA quantification was performed using ND-2000 Spectrophotometer (Thermo Inc., USA). Levels of mRNA of select genes were analyzed with appropriate oligonucleotide pairs (Supplementary Table S1). For RNA-seq analyses, 3 day-old cultures of WT and ∆rgsA strains were harvested from solid MMY. Quantitative RT-PCR assays were performed as previously described [13,44]. Briefly, conidia (5 × 10 5 conidia/mL) of three strains were inoculated into liquid MMY with appropriate supplements and incubated at 37 • C, 250 rpm. Individual mycelial samples, which were collected at designated time points, were homogenized using a Mini Bead beater in the presence of 1 mL of TRIzol ® reagent (Invitrogen, USA) and 0.3 mL of silica/zirconium beads (BioSpec Products, USA). QRT-PCRs were performed according to the manufacturer's instructions 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 S1). Reverse transcription was carried out at 42 • C for 30 min. PCR conditions were 95 • C/5 min for one cycle, followed by 95 • C and 55 • C/30 s for 40 cycles. Amplification of one specific target DNA was checked by melting curve analysis. The expression ratios were normalized to ef1α expression and calculated according to the ∆∆Ct method [44].

Phenotypic Analyses
Germination rates were determined with a previously-established method with some modifications [45]. The conidia of WT and ∆rgsA were inoculated in 5 mL of MMY, or media without a carbon source, and incubated at 37 • C. Levels of both isotopic growth and germ tubing were examined every 2 h after inoculation under a microscope. To assess the role of RgsA in stress responses, WT and ∆rgsA strains were grown on various media, the base being MMG as described above. To test for oxidative stress, PQ (200 µM) and hydrogen peroxide (2 mM) were added to MMG after autoclaving. All experiments were performed in triplicate. To determine the production of gliotoxin (GT), 1 × 10 7 conidia/mL of each strain were inoculated into 5 mL complete medium in a test tube and incubated at 37 • C slanted for 7 days. GT was extracted with chloroform as described previously [46]. Individual chloroform extracts were air-dried and resuspended in 100 µL of methanol. Aliquots (10 µL) of each sample were applied to a TLC silica plate containing a fluorescence indicator (Kiesel gel 60, E. Merck, USA). The TLC plate was developed with chloroform:methanol (9:1, v/v).

Enzyme Assay
To measure the PKA activity, the Non-Radioactive cAMP-Dependent Protein Kinase Assay kit (Promega, USA) was used. Homogenized mycelia of each strain were suspended in extraction buffer [47] and incubated in ice for 15 min. After centrifugation, 10 µL of supernatant (3 mg/mL protein) was used for determining PKA activity. To determine catalase and SOD activities, the conidia (1 × 10 5 ) of relevant strains were inoculated into liquid MMY with appropriate supplements and incubated at 37 • C, 250 rpm, for 24 h. Then, oxidative stress agents (200 µM PQ or 2 mM H 2 O 2 ) were added, and the conidia were further incubated for 24 h. The mycelia were disrupted with glass beads in 20 mM pH 7.5 phosphate buffer supplemented with a protease inhibitor cocktail. Protein contents were quantified with the Bradford reagent (Bio-Rad, USA), using bovine serum albumin as the standard.
Catalase activity was visualized by negative stain with ferricyanide [48] and SOD activity was detected as inhibition of the reduction of NBT [49]. For testing endoglucanase activity, fungal strains were grown on MMY with 1% barley β-glucan which contained both β-1,3 and β-1,4-glucan. The enzyme activities of culture filtrates and mycelial extracts were measured by DNS method [50] using CMC (for β-1,4-glucanase) and laminarin (for β-1,3-glucanase) as the substrates. One unit of endoglucanase activity was defined as the amount of enzyme that liberated reducing sugars equivalent to 1 µmol of glucose under the assay conditions.

Murine Virulence Assay
For the immunocompromised mouse model, we used outbred CrlOri:CD1 (ICR) (Orient Bio Inc, Korea) female mice (30 g in body weight, 6 to 8 weeks old), which were housed five per cage and had access to food and water ad libitum. Mice were immunosuppressed with subcutaneous injections of cortisone acetate (Sigma-Aldrich, USA) at 10 mg/mouse for 4 days prior to infection and cortisone acetate injected subcutaneously, 10 mg/mouse, 2 days prior to infection. At days 0, 3, and 6 post-infection, administrations were repeated with cortisone acetate (10 mg/mouse). For conidia inoculation, mice were anesthetized with isoflurane, and then intranasally infected with 1 X 10 7 conidia of A. fumigatus strains (10 mice per fungal strain) in 30 µL of 0.01% Tween 80 in PBS. Mice were monitored every 12 h for survival for 8 days after the challenge. Mock mice included in all experiments were inoculated with sterile 0.01% Tween 80 in PBS. Mice were checked every 12 h for survival and Kaplan-Meier survival curves were analyzed using the log-rank (Mantel-Cox) test for significance (p < 0.01).

Fungal Burden in Mice
Fungal burdens were obtained by the qPCR approach from a previous study [51] with A. fumigatus-specific primers [52]. Briefly, a complete lung from each mouse was lyophilized and ground to a fine powder. Approximately 50 mg of each powdered lung was used for DNA extraction. The amounts of DNA from the lung tissues were quantified by spectrophotometer and all samples were diluted to 100 ng/µL. Standard C T -versus-concentration plots were constructed using A. fumigatus DNA (20 to 0.2 ng per reaction, R = 0.992) and the concentration of fungal DNA in each sample was calculated.

Ethics Statement
All of the animal procedures in this study were reviewed and approved by the Institutional Animal Care and Use Committee of Daejeon University (DJUARB2019-024).

RNA-seq Experiment and Analyses
Construction of the RNA-seq library was performed using QuantSeq 3 mRNA-Seq Library Prep Kit (Lexogen, Inc., Austria) according to the manufacturer's instructions. In brief, each 500 ng total RNA was prepared and an oligo-dT primer containing an Illumina-compatible sequence at its 5 end was hybridized to the RNA, and reverse transcription was performed. After degradation of the RNA template, second strand synthesis was done by a random primer containing an Illumina-compatible linker sequence at its 5 end. The double-stranded library was purified by using magnetic beads and amplified to add the complete adapter sequences required for cluster generation. The finished library was purified from PCR components. High-throughput sequencing was performed as single-end 75 sequencing using NextSeq 500 (Illumina, Inc., San Diego, CA, USA). QuantSeq 3 mRNA-Seq reads were aligned using Bowtie2 [53]. Bowtie2 indices were either generated from genome assembly sequences or the representative transcript sequences for aligning the genome and transcriptome. The alignment file was used to assemble transcripts, estimating their abundances and detecting the differential expressions of genes. Differentially-expressed genes were determined based on counts from unique and multiple alignments using coverage in Bedtools [54]. The RT (read count) data were processed based on Quantile normalization method using EdgeR within R using Bioconductor [55]. Gene classification was based on searches done by DAVID (http://david.abcc.ncifcrf.gov/) and Medline databases (http://www.ncbi.nlm.nih.gov/).

Data Analysis
Comparisons among groups of two and larger groups were performed by the unpaired Student's t-test and one-way ANOVA, respectively. GraphPad Prism 4 (GraphPad Software, Inc., San Diego, CA, USA) was used for the statistical analyses and graphical presentation of survival curve. p < 0.05 was considered to be statistically significant.

Data Availability
The RNA-seq data are available from NCBI Gene Expression Omnibus (GEO) database (the accession number is GSE123744).

Conclusions
Canonical regulator of G protein signaling (RGS) proteins accelerate the hydrolysis of GTP bound to Gα, leading the formation of the inactive hetero-trimer and turning off G protein signaling [9]. A better understanding of individual RGS proteins in A. fumigatus may lead to the identification of novel antifungal drug targets and the control of human-pathogenic fungi.
Our study identified that RgsA can negatively regulate the cAMP-PKA and stress-activated kinase signaling pathways, which leads to a disturbed lifecycle; i.e., enhanced asexual development, increased GT production, increased virulence, and a higher oxidative stress response. However, the precise mechanisms of RgsA-mediated regulation of hetero-trimeric G protein signaling and downstream signaling pathways are not understand yet. To fully understand the function of RgsA in A. fumigatus, protein-protein interactions between RgsA and G proteins can be studied to identify the targets, and the outcomes of those studies may provide an expanded model for understanding G protein signaling in other human-pathogenic fungi.  Table S1. Oligonucleotides used in this study. Supplementary Table S2. Up-regulated genes in ∆rgsA relative to WT (> 2.0-fold, p < 0.05). Supplementary Table S3. Down-regulated genes in ∆rgsA relative to WT (> 2.0-fold, p < 0.05).