Transcriptomic and Functional Studies of the RGS Protein Rax1 in Aspergillus fumigatus

In the comparative transcriptomic studies of wild type (WT) and rax1 null mutant strains, we obtained an average of 22,222,727 reads of 101 bp per sample and found that 183 genes showed greater than 2.0-fold differential expression, where 92 and 91 genes were up-and down-regulated in Δrax1 compared to WT, respectively. In accordance with the significantly reduced levels of gliM and casB transcripts in the absence of rax1, the Δrax1 mutant exhibited increased sensitivity to exogenous gliotoxin (GT) without affecting levels of GT production. Moreover, Δrax1 resulted in significantly restricted colony growth and reduced viability under endoplasmic reticulum stress condition. In summary, Rax1 positively affects expression of gliM and metacaspase genes.


Introduction
The G protein signaling pathway plays central roles in transmitting environmental stimuli into intrinsic signals leading to appropriate gene expression changes. Binding of an external ligand to an associated G protein coupled receptor (GPCR) causes GDP-to-GTP exchange of a Gα, which then results in dissociation of Gα-GTP and Gβγ subunits. Dissociated (activated) G protein subunits interact with downstream effectors leading to activation of various signaling pathways involving cAMP-dependent protein kinase A (PKA), mitogen activated protein kinases (MAPKs), or protein kinase C (PKC) [1][2][3]. One of the key control points of these signaling pathways is at the G protein level exerted by regulators of G protein signaling (RGSs). Normally, an RGS protein interacts with its target Gα-GTP subunit and activates intrinsic GTPase activity of the Gα-GTP, leading to accelerated GTP to GDP hydrolysis. Gα-GDP then associates with its Gβγ subunit, leading to turn-off of the G protein-mediated signaling pathway [2,4].
In the genome of the opportunistic human pathogenic fungus Aspergillus fumigatus, six RGS proteins that contain the conserved RGS domain have been identified based on homology searches, and the roles of them have been characterized [5][6][7][8][9][10]. Previously, we reported that Rax1 of A. fumigatus positively controls vegetative growth and asexual development. In addition, the rax1 null mutant conidia were more resistant against exogenous H 2 O 2 via elevated accumulation of intracellular trehalose and cell wall melanin [7]. In the present study, to further understand the functions of Rax1, we carried out RNA-Seq-based transcriptome analyses of wild types (WT) and rax1 null mutants. In addition, follow-up functional studies corroborating the findings of genome-wide expression analyses are reported.

Summary of the RNA-Seq Data
For the comparative RNA-seq analysis between WT and ∆rax1 strain, cells were collected at 12 h cultures as described in Materials and Methods. Correlation of overall gene expression between WT and ∆rax1 strain was highly correlated (R = 0.95, Figure 1A). The A. fumigatus genome contains 9840 protein, 205 tRNA, 34 rRNA, 29 snoRNA, 12 ncRNA, and 9 snRNA coding genes with 1 pseudogene (AspGD: http://www.aspergillusgenome.org). We obtained an average of 22,222,727 reads and 9,836 different transcripts for ∆rax1 strain. Of the 9,836 genes, 1,300 genes (13.2%) displayed more than 2.0-fold (log 2 FC = 1.0) changes in mRNA levels ( Figure 1B), of which 957 (9.7%) exhibited higher transcript levels in the ∆rax1 strain than in the WT strain whereas 343 (3.5%) were down-regulated in the ∆rax1 strain. Among them, statistically significant (p < 0.05) genes were 92 and 91, respectively (Tables S1 and S2).

Rax1 Functions in the Resistance against Exogenous Gliotoxin
The most down-regulated gene by Δrax1 was the one encoding the O-methyltransferase GliM (Log2FC = −10.83). To assess RNA-seq result, we estimated the mRNA levels of gliM by qRT-PCR. As shown in Figure 2A, level of gliM transcript in mutant strain was significantly lower (about 0.025fold) than that of WT. GliM catalyzes the conversion of reduced dithiol gliotoxin (GT) to bisthiomethyl gliotoxin and plays a role in protecting fungi against dithiol containing toxins [11]. The production of GT analyzed by TLC and the relative intensity of GT of WT, Δrax1, and complemented (C′) strains was 16.2, 16.1, and 17.2, respectively, implying that GT production was not affected by the absence of rax1 ( Figure 2B). We also checked the effect of Rax1 on the virulence in the CD1 (ICR) female mice. Conidia of three strains were inoculated in mice, and monitored every 12 h for survival for 4 days after challenge. The virulence of Δrax1 strains was avirulent compared to WT and C′ strains ( Figure 2C). Comparison of survival curves analyzed by a log rank test revealed that there was no meaningful difference between the three strains (p = 0.7919). To test the effect of reduced expression of gliM in the Δrax1 strain, we checked the tolerance of the mutant to exogenous GT in comparison to WT. When conidia of three strains were inoculated in the presence of GT, the radial growth of all strains was decreased compared to untreated controls. While the growth of WT and C' strains were decreased about 30%, which of Δrax1 strain was decreased about 40% ( Figure 2D), indicating that the Δrax1 mutant was more sensitive than WT and C' strains against exogenous GT. These results suggest that Rax1 is required for proper expression of gliM leading to the self-protection against exogenous GT, but not de novo biosynthesis of GT.

Rax1 Functions in the Resistance against Exogenous Gliotoxin
The most down-regulated gene by ∆rax1 was the one encoding the O-methyltransferase GliM (Log 2 FC = −10.83). To assess RNA-seq result, we estimated the mRNA levels of gliM by qRT-PCR. As shown in Figure 2A, level of gliM transcript in mutant strain was significantly lower (about 0.025-fold) than that of WT. GliM catalyzes the conversion of reduced dithiol gliotoxin (GT) to bisthiomethyl gliotoxin and plays a role in protecting fungi against dithiol containing toxins [11]. The production of GT analyzed by TLC and the relative intensity of GT of WT, ∆rax1, and complemented (C ) strains was 16.2, 16.1, and 17.2, respectively, implying that GT production was not affected by the absence of rax1 ( Figure 2B). We also checked the effect of Rax1 on the virulence in the CD1 (ICR) female mice. Conidia of three strains were inoculated in mice, and monitored every 12 h for survival for 4 days after challenge. The virulence of ∆rax1 strains was avirulent compared to WT and C strains ( Figure 2C). Comparison of survival curves analyzed by a log rank test revealed that there was no meaningful difference between the three strains (p = 0.7919). To test the effect of reduced expression of gliM in the ∆rax1 strain, we checked the tolerance of the mutant to exogenous GT in comparison to WT. When conidia of three strains were inoculated in the presence of GT, the radial growth of all strains was decreased compared to untreated controls. While the growth of WT and C' strains were decreased about 30%, which of ∆rax1 strain was decreased about 40% ( Figure 2D), indicating that the ∆rax1 mutant was more sensitive than WT and C' strains against exogenous GT. These results suggest that Rax1 is required for proper expression of gliM leading to the self-protection against exogenous GT, but not de novo biosynthesis of GT.

Rax1 Positively Regulates Metacaspase Activity
In RNA-seq analysis, we found that the transcript encoding the metacaspase CasB was significantly down-regulated (Log2FC = −2.90) in the Δrax1 mutant compared to WT (Table S3). To test whether Rax1 affects metacaspases expression, qRT-PCR analysis for the casA and casB transcript was carried out. The deletion of rax1 led to reduced mRNA levels of metacaspase encoding genes ( Figure 3A). To test sensitivity against endoplasmic reticulum (ER) homeostasis disrupting agents, radial growth and survival rates of three strains were determined in the presence of dithiothreitol (DTT), 2-deoxy-D-glucose (2-DG), and brefeldin A (BFA). The radial growth of the mutant was reduced by the treatment of DTT and 2-DG ( Figure 3B). Markedly, the Δrax1 mutant exhibited significantly restricted vegetative growth on the medium containing 0.1% 2-DG as the sole carbon source (about 50% of WT and C′ strains) ( Figure 3B). Against another ER stressor BFA, as shown in Figure 3C, while viability of WT and C′ strains was about 47%, but that of Δrax1 strain was 35% indicating that it is likely that Δrax1 might lead to reduced tolerance to BFA ( Figure 3C).

Rax1 Positively Regulates Metacaspase Activity
In RNA-seq analysis, we found that the transcript encoding the metacaspase CasB was significantly down-regulated (Log 2 FC = −2.90) in the ∆rax1 mutant compared to WT (Table S3). To test whether Rax1 affects metacaspases expression, qRT-PCR analysis for the casA and casB transcript was carried out. The deletion of rax1 led to reduced mRNA levels of metacaspase encoding genes ( Figure 3A). To test sensitivity against endoplasmic reticulum (ER) homeostasis disrupting agents, radial growth and survival rates of three strains were determined in the presence of dithiothreitol (DTT), 2-deoxy-D-glucose (2-DG), and brefeldin A (BFA). The radial growth of the mutant was reduced by the treatment of DTT and 2-DG ( Figure 3B). Markedly, the ∆rax1 mutant exhibited significantly restricted vegetative growth on the medium containing 0.1% 2-DG as the sole carbon source (about 50% of WT and C strains) ( Figure 3B). Against another ER stressor BFA, as shown in Figure 3C, while viability of WT and C strains was about 47%, but that of ∆rax1 strain was 35% indicating that it is likely that ∆rax1 might lead to reduced tolerance to BFA ( Figure 3C).

Discussion
RGSs are a group of proteins including conserved RGS box that interacts with a GTP-Gɑ subunit and turn-off the G protein signaling pathways [4]. One of the A. fumigatus RGS, Rax1 (RgsB) protein, is similar to ScRax1P of the yeast Saccharomyces cerevisiae, where ScRax1P is required for the establishment of the bipolar budding [12]. However, the function of its homologous proteins is poorly understood in filamentous fungi. Previously, we reported that Rax1 of A. fumigatus positively controls growth, development, and oxidative stress response [7]. To further understand the functions of Rax1, we carried out RNA-Seq-based genome-wide expression studies and additional functional investigation.
Based on the transcriptome analysis, we found that a large number of genes are down-regulated in the Δrax1 strain compared to the WT strain. At first, we focused on the most down-regulated gene, gliM. The gliM gene is located in the GT biosynthetic gene cluster that comprises 13 genes [13,14] and encodes an O-methyltransferase. Disruption of gliM completely abrogated bis-methyl gliotoxin biosynthetic ability, while GT production and secretion was unaffected [11]. By modification of the active dithiol form of GT to an inactive bis-thiomethylated form, the GT producing fungi are protected from dithiol end products and/or toxic biosynthetic intermediates. In accordance with these, while there were no significant differences in the production of GT between strains, growth of the Δrax1 strain was more inhibited by exogenous GT compared to WT and C′ strains (Figure 2).

Discussion
RGSs are a group of proteins including conserved RGS box that interacts with a GTP-Gα subunit and turn-off the G protein signaling pathways [4]. One of the A. fumigatus RGS, Rax1 (RgsB) protein, is similar to ScRax1P of the yeast Saccharomyces cerevisiae, where ScRax1P is required for the establishment of the bipolar budding [12]. However, the function of its homologous proteins is poorly understood in filamentous fungi. Previously, we reported that Rax1 of A. fumigatus positively controls growth, development, and oxidative stress response [7]. To further understand the functions of Rax1, we carried out RNA-Seq-based genome-wide expression studies and additional functional investigation.
Based on the transcriptome analysis, we found that a large number of genes are down-regulated in the ∆rax1 strain compared to the WT strain. At first, we focused on the most down-regulated gene, gliM. The gliM gene is located in the GT biosynthetic gene cluster that comprises 13 genes [13,14] and encodes an O-methyltransferase. Disruption of gliM completely abrogated bis-methyl gliotoxin biosynthetic ability, while GT production and secretion was unaffected [11]. By modification of the active dithiol form of GT to an inactive bis-thiomethylated form, the GT producing fungi are protected from dithiol end products and/or toxic biosynthetic intermediates. In accordance with these, while there were no significant differences in the production of GT between strains, growth of the ∆rax1 strain was more inhibited by exogenous GT compared to WT and C strains (Figure 2).
The caspase plays a central role in apoptosis and can be classified into the metazoan caspase family, paracaspase family, and metacaspase family [15]. The metacaspase family is specific to plant, fungi, and protozoa. A. fumigatus has two metacaspases, CasA and CasB, which function in both apoptotic morphology and resistance to ER stress [16]. The metacaspase deletion mutants showed a growth detriment in ER stress conditions [16]. In RNA-seq and qRT-PCR analyses, the expression of both metacaspases was significantly decreased by loss of rax1 ( Figure 3A, Table S2). Moreover, growth of the ∆rax1 strain was impaired on medium containing DTT and 2-DG, which induces ER stress [17,18]. The sensitivity to ER stressor was confirmed using BFA, which generates ER stress by blocking the transport of proteins to the Golgi apparatus [19]. Viability of the ∆rax1 strain was significantly reduced compared to WT and C strains in the presence of BFA ( Figure 3C). Metacaspase-deficient mutants showed similar results. Loss of either casA or casB resulted in impaired radial growth on medium containing 2-DG and tunicamycin that induces ER stress in fungi [18,20,21], and loss of both genes had an additive effect. From these results, it can be concluded that Rax1 is required for normal growth of A. fumigatus in the ER homeostasis perturbation conditions by positively regulating CasA and CasB.

Strains and Culture Conditions
Glucose minimal medium (MMG) and MMG with 0.1% yeast extract (MMY) with appropriate supplements were used for culture of wild type (AF293) and the rax1 null mutant strains [7]. Conidia were propagated MMY for 5 days at 37 • C and were collected with 0.02% Tween 80. The suspension was passed through a Miracloth (Millipore, Burlington, MA, USA) to remove hyphae and counted using a hemocytometer. To examine production of secondary metabolites, spores of relevant strains were inoculated at a final concentration of 5 × 10 5 conidia/mL to 50 mL of liquid MMY with appropriate supplements and incubated at 250 rpm at 37 • C for 4 days.

Nucleic Acid Manipulation
Total RNA isolation and quantitative RT-PCR (qRT-PCR) assays were was carried out as previously described [5,22]. Briefly, conidia (5 × 10 5 conidia/mL) of WT, ∆rax1, and C strains were inoculated into MMY liquid media with appropriate supplements and incubated at 37 • C, 250 rpm. Mycelial samples were collected and squeeze-dried at indicated time points. The sample was homogenized using a Mini Bead beater in the presence of silica/zirconium beads (0.3 mL, BioSpec Products, Bartlesville, OK, USA) and TRIzol ® reagent (1.0 mL, Invitrogen, Carlsbad, CA, USA). Quantitative RT-PCR (qRT-PCR) assays were carried out according to the manufacturer's instruction using a Rotor-Gene Q (Qiagen, Hilden, Germany). Each run was assayed in triplicate with the RNA template, One Step RT-PCR SYBR Mix (MGmed, Seoul, Korea), reverse transcriptase (MGmed, Korea), and 10 pmole of each primer (Table S3). The data were normalized to ef1α expression [10] and calculated according to the ∆∆Ct method [22].

RNA-Seq Experiment
For RNA-seq analyses, cellophane sheet was spread out on the agar medium and 50 µL of a conidial suspension (10 7 /mL) was inoculated onto the cellophane. The plates were incubated in the dark at 37 • C for 12 h. We collected fungal colonies in triplicate and they were immediately frozen in liquid nitrogen and ground into a fine powder with a mortar and pestle. Total RNA was extracted as in the above described method. After analysis of RNA quality, three RNA samples from the same strain were combined and analyzed. This was done in duplicate. Total RNA was submitted to eBiogen Inc. (Seoul, Korea) for further analyses. RNA-Seq reads were mapped using TopHat software v2.1.0 tool in order to obtain the alignment file and Aspergillus fumigatus Database (NCBI) was used as the template for mapping. The alignment file was used to assemble transcripts, estimate their abundances, and detect differential expression of genes or isoforms using Cufflink v2.2.1. Gene classification was done based on searches done by BioCarta (http://www.biocarta.com/), GenMAPP (http://www.genmapp.org/), DAVID (http://david.abcc.ncifcrf.gov/), and Medline databases (http://www.ncbi.nlm.nih.gov/).

Phenotype Analyses
The tolerance to gliotoxin (GT, Sigma-Aldrich, Saint Louis, MO, USA) was tested with conidia of WT and mutant strains. Each conidia was inoculated onto MMG plates with the presence or absence of GT (10 µg/mL) and incubated at 37 • C for 48 h [14]. GT was extracted with chloroform as described previously [23] and analyzed by a thin-layer chromatography (TLC) silica plate (Kiesel gel 60, E. Merck). The TLC plate was developed with chloroform:methanol (9:1, v/v). For test of ER stress responses, DTT (5 mM) was added into MMG and 2-DG was used in place of glucose at 0.1% (w/v) in MMG solid medium. To assess sensitivity to BFA, conidia (1 × 10 4 ) were inoculated into 2 mL of MMG containing BFA (1×) and incubated at 37 • C for 48 h. Cell viability was determine using alamarBlue (AB) assay based on the percent reduction of AB as described previously [24].

Murine Virulence Assay
For the persistently neutropenic mouse model, we used outbred ICR (Orient Bio Inc., Seongnam-si, 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 intraperitoneal injections (i.p.) of cyclophosphamide (Sigma-Aldrich, USA) at 250 mg/kg for 4 days prior to infection and with cyclophosphamide at 250 mg/kg and cortisone acetate (Sigma-Aldrich, USA) injected subcutaneously at 250 mg/kg 1 day prior to infection. At day 1 and 3 day post-infection, administrations were repeated with cyclophosphamide (125 mg/kg i.p.). Mice were anesthetized with isoflurane and then intranasally infected with 1 × 10 7 conidia of A. fumigatus strains (10 mice per each strain) in 30 µL of 0.01% Tween 80 in PBS. Mice were observed every 12 h for survival for 4 days after challenge. Mock mice were inoculated with sterile 0.01% Tween 80 in PBS.

Statistical Analyses
Comparison of mRNA expressions, radial growth, and viability within the different strains were performed by one-way ANOVA tests. All results were expressed as means ± standard deviation (SD) and a p value less than 0.05 was considered statistically significant. Analyses of survival rates were performed by creating Kaplan-Meier plots and then performing log rank tests (GraphPad Prism 7.0). And comparisons for survival studies were considered significant if the p value was <0.05.

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).

Data Availability
The RNA-seq data are available from NCBI Gene Expression Omnibus (GEO) database under the accession number GSE100101.