Cytosol Peroxiredoxin and Cell Surface Catalase Differentially Respond to H2O2 Stress in Aspergillus nidulans

Both catalase and peroxiredoxin show high activities of H2O2 decomposition and coexist in the same organism; however, their division of labor in defense against H2O2 is unclear. We focused on the major peroxiredoxin (PrxA) and catalase (CatB) in Aspergillus nidulans at different growth stages to discriminate their antioxidant roles. The dormant conidia lacking PrxA showed sensitivity to high concentrations of H2O2 (>100 mM), revealing that PrxA is one of the important antioxidants in dormant conidia. Once the conidia began to swell and germinate, or further develop to young hyphae (9 h to old age), PrxA-deficient cells (ΔprxA) did not survive on plates containing H2O2 concentrations higher than 1 mM, indicating that PrxA is an indispensable antioxidant in the early growth stage. During these early growth stages, absence of CatB did not affect fungal resistance to either high (>1 mM) or low (<1 mM) concentrations of H2O2. In the mature hyphae stage (24 h to old age), however, CatB fulfills the major antioxidant function, especially against high doses of H2O2. PrxA is constitutively expressed throughout the lifespan, whereas CatB levels are low in the early growth stage of the cells developing from swelling conidia to early growth hyphae, providing a molecular basis for their different contributions to H2O2 resistance in different growth stages. Further enzyme activity and cellular localization analysis indicated that CatB needs to be secreted to be functionalized, and this process is confined to the growth stage of mature hyphae. Our results revealed differences in effectiveness and timelines of two primary anti-H2O2 enzymes in fungus.


Introduction
Catalases and peroxiredoxins (Prxs) are the two key antioxidants involved in H 2 O 2 detoxification. Catalases decompose two molecules of H 2 O 2 into H 2 O and oxygen [1]. Most Prxs reduce H 2 O 2 to H 2 O with the concomitant oxidation of two cysteine residues to form a disulfide bond [2,3]. The contribution of Prxs and catalases to H 2 O 2 resistance varies across organisms. In animals, almost all of H 2 O 2 is enzymatically detoxified by glutathione peroxidase (Gpx) and catalase is of little import [4]. Similarly, loss of Gpx, but not catalase, induced sensitivity to H 2 O 2 in Saccharomyces cerevisiae [5][6][7]. In Escherichia coli, the Prx AhpC efficiently scavenges low concentrations of H 2 O 2 , whereas catalase predominantly protects the cells at high H 2 O 2 concentrations [8,9]. In Schizosaccharomyces pombe, the Prx Tpx1 is the first line of defense, controlling H 2 O 2 generation during aerobic metabolism, whereas catalase does the same at high levels of H 2 O 2 [10]. Contrary to all these organisms, the catalase KatG is the primary detoxifier of H 2 O 2 produced during aerobic metabolism in Bradyrhizobium japonicum [11]. Clearly, the primary H 2 O 2 detoxifying enzyme varies in a species-specific manner.
The antioxidant systems in filamentous fungi have been intensively studied, especially in Aspergillus species. PrxA has been identified as an indispensable antioxidant among the Table S2 lists the plasmids used in this study. Genomic DNA was extracted from A. nidulans A6 using the Genomic DNA Purification Kit (Promega, Madison, WI, USA). All plasmids were constructed using the ClonExpress MultiS One Step Cloning Kit (Vazyme, Nanjing, China), and PCR was performed using the PrimeSTAR HS DNA Polymerase (Takara, Osaka, Japan). The primers designed for constructing recombinant plasmids are listed in Table S3. pUC19-pyrG is a modified pUC19 plasmid containing the selective marker gene pyrG, cloned from the genomic DNA of A. nidulans A6 into the multiple cloning site of XbaI. pUC19-GFP-trpC.T-pyrG is a pUC19-pyrG-based plasmid harboring the following sequences at the multiple cloning site: 5GA-linker-tagged green fluorescent protein (GFP) encoding sequence, trpC terminator, and pyrG. The pUC19-pyrG was linearized via an inverse PCR with the primer pair Inverse-1F/Inverse-1R. The DNA fragment for 5GA-linker-tagged GFP expression was cloned from our previously constructed plasmid pUC-GFP [14] with the primer pair GFP-F/GFP-R. The trpC terminator was cloned from the genomic DNA of A. nidulans A6 with the primer pair trpC.T-1F/trpC.T-1R. The three resulting DNA fragments were cyclized to construct pUC19-GFP-trpC.T-pyrG using the One Step Cloning Kit. pUC19-GFP-trpC.T-pyroA was constructed by an inverse PCR using the primer pair Inverse-2F/Inverse-2R and the plasmid template pUC19-GFP-trpC.T-pyrG. The marker gene pyroA was amplified using the primer pair pyroA-F/pyroA-R and the A6 genomic DNA as template. The two resulting PCR fragments were cyclized to construct pUC19-GFP-trpC.T-pyroA. pUC19-pyrG-catB.P-uidA-trpC.T is a plasmid harboring the following sequences at the multiple cloning site: pyrG, catB gene promoter (catB.P), reporter gene uidA, and trpC.T. catB.P and trpC.T were cloned from the genomic DNA of A. nidulans A6 with the primers catB.P-F/catB.P-R and trpC.T-2F/trpC.T-2R, respectively. uidA was cloned from E. coli genomic DNA with the primers uidA-F/uidA-R [19]. The resulting DNA fragments, with the linearized pUC19-pyrG, were used to construct pUC19-pyrG-catB.P-uidA-trpC.T. pUC19-pyrG-prxA.P-uidA-trpC.T and pUC19-pyrG-gpdA.P-uidA-trpC.T are two plasmids with an arrangement similar to pUC19-pyrG-catB.P-uidA-trpC.T except for the promoter components in the inserted DNA fragments. The two plasmids were constructed in the same way.
Plasmids for the generation of prxA and catB-complemented strains were constructed as follows: The prxA expression cassette (prxA com ) and catB expression cassette (catB com ) individually encompassing 1.5 kb of 5 -UTR and 1 kb of 3 -UTR of prxA and catB were amplified by PCR using the genomic DNA of A. nidulans A6 and the primer pairs prxA com -F/prxA com -R and catB com -F/catB com -R, respectively. The linearized pUC19-pyrG and prxA com as well catB com were individually cyclized to construct pUC19-prxA com -pyrG and pUC19-catB com -pyrG using the One Step Cloning Kit.

Construction of Recombinant Strains
The single guide RNAs (sgRNAs) were synthesized using the GeneArtTM Precision gRNA Synthesis Kit (Invitrogen, Carlsbad, CA, USA). A series of primer pairs, shown in Supplementary Table S4, were designed for the individual gRNA DNA templates assembled by PCR. Table S5 lists the primers designed for recombinant DNA cassette construction.
To construct the catB-GFP strain containing the catB-GFP expression DNA cassette at the original locus of catB, the DNA donor was amplified by PCR using pUC19-GFP-trpC.T-pyrG as a template with the primers catB-GFP-F/catB-GFP-R. The PCR product, the corresponding sgRNA, and 1 µg Cas9 were co-transformed into WT to obtain the catB-GFP strain. The prxA-GFP and catB-flag strains were constructed using the same method.

Confirmation of the Genomic Integration of the Target Genes in the Recombinant Strains
All recombinant strains were confirmed using PCR ( Figures S2, S3, S5-S7, S9 and S11). The corresponding primers are listed in Table S6. The copy numbers of the fusion genes catB.P-uidA, prxA.P-uidA, catB-GFP and prxA-GFP integrated in genomic DNA of the individual strains were investigated by Southern blotting as previously described [20,21]. The total genomic DNA of the fungal strains was isolated on 0.8% agarose gel via electrophoresis after digestion with the indicated restriction enzymes ( Figures S4 and S8) and subsequently transferred to a Hybond N+ membrane (GE Healthcare, Little Chalfont, UK). Hybridization probes were amplified with the primer sets listed in Table S7 and labeled using digoxigenin (DIG). Hybridization and signal detection were performed using the DIG High Prime DNA Labeling and Detection Starter Kit I (Roche Diagnostics, Basel, Switzerland), according to the manufacturer's instructions. The presence of a single copy of the integrated target gene was determined based on the estimated sizes of the DNA fragments ( Figures S4 and S8).

Conidia Prepared for H 2 O 2 Resistance Assays
Conidia from the 2-day cultures of each fungus on MM plates were collected in 15 mL sterilized centrifuge tube containing Tween saline (0.1% Tween-80 in 0.8% NaCl). After standing for 15 min, the supernatant conidial suspension was collected and concentrated by centrifugation at 5000 rpm/min for 10 min. The precipitate was resuspended in 2 mL of Tween-saline solution and placed in a hemocytometer (Neubauer chamber) to estimate the conidia concentration. Once the number of conidia had been estimated, they were diluted to the appropriate final concentrations and 10 µL of conidial suspension was plated onto plates containing the indicated concentrations of H 2 O 2 for resistance testing.

Quantification of Intracellular GFP Levels and β-Glucuronidase (GUS) Activity
Conidia were collected and suspended in Tween-saline solution and the suspension was standardized to a final concentration of about 1 × 10 7 conidia/mL. A conidia suspension of 100 µL (1 × 10 6 conidia) was spread on cellophane-coated minimal medium (MM) plates containing 1 mM H 2 O 2 and incubated for the indicated times. Mycelia were harvested along with cellophane, ground in liquid nitrogen, and resuspended in 500 µL PBS (pH 7.4). The supernatant of the disrupted mycelia was used for fluorescence detection. Protein concentration was determined using the Bradford assay, with bovine serum albumin as the standard. The total protein concentration was diluted to 1 mg/mL, and the GFP fluorescence intensity was measured using a fluorescence spectrophotometer (Hitachi, Tokyo, Japan) with excitation/emission wavelengths of 488/512 nm. For the GUS assay, cell lysates were resuspended in GUS assay buffer (50 mM sodium phosphate Antioxidants 2023, 12, 1333 5 of 17 buffer pH 7.0, 10 mM β-mercaptoethanol, 10 mM Na 2 EDTA, and 0.1% Triton X-100). After centrifugation, the supernatant was subjected to fluorometric analysis of GUS activity, as described previously [19]. Fluorometric analysis of GUS activity was performed using 4-methylumbelliferyl-b-glucuronide (4-MUG) as a substrate. GUS activity was determined upon the detection of 4-methylumbelliferone (4-MU) fluorochrome generated by the GUSmediated catalysis of 2 mM 4-MUG hydrolysis using a fluorescence spectrophotometer (HITACHI, Japan) with an excitation wavelength of 365 nm and an emission wavelength of 455 nm. GUS activity was defined as nmol of 4-MU per mg protein per min (abbreviated as mU/mg).

Quantitative Real-Time PCR
Total RNA and cDNA were extracted and prepared as previously described [14]. Quantitative PCR was performed using the SYBR Green PCR Kit (Toyobo, Osaka, Japan) on a CFX-96 Real-Time PCR system (Bio-Rad, Hercules, CA, USA). The primer pairs RT-catB-F/RT-catB-R and RT-actA-F/RT-actA-R (Table S8) were used to quantify the catB and actA genes, respectively. The relative mRNA levels were normalized to that of the reference gene actA using the 2 −∆∆Ct method of relative quantification [14,19]. The experiment was repeated thrice. The mean values ± SD of the three independent experiments were calculated based on one-way analysis of variance with Dunnett's post hoc test, which were used to identify the statistical differences (* p < 0.05; ** p < 0.01; and *** p < 0.001).

Cell Lysate Catalase Activity Assay
Conidia were spread on cellophane-coated solid plates containing 1 mM H 2 O 2 and incubated for the indicated times. Fungal cells were collected along with cellophane and ground by liquid nitrogen. The resulted cell debris was dissolved in Catalase Assay Buffer (Beyotime, Shanghai, China) for catalase assay. The crude cell lysate was quantified by the total soluble protein using the Bradford assay. The crude cell lysate containing the protein at the concentration of 0.2 mg/mL was used to determine the enzyme activity using the Catalase Assay Kit (Beyotime, Shanghai, China). This assay is based on colorimetrically measuring the hydrogen peroxide substrate remaining after the action of catalase. The colorimetric method uses a substituted phenol (3,5-dichloro-2-hydroxybenzenesulfonic acid), which couples oxidatively to 4-aminoantipyrine in the presence of hydrogen peroxide and horseradish peroxidase, to give a red quinoneimine dye (N-(4-antipyryl)-3-chloro-5sulfonatep-benzoquinone-monoimine) that absorbs at 520 nm. One unit of catalase activity was defined as the decomposition of 1 µmol of H 2 O 2 per minute at pH 7.0 and 25 • C.

Fluorescence Microscopy
Cells were prepared as described previously [13,14]. After cultivation, cells were washed twice with phosphate-buffered saline (PBS; pH 7.4) and observed under a confocal laser scanning microscope (TCS SP8; Leica, Wetzlar, Germany). The fluorescence of the GFP was excited using a 488 nm laser and detected at 510 nm. For imaging endoplasmic reticulum, cells were stained with ER-Tracker Red (Beyotime, Shanghai, China) for 20 min at 37 • C in the dark. The fluorescence of ER-Tracker was excited at 587 nm and emitted at 615 nm.

PrxA Plays a Role in H 2 O 2 Resistance in Dormant Conidia
Among the catalase family members, CatA appears to be a unique and indispensable enzyme for protecting conidia against H 2 O 2 [16]. However, it is unclear whether PrxA contributes to H 2 O 2 resistance in dormant conidia. Conidia from WT and PrxA gene disruption strain (∆prxA) were treated with a series of H 2 O 2 concentrations from 0 to 300 mM to investigate the performance of conidial PrxA. Very high concentrations of H 2 O 2 were used here because dormant conidia are extremely resistant to H 2 O 2 as described previously, but in fact the environmental H 2 O 2 generally does not exceed a few millimolar [22][23][24]. After a Antioxidants 2023, 12, 1333 6 of 17 20-min exposure, the dormant conidia were plated on MM plates and the colony numbers were used to determine the fungal H 2 O 2 tolerance. Conidia from ∆catA have been reported to be killed by a 20-min exposure of 100 mM H 2 O 2 [16]. The same treatment also caused adverse effects on the conidial survival of ∆prxA ( Figure 1A). With the H 2 O 2 concentrations increasing to 200 mM and 300 mM, the inhibitory effects were more aggravated by ∆prxA than WT ( Figure 1A). The H 2 O 2 -resistance ability of ∆prxA was restored to a level identical to that of WT via prxA re-complementation ( Figure S1A), indicating that conidial PrxA is involved in the resistance to extremely high H 2 O 2 concentrations. As a control, catB disruption strain (∆catB) and prxAcatB double disruption strain (∆prxA∆catB) were constructed ( Figure S2). H 2 O 2 sensitivities of both strains were also investigated, and showed a similar performance to WT and ∆prxA, respectively ( Figure 1A), excluding conidial CatB in H 2 O 2 resistance participation, which is in agreement with the experimental results from the previous literature [15,17]. Thus, this study's results revealed that, besides the previously characterized CatA, PrxA functions as another H 2 O 2 defender in the dormant conidia. ruption strain (prxA) were treated with a series of H2O2 concentrations from 0 to 300 mM to investigate the performance of conidial PrxA. Very high concentrations of H2O2 were used here because dormant conidia are extremely resistant to H2O2 as described previously, but in fact the environmental H2O2 generally does not exceed a few millimolar [22][23][24]. After a 20-min exposure, the dormant conidia were plated on MM plates and the colony numbers were used to determine the fungal H2O2 tolerance. Conidia from catA have been reported to be killed by a 20-min exposure of 100 mM H2O2 [16]. The same treatment also caused adverse effects on the conidial survival of prxA ( Figure 1A). With the H2O2 concentrations increasing to 200 mM and 300 mM, the inhibitory effects were more aggravated by prxA than WT ( Figure 1A). The H2O2-resistance ability of prxA was restored to a level identical to that of WT via prxA re-complementation ( Figure S1A), indicating that conidial PrxA is involved in the resistance to extremely high H2O2 concentrations. As a control, catB disruption strain (catB) and prxAcatB double disruption strain (prxAcatB) were constructed ( Figure S2). H2O2 sensitivities of both strains were also investigated, and showed a similar performance to WT and prxA, respectively ( Figure 1A), excluding conidial CatB in H2O2 resistance participation, which is in agreement with the experimental results from the previous literature [15,17]. Thus, this study's results revealed that, besides the previously characterized CatA, PrxA functions as another H2O2 defender in the dormant conidia.

PrxA and CatB Play Nonredundant Roles in Defense against H 2 O 2 at Different Growth Stages
Our previous study has reported that ∆prxA conidia spotted on MM plates containing H 2 O 2 exceeding 1 mM could not form surviving colonies [12,14], indicating that PrxA Antioxidants 2023, 12, 1333 7 of 17 may also protect against H 2 O 2 in the post dormant conidia stage. CatB activity has been reported to be developmentally regulated, was barely detectable in conidia, and started to accumulate 10 h after conidia inoculation [15,17]. Accordingly, it can be speculated that PrxA and CatB may functionally or cooperatively overlap in protecting fungal cells from H 2 O 2 damage during the post dormant conidia stage, including conidia swelling and germination, young hyphae development, hyphae maturation, and conidiation. The H 2 O 2 strains sensitivities lacking either or both PrxA and CatB at different growth stages were investigated to discriminate PrxA and CatB functions during the post dormant conidia stage.
To apply H 2 O 2 stress to swelling and germinating conidia, we added the dormant conidia into top agar premixed with the indicated concentrations of H 2 O 2 and plated them onto the precast MM plates to observe the surviving colonies ( Figure 1B). Considering the extreme tolerance of dormant conidia to high-dose H 2 O 2 as shown in Figure 1A and other reports [16], we assumed that this operation could make the conidia encounter a relatively low dose of H 2 O 2 stress only when swelling and germination occur. After 60 h of cultivation, we found that ∆prxA strains could not survive on the plates containing >1 mM H 2 O 2 , whereas ∆catB strains showed the same H 2 O 2 resistance as WT under both low (1 mM) and high (5 mM) stress ( Figure 1B), indicating the central role of PrxA, but not CatB, in H 2 O 2 defense at this stage. ∆prxA∆catB strains showed the same H 2 O 2 sensitivity as ∆prxA strains ( Figure 1B), further confirming the nonfunctional role of CatB during the conidia swelling and germination stages.
Next, we evaluated the contributions of PrxA and CatB in the growth phase of young hyphae (6-12 h). Conidia were mixed in H 2 O 2 -free top agar and poured on MM plates followed by another layer of top agar containing the indicated doses of H 2 O 2 ( Figure 1C). This operation ensured only the young hyphae just extended from the lower agar in to the upper top agar to be exposed to H 2 O 2 stress. After 60 h of cultivation, the ∆prxA and ∆prxA∆catB strains did not form any colonies under a stress higher than 1 mM H 2 O 2 , whereas the ∆catB strains were insensitive to 1 mM or 2 mM H 2 O 2 ( Figure 1C), indicating that PrxA was indispensable in protecting young hyphae against H 2 O 2 . Once H 2 O 2 concentrations exceeding 5 mM, ∆catB strains began showing higher sensitivities to H 2 O 2 than WT ( Figure 1C), suggesting the emergence of a functional CatB along in the young hyphae under high H 2 O 2 stress.
Next, we examined the protection offered by PrxA and CatB in mature hyphae (>18 h). The conidia in H 2 O 2 -free top agar were incubated on MM plates for 24 h, and new H 2 O 2 -containing top agar was poured to provide oxidative stress to the growing hyphae ( Figure 1D). After incubating for another 24 h, colonies of the individual strains were compared. The H 2 O 2 tolerance of all strains was significantly enhanced at this stage compared with the two preceding stages; notably, the WT could even partially survive under 15 mM H 2 O 2 stress ( Figure 1D). ∆catB strains exhibited slightly decreased resistance to oxidative stress at 2 mM H 2 O 2 , and the adverse effects became more significant at 5 mM and 15 mM H 2 O 2 , indicating that the role of CatB in H 2 O 2 resistance increased in significance as the hyphae grew. Re-introduction of catB into ∆catB (∆catB-catB com ) restored the H 2 O 2 -resistance ability of the mature hyphae of ∆catB ( Figure S1B), further confirming the functions of CatB in the cells at the mature hypha stage. Interestingly, compared with WT, ∆prxA strains did not show more sensitivity to any concentrations of H 2 O 2 at this stage ( Figure 1D), which is contrary to the findings at the earlier growth stages ( Figure 1B,C). Therefore, we demonstrated that CatB primarily protected against H 2 O 2 in mature hyphae, whereas PrxA only played a minor role, especially at H 2 O 2 concentrations higher than 5 mM. The ∆prxA∆catB strains, such as WT, showed robust tolerance to 1 mM H 2 O 2 stress, and partially survived even at 2 mM and 5 mM H 2 O 2 ( Figure 1D), which is quite different from the sensitivities of the ∆prxA∆catB observed during swelling and germinating stages at those concentrations of H 2 O 2 as shown in Figure 1B,C. Thus, besides PrxA and CatB, an additional, unknown H 2 O 2 tolerance system must be supporting cellular growth at this stage. Based on the phenotypes of these strains, we concluded that at the growth stage of mature hyphae, both PrxA and CatB contribute to resistance to H 2 O 2 , with CatB playing the leading role.
After the mature hyphae growth stage, conidia will be formed on the specialized aerial hyphae, confounding the hyphae and conidia. In consideration of the difference in performance between the hyphae and the conidia, the cellular H 2 O 2 tolerance after the mature hyphae stage was no longer characterized.

Expression Profiles of PrxA and CatB Were Correlated with Their H 2 O 2 Resistance
We predicted that the different intracellular levels of PrxA and CatB may be responsible for their varying contribution to H 2 O 2 resistance at different stages of cell growth. To verify this, we constructed catB.P-uidA and prxA.P-uidA strains in which uidA, the β-glucuronidase (GUS) encoding gene [19], was expressed under the promoters of catB and prxA, respectively ( Figure S3). GUS activities of catB.P-uidA and prxA.P-uidA strains were used to investigate the promoter activities of both genes during the growth stages from dormant conidia to mature hyphae. The single copy of uidA was precisely inserted into the indicated genomic locus of catB.P-uidA and prxA.P-uidA strains as confirmed via Southern blotting ( Figure S4). As shown in Figure 2A, the cell extracts of H 2 O 2 -induced catB.P-uidA expression strains exhibited fluctuating GUS activities: they dropped to a minimum at about 6 h, increased to a high level within the following 10 h, and remained stable for the next several hours, which is consistent with the CatB transcription pattern observed in previous literature [17]. Conversely, the prxA.P-uidA strain extracts showed a relatively constant and high level of GUS activity during all growth stages ( Figure 2B). Together with the different H 2 O 2 -resistance abilities of ∆catB and ∆prxA at different growth stages (Figure 1), it seemed possible that the expression levels of PrxA and CatB directly determined their H 2 O 2 resistance. different from the sensitivities of the prxAcatB observed during swelling and germinat ing stages at those concentrations of H2O2 as shown in Figure 1B,C. Thus, besides PrxA and CatB, an additional, unknown H2O2 tolerance system must be supporting cellula growth at this stage. Based on the phenotypes of these strains, we concluded that at th growth stage of mature hyphae, both PrxA and CatB contribute to resistance to H2O2, with CatB playing the leading role.
After the mature hyphae growth stage, conidia will be formed on the specialized aer ial hyphae, confounding the hyphae and conidia. In consideration of the difference in per formance between the hyphae and the conidia, the cellular H2O2 tolerance after the matur hyphae stage was no longer characterized.

Expression Profiles of PrxA and CatB Were Correlated with Their H2O2 Resistance
We predicted that the different intracellular levels of PrxA and CatB may be respon sible for their varying contribution to H2O2 resistance at different stages of cell growth. To verify this, we constructed catB.P-uidA and prxA.P-uidA strains in which uidA, the β-glu curonidase (GUS) encoding gene [19], was expressed under the promoters of catB and prxA, respectively ( Figure S3). GUS activities of catB.P-uidA and prxA.P-uidA strains wer used to investigate the promoter activities of both genes during the growth stages from dormant conidia to mature hyphae. The single copy of uidA was precisely inserted into the indicated genomic locus of catB.P-uidA and prxA.P-uidA strains as confirmed vi Southern blotting ( Figure S4). As shown in Figure 2A, the cell extracts of H2O2-induced catB.P-uidA expression strains exhibited fluctuating GUS activities: they dropped to a min imum at about 6 h, increased to a high level within the following 10 h, and remained stabl for the next several hours, which is consistent with the CatB transcription pattern ob served in previous literature [17]. Conversely, the prxA.P-uidA strain extracts showed relatively constant and high level of GUS activity during all growth stages ( Figure 2B) Together with the different H2O2-resistance abilities of catB and prxA at differen growth stages (Figure 1), it seemed possible that the expression levels of PrxA and CatB directly determined their H2O2 resistance. The conidia of the strain were cultivated at 37 °C on cellophane-coated plates containing 1 mM H2O2. The strains were col lected at the indicated time and disrupted using liquid nitrogen. The cell lysate (1 mg/mL) was used to measure the GUS activity (n = 3). Results were presented as mean ± SD and analyzed using one way ANOVA (N.S., not significant, * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001). The strains were collected at the indicated time and disrupted using liquid nitrogen. The cell lysate (1 mg/mL) was used to measure the GUS activity (n = 3). Results were presented as mean ± SD and analyzed using one-way ANOVA (N.S., not significant, * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001).

Overexpressing CatB Cannot Enhance H 2 O 2 Resistance of the Swelling and Germinating Conidia
The above results indicate that the nonfunction of CatB in H 2 O 2 resistance during the conidia swelling and germination stages would be due to its extremely low expression level, raising an intriguing question: would fungal H 2 O 2 resistance increase if CatB expression was promoted during these stages? To answer this question, we overexpressed CatB by replacing its native promoter with the stronger and constant prxA promoter and investigated the H 2 O 2 resistance of the swelling and germinating conidia of CatB-overexpression strains by spotting the conidia onto MM plates containing the indicated concentrations of H 2 O 2 ( Figure 3A). Using ∆prxA∆catB as the background strain, the prxA promoter-dependent CatB expression strain (prxA.P-catB/∆∆) was constructed ( Figure S5). Unexpectedly, this strain exhibited only a slightly higher resistance to 0.5 mM H 2 O 2 than its parent strain ∆prxA∆catB ( Figure 3A), despite catB transcriptional level being markedly enhanced (Figure 3B). Next, we tried to further overexpress CatB by placing the gene under the gpdA promoter, a strong constitutive promoter derived from the A. nidulans gpdA gene, which encodes glyceraldehyde-3-phosphate dehydrogenase ( Figure S5) [25]. The catB transcriptional level was further enhanced in the gpdA.P-catB/∆∆ strain, irrespective of the presence of H 2 O 2 ( Figure 3B). Nevertheless, as with the prxA.P-catB/∆∆ strain, the overexpression of CatB in the gpdA.P-catB/∆∆ strain could not compensate for the lack of H 2 O 2 resistance caused by the absence of PrxA ( Figure 3A). Moreover, overexpressing CatB by gpdA promoter in WT did not improve the H 2 O 2 resistance of the strain in its swelling and germinating conidia state ( Figures S5 and 3A), clearly indicating that CatB does not protect the swelling and germinating conidia against H 2 O 2 . To investigate whether CatB functionalization was associated with the growth phase, we compared the H 2 O 2 -tolerance of the CatB-overexpression strain and WT in their mature hyphae stage ( Figure 3C). We observed that both WT and ∆prxA∆catB mutants, under 15 mM H 2 O 2 stress, survived significantly better when CatB was overexpressed ( Figure 3C), confirming that CatB functionalization confined to the mature hyphae stage.
The above results indicate that the nonfunction of CatB in H2O2 resistance during the conidia swelling and germination stages would be due to its extremely low expression level, raising an intriguing question: would fungal H2O2 resistance increase if CatB expression was promoted during these stages? To answer this question, we overexpressed CatB by replacing its native promoter with the stronger and constant prxA promoter and investigated the H2O2 resistance of the swelling and germinating conidia of CatB-overexpression strains by spotting the conidia onto MM plates containing the indicated concentrations of H2O2 ( Figure 3A). Using prxAcatB as the background strain, the prxA promoterdependent CatB expression strain (prxA.P-catB/) was constructed ( Figure S5). Unexpectedly, this strain exhibited only a slightly higher resistance to 0.5 mM H2O2 than its parent strain prxAcatB ( Figure 3A), despite catB transcriptional level being markedly enhanced ( Figure 3B). Next, we tried to further overexpress CatB by placing the gene under the gpdA promoter, a strong constitutive promoter derived from the A. nidulans gpdA gene, which encodes glyceraldehyde-3-phosphate dehydrogenase ( Figure S5) [25]. The catB transcriptional level was further enhanced in the gpdA.P-catB/ strain, irrespective of the presence of H2O2 ( Figure 3B). Nevertheless, as with the prxA.P-catB/ strain, the overexpression of CatB in the gpdA.P-catB/ strain could not compensate for the lack of H2O2 resistance caused by the absence of PrxA ( Figure 3A). Moreover, overexpressing CatB by gpdA promoter in WT did not improve the H2O2 resistance of the strain in its swelling and germinating conidia state ( Figures S5 and 3A), clearly indicating that CatB does not protect the swelling and germinating conidia against H2O2. To investigate whether CatB functionalization was associated with the growth phase, we compared the H2O2-tolerance of the CatB-overexpression strain and WT in their mature hyphae stage ( Figure 3C). We observed that both WT and prxAcatB mutants, under 15 mM H2O2 stress, survived significantly better when CatB was overexpressed ( Figure 3C), confirming that CatB functionalization confined to the mature hyphae stage. stress conditions. After preculturing conidia on MM plates containing the indicated concentrations of H 2 O 2 for 6 h, the resulting cells on the cellophane were collected for real-time PCR. The transcription level of catB in WT without H 2 O 2 treatment was set to 1 (mean ± SD; n = 3, *** p < 0.001, **** p < 0.0001, one-way ANOVA). (C) Overexpression of CatB conferred resistance to mature mycelia against high concentrations of H 2 O 2 . The newly constructed strain is as follows: catB.P-catB/∆∆, introducing the catB gene under its native promoter into ∆prxA∆catB. Conidia of all strains were premixed with top agar without H 2 O 2 and poured on MM plates. After mycelia were visible (approximate 24 h), new top agar containing the indicated concentrations of H 2 O 2 was poured to subject the mycelia to oxidative stress. The colonies formed were used to evaluate their H 2 O 2 resistance.

Extracellular Secretion of CatB Is Necessary for H 2 O 2 Resistance
Next, we investigated the reasons underlying the distinct outcomes of overexpressing CatB during the different growth stages. We first tested whether CatB exhibited different activities at the different growth stages by comparing the total catalase activities in the crude cell lysates of CatB-overexpression strains cultivated on H 2 O 2 -containing MM plates for 0, 3, 6, 12, and 24 h. As a control, catalase activity in WT strains was also measured. As shown in Figure 4A, the level of catalase activity in WT between 0 and 24 h had a special change of descend first then ascend, consistent with the promoter activity change of CatB during this time period (Figure 2A). However, in cell lysates of CatB-overexpression strains, the catalase activities were relatively constant and far higher than in the WT cells at any growth stages ( Figure 4A). These results suggest that the activity change of CatB was not responsible for its distinct contribution to H 2 O 2 resistance during different growth stages.
We observed that bubbles were produced upon dipping gpdA.P-catB/WT cells of different growth stages in water containing 15 mM H 2 O 2 ( Figure 4B). The bubbles should be due to the rapidly released O 2 derived from the decomposition of H 2 O 2 catalyzed by catalase, which can be monitored and calculated by the absorbance decreases at 240 nm ( Figure 4B, on the bottom). Interestingly, many more bubbles were produced by the hyphae of 12-h strain age than the cells of 3-h strain age ( Figure 4B), though the total catalase activities of the two cells' lysate showed the same levels ( Figure 4A). Thus, more O 2 bubbles may suggest that more CatB was secreted extracellularly in these cells. Rinsing the 12-h strain age mycelia in phosphate-buffered saline ended the rapid decomposition of H 2 O 2 , whereas the eluent showed high catalase activity ( Figure 4B, on the bottom). Based on these observations, we deduced that the functional catalase would be distributed on the cell surface and linked to the cells via noncovalent bonds. To verify this, we constructed a strain overexpressing CatB*, which is a truncated CatB lacking the putative secretory signal peptide (the first 26 amino acids at the N-terminus) ( Figure S6) [26,27]. As predicted, the mature hyphae of the truncated CatB overexpressing strains decomposed H 2 O 2 only to the extent of WT strains, losing the high catalase activity ( Figure 4C). In accordance, overexpressing the truncated CatB did not promote the fungal H 2 O 2 resistance ( Figure 4D). These results emphasize that CatB must be secreted for its functionalization and also imply that CatB secretion is confined in the late growth stage.

Functionalized PrxA and CatB Localize in Cytosol and Cell Surface, Respectively
Based on the above results, we speculated that PrxA and CatB defend against H 2 O 2 intracellularly and extracellularly, respectively. To verify this, the sensitivities of ∆prxA and ∆catB to endogenous H 2 O 2 were investigated. DAAO catalyzes the conversion of D-amino acids to their corresponding alpha-keto acids, producing H 2 O 2 in the process. DAAO from the yeast R. gracilis has been successfully applied to specifically generate endogenous H 2 O 2 in mammalian cells [28,29]. Here, we attempted to use this system in A. nidulans for the generation of endogenous H 2 O 2 . Expression of DAAO in WT (DAAO/WT) and ∆catB (DAAO/∆catB) had no influence on the growth of the spotted conidia of both strains exposed to various concentrations of D-Ala ( Figure 5A). In contrast, growth inhibition was detected on the DAAO-expressing ∆prxA (DAAO/∆prxA) and ∆prxA∆catB (DAAO/∆prxA∆catB) along with the increase in D-Ala concentrations in the media, confirming our speculation that PrxA rather than CatB counters the production of endogenous H 2 O 2 ( Figure 5A). Next, we investigated the cellular locations of the functionalized PrxA and CatB by the newly constructed GFP-tagged PrxA and CatB expression strains (prxA-GFP and catB-GFP) ( Figure S7). The single copy of GFP was precisely inserted into the indicated genomic locus of prxA-GFP and catB-GFP strains as confirmed via Southern blotting ( Figure S8). The fluorescence of PrxA-GFP diffused through all cells of the swelling and germinating conidia ( Figure 5B), indicating that PrxA localizes to cytosol. The fluorescence of CatB-GFP in the mature hyphae was distributed as discontinuous spots ( Figure 5C), indicating that it should localize to specific organelles. CatB-GFP partially coincided with the dyed endoplasmic reticulum (ER), indicating its ER retention ( Figure 5C). However, no fluorescence was observed at the surface of hyphae, suggesting that CatB-GFP was not secreted to the cell wall, which does not agree with the predicted location of functional CatB. We speculated that the GFP tag may sterically hinder the secretion of CatB-GFP, leading to its accumulation in the ER. The CatB-GFP expression strains could not survive as did WT on plates containing 15 mM H 2 O 2 ( Figure 5C), further supporting our speculation. Contrary to CatB-GFP, the cytosol should be the right location of PrxA-GFP, since cytosolic PrxA-GFP defended against H 2 O 2 just as PrxA did ( Figure 5B).
To confirm the secretion and cell surface immobilization of CatB, we replaced the C-terminal GFP with a FLAG tag ( Figure S7), which was expected to not impede CatB secretion and also make the tagged protein detectable via western blotting. As shown in Figure 5D, CatB-FLAG and CatB similarly protected mature hyphae against 15 mM H 2 O 2 , suggesting the extracellular secretion of CatB-FLAG. To directly detect the secreted CatB-FLAG on the cell surface, the mature hyphae were rinsed, and the washed eluate was subjected to western blotting using an anti-FLAG antibody. As expected, CatB-FLAG was detected in the washed eluate ( Figure 5D). The cell lysate from the rinsed cells was also analyzed. We found that the cell lysate had slightly lower levels of CatB-FLAG than the same volume of washed eluate ( Figure 5D), indicating that more CatB was secreted. were covered with top agar containing 0 or 15 mM H2O2 for comparing the H2O2 resistance. The intracellular residual CatB-FLAG and the extracellularly secreted CatB-FLAG were confirmed by western blotting using an anti-FLAG monoclonal antibody. The mycelia from a plate were collected and well-rinsed with PBS. The eluent was centrifuged to 1 mL and 10 μL of the eluent was loaded for analysis. The rinsed mycelia were disrupted by liquid nitrogen and the cell extract was modulated up to 1 mL, of which 10 μL was loaded for analysis. The samples from WT were used to test the specificity of the anti-FLAG antibody. The intracellular residual CatB-FLAG and the extracellularly secreted CatB-FLAG were confirmed by western blotting using an anti-FLAG monoclonal antibody. The mycelia from a plate were collected and well-rinsed with PBS. The eluent was centrifuged to 1 mL and 10 µL of the eluent was loaded for analysis. The rinsed mycelia were disrupted by liquid nitrogen and the cell extract was modulated up to 1 mL, of which 10 µL was loaded for analysis. The samples from WT were used to test the specificity of the anti-FLAG antibody.

Cytosolic Localized CatB Cannot Functionally Substitute PrxA at Germinating Conidia Stage
As shown in Figure 3A, CatB's intracellular abundance in gpdA.P-catB/∆∆ did not show any improved antioxidant ability comparing to its parent strain ∆prxA∆catB (∆∆) at the swelling and germinating conidia stages, leading us to hypothesize the specific CatB organelle-location may be a barrier for its antioxidant function. To verify that, we attempted translocating CatB from ER to cytosol to observe fungal resistance changes to H 2 O 2 . We overexpressed a GFP-fused N-terminal signaling peptide absent CatB (CatB*-GFP) in ∆prxA∆catB to construct the strain gpdA.P-catB*-GFP/∆∆ ( Figure S9). Compared with the WT strain expressing GFP-tagged CatB (CatB-GFP), the fluorescence in the swelling and germinating conidia of gpdA.P-catB*-GFP/∆∆ strain was greatly enhanced and diffused in the whole fungal cell ( Figure S10A), indicating the truncated CatB was successfully overexpressed in the germinating conidia's cytosol. However, CatB*-GFP expression strain H 2 O 2 tolerance was similar to the signal peptides retaining CatB-overexpression strains and was far inferior to that of the PrxA-overexpression strains (gpdA.P-prxA-GFP/∆∆) ( Figure S10B,C), which was constructed by replacing prxA promoter with gpdA promoter in ∆catB ( Figure S11). These results suggested that the reason CatB cannot substitute PrxA for H 2 O 2 resistance is not due to their different cellular locations.

Discussion
Both Prxs and catalases possess H 2 O 2 -decomposing activities. However, the exact reason underlying the existence of two functionally duplicated systems in one organism remains unclear. Our present study successfully discriminated the antioxidant functions of A. nidulans PrxA and CatB. During the early growth stages, including conidial swelling, germination, and the young hyphae, the fungus can only resist low concentrations of H 2 O 2 and PrxA acts as an indispensable antioxidant in this period. The mature hyphae can resist high concentrations of H 2 O 2 , mainly due to CatB secreted and immobilized on the cell wall. We verified CatB secretion is confined at the mature hyphae growth stage. Therefore, the antioxidant network in the model fungus A. nidulans involves an asynchronous collaboration between Prx and catalase to defend against H 2 O 2 .
The abundant expression of CatB is highly correlated with its protective function under H 2 O 2 stress. However, fungal H 2 O 2 resistance is not determined by CatB expression levels exclusively, because constant high expression of CatB conferred high H 2 O 2 resistance upon mature hyphae but not young hyphae ( Figure 3A,C). Furthermore, our results suggest that the rapid decomposition of H 2 O 2 around the mature hyphae by the secreted CatB relieved the oxidative stress, and this secretion ability may be lacking in younger cells ( Figure 4B). The timing of CatB secretion seems to be important for its function, which must be tightly regulated. Protein secretion is a complicated and energy-consuming process that involves vesicle biogenesis, cargo loading, concentration and processing, vesicle transport and targeting, vesicle docking, and Ca 2+ -dependent vesicular fusion with the plasma membrane [30,31]. The forced secretion of CatB during the early growth phase might exhaust energy and occupy the secretory routes of other essential proteins involved in nutrition intake or cell wall synthesis, eventually leading to an intracellular traffic jam of protein cargo. Since the cellular machinery for CatB secretion is not fully equipped, suppressing CatB expression during the early growth stage might be a behavior favored by natural selection for A. nidulans.
The Prx and catalase of S. pombe have also been intensively investigated and functionally discriminated for their roles in H 2 O 2 scavenging. The yeast Prx Tpx1 defends against low levels of intracellular H 2 O 2 produced during aerobic growth, whereas the yeast catalase chiefly scavenges extracellular H 2 O 2 [10,32]. By contrast, A. nidulans ∆prxA does not show O 2 sensitivity and both PrxA and CatB are required to defend against exogenous H 2 O 2 . Another notable antioxidant system distinction between A. nidulans and S. pombe is the different activation periods of their individual catalase. The A. nidulans CatB activation is confined in the mature hyphae growth stage; whereas the S. pombe catalase may stay active in the early growth stages, since catalase-deficient S. pombe seeded on the plate containing 1 mM H 2 O 2 were nonviable [10]. The close resemblance between the A. nidulans and S. pombe antioxidant systems is that the catalases' overexpression cannot suppress the growth defects caused by the absences of their individual Prx under the exogenous H 2 O 2 stress conditions ( Figure S2B) [10]. Prxs' nonsubstitutability by catalases in H 2 O 2 tolerance in the two fungal species suggested the other functions of Prx, such as molecular chaperoning and signaling, may also be indispensable during the antioxidant process.
We observed an interesting phenomenon when we compared the phenotypes of ∆prxA∆catB strains at different growth stages: 1 mM H 2 O 2 was lethal to cells in the early stage but did not exert any stress on the late-growth hyphae, which could even partially survive 5 mM H 2 O 2 ( Figure 1B-D). This suggests the possibility of other important H 2 O 2resistant systems that are activated only at the later growth stage. These unknown systems may include the other previously reported multiple peroxidases and catalases. Although single and multiple gene mutants of these enzymes had no effect on their function in the early growth stage [33], they may function in the late growth stage, especially cooperatively. The distinct cellular permeability to H 2 O 2 exhibited by hyphae in the two growth stages may also result in their different H 2 O 2 resistance. H 2 O 2 does not freely diffuse into human, E. coli, and S. cerevisiae cells; it is a regulated biological process [9,[34][35][36]. In S. cerevisiae, limiting H 2 O 2 diffusion by actively decreasing the cellular permeability is an important protective strategy against extracellular H 2 O 2 [35,37]. The cellular permeability to H 2 O 2 is five-fold lower in the stationary phase than in the exponential phase. Consequently, when challenged with H 2 O 2 , more S. cerevisiae cells in the stationary phase survive than those in the exponential phase, similar to A. nidulans. The cell wall and plasma membrane of yeast constitute the permeability barriers to H 2 O 2 [35,36], which may be common for fungi due to their similar cell structures. Future studies should focus on identifying the key components contributing to the barrier function against H 2 O 2 as well as the molecular mechanisms by which these factors are regulated in different growth phases.

Conclusions
The major peroxiredoxin (PrxA) and the key catalase (CatB) of Aspergillus nidulans play nonredundant roles in H 2 O 2 resistance. PrxA is indispensable for protecting against H 2 O 2 -induced damage in the conidia or young hyphae but not in mature hyphae. By contrast, CatB only performs an antioxidant role in mature hyphae. Functionalization of CatB requires it to be secreted and immobilized on the cell wall, both of which are confined to the mature hyphae stage. Therefore, fungal catalase and peroxiredoxin have a distinct division of labor in counteracting H 2 O 2 .
Supplementary Materials: The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/antiox12071333/s1. Figure S1: Re-introduction of prxA and catB to their gene-deletion strains recovered their H 2 O 2 -resstance abilities identical to WT level; Figure S2: Disruption of catB gene in ∆prxA; Figure S3: Construction of the promoter reporter strains of A. nidulans; Figure S4: Identification by Southern blotting analyses of A. nidulans transformants carrying the catB.P-uidA or prxA.P-uidA genotype; Figure S5: Construction of promoter substitution strains in A. nidulans; Figure S6: Construction of overexpression strains lacking signal peptides in A. nidulans; Figure S7: Construction of catB-GFP, catB-flag, and prxA-GFP expression strains of A. nidulans; Figure S8: Identification by Southern blotting analyses of A. nidulans transformants carrying the catB-GFP or prxA-GFP genotype; Figure S9: Construction of the strains overexpressing CatB* or CatB*-GFP in ∆prxA∆catB; Figure S10: Cytosolic CatB cannot functionally substitute PrxA at the swelling and germinating conidia stage; Figure S11: Construction of prxA promoter was replaced by gpdA promoter strains in A. nidulans; Table S1: Strains used in this study; Table S2: Plasmids used in this study; Table S3: Primers used for plasmids construction; Table S4: Primers used for sgRNA synthesis; Table S5: Primers used for strains construction; Table S6: Primers used for strains verification; Table S7: Primers used for probe synthesis; Table S8