The Arf-GAP Proteins AoGcs1 and AoGts1 Regulate Mycelial Development, Endocytosis, and Pathogenicity in Arthrobotrys oligospora

Small GTPases from the ADP-ribosylation factor (Arf) family and their activating proteins (Arf-GAPs) regulate mycelial development, endocytosis, and virulence in fungi. Here, we identified two orthologous Arf-GAP proteins, AoGcs1 and AoGts1, in a typical nematode-trapping fungus Arthrobotrys oligospora. The transcription of Aogcs1 and Aogts1 was highly expressed in the sporulation stage. The deletion of Aogcs1 and Aogts1 caused defects in DNA damage, endocytosis, scavenging of reactive oxygen species, lipid droplet storage, mitochondrial activity, autophagy, serine protease activity, and the response to endoplasmic reticulum stress. The combined effects resulted in slow growth, decreased sporulation capacity, increased susceptibility to chemical stressors and heat shock, and decreased pathogenicity of the mutants compared with the wild-type (WT) strain. Although deletion of Aogcs1 and Aogts1 produced similar phenotfypic traits, their roles varied in conidiation and proteolytic activity. The ΔAogts1 mutant showed a remarkable reduction in conidial yield compared with the WT strain but not in proteolytic activity; in contrast, the ΔAogcs1 mutant showed an increase in proteolytic activity but not in sporulation. In addition, the growth of ΔAogcs1 and ΔAogts1 mutants was promoted by rapamycin, and the ΔAogts1 mutant was sensitive to H-89. Collectively, the ΔAogts1 mutant showed a more remarkable difference compared with the WT strain than the ΔAogcs1 mutant. Our study further illustrates the importance of Arf-GAPs in the growth, development, and pathogenicity of nematode-trapping fungi.


Introduction
The superfamily of small G-proteins (GTPases) that mediate signal transduction in eukaryotic cells can be classified into at least five subfamilies, including Rho, Rab, Ras, Ran, and Arf [1,2]. These proteins are known for their prime roles in vesicle trafficking and actin cytoskeleton remodeling [3]. These superfamilies serve as hydrolases that bind and hydrolyze GTP to form GDP and act as molecular switches for small G-proteins [4]. The cycling of small GTPases is accomplished by switching between an activated state (GTP-bound state) and an inactive state (GDP-bound state), and GTPase-activating proteins (GAPs) are important proteins involved in this transition. Upon the action of GAPs, the hydrolase activity of the small GTPase itself is activated, and the phosphate group is released, thereby hydrolyzing the activated small GTPase into an inactive state [5].
ADP ribosylation factors (Arfs) belong to the small GTPases subfamily and interact with many proteins and other molecules that regulate their state of activation. The function of an Arf depends on switching between inactive Arf-GDP and active Arf-GTP. Thus,

Fungal Strains and Culture Conditions
Arthrobotrys oligospora (ATCC24927) purchased from the American Type Culture Collection (ATCC) and deletion mutants generated during this study were routinely cultured on potato dextrose agar (PDA) medium at 28 • C. The Saccharomyces cerevisiae FY834 strain was grown in yeast extract-peptone-dextrose (YPD) broth, and SC-Ura medium was used to screen recombinant transformants [36]. Escherichia coli DH5α was used to clone the constructed vector and grown in LB medium at 37 • C. Oat medium was used to grow Caenorhabditis elegans (strain N2) at 26 • C in the dark for a week. PDA, tryptone glucose (TG), and tryptone yeast extract glucose agar (TYGA) were used for the phenotypic analysis as previously described [37] and corn-maizena yeast extract (CMY) was used for sporulation as previously described [38].

Deletion of Aogcs1 and Aogts1
The replacement fragments for the target gene were constructed as previously described [42]. Briefly, the 5 and 3 DNA sequences of the target genes were amplified from A. oligospora using paired primers (Table S1), and the hygromycin resistance gene (hph) cassette was amplified from pCSN44. Next, all fragments including the 5 sequence, 3 sequence, hph sequence, and the linearized pRS426 plasmid digested with EcoRI and XhoI were transformed into the S. cerevisiae FY834 strain [43]. Then, the yeast plasmid with AoGcs1-hph-pRS426 and AoGts1-hph-pRS426 were isolated and used as DNA templates, and the gene replacement fragment was amplified using paired primers (Table S1). Finally, the replacement fragments were chemically introduced into A. oligospora protoplasts and spread on PDAS medium with 200 µg/mL hygromycin B [44]. Each transformant for genes Aogcs1 and Aogts1 was verified by a PCR method using the specific primers YZ-F and YZ-R (Table S1) as previously described [42], and the positive transformants were further identified by Southern blot analysis, which was performed using the North2South Chemiluminescent Hybridization Detection Kit (Pierce, Rockford, IL, USA).

Comparison of Vegetative Growth and Conidia
The wild-type (WT) and mutant strains were cultured on PDA medium for 5 days, then 7 mm discs were collected from each WT and mutant strain using a hole puncher and inoculated onto PDA, TG, and TYGA medium for incubation at 28 • C for 7 days to determine the colonial diameter for each fungal strain. To assess the sporulation of each strain, 7 mm equal-sized fungal colonies from each strain were inoculated in the center of the CMY medium nad incubated at 28 • C for 14 days as previously described [45]. The number and morphology of conidia and conidiophores were observed by the side-shot method on the WA plate [46,47]. Conidial germination rates were determined as previously described [45]. Briefly, equal amounts of conidia from the WT and mutant strains were incubated on MM medium at 28 • C, and germinated conidia were calculated at specific time points (i.e., 4, 8, and 12 h). The experiment was repeated three times for each strain.

Analysis of DNA Damage and Cell Apoptosis
Colonies from the WT and mutant strains were cultured on PDA with sterile coverslips for 7 days at 28 • C, and then the nuclei and morphology of mycelia were stained with DAPI and CFW, respectively. DNA damage and apoptosis were assessed by the terminal deoxynucleotidyl transferase-mediated dUTP nick end-labeling (TUNEL) assay [48]. The nuclei of the hyphae were stained with propidium iodide (PI), and fragmented DNA in the hyphae was re-stained with FITC-dUTP. The ratio of green to red fluorescence intensities was determined for at least 30 fields under a microscope to quantify the extent of DNA damage.

Comparison of Stress Tolerance
Seven millimeter fungal discs from WT and mutant strains were cultured on TG medium (control) supplemented with different concentration gradients of chemical stressors, including osmotic stressors NaCl (0.1, 0.2, and 0.3 M) and sorbitol (0.25, 0.50, and 1 M), oxidative stressor H 2 O 2 (5, 10, and 15 mM), menadione (0.01, 0.03, and 0.05 mM), or the cell wall stress agents sodium dodecyl sulfate (SDS) (0.01, 0.02, and 0.03%) and Congo red (0.02, 0.05, and 0.1 mg/mL). Inhibitors of the mTOR and cAMP signaling pathway, rapamycin (10 µM) and H-89 (10 µM), were used at 28 • C for 7 days to test the response of the WT and mutant strains to intracellular signals [29]. In addition, fungal mycelium discs of the same size (i.e., 7 mm) from different strains were inoculated onto TG medium and incubated at different temperatures (28,34,38, and 42 • C) for 8 h and then returned to 28 • C until the seventh day to examine the sensitivity to heat shock [49]. Colonial morphology was inspected, the diameter of each colony was measured, and the relative growth inhibition (RGI) value was calculated as previously described [29,50]. The experiment was repeated three times for each strain.

Trap Formation, Nematocidal, and Proteolytic Activity
Approximately 1 × 10 5 conidia from each strain were spread on WA plates and incubated at 28 • C for 3 days. Then, 200 nematodes were added to the center of each plate to induce trap formation and to observe and quantify the number of traps and captured nematodes at 12 h intervals [51]. The experiment was repeated three times for each strain.
To determine proteolytic activity, equal-sized fungal colonies were added to PD broth supplemented with 20 mL skim milk (5%) and cultured at 180 rpm at 28 • C for 7 days. The fermentation liquids were collected in an Erlenmeyer flask and the protease activity was evaluated on casein plates as previously described [52]. Quantitative analysis of protease activity was performed as previously described [52,53].

Analyses of Lipid Droplets (LDs), Endocytosis, ROS Accumulation, and Autophagy
To observe LDs, each strain was incubated on PDA plates with sterile coverslips at 28 • C for 7 days. Then, the mycelia were treated with 30 µL of 10 µg/mL boron dipyrromethene (BODIPY) dye at 37 • C in the dark, and LDs were observed by fluorescence microscope. The morphology of LDs was also observed by transmission electron microscope (TEM; Hitachi, Tokyo, Japan). To compare the endocytosis of fungal strains, the mycelial samples were stained with FM4-64, and the changes in FM4-64 at 0 s, 10 s, 1 min, and 3 min after entering the hyphae were observed by fluorescence microscope.
In addition, the fluorescent dyes dihydroethidium (DHE) and monodansylcadaverine (MDC) were used at a concentration of 10 µg/mL to analyze the accumulation of ROS and autophagy in hyphae, respectively. Subsequently, autophagy activity was analyzed by estimating the fluorescence intensity values of MDC observed under a fluorescence microscope for at least 30 fields, and ROS levels were estimated from the intensity of DHE fluorescence. Meanwhile, autophagosomes were observed by TEM.

Analysis of ER Stress and Mitochondrial Activity
Hyphal discs (7 mm) were cultured on TG medium supplemented with 10 mM dithiothreitol (DTT) and 10 mM tunicamycin for 7 days at 28 • C to test the ER stress response of WT and mutant strains. Then, the colony diameter was measured, and the RGI value was calculated [29,50]. For the analysis of mitochondrial morphology and mitochondrial membrane potential (MMP), hyphal discs were cultured on PDA plates with a glass slide. Afterwards, the hyphae were treated with 10 µg/mL MitoTracker Green or 10 µg/mL tetraethylbenzimidazolyl-carbocyanine iodide (JC-1) dye and incubated in the dark for 30 min, and images were collected with a fluorescence microscope. The ratio of green/red fluorescence intensity was calculated using ImageJ software (https://imagej.nih.gov/ij, accessed on 16 January 2022) for at least 30 randomly selected fields of view per slide. The ratio of green to red fluorescence intensities of JC-1 staining was used to assess MMP levels [29].

Quantitative Real-Time PCR (RT-PCR)
To analyze transcription levels for the genes involved in sporulation, cell wall synthesis, oxidative stress, cAMP signaling, and autophagy at different stages of growth and development in A. oligospora, the WT and mutant strains were cultivated on TYGA medium at 28 • C for 3, 5, and 7 days, and hyphae were collected for RT-PCR analysis. In addition, the WT strain was induced with nematodes for 0, 12, 24, 36, and 48 h, and hyphae were collected to determine transcription levels of the Aogcs1 and Aogts1 genes during trap formation. The total RNA from all mycelia samples was isolated using an RNA Extraction Kit (Axygen, Suzhou, China) and reverse-transcribed cDNA was prepared using a FastQuant RT Kit (Takara, Kusatsu, Japan). The transcription levels for candidate genes were then analyzed by RT-PCR assays with specific paired primers (Table S2). The β-tubulin gene (AOL_s00076g640) from A. oligospora was used as an internal standard. RT-PCR was performed in triplicate for each gene as previously described [54]. The transcription levels for the target genes were calculated using the 2 −∆∆Ct method; briefly, fold changes were calculated using the following formula: 2 −∆∆Ct , where ∆∆Ct is ∆Ct (treatment) − ∆Ct (control), ∆Ct is Ct (target gene) − Ct (tub), and Ct is the threshold cycle [55]. In addition, the relative transcript level (RTL) for each gene was quantified as the transcript level of each mutant versus WT.

Statistical Analyses
GraphPad Prism version 9.3.1 (GraphPad Software, San Diego, CA, USA) was used for photographic and statistical analysis. Data from triplicate experiments were examined using a one-way analysis of variance (ANOVA) and ranked by Tukey's multiple comparison test. p < 0.05 was considered a significant difference.

Sequence Analysis, Phylogenetic Relationship, and Transcription of Aogcs1 and Aogts1
Two Arf-GAP proteins, namely, AoGcs1 and AoGts1, were retrieved from A. oligospora using the homologous sequences of S. cerevisiae as queries. AoGcs1 and AoGts1 were found to contain 366 and 622 amino acid residues with predicted isoelectric points of 7.30 and 10.16 and molecular weights of 39.2 and 72.8 kDa, respectively. Gcs1 and Gts1 contained a conserved Arf-GAP domain in different fungi, and Gts1 also contained a ubiquitinassociated (UBA) domain ( Figure S1A). AoGcs1 shared a high degree of similarity to orthologs from the NT fungi Dactylellina haptotyla (81.7%) and Duddingtonia flagrans (93.7%), and it shared a middle degree of similarity (44.8-47.9%) to orthologs from other filamentous fungi; similarly, AoGts1 shared a high degree of similarity to orthologs from NT fungi D. haptotyla (75.5%) and D. flagrans (95.5%), and it shared a similarity of 28.3-31.7% to orthologs from other filamentous fungi (Table S3). The phylogenetic tree of homologous Gcs1 and Gts1 from different fungi was constructed, and they were separated into two clades ( Figure S1B). The transcription of Aogcs1 and Aogts1 were downregulated during trap formation (compared to zero time), but significantly upregulated (p < 0.05) during the conidiation stage (from 3 to 9 days) compared with those in the vegetative growth stage (day 2) (Figure S1C).

AoGcs1 and AoGts1 Regulate Vegetative Growth and DNA Damage
The genes Aogcs1 and Aogts1 were deleted ( Figure S2) and two mutants of gene Aogcs1 and three mutants of gene Aogts1 were obtained for further study. Two mutants (i.e., ∆Aogcs1 and ∆Aogts1) showed significantly reduced mycelial growth (p < 0.05) on PDA, TG, and TYGA media ( Figure 1A), and the average colony diameters for the WT, ∆Aogcs1, and ∆Aogts1 strains on PDA medium were 8.3, 6.5, and 4.8 cm, respectively. The corresponding colony diameters for each strain when grown on TG medium were 8.5, 5.2, and 4.9 cm compared to 8.5, 6.5, and 5.1 cm when grown on TYGA medium ( Figure 1B). The colonial morphology of the ∆Aogts1 mutant was irregular ( Figure 1A). After the mycelia and cell nuclei of the WT and mutant strains were stained with CFW and DAPI, the mycelia of the ∆Aogts1 mutant were more swollen and irregular than the WT strain ( Figure 1C). Additionally, the nuclei of the mutants were difficult to stain as they appeared diffused and fragmented ( Figure 1C). Fragmentation of chromosomal DNA was detected in the mycelia of the mutant strains using the TUNEL assay, and the nuclei were in a diffuse form when stained with PI ( Figure 1C). Since most of the DNA in the mutants was in a diffuse state, the FI ratios of FITC-dUTP and PI were calculated to analyze the differences in the extent of DNA damage and cell apoptosis between the WT and mutant strains. The FI ratios for the WT, ∆Aogcs1, and ∆Aogts1 strains were 0.09, 0.33, and 0.55, respectively ( Figure 1D).

AoGcs1 and AoGts1 Regulate Conidiation, Conidial Morphology, and Conidial Germination
The WT strain produced many conidiophores, and multiple spores formed in the denticle node of the conidiophore. However, few conidiophores were observed for the ∆Aogts1 mutant (Figure 2A). The spore yields for the WT, ∆Aogcs1, and ∆Aogts1 were 30.9, 28.8, and 9.7 × 10 5 spores/mL, respectively ( Figure 2C). After staining the spore morphology and cell nucleus with CFW and DAPI, the conidia produced by the WT strain were usually obovoid in shape and formed one septum near the base, whereas most of the conidia from the ∆Aogcs1 (48%) and ∆Aogts1 (63%) mutants did not feature a septum or a change in morphology ( Figure 2B,D). However, no difference in conidial germination of the ∆Aogcs1 mutant was observed when compared with the WT strain. At 4 h, approximately 73.8%, 77.3%, and 57.8% of spores had germinated in the WT strain, ∆Aogcs1, and ∆Aogts1, respectively ( Figure 2E). Compared with the WT strain, the transcript levels of all five genes involved in sporulation, namely, abaA, veA, velB, vosA, and fluG, were downregulated on day 5 in the ∆Aogcs1 mutant. In addition, expression levels of abaA, velB, vosA, and fluG were also downregulated in the ∆Aogts1 mutant on day 5. In contrast, the expression of abaA, vosA, and fluG was significantly upregulated (p < 0.05) in the ∆Aogcs1 mutant on day 7. Furthermore, the expression of veA was significantly upregulated (p < 0.05) in the ∆Aogts1 mutant at all tested time points, whereas the expression levels of other genes were upregulated at only some time points ( Figure 2F).

AoGcs1 and AoGts1 Are Involved in Resistance to Chemical Stresses
The WT strain and mutants were compared for their responses to different chemical agents ( Figure 3A   To probe the regulatory mechanism of Arf-GAP proteins in stress resistance, the transcription levels of fifteen putative genes involved in oxidative stress, hyperosmolarity, and cell wall synthesis were analyzed. The expression of five genes involved in osmotic pressure was downregulated in the ∆Aogcs1 mutant on day 5. However, the expression of all five genes was upregulated on day 3 and day 7, especially the level of the msn2 gene (p < 0.05). Similar to the observations for the ∆Aogcs1 mutant, the expression levels of three genes, namely, sho1, sln1, and ssk1, were downregulated in the ∆Aogts1 mutant on days 3 and 5, and all genes were significantly upregulated on day 7 (p < 0.05) ( Figure 3F). Transcription levels of five genes involved in cell wall synthesis (i.e., chs1, chs2, gfpa, gls, and glu) were analyzed in mutants and WT strains, and all five genes were downregulated in the ∆Aogcs1 mutant on day 5. Similarly, the transcript levels of four genes (i.e., chs1, gfpa, gls, and glu) were also downregulated on day 5 in the ∆Aogts1 mutant, whereas the chs2 gene was upregulated. In contrast, the expression of chs1, chs2, gfpa, and gls was significantly upregulated (p < 0.05) in the ∆Aogts1 mutant on days 3 and 7. Similarly, the expression of chs2 and gfpa was also significantly upregulated (p < 0.05) in the ∆Aogcs1 mutant on days 3 and 7 ( Figure S3D). When the expression of five genes related to oxidative stress, namely, per, thr, trxR, trx, and glr, was analyzed, the results showed that the thr gene was significantly downregulated (p < 0.05) in the ∆Aogcs1 mutant at all tested time points, whereas the expression levels of the other genes were downregulated at only some time points. The expression of two genes, trx and glr, was significantly downregulated (p < 0.05) in the ∆Aogts1 mutant at all tested time points, and the expression levels of the other three genes were downregulated on day 5. In contrast, the expression of three genes (i.e., trxR, trx, and glr) was upregulated in the ∆Aogcs1 mutant on days 3 and 7. Noticeably, the expression of trxR was significantly upregulated (p < 0.05) in the ∆Aogts1 mutant on days 3 and 5, implying that this gene could be important for the ability of AoGts1 to regulate the fungal response to oxidants ( Figure 3G).
Moreover, compared with the WT strain (49.8 U/g), disruption of the Aogts1 gene (45.5 U/g) had little influence on the proteolytic activity of the mutant. In contrast, the ∆Aogcs1 mutant (71.7 U/g) displayed increased proteolytic activities ( Figure 4D,E). The transcriptional levels of five serine protease genes in A. oligospora were determined using RT-PCR. Of these, 54g992, 75g8, 78g136, and 215g702 were significantly downregulated (p < 0.05) in the ∆Aogts1 mutant on days 3 and 5, whereas the transcriptional levels of those genes were significantly upregulated (p < 0.05) on day 7. For the ∆Aogcs1 mutant, the expression levels of all five genes were significantly downregulated (p < 0.05) on day 5 compared to WT, whereas very few genes were downregulated at the other time points ( Figure 4F).
The role of Arf-GAPs in cAMP-PKA signaling in A. oligospora was tested, and the ∆Aogts1 mutant showed sensitivity to the PKA inhibitor, H-89, at a concentration of 10 µM ( Figure S4). In addition, the transcriptional levels of the downstream genes of the cAMP/PKA, TOR, and MAPK pathways were significantly downregulated (p < 0.05) in the ∆Aogcs1 and ∆Aogts1 mutants on day 5 ( Figure 5C).

AoGcs1 and AoGts1 Regulate Intracellular Lipid Storage and Endocytosis
To examine the roles of the Aogcs1 and Aogts1 genes in lipid utilization and endocytosis, we visualized LDs and endocytosis in fungal mycelia using BODIPY and an endocytosis tracer, the lipophilic styryl dye FM4-64, respectively. Compared with their distribution in the WT, LDs were remarkably reduced and unevenly distributed in the hyphae of the ∆Aogcs1 mutant, whereas there was no difference in the distribution between the ∆Aogts1 mutant and the WT strain. Moreover, it can be observed that the volumes of LDs in the two mutant strains were larger than the WT strain ( Figure 6A). The FM4-64 dye was gradually internalized by the cells over time, and after 3 min of incubation, endosomes and vacuoles in the hyphae of the WT strain were almost completely stained. In contrast, only the plasma membrane near the hyphal wall and septa were stained by FM4-64 in the ∆Aogcs1 and ∆Aogts1 mutants after 3 min of incubation ( Figure 6B). When the LDs in the WT and mutant cells were observed by TEM, we found that the LDs in the ∆Aogcs1 and ∆Aogts1 mutant strains were larger ( Figure 6C). Statistical analysis showed that the diameter of LDs in ∆Aogcs1 and ∆Aogts1 mutant cells were approximately 2-3 times larger than the WT strain ( Figure 6D).

Loss of Aogcs1 and Aogts1 Affects the ER Stress Response and Mitochondrial Activity
We examined the sensitivity of the ∆Aogcs1 and ∆Aogts1 mutant to DTT and tunicamycin, which induces ER-stress, and found that the ∆Aogcs1 and ∆Aogts1 mutants were hypersensitive to DTT ( Figure 7A,B). In addition, mitochondria were unevenly distributed in the ∆Aogcs1 and ∆Aogts1 mutants and their morphology had changed ( Figure 7C). Moreover, we determined mitochondrial activity using JC-1, a cationic dye that accumulates in energized mitochondria [56], and the mycelia of ∆Aogcs1 and ∆Aogts1 mutants displayed stronger green FI than the WT strain ( Figure 7C). The ratio of the green/red FI ratio was significantly increased in the ∆Aogcs1 and ∆Aogts1 mutants when compared with the WT strain. The FI ratios of WT and mutant strains (i.e., ∆Aogcs1 and ∆Aogts1) were 0.23, 0.34, and 0.45, respectively ( Figure 7D).

Deletion of Aogcs1 and Aogts1 Results in ROS Accumulation and Enhanced Autophagy in Hyphae
The hyphal production of ROS was compared between WT and mutant strains. ROS were aggregated in certain regions of the mycelium in the WT strain, whereas ROS were scattered and dispersed in the mycelia of mutants ( Figure 8A). The FI values for DHE in the WT and mutant strains (i.e., ∆Aogcs1 and ∆Aogts1) were 44.6, 59.7, and 54.7, respectively ( Figure 8B). When MDC-stained hyphae of the WT and mutant strains were observed, the ∆Aogcs1 and ∆Aogts1 mutants produced more autophagosomes than the WT mutant strains ( Figure 8A).
The FI values for the WT, ∆Aogcs1, and ∆Aogts1 strains were 40.4, 62.7, and 82.9, respectively ( Figure 8C). Moreover, we observed more autophagosomes in the ∆Aogcs1 and ∆Aogts1 mutants than in the WT strain by TEM ( Figure 8D). The transcriptional levels of five autophagy-related genes (i.e., atg1, atg8, atg9, atg13, and atg17) were upregulated in the mycelium of the ∆Aogts1 mutant on day 7, whereas expression was upregulated in the ∆Aogcs1 mutant on day 3 and not significantly different (p < 0.05) from the WT strain on day 7 ( Figure 8E).

Discussion
Arf is a subfamily of small GTPase proteins that are ubiquitous and highly conserved in all eukaryotic cells, with more than one member present in every organism studied to date [57]. Arf-GAPs can switch from active Arf-GTP to inactive Arf-GDP by hydrolyzing GTP bound to Arf, and it plays a role in vegetative growth, sporulation, and the pathogenicity of fungi [18]. However, little is known about the functions of Arfs and Arf-GAPs in filamentous fungi. Here, two orthologous Arf-GAP proteins, AoGcs1 and AoGts1, were characterized in the NT fungus A. oligospora, and our results showed that AoGcs1 and AoGts1 are key regulators involved in diverse cellular functions in A. oligospora.
Previous studies have shown that Arf-GAPs are involved in the vegetative growth and development of hyphae [14,19,58]. In C. albicans, deletion of age3 (homologous to gcs1) results in defective hyphal growth and loss of invasive hyphae [17]. Similarly, deletion of Moglo3 caused distinct growth defects on different media, suggesting that this protein was required for normal fungal growth of M. oryzae [18]. Recently, an orthologous Glo3 (AoGlo3) was identified for A. oligospora, and deletion of the Aoglo3 gene resulted in growth defects and an increase in the hyphal septum. Furthermore, deletion of gts1 in A. gossypii resulted in severe growth retardation as well as hyphae with spherical swelling and irregular branching [19]. In this study, we also found that ∆Aogcs1 and ∆Aogts1 mutants exhibited growth defects, and hyphae of the ∆Aogts1 mutant became swollen compared to the WT; additionally, deletion of Aogcs1 and Aogts1 resulted in DNA damage and cell apoptosis. Thus, Arf-GAPs play a conserved role in hyphal growth and development in A. oligospora and other fungi.
The growth of the ∆Aogcs1 and ∆Aogts1 mutants was significantly inhibited by chemical stressors such as oxidants, cell wall-disturbing agents, and hypertonic reagents. Specifically, deletion of Aogts1 caused severe damage to the cell wall of A. oligospora, which manifested as increased sensitivity to low concentrations of SDS (0.01%) and Congo red at any concentration, and was associated with a phenotype of altered hyphal and spore morphology. This result was further confirmed by increased expression of the genes involved in cell wall synthesis, especially chs2. Similarly, ∆Moglo3 mutants have increased sensitivity to three cell wall inhibitors, namely, CFW, Congo red, and SDS [18], and the ∆age3 mutant strain of C. albicans was susceptible to SDS [17]. In addition, ∆Aogcs1 and ∆Aogcs1 mutants were more sensitive to the oxidants menadione and H 2 O 2 , which may be related to the downregulation of some genes related to oxidative stress (thr, glr, and thxR) during the vegetative growth stage. Interestingly, different concentrations of the hypertonic reagents NaCl and sorbitol had different effects on the growth of the ∆Aogcs1 and ∆Aogts1 mutants, and the expression of osmotic stress-related genes was significantly upregulated, especially msn2. Moreover, deletion of Aogcs1 and Aogcs1 caused defective growth after heat shock. MSN2/4 are zinc finger protein transcription factors, which are involved in transcriptional induction via stress/cAMP-response elements, functioning in temperature, osmotic, oxidative, and hydrostatic pressure stress responses in S. cerevisiae [59,60]. In the entomopathogenic fungi Beauveria bassiana and Metarhizium robertsii, both ∆Bbmsn2 and ∆Mrmsn2 mutants lost 20-65% of their tolerance to hyperosmolarity, oxidation, carbendazim, cell wall perturbation, and high temperature at 34 • C during colony growth [61,62]. These results show that the orthologs of Gts1 and Gcs1 play a crucial role in the response of yeasts and filamentous fungi to environmental stress.
Asexual sporulation is crucial for the reproduction and survival of A. oligospora and other NT fungi [42]. In this study, the deletion of Aogts1 caused a remarkable reduction in conidial yield, spore germination rate, and altered conidial morphology, whereas the ∆Aogcs1 mutant showed a WT-like phenotype for conidiation. The marked reduction in conidia yield in the ∆Aogts1 mutant was consistent with the altered expression of Aogts1 during the sporulation stage, suggesting that Aogts1 is critical for conidia production. In addition, the expression of several conidia-related genes, such as abaA, fluG, and velB, were significantly downregulated (p < 0.05) in the ∆Aogts1 mutant during the sporulation stage. Previous studies have proved that abaA, fluG, and velB are important genes involved in conidiation of Aspergillus nidulans and other filamentous fungi [42,63,64]. Deletion of velB in A. oligospora abolished sporulation [42]. Similarly, conidia production was completely abolished in the abaA deletion mutant of Fusarium graminearum [65]. Additionally, both arf1 and arf2 are required for sporulation in M. circinelloides [66]. Moreover, similar results were reported in M. oryzae, where the ∆Moglo3 mutant produced significantly fewer conidia compared to the WT strain [18]. These results suggest a conserved role for Arf and Arf-GAPs in regulating asexual development in A. oligospora and other filamentous fungi.
The NT fungi capture nematodes by producing traps and then decompose them for nutrients by secreting extracellular proteases [28]. Disruption of Aogcs1 and Aogts1 in A. oligospora resulted in a significant reduction in the number of traps followed by a significant decrease in nematode predation efficiency. The extracellular proteolytic activity of the ∆Aogcs1 mutant was significantly increased (p < 0.05), but the proteolytic activity of the ∆Aogts1 mutant was similar to WT, a result that was confirmed by significantly decreased (p < 0.05) transcription of the genes encoding serine proteases on day 5. These findings are consistent with results for C. albicans and M. oryzae. Knockout of Caage3 results in the disruption of hyphal cell wall integrity and produces serious defects in invasive growth [17]. Likewise, the deletion of glo3 in M. oryzae results in severe invasive growth defects and reduced virulence in the host [18]. These results suggest a conserved role for Arf-GAPs in infectious structure development, extracellular protease production, and fungal pathogenicity.
Arfs and Arf-GAPs are key components during vesicle trafficking and have been implicated in fungal growth and virulence [16,18,67]. Earlier studies found that the Arf-like protein yArl3 is required for protein transport from the ER to the Golgi or from the Golgi to the vacuole in S. cerevisiae [68]. Glo3 and Gcs1 mediate retrograde vesicular trafficking from the Golgi to the ER in S. cerevisiae, so mutant strains lacking both Glo3 and Gcs1 failed to proliferate, experienced a dramatic accumulation of ER, and were defective in protein trafficking between the ER and the Golgi [14]. The deletion of gts1 in A. gossypii results in defective actin cytoskeleton organization and endocytosis [19]. In addition, ∆Moglo3 mutants are defective in endocytosis, as well as in response of M. oryzae to ER stress and scavenging ROS [18]. Consistent with previous findings, ∆Aogcs1 and ∆Aogts1 mutants delayed uptake of the lipophilic dye FM4-64, exhibited increased sensitivity to ER stressors, and had significantly increased ROS levels in hyphae. These results suggest that Arf and Arf-GAPs play conserved functions in vesicle trafficking, endocytosis, ROS scavenging, and ER stress responses in fungi.
Previous studies have shown that Arf-GAP1 acts as a lipid-packaging sensor and helps anchor Arf-GAP1 to the highly curved membrane surface, allowing GTP hydrolysis on Arf1 [69]. In addition, Arf-GAPs are essential components of cellular signaling cascades that regulate the actin cytoskeletal network and gene expression changes, normal cell proliferation, survival through cell adhesion, and growth factor stimulation [70]. In this study, we found that the LDs in the mutants had increased in volume and were unevenly distributed in hyphal cells. Both ∆Aogcs1 and ∆Aogts1 mutants showed sensitivity to H-89, suggesting that the activities of PKA signaling pathways were reduced or lost in the mutant strains. The transcriptional downregulation of the genes related to the PKA and TOR signaling pathway further verified the result that Arf-GAPs are involved in intracellular signal transduction. In addition, cAMP/PKA signaling has been proved to be involved in conidiation, secondary metabolism, and pathogenicity in diverse fungi [71]. Taken together, these results indicate that Arf-GAPs are important multifunctional regulators involved in membrane trafficking, coordination of lipid homeostasis, and intracellular signaling in A. oligospora and other fungi.
Arf regulates various membrane functions, including ER stress and the ER-mitochondrial encounter structure [11,72]. Mitochondria and ER are highly interacting organelles that contribute to hyphal growth, mitochondrial morphology, and virulence in the pathogenic mold Aspergillus fumigatus [73]. ATP production, signaling, apoptosis, and autophagy are all regulated by multifunctional regulators in the mitochondria [74]. In S. cerevisiae overexpressing Gcs1, the mitochondria were morphologically abnormal and functionally impaired, with unbranched tubules and large spherical structures; therefore, it is thought that Gcs1 may participate in the maintenance of mitochondrial morphology by organizing the actin cytoskeleton [75]. Our previous study demonstrated that deletion of glo3 enhanced autophagy in A. oligospora [51]. Similarly, our present study found that knockout of Aogts1 and Aogcs1, apart from affecting the ER stress response, DNA damage, and apoptosis, also significantly reduced MMP, which indicates impaired mitochondrial activity. In addition, the enhanced MDC fluorescence intensity, a significantly reduced green-to-red fluorescence ratio after JC-1 staining, upregulated autophagy-related gene expression, and increased autophagosomes observed by TEM in the mutants further indicate that Arf-GAPs play an essential role in the regulation of fungal autophagy and mitochondrial activity.
In this study, we identified two orthologous Arf-GAPs in A. oligospora. Our findings suggest that AoGcs1 and AoGts1 play multifunctional roles in the growth, development, and pathogenicity of this fungus. Knockout of Aogcs1 and Aogts1 in A. oligospora affected intracellular signaling associated with cAMP/PKA signaling, which regulates downstream transcription factors (such as Msn2), resulting in increased sensitivity to intracellular and extracellular stressors. The deletion of Aogcs1 and Aogts1 also altered the morphology and function of the nucleus and inhibited intracellular vesicle trafficking. Abnormal nuclear function causes an ER stress response, and defective ER function impairs the exchange and renewal of the membrane system between ER-Golgi and ER-mitochondria, resulting in a decrease in MMP and an increase in intracellular ROS and enhanced autophagy. In addition, the remarkable accumulation of LDs is indicative of Golgi deficiency, and the abnormally increased ROS will aggravate DNA damage. This combination results in slow growth of the fungus, increased sensitivity to stressors, and decreased sporulation and pathogenicity ( Figure 9). Figure 9. Schematic diagram of the regulation of AoGcs1 and AoGts1 in A. oligospora. In A. oligospora, AoGcs1 and AoGts1 are involved in intracellular signaling related to cAMP/PKA, which further regulates related transcription factors and plays a key role in vesicle trafficking, nuclear morphology, apoptosis, ER stress, mitochondrial activity, ROS scavenging, autophagy, and lipid droplet storage, ultimately regulating fungal phenotypic traits. LD, lipid droplet; Mito, mitochondria; ER, endoplasmic reticulum; MMP, mitochondrial membrane potential.

Conclusions
Our work showed that AoGcs1 and AoGts1 play a similar role in the growth, development, and pathogenicity of A. oligospora. These Arf-GAP proteins are involved in the regulation of DNA damage, endocytosis, ROS degradation, ER stress response, lipid storage, mitochondrial activity, and autophagy, and their deletion impairs mycelial growth, conidiation, stress responses, and nematode predation efficiency. Our results have expanded the understanding of the biological functions of Arf-GAP proteins in A. oligospora and provide novel insights into the role of Arf-GAPs in pathogenic fungal growth and development.
Supplementary Materials: The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/jof8050463/s1, Figure S1: Multiple sequence alignment, phylogenetic analysis, and transcript-level analysis of Gcs1 and Gts1; Figure S2: Knockout and verification of the genes Aogcs1 and Aogts1 in A. oligospora; Figure S3: Comparison of stress tolerance to cell wall stress agents; Figure S4: Colony diameters and RGI values of the WT and mutants cultured in the presence of 10 µM H-89 and 10 µM rapamycin; Table S1: List of primers for gene disruption used in this study; Table S2: Paired primers for RT-PCR analysis of genes associated with phenotypes such as conidiation, stress resistance, serine proteases, and autophagy in A. oligospora; Table S3: The sequence similarity between AoGsc1/AoGts1 and orthologs from different fungi.