We analyzed various strains of Aspergillus nidulans (Supplementary Table S1). Mitochondria in the SRS29 (Suelmann and Fischer, 2000) strain were assessed using fluorescence microscopy. Aspergillus nidulans strains were cultivated in Minimal Medium (MM; 10 g/L glucose, 6 g/L NaNO₃ or 1.84 g/L ammonium tartrate, 0.52 g/L KCl, 0.52 g/L MgSO₄·7H₂O, 1.52 g/L KH₂PO₄, 1 mL/L Hutner’s trace element solution, pH 6.5). The pH of the medium was adjusted using 0.1 M NaOH. We added 0.05 mg/L pyridoxine, 1.22 g/L uracil +1.21 g/L uridine to the medium to cultivate the TN02A3 strain (Nayak et al., 2006). Sodium nitrite was added to sterilized medium, then the pH was adjusted to 5.5. Plasmids were manipulated and proteins were generated in Escherichia coli DH5α and BL21 (DE3) strains cultured in LB medium (5 g/L yeast extract, 10 g/L tryptone, 10 g/L NaCl).

The gpdC (Gene ID, AN2583), pdcA (AN4888), alcC (AN2286), and dldA (AN9066; S. cerevisiae DLD1 ortholog) genes were replaced with the pyrG gene in A. nidulans TN02A3. We amplified DNA fragments encoding the 5′-and 3′-regions of the gpdC, pdcA, alcC, and dldA genes by PCR using the primer sets xxxX-FC/xxxX-R1, and xxxX-F3/xxxX-RC (“xxxX” represent gene names; Supplementary Table S2). Fragments of pyrG DNA were amplified by PCR using the primers pyrG-F and pyrG-R. The PCR template was genomic DNA of A. nidulans FGSC A4. Amplified fragments were fused by PCR using the primers xxxX-F1 and xxxX-R3, and transformed into A. nidulans TN02A3. Transformants were grown on MM agar. Gene disruption was confirmed by PCR using the primers xxxX-FC and xxxX-RC. Resulting disruptants were designated GpdCΔ, PdcAΔ, AlcCΔ, and DldAΔ (Supplementary Figure S1). The NPC1 strain constructed by introducing a pyrG gene fragment at the chromosomal pyrG89 loci of the TN02A3 strain was the control (Supplementary Table S1).

We generated DNA fragments encoding the 5′-regions of gpdC, ORF of gpdC, and the 3′-region of pyrG by PCR using the primer sets; gpdC-FC/gpdCcomp-R1, gpdCcomp-F2/gpdCcomp-R2, and gpdCcomp-F4/gpdCcomp-RC, respectively (Supplementary Table S2). Genomic DNA of A. nidulans FGSC A4 was the template. The ptrA gene was a selection marker and corresponding DNA fragments were amplified using the primers gpdCcomp-F3/gpdCcomp-R3, and pPTRI (Kubodera et al., 2002). Resultant amplicons were fused by PCR using the primers gpdC-F1 and gpdCcomp-R4 (Supplementary Table S2), and transformed into the GpdCΔ strain. Transformants were cultivated on MM agar containing 0.1 μg/mL pyrithiamine. That gpdC was transduced into the target locus to generate the GpdCΔ+gpdC strain was confirmed by PCR using the primers gpdC-FC and gpdCcomp-RC (Supplementary Figure S2).

Conidia (2 × 10⁷) of the A. nidulans FGSC A4 strain were rotary shaken in 100 mL of MM at 60 rpm and 37°C for 14 h. Total RNA was extracted from powdered mycelia frozen in liquid nitrogen using RNeasy Plant Mini kits (Qiagen, Hilden, Germany), and cDNA was synthesized using PrimeScript RT Master Mix (Takara Bio, Shiga, Japan) as described by the manufacturer. Complementary DNA fragments for gpdA were generated by PCR using fungal cDNA and pET21a-gpdA-F and pET21a-gpdA-R primers (Supplementary Table S2). Complementary DNA fragments for gpdC (synthesized at Eurofins Tokyo, Japan) with optimized codons for E. coli, were digested by Nde I and Hind III, then cloned into pET-21a using Ligation High ver. 2 (Toyobo, Osaka, Japan).

Transformants harboring pET21a-gpdA and pET21a-gpdC were rotary-shaken in 3 mL of LB medium at 180 rpm at 37°C for 14 h, then 1 mL portions were inoculated into 100 mL LB medium, and rotary shaken at 160 rpm at 37°C. When the optical density reached 0.6, 0.5 mM isopropyl-β-D(−)-thiogalacto-pyranoside (IPTG) was added to the medium and incubated for 3 h. Harvested cells were resuspended and ultrasonicated in 10 mL of 20 mM HEPES-NaOH buffer (pH 7.5) containing 0.5 M NaCl and 20 mM imidazole, followed by centrifugation at 8,000 × g for 30 min. Supernatants were passed through a HisTrap FF crude column (Cytiva, North Logan, UT, United States). Thereafter, recombinant GpdA (rGpdA) and GpdC (rGpdC) were eluted with the same buffer containing 200 mM imidazole, and their homogeneity was confirmed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE).

We added 3 mM GAP to cuvettes containing rGpdA or rGpdC (10 μg each) in 90 μL of 1 mM NAD⁺, 10 mM potassium phosphate, 30 mM sodium pyrophosphate (pH 8.4), then changes in absorbance at 340 nm were monitored at 25°C for 2 min using a U-3900 spectrophotometer (Hitachi, Tokyo, Japan). The K_m and k_cat values were determined by using 0.01–1 mM NAD⁺, 0.01–10 mM NADP⁺, and 0.015–3 mM GAP. We incubated rGpdA and rGpdC (10 μg each) dissolved in 20 mM HEPES-NaOH buffer (pH 7.5) containing 150 mM NaCl) with NOC-5 and GSNO (Dojindo, Kumamoto, Japan) (0–1 mM each) at 25°C for 1 h, then we analyzed the susceptibility of GAPDH to NO donors.

Aspergillus nidulans FGSC A4 conidia (2 × 10⁷) were inoculated into 100 mL of liquid MM, rotary shaken at 120 rpm and 37°C for 14 h, followed by 10 mM NaNO₂, 3 mM H₂O₂, 50 μM menadione, 1.5 mM diamide, or 1 mM tert-butylhydroperoxide (t-BOOH) under the same conditions for 1 h. Mycelial cDNA was prepared as described in Section 1.4, then qRT-PCR proceeded using the THUNDERBIRD® SYBR® qPCR Mix (Toyobo), and the Thermal Cycler Dice Real-Time System MRQ (Takara Bio). Supplementary Table S2 shows the primer sequences.

Aspergillus nidulans NPC1 and GpdCΔ conidia (2 × 10⁷) inoculated into 100 mL of liquid MM (pH 5.5) were rotary shaken at 120 rpm and 37°C for 14  h, followed by 120 rpm and 37°C for 0–2  h after adding 10 mM NaNO₂. Powdered mycelia (as described in Section 1.4) were dissolved in 1 mL of 20 mM Tris–HCl containing 150 mM NaCl/0.1 g mycelia, vortex-mixed, then centrifuged for 30  min at 18,800 × g and 4°C. We assessed GAPDH activity in the supernatant as described (Ralser et al., 2007) with the following modification. Supernatants (10 μL) were added to cuvettes containing 1 mM NAD⁺ in 90 μL of 30 mM sodium pyrophosphate (pH 8.4). The reaction was started by adding 3 mM GAP, then changes in absorbance at 340 nm were monitored at 25°C for 2 min (as described in section 1.5). We then determined the dry weight of lyophilized mycelia.

Aspergillus nidulans conidia (2 × 10⁷) were cultured and mycelia were prepared as described in section 1.7. Ground mycelial powder (0.1 g) dissolved in 1 mL of 30 mM pyrophosphate buffer (pH 8.4) was vortex-mixed, and centrifuged for 30 min at 18,800 × g and 4°C. Intracellular GAP was quantified in supernatants as described by Garrigues et al. (1997) and modified as follows. Recombinant GpdA (50 μg) was added to cuvettes containing 100 μL of supernatants. Reactions were started by adding 1 mM NAD⁺, and changes in absorbance at 340 nm were monitored at 25°C for 5 min. Ethanol in culture media was quantified using F-kit ethanol (J.K. International, Tokyo, Japan) as described by the manufacturer.

Recombinant GpdA and GpdC were incubated with 0–0.2 M GSNO at 25°C for 1 h, precipitated using an excess of acetone to remove GSNO, then resuspended in 100 μL of 20 mM HEPES-NaOH buffer (pH 7.5) containing 150 mM NaCl. Nitrosothiols were modified as described (Li and Kast, 2017) with the following modifications. Suspensions were mixed with equal volumes of 10 mg/mL N-ethylmaleimide, incubated at 37°C for 1 h, then proteins were precipitated using acetone. The precipitates were suspended in 100 μL of 20 mM HEPES-NaOH (pH 7.5) containing 150 mM NaCl, 1 mM biotin-HPDP solution (Dojindo) and 1 mM ascorbic acid, then incubated at 25°C for 1 h. Proteins were resolved by non-reducing SDS-PAGE, then S-nitrosothiol detected using horse radish peroxidase-conjugated streptavidin was analyzed by immunoblotting (Cosmo Bio, Tokyo, Japan).

Amino acid sequences of fungal GAPDH isozymes were downloaded from the Fungal and Oomycete genomics resource (FungiDB) database (https://fungidb.org/fungidb/app/). We analyzed their phylogenies using Molecular Evolutionary Genetics Analysis (MEGA) version 11.0.10 (Tamura et al., 2021) and the neighbor-joining method with complete gap deletion. Numbers and branches indicate values calculated from 1,000 bootstrap resampling replicates. The amino acid sequence of human GAPDH was obtained from the National Center for Biotechnology, and aligned with the fungal GAPDH using CLUSTALW version 2.1 (Larkin et al., 2007).

The A. nidulans genome consists of gpdA (Gene ID, AN8041) that encodes ubiquitous glycolytic GAPDH, and gpdC (AN2583) with unknown physiological function. The amino acid sequence identity between GpdA and predicted GpdC proteins was 47.0%. A phylogenetic analysis identified GpdA, B, and C isozymes in all Aspergillus species except A. nidulans, which lacks a gene encoding GpdB (Figure 1A). Here, we investigated the role of A. nidulans GpdC in nitrosative and oxidative stress responses because our transcriptome analysis suggested that gpdC expression is upregulated upon exposure to RNS (Amahisa et al., 2024). Transcripts of gpdC in A. nidulans mycelia were quantified by PCR after exposure to acidified nitrite (1 mM NaNO₂, pH 5.5), H₂O₂, menadione, diamide, and t-BOOH as ROS donors for 1 h. We verified that acidified nitrite equilibrated with nitrosonium cations (NO⁺), imposes RNS stress on A. nidulans (Zhou et al., 2012). These results indicated that the fungus accumulated 4.9- and 1.6-fold more gpdC transcripts after exposure to acidified nitrite and diamide, respectively (Figure 1B). None of the other reagents altered the quantity of intracellular gpdC transcripts, whereas that of gpdA transcripts was decreased 0.34- and 0.20-fold by H₂O₂ and t-BOOH, respectively, but not by acidified nitrite. These results indicated that RNS stress induced gpdC expression.

Recombinant GpdA (rGpdA) and GpdC (rGpdC) were generated in E. coli. Purified proteins migrated as single bands on SDS-PAGE, indicating their homogeneity (Figure 2A). The specific activities of rGpdC on GAP-dependent NAD⁺ and NADP⁺ reduction were 4.1 ± 1.6 and 12 ± 2 μmol min⁻¹ mg⁻¹, respectively. The activities of rGpdA on GAP-dependent NAD⁺ and NADP⁺ reduction were 110 ± 10 and < 0.01 μmol min⁻¹ mg⁻¹, indicating that rGpdC and rGpdA oxidize GAP by using NADP⁺ and NAD⁺, and NAD⁺, respectively, as electron acceptors. Steady-state kinetics for the GAP-dependent NAD⁺- and NADP⁺-reduction were analyzed. The apparent Michaelis–Menten (K_m) constant of the GpdC reaction for GAP was 290 ± 20 μM, which was similar to that of GpdA (Table 1). The K_m value for NAD⁺ was 420 ± 20 μM and comparable to that of rGpdA (410 ± 90 μM). These values were higher than that of NADP⁺ reduction by rGpdC (47 ± 2 μM), indicating that GpdC preferentially uses NADP⁺ over NAD⁺. These results indicated that GpdC is a novel NADP⁺-and NAD⁺-dependent GAPDH. The preference for NADP⁺ over NAD⁺ distinguished GpdC from GpdA, which has very low levels of NADP⁺-dependent GAPDH activity.

The enzymatic activity of mammalian GAPDH is susceptible to oxidative stress due to a ROS-labile catalytic thiol residue (Hara et al., 2006; Sirover, 2014) that reacts with the nitrosating reagent GSNO to generate S-nitrosothiol (Broniowska and Hogg, 2010). Amino acid sequence alignment showed that this thiol residue is conserved in GpdC (Supplementary Figure S4). We investigated the effects of NOC-5 (NO donor) and GSNO on the activities of the fungal GAPDH. After incubation at 25°C for 1 h, NOC-5 concentration-dependently decreased the activity of rGpdA and rGpdC (Figure 2B). At 0.2 mM, NOC-5 impaired the activity of rGpdA more than rGpdC, indicating that rGpdC is more resistant to NO. Adding GSNO also decreased GAPDH activity, but that of rGpdC was more susceptible. Biotin-switch methods visualized S-nitrosothiol of GSNO-treated rGpdA and rGpdC on blots, and this was elevated by increasing the GSNO concentration (Supplementary Figure S3). This confirmed that GSNO modified the thiol residues of rGpdA and rGpdC to S-nitrosothiol and impaired enzyme activity. The multiple alignment of GAPDH isozymes found five unique Cys residues in GpdC that were not found in human GAPDH and GpdA (Supplementary Figure S4). We consider that this finding complements the higher levels of nitrosation in GpdC (Supplementary Figure S3) because these residues provided extra thiol targets of the nitrosating reagents in GpdC.

Intracellular GAPDH activity produced by the NPC1 (control gpdC⁺) strain and GpdCΔ that lacks intact gpdC (Supplementary Figure S1) were determined. Mycelia of the strains produced similar levels of activity in the absence of RNS stress, indicating that GpdC is not required for cellular GAPDH activity under these conditions (Figure 3A). This agreed with decreased gpdC expression in the absence of stress (Figure 2). Exposure to acidified nitrite (pH 5.5) for 2 h obviously decreased GAPDH activity, indicating impairment by RNS probably due to an oxidation-labile thiolate residue at the catalytic cysteine. This decrease was more prominent in the GpdCΔ strain that generated little GAPDH activity after exposure to acidified nitrite, which was consistent with the finding that this strain accumulated 1.7-fold more GAP than the NPC1 strain (Figure 3B). These results indicated that GpdC maintains cellular GAPDH activity, and that GpdC replaces RNS-sensitive GpdA under RNS stress.

The GpdCΔ strain grew slowly on agar containing acidified nitrite and H₂O₂ (Figure 3C). The growth rate was more rapid for the GpdCΔ, than the NPC1 strain in the presence of the NO donor diethylenetriamine (DETA)/NO adduct (NOC-18). Under these conditions, GpdCΔ produced a reddish brown pigment in the medium, which was not evident in the NPC1 strain (Figure 3D). Introducing gpdC into the GpdCΔ strain restored the growth rate of the strain (GpdCΔ + gpdC; Figure 3D), indicating that gpdC complemented the growth deficiency of the ΔgpdC strains. These results indicated that GpdC is a GAPDH isozyme, the activity of which confers A. nidulans growth tolerance under RNS stress.

Reactive nitrogen species react with cellular thiols and metal ions, and mostly abrogate the respiration mechanisms of eukaryotic (Poderoso et al., 2019) and A. nidulans mitochondria (Amahisa et al., 2024; Zhou et al., 2012). These conditions might lead to increased dependence of the fungal growth on the glycolytic system. Figure 4A indicate that the acidified nitrite inhibited NPC1 growth more on agar media containing non-glycolytic glycerol, acetate, and glutamate as the carbon source compared with glucose. This suggested the importance of the glycolytic mechanism for fungal growth under RNS stress. The gene disruptants PdcAΔ (without pdcA encoding glycolytic pyruvate decarboxylase), AlcCΔ (alcC, alcohol dehydrogenase), and DldAΔ (dldA, putative lactate dehydrogenase) (Grahl et al., 2011) grew at comparable rates on medium containing glucose in the absence of RNS stress (Figure 5A). Acidified nitrite retarded the growth rates of PdcAΔ and AlcCΔ whereas these rates were comparable between DldAΔ and the control strain. These results indicated that the glycolytic ethanol fermentation mechanism is important for fungal growth under RNS stress.

Mitochondria were visualized using mitochondrial citrate synthase (CitA) fused to green fluorescent protein (GFP). Fluorescence shaped filaments, which are typical of fungal mitochondria under stress-free conditions (Figure 4B) dispersed into dot-like structures upon exposure to acidified nitrite. This dot-like morphology is also similar in fungal mitochondria under oxidative stress (Ruf et al., 2018). These results suggested that the RNS stress causes mitochondrial dysfunction and increased fungal metabolic dependence on glycolysis for proliferation.

We analyzed ethanol fermentation to understand the glycolytic role of GpdC under RNS stress. The NPC1 and GpdCΔ strains accumulated the same amounts of ethanol in culture media without acidified nitrite, whereas PdcAΔ and AlcCΔ accumulated 0.69- and 0.25-fold less ethanol than the NPC1 strain (Figure 5B). In contrast, GpdCΔ produced 0.16-fold more ethanol than the control strain, whereas PdcAΔ and AlcCΔ produced very little under RNS stress. Together with the defective growth under RNS stress (Figure 3), these results indicated that A. nidulans GpdC plays a significant role in energy conservation via ethanol fermentation as a mechanism to survive RNS stress.

Figure 6 shows the mechanism of how GpdC renders confers fungal tolerance of RNS stress. The activity of canonical GpdA is largely impaired under RNS stress. Aspergillus nidulans induces expression of the gpdC gene and maintains intracellular levels of GAPDH activity. Increased tolerance against NO (Figure 2B) agrees with the notion that GpdC replaces GpdA. The NADP⁺-dependent activity of GpdC might contribute to maintain fungal glycolysis because NADPH production is associated with fungal oxidative and RNS stress responses (Yoshikawa et al., 2021). Thus, GpdC functions as an alternative to GpdA under RNS stress, whereas respiratory/mitochondrial defects increase the dependence of cellular energetic metabolism on a glycolytic mechanism.