Skip to main content

Increasing NADPH impairs fungal H2O2 resistance by perturbing transcriptional regulation of peroxiredoxin


NADPH provides the reducing power for decomposition of reactive oxygen species (ROS), making it an indispensable part during ROS defense. It remains uncertain, however, if living cells respond to the ROS challenge with an elevated intracellular NADPH level or a more complex NADPH-mediated manner. Herein, we employed a model fungus Aspergillus nidulans to probe this issue. A conditional expression of glucose-6-phosphate dehydrogenase (G6PD)-strain was constructed to manipulate intracellular NADPH levels. As expected, turning down the cellular NADPH concentration drastically lowered the ROS response of the strain; it was interesting to note that increasing NADPH levels also impaired fungal H2O2 resistance. Further analysis showed that excess NADPH promoted the assembly of the CCAAT-binding factor AnCF, which in turn suppressed NapA, a transcriptional activator of PrxA (the key NADPH-dependent ROS scavenger), leading to low antioxidant ability. In natural cell response to oxidative stress, we noticed that the intracellular NADPH level fluctuated “down then up” in the presence of H2O2. This might be the result of a co-action of the PrxA-dependent NADPH consumption and NADPH-dependent feedback of G6PD. The fluctuation of NADPH is well correlated to the formation of AnCF assembly and expression of NapA, thus modulating the ROS defense. Our research elucidated how A. nidulans precisely controls NADPH levels for ROS defense.

Graphical Abstract


Reactive oxygen species (ROS) are produced in a wide range of physiological processes, including aerobic respiration (Kalyanaraman et al. 2018), exposure to environmental agents such as UV irradiation (Kuehne et al. 2015) and drugs/xenobiotics (Van Acker et al. 2017), and immune response in higher organisms (Nathan et al. 2013; Yang et al. 2013). The resultant ROS can lead to oxidative stress by directly or indirectly damaging DNA, proteins, and lipids (Nathan et al. 2013). Cells have evolved a variety of enzymes, such as glutathione reductase, superoxide dismutase, and peroxidases, to defend against ubiquitous ROS toxicity. Among such defense mechanisms, the thioredoxin peroxidase enzymes, peroxiredoxins (Prx), are the most abundant antioxidants and is widespread among archaea, bacteria, and eukaryotes (Poole et al. 2011; Rhee 2016). Prx are highly reactive with H2O2, using the reversible oxidation of cysteine residues to reduce peroxides. The resulting Prx disulfides are reduced by thioredoxin (Trx) using electrons from NADPH in the presence of Trx reductase (TrxR). We have previously shown that Aspergillus nidulans Prx (PrxA) is not essential to sustain normal growth, but lack of PrxA results in hypersensitivity of strains toward oxidative stress, indicating a specific biological function for exogenous ROS detoxification (Xia et al. 2018). A similar result was also obtained with the Prx ortholog in A. fumigatus (Hillmann et al. 2016; Rocha et al. 2018), suggesting PrxA has a key role in H2O2 resistance in Aspergillus species.

The CCAAT-binding complex is an important general transcriptional regulator for that transcription of numerous genes, including prosurvival and cell cycle-promoting genes (Oldfield et al. 2014; Hortschansky et al. 2017). The corresponding A. nidulans CCAAT-binding factor (AnCF) consists of the subunits HapB, HapC, and HapE (Thon et al. 2010). AnCF is regulated at the posttranscriptional level by the redox status of the cell, thereby serving as a redox sensor coordinating the cellular oxidative stress response. AnCF senses the cellular redox status via oxidative modification of thiol groups in HapC. Oxidized HapC is unable to participate in AnCF assembly, leading to abolishment of the regulation role, while, the invalid HapC can be revived via the reduction by Trx system (Thon et al. 2010). AnCF acts as a repressor of napA, which then encodes the ROS-specific transcriptional activator NapA whose target genes include A. nidulans Trx-encoding gene (trxA), catalase-encoding gene (catB), and PrxA-encoding gene (prxA) (Thon et al. 2010). Therefore, ROS deactivates AnCF and then increases expression of napA via release of AnCF repression for activation of oxidative defense mechanisms.

NADPH serves as an important reducing equivalent and is essential in cellular defense against oxidative damage. In many biological systems, the antioxidant functions of NADPH are exerted via regeneration of Trx by TrxR. NADPH-reduced Trx may provide reducing equivalents to Prx, as well the other proteins containing oxidized cysteine groups, including NapA and HapC, through a thiol–disulfide interaction (Thon et al. 2010). To replenish exhausted inventories of intracellular NADPH, numerous pathways are known to be involved in NADPH regeneration (Wang et al. 2014; Zhou et al. 2016). Among them, the pentose phosphate pathway (PPP) is believed to be the major one, in which NADPH is produced by two enzymes, namely glucose-6-phosphate dehydrogenase (G6PD), 6-phosphogluconate dehydrogenase (6PGD). Although NADPH is considered an indispensable reducing agent for ROS elimination, how NADPH manipulates cellular ROS resistance remains obscure, and many contradictory results can be gleaned from previous reports. For example, animal cell studies demonstrated that increases in NADPH levels by overexpressing of the corresponding enzymes could protect cells or tissues against oxidative damage (Xiao et al. 2020). The advantages brought by excess production of NADP(H) are due to its functional roles as an indispensable cofactor for glutathione reductase (GR) and TrxR that are essential for GPx- and Prx-mediated peroxide removal, respectively. However, many other studies have supported the opposite notion that excess levels of intracellular NADPH can induce reductive stress and cellular dysfunction. This opposite conclusion was drawn from the fact that excess NAD(P)H may be used by NADPH oxidases (NOXs) to produce ROS (Brandes et al. 2014; Yu et al. 2014; Xiao et al. 2020). Therefore, NAD(P)H serves as dual-function participants, either an antioxidant cofactor or a pro-oxidant, to maintain cellular redox homeostasis in many animal cells. Notably, the effect of excess NADPH on microbes is yet to be investigated, though overexpression of NADPH-producing enzymes has been regarded as a promising strategy in NADPH-required metabolic engineering (Xue et al. 2017).

Here, we used the filamentous fungus A. nidulans, a classical model of pathogenic and commercial fermentation Aspergillus species, to investigate the relationship between NADPH generation and fungal ROS defense. Our findings indicated that although NADPH is an indispensable reducing agent for ROS elimination, increasing NADPH levels by modulating the expression strength of typical NADPH-generating enzymes produced an adverse effect on fungal resistance to H2O2. We found that excess NADPH promoted the assembly of AnCF, which in turn suppressed NapA, leading to low levels of PrxA and the eventual impairment of antioxidant ability. Our results provide new insights into the dual-function role of NADPH in maintaining cellular redox homeostasis.

Materials and methods

Strains and growth conditions

Aspergillus nidulans is the experimental model used in this study. Genotypes of strain used in this study are listed in Additional file 1: Table S1. All fungi were grown at 37 °C in minimal medium (MM) (1% glucose, 10 mM NaNO3, 10 mM KH2PO4, 7 mM KCl, 2 mM MgSO4, 2 mL L−1 Hunter’s trace metals, pH 6.5) (Kadooka et al. 2016), supplemented appropriately (0.4 mg/L biotin, 0.5 g/L uracil, 0.6 g/L uridine, 0.4 mg/L pyridoxine). E. coli DH5α and BL21 (DE3) were used for gene cloning and protein expression, respectively. Sodium nitrate (10 mM), proline (10 mM) and ammonium tartrate (5 mM) were used as sole nitrogen sources for niaD promoter replacement strains according to the experimental requirement.

Construction of gene disruption strains

Primers are listed in Additional file 1: Table S2. A. nidulans ABPUN genomic DNA was obtained using Wizard Genomic DNA Purification Kit (Promega, USA), and used as template to produce the gene deletion constructs. All PCR was performed using PrimeSTAR HS DNA Polymerase (Takara, Japan). CRISPRdirect ( was used for sgRNA protospacer selection. The sgRNAs were synthesized using A. nidulans ABPUN genomic DNA; primers pairs AN4034-sgF1/AN4034-sgR1 (for ΔhapC), AN2981-sgF1/AN2981-sgR1 (for ΔAN2981), AN2981-sgF2/AN2981-sgR2 (for ΔAN2981), AN3954-sgF1/AN3954-sgR1 (for ΔAN3954), AN3954-sgF2/AN3954-sgR2 (for ΔAN3954) were prepared using a GeneArtTM Precision gRNA Synthesis Kit (Invitrogen, USA) in accordance with the manufacturer’s instructions. In vitro cleavage activity tests were performed using the PC1400 Kit (Inovogen Tech. Co., China). The pyrG marker gene was amplified using A. oryzae RIB40 genomic DNA and the primers, pyrG-F1 and pyrG-R primers. The marker gene argB was amplified using A. nidulans A6 genomic DNA and the primers, argB-F and argB-R. To delete hapC (AN4034) gene, primer pairs AN4034-uF/AN4034-uR and AN4034-dF/AN4034-dR were used to amplify the 1 kb of the 5′ and 3′ untranslated regions (UTR) of AN4034, respectively. Primer AN4034-nested-F and AN4034-nested-R were used to amplify the final fusion product. The resultant DNA cassette 5′AN4034-argB-3′AN4034 (400 ng), together with the corresponding sgRNA (100 nM) and 1 μg purified Cas9 (Inovogen Tech. Co.), was introduced into A. nidulans ABPUN strain as previously described method (Kitamoto 2002; Pohl et al. 2016) to create an hapC deletion strain ∆hapC. Disruptions of the gsdA (AN2981) gene were performed using the same methods except that the marker gene pyrG was employed for gsdA disruption. Transformants were selected from the plates based on their auxotrophy and confirmed by colony PCR (Additional file 1: Fig. S1) using KOD FX polymerase (Toyobo, Japan). The genotype of the resultant disruptants is shown in Additional file 1: Fig. S1.

Construction of promoter substitution strains

The corresponding primers are listed in Additional file 1: Table S2. nP.gsdA was constructed as follows. niaD promoter (niaD.P) was cloned from A. nidulans ABPUN genomic DNA using the primers niaD-F and niaD-R. Marker gene pyrG from A. oryzae was amplified using A. oryzae RIB40 genomic DNA with primers pyrG-F2 and pyrG-R. The resultant niaD.P was fused with pyrG was to generate pyrG-niaD.P fragment by an overlapping PCR using the two primers pyrG-F1 and niaD-nested-R. Approximately 1 kb of 5′ UTR and 1 kb of gsdA open reading frame (ORF) were cloned from A. nidulans ABPUN genomic DNA using primer pairs AN2981-5′-F/AN2981-5′-R and AN2981-3′-F/AN2981-3′-R, respectively. To construct the pyrG-niaD.P-gsdA cassette, pyrG-niaD.P was flanked with the resulted 5′ UTR and ORF of gsdA by fusion PCR using the AN2981-nested-F and AN2981-nested-R primers. The resultant PCR product was transformed to ABPUN strain to obtain nP.gsdA strain. The strains, gP.gsdA and gP.gndA were constructed using the same strategy with their corresponding primers (Additional file 1: Table S2) and shown in Additional file 1: Fig. S2.

A recombinant pUC19-pyroA-gpdA.P-prxA-TtrpC plasmid was constructed to obtain the gP.prxA strain. In this plasmid, the gpdA promoter (gpdA.P) was cloned from A. nidulans ABPUN genomic DNA using primers gpdA-F and gpdA-R and then fused to the ORF of A. nidulans prxA and E. coli terminator TtrpC with primer pair pUC19-pyroA-gPprxA-F/pUC19-pyroA-gPprxA-R. Then, the fused fragment was inserted into pUC19-pyroA plasmid (our lab) using the ClonExpress II One Step Cloning Kit (Vazyme, China). The resultant gP.prxA strain was further transformed with pyrG-niaD.P-gsdA cassette to construct nP.gsdA/gP.prxA strain. The pyrG-niaD.P-gsdA cassette was introduced into ΔhapC strains to generate the strain nP.gsdAhapC. The successful disruptants were confirmed using colony PCR with the corresponding primers indicated in Additional file 1: Fig. S2.

Construction of GFP-tagged NapA, GFP-tagged PrxA and Flag-tagged HapC expression strains

Primers are listed in Additional file 1: Table S1. The cassette for expressing C-terminal tagged GFP of NapA was constructed as follows. The marker gene pyroA was amplified by PCR using A. nidulans A6 genomic DNA with primers pyroA-F and pyroA-R. The gene for expressing GFP with an N-terminal 5GA linker was cloned from pUC19-gfp plasmid (our lab) with primers gfp-F and gfp-R. An overlapping PCR was performed to obtain the gfp::pyroA fragment with the primers, gfp-pyroA-F and gfp-pyroA-R. Approximately 2.8 kb of 5′ UTR plus the ORF and 1 kb of 3′ UTR of napA were cloned from A. nidulans A6 genomic DNA with the corresponding primer pairs napA-5′-F/napA-5′-R and napA-3′ UTR-F/napA-3′ UTR-R. The cassette (napA-gfp-3′napA) containing gfp::pyroA flanked with 5′ UTR plus the ORF and 3′ UTR of napA was constructed by fusion PCR using the above-mentioned three resultant PCR products with the nested primers napA-5′-nested-F/napA-3′ UTR-nested-R. The resultant cassette was transformed into the strain WT_argB (our lab) to construct GFP-tagged NapA-expression strain (N_Gfp) as shown in Additional file 1: Fig. S3. One of the successful transformants was verified using colony PCR and sequenced at Tsingke Biotechnology Co.. The resultant N_Gfp strain was further introduced with pyrG-niaD.P-gsdA cassette to obtain the strain nP.gsdA/N_Gfp.

For the N-terminal tagged GFP of PrxA expression strain, a pUC19-pyroA-gfp-prxA plasmid was first constructed. Approximately 1 kb of 5′ UTR and 1.6 kb of 3′ UTR plus ORF of prxA was cloned from A. nidulans A6 genomic DNA with the corresponding primer pairs prxA-uF/prxA-uR and prxA-dF/prxA-dR. The gene expressing GFP with a C-terminal 5GA linker was cloned from a pUC19-gfp plasmid (our lab) with primers gfp-cF and gfp-cR and then fused with the two PCR products with the nested primers pUC19-prxA-nested-F/pUC19-prxA-nested-R. Next, the pUC19-pyroA plasmid was digested by Sma I and ligated with the above fusion PCR product using a ClonExpress II One-Step Cloning Kit (Vazyme). The resulting plasmid pUC19-pyroA- gfp-prxA was introduced into the ΔprxA strain to generate P_Gfp. One of the successful transformants was verified using colony PCR, as indicated in Additional file 1: Fig. S3. The pyrG-niaD.P-gsdA cassette was also introduced into P_Gfp to obtain the strain nP.gsdA/P_Gfp.

For the Flag-tagged HapC expression strain, a pUC19-pyroA-hapC-Flag plasmid was first constructed. A DNA fragment containing approximately 1.5 kb 5′ UTR followed by HapC-Flag-encoding DNA was amplified using A. nidulans ABPUN genomic DNA as template with the primers hapC-uF and hapC-Flag-R. Approximately 1 kb of 3′ UTR of hapC was cloned using the same template with primer pairs Flag-hapC-F and hapC-dR. The two resultant PCR products were fused by overlapping PCR with the primers pUC19-pyroA-hapC-uF and pUC19-pyroA-hapC-dR. Plasmid pUC19-pyroA was digested by Hind III and Sma I and ligated with the above fusion PCR product using a ClonExpress II One Step Cloning Kit (Vazyme). The resulting plasmid pUC19-pyroA-hapC-Flag was introduced into ΔhapC and nP.gsdAhapC strain to generate H_Flag and nP.gsdA/H_Flag strains, respectively. One of the successful transformants was verified using colony PCR as shown in Additional file 1: Fig. S4, and the resultant PCR products were sequenced for further confirmation at Tsingke Biotechnology Co.

Recombinant AnG6PD preparation

The DNA encoding AnG6PD was amplified from A. nidulans cDNA using the primers pET28a-gsdA-F/pET28a-gsdA-R (Additional file 1: Table S2). The DNA fragment was inserted between Nde I and Xho I restriction sites of pET28a (+) vector. E. coli BL21 (DE3) cells transformed with expression plasmid was cultured in LB medium supplied with 50 μg/mL kanamycin. The protein expression was induced at 30 °C with 0.2 mM isopropyl-β-d-thiogalactoside (IPTG). The recombinant protein was purified by affinity chromatography with HisTrap FF column (GE Healthcare, USA) and confirmed using SDS-PAGE analysis (Additional file 1: Fig. S5).

Sensitivity of A. nidulans to H2O2

Serial dilutions of 48-h cultivated A. nidulans conidia were spotted onto MM plates containing indicated concentrations of H2O2, and then incubation at 37 °C for 2 days. The morphology of the colonies was examined to determine their sensitivity to H2O2. For the conidia survival assay, conidia (1 × 103 mL−1, 10 μL) were suspended in MM containing top agar (0.75% agar), and the indicted concentrations of H2O2 were then spread on MM plates containing the same concentrations of H2O2. Colonies were counted after a 48-h incubation, and CFU were expressed as percentages of the CFU for strains incubated without H2O2.

G6PD activity assay

Intracellular G6PD enzyme activity was determined in cell lysates by measuring the rate of increase of NADPH at 340 nm (UV-5100 Spectrophotometer, Hitachi, Japan). The assay was performed at 25 °C in 1 mL containing 100 mM Tris–HCl buffer (pH 8.0), 20 mM MgCl2, 5 mM D-glucose-6-phosphate, 0.3 mM NADP+, and 50 mg cell lysates. G6PD activity was expressed in U/mg protein; 1 unit (1 U) of G6PD was defined as 1 mg/mL of enzyme required to produce 1 µmol NADPH in 1 min at 25 °C.

Fluorescence microscopy imaging of NapA-GFP strains

Approximately 105 conidia were suspended in 200 µL MM medium, seeded in a 35-mm confocal dish, and incubated at 37 °C for 10 h. Samples were treated with or without 2 mM H2O2 and incubated for 20 min, and then the fluorescent mycelia were imaged by laser confocal microscope (TCS SP8, Leica, Germany). Nuclei were stained with Hoechst 33258 for 15 min, and then washed by PBS for three times before the observation.

Native and denaturing PAGE and western blot analysis

Strains expressing Flag-tagged HapC (H_Flag, nP.gsdA/H_Flag) were precultured in MM medium with nitrate as sole nitrogen source for 16 h, and then incubated with 1 mM H2O2 at 37 °C for 30, 90, and 120 min. Protein was extracted as described (Thon et al. 2010), using a non-denaturing procedure in lysis buffer (100 mM pH 8.0 Tris–HCl, 1 mM EDTA and 1% 1 × protease inhibitor mixture). For denaturing PAGE analysis, protein was extracted using lysis buffer supplied with 1% SDS. Protein concentrations of the cell tracts were measured by Bradford assay (Sangon), and then samples were diluted to a protein concentration of 1 mg/mL. Native PAGE was performed using a running buffer of 25 mM Tris–HCl (pH 8.0) and 195 mM glycine. Samples (100 µg per lane) were run for 2 h using a commercial 4–20% gradient gels (Beyotime) and then analyzed by western blotting with a PVDF membrane. For denaturing conditions, SDS-PAGE was performed using SDS running buffer containing 0.1% SDS and 12% SDS-PAGE gels. The AnCF-Flag was detected using an anti-Flag antibody (Transgen). The secondary antibody for Flag was anti-mouse IgG HRP conjugate (Transgen). The blots were subsequently reprobed for Actin using an anti-Actin antibody (Sigma-Aldrich) and anti-rabbit IgG HRP conjugate (Transgen, China). The ECL detection system (Tanon) was used to visualize proteins. Quantitative densitometric analyses of western blots were conducted using ImageJ.

Quantification real-time PCR analysis

Total RNA was isolated using EZ-10 DNAaway RNA Mini-Preps Kit (Sangon, China). cDNAs were then reverse-transcribed with ReverTra Ace qPCR RT Master Mix with gDNA Remover (Toyobo, Japan). Quantitative PCR was preformed using a SYBR Green PCR Kit (Toyobo) and conducted on a CFX-96 Real-Time PCR system (Bio-Rad, USA). Primer pairs q-RT-prxA-F/q-RT-prxA-R, q-RT-gsdA-F/q-RT-gsdA-R, q-RT-gndA-F/q-RT-gndA-R, q-RT-napA-F/q-RT-napA-R, and q-RT-actA-F/q-RT-actA-R (Additional file 1: Table S2) were designed to amplify prxA, gsdA, gndA, napA, and actA, respectively. Relative mRNA levels were normalized to reference gene actA.

Measurement of NADPH/NADP+ ratio

Mycelium was collected and ground in liquid nitrogen and then resuspended in 300 μL extraction buffer as described in the instruction manual (Sigma-Aldrich, USA). Cell lysates were filtered using a 10-kDa Ultra filter (Millipore, Sigma) to minimize the influence of NADPH-consume proteins. The NADPH/NADP+ ratio was calculated as [NADPH/ (total NADPH–NADPH)].

Quantification analyses of intracellular GFP

Strains expressing GFP-tagged Prx (P_Gfp, nP.gsdA/P_Gfp) were precultured in MM medium with nitrate as a sole nitrogen source for 16 h and then incubated with or without 1 mM H2O2 at 37 °C for 2 h. Cells were collected and then disrupted with liquid nitrogen. The fluorescence values of the supernatant of cell lysates were measured using a fluorescence spectrophotometer (F-4600, Hitachi, Japan) at an excitation wavelength of 488 nm and an emission wavelength of 509 nm.

Quantification analyses of intracellular ROS

Cell-permeable BES-H2O2-Ac (Wako, Japan) and BES-So-AM (Wako, Japan) were used as H2O2- and O2·−-specific fluorescent probes, respectively. Precultivated fungal cells (16 h) were incubated with individual probes for 30 min before exposing these cells to H2O2 (1 mM) for 30 min. The ROS scavenger N-acetyl-l-cysteine (NAC) (Sigma-Aldrich, USA) was added to block H2O2 generation, cells were pretreated with or without 10 mM NAC for 1 h at 37 °C prior to incubation with BES-H2O2-Ac probes. Mycelia were then washed thrice using PBS, immediately ground into powder with liquid nitrogen, and then suspended in 50 mM PBS. The supernatant of the disrupted mycelia was analyzed using a fluorescence spectrophotometer (F-4600, Hitachi, Japan) at an excitation wavelength of 485 nm and an emission wavelength of 515 nm for H2O2 detection and an excitation wavelength of 505 nm and an emission wavelength of 544 nm for O2·− detection.

Statistical analysis

All experiments were repeated at least three times on independently generated samples with similar results. Representative experiments or the quantitative densitometric analyses of several experiments are shown, data are represented as mean ± SD. P < 0.05 was considered significant.


NADPH-consuming PrxA is essential to cell survival under H2O2 stress conditions

We have previously shown that A. nidulans lacking PrxA displayed pronounced sensitivity to H2O2 (Xia et al. 2018), whereas, in many living beings, catalases act as the key H2O2-detoxifying enzyme (Rodriguez-Segade et al. 1985). Considering that catalases are abundant in A. nidulans (Kawasaki et al. 1997; Kawasaki et al. 2001), we directly compared the H2O2 protection functions exerted by PrxA and those of catalase B (a major catalase in A. nidulans) (Fig. 1A). Mutants carrying deletions in these genes (∆prxA or ∆catB) were viable with identical growth on agar plates to wild-type A. nidulans (WT) under normal growth conditions. Growth of ∆prxA was completely inhibited with 0.5 mM H2O2, whereas ∆catB exhibited little sensitivity to H2O2, clearly indicating that PrxA, rather than catalase B, is the indispensable enzyme that protects A. nidulans against H2O2 stress.

Fig. 1
figure 1

A. nidulans PrxA in tandem with G6PD is essential to fungal ROS defense. A Effects of disrupting prxA and catB on protection against H2O2. Fresh conidia (1 × 105) of WT (WT_argB), ∆prxA, and ∆catB were inoculated on MM plates containing H2O2 at indicated concentrations and then incubated at 37 °C for 2 days. B Lack of PrxA resulted in intracellular accumulation of NADPH. After culture in liquid MM for 16 h, strains were treated with or without 1 mM H2O2 for 30 min before collection for determination of the NADPH/NADP+ ratio in cell lysates. NADPH/NADP+ is presented in relative quantitation; NADPH/NADP+ value of unstressed WT was set to 1 (mean ± SD; n = 3, *P < 0.05, **P < 0.01, one-way ANOVA). CE Phenotypes of WT (WT_pyrG) and nP.gsdA strains on MM plates under H2O2 conditions using ammonium tartrate (left), proline (middle), or nitrate (right) as the sole nitrogen source. F NADPH/NADP+ ratios in WT (WT_pyrG) and nP.gsdA strains. Strains were cultivated in liquid MM with ammonium tartrate (left), proline (middle), or nitrate (right) as the sole nitrogen source for 16 h, and then the NADPH/NADP+ ratios in cell lysates were quantified and compared; NADPH/NADP+ value of WT cultivated with each nitrogen source was set to 1 (mean ± SD; n = 3, *P < 0.05, **P < 0.01, t-test)

To confirm the process where PrxA employs NADPH to decompose H2O2 in vivo, we calculated the changes of intracellular NADPH/NADP+ ratio in WT and ∆prxA under oxidative stress conditions caused by H2O2. As expected, exposing WT to 1 mM H2O2 significantly decreased the NADPH/NADP+ ratio, which is considered to be the result of the NADPH being used for H2O2 decomposition (Fig. 1B). The ∆prxA strain was determined to have a slightly increased NADPH/NADP+ ratio under normal conditions in comparison with that of WT. In sharp contrast to the WT, H2O2 exposure further increased the NADPH/NADP+ ratio in ∆prxA (Fig. 1B), indicating that PrxA consumes NADPH to decompose H2O2. Taken together, we concluded that the NADPH-consuming PrxA plays an essential role in H2O2 detoxification.

Decreasing NADPH impairs antioxidant ability

One of the major NADPH-producing enzymes in A. nidulans is identified as glucose-6-phosphate dehydrogenase (G6PD, encoded by AN2981) (Wennekes et al. 1993). To reveal the direct link between intracellular NADPH and cellular defense against oxidative stress in A. nidulans, we attempted to construct and phenotypically characterize the G6PD deficiency strain (∆gsdA). However, only heterokaryon mutants were obtained (data not shown), suggesting that gsdA may be essential for cell development and growth in A. nidulans. To analyze the functions of the potentially essential gene on oxidative stress resistance, we used the conditional promoter replacement strategy (Marchegiani et al. 2015). This strategy uses the niaD promoter (niaD.P), a nitrogen-regulated promoter from A. nidulans, to replace the endogenous promoter of a target gene to enable strict regulation. Up- and down-regulated expression can be achieved in the presence of NO3 and NH4+ as the sole nitrogen source, respectively. Additionally, proline can be used as a neutral nitrogen source to partially derepress the activity of niaD.P from NH4+ suppression. Using this strategy, the conditional mutant nP.gsdA was successfully constructed (Additional file 1: Fig. S2).

In the absence of H2O2, nP.gsdA exhibited drastically attenuated growth under NH4+ repression conditions (Fig. 1C). The addition of proline partially relieved growth inhibition from NH4+ repression, whereas the addition of NO3 almost recovered the growth rate compared with that of WT under unstressed conditions (Fig. 1D–E). These diverse phenotypes of the conditional mutant responding to the three nitrogen sources further supported the deduction that G6PD is important for fungal development. To provide insights into how intracellular NADPH levels affect the cell growth rate, we measured the intracellular NADPH/NADP+ ratios of nP.gsdA and found profound fluctuation of the NADPH/NADP+ ratio in response to different nitrogen sources (Fig. 1F). NO3 induced a sixfold higher NADPH/NADP+ ratio in nP.gsdA than that in WT, whereas proline and NH4+ decreased the ratio to 2/3 and 1/5 of that of WT, respectively. Obviously, depressing of G6PD decreased intracellular NADPH, which should be responsible for fungal growth retardation.

Next, we investigated how NADPH decrease affects resistance ability of the fungus to oxidative stress. Although nP.gsdA remained alive under NH4+ conditions, the poor cellular growth should make it difficult to estimate the severity of the H2O2 damage under these conditions (Fig. 1C); therefore, we compared conidial viabilities in response to H2O2 treatment between WT and nP.gsdA strains by counting the colonies formed. The survival of nP.gsdA was poorer than that of WT under the oxidative stress conditions induced by 1 mM H2O2 (Additional file 1: Fig. S6A). Consistent with this result, the activity of G6PD and the corresponding NADPH/NADP+ ratio were significantly repressed by NH4+ (Additional file 1: Fig. S6B–C). Together with the fact that the slight derepression of gsdA by proline partially alleviated the H2O2 resistance defect of the mutant (Fig. 1D), we may conclude that the artificial down-regulation of NADPH levels impairs fungal H2O2 resistance ability. This is in agreement with the above-mentioned finding that the indispensable antioxidant PrxA employs NADPH for ROS elimination.

Increasing NADPH also impairs cell antioxidant ability

Given that the intracellular NADPH level is crucial for fungal antioxidant ability, increasing intracellular NADPH levels may be beneficial for fungal oxidative defense, as in Drosophila melanogaster and some other animal cells (Salvemini et al. 1999; Leopold et al. 2003; Legan et al. 2008; Zhang et al. 2012; Xiao et al. 2018, 2020). In our study, we found that NO3 significantly induced gsdA expression (Fig. 2A) and accelerated G6PD activity (Fig. 2B), which resulted in at least a fivefold higher NADPH/NADP+ ratio in nP.gsdA than that in WT under either unstressed or stressed conditions (Fig. 2C). However, unexpectedly, nP.gsdA showed higher H2O2 sensitivity than that of WT (Fig. 1E), which led us to consider that increasing NADPH did not promote and, on the contrary, impaired cell antioxidant ability.

Fig. 2
figure 2

NO3 induction promoted transcription (A), activity (B), NADPH production (C) of G6PD in nP.gsdA strains. AC Fresh conidia (1 × 108) of WT (WT_pyrG) and nP.gsdA strains were cultivated in MM medium using NO3 as the nitrogen source for 16 h and then exposed to the indicated concentrations of H2O2 for 30 min for the following analysis (mean ± SD; n = 3, *P < 0.05, **P < 0.01, one-way ANOVA)

For further verification of this hypothesis, we constructed two other NADPH-high producing strains, gP.gsdA and gP.gndA. A constant and high-yield of NADPH was expected to be realized by replacing the native promoters of gsdA and gndA (6PGD encoding gene) with gpdA promoters, which is a strong constitutive promoter derived from the A. nidulans gpdA gene that encodes glyceraldehyde-3-phosphate dehydrogenase (Umemura et al. 2020). In gP.gsdA, the gpdA promoter produced approximately 100-fold more gsdA mRNA than that produced by the native gsdA promoter, but only half of that was produced by the niaD promoter (Additional file 1: Fig. S7A). The intracellular NADPH levels ranged from high to low across the nP.gsdA, gP.gsdA, and WT strains (Additional file 1: Fig. S7B), which was contrary to the orders of fungal H2O2 resistance (Additional file 1: Fig. S7C). In gP.gndA, both gndA mRNA and intracellular NADPH levels were significantly elevated by the gpdA promoter, which also lowered its antioxidant ability (Additional file 1: Fig. S8A–C). These results strengthened the fact that artificial increasing NADPH levels has adverse effects on fungal H2O2 resistance.

To investigate whether excess NADPH increased the levels of oxidants, we used fluorescent probes to measure and compare superoxide and H2O2 accumulated in WT and nP.gsdA strains. Although excess NADPH theoretically can be utilized by NOXs to produce superoxide (Leopold et al. 2003; Gupte et al. 2007; Lee et al. 2011), overexpression of A. nidulans gsdA did not lead to an increase in intracellular superoxide under both stressed and unstressed conditions (Fig. 3A). However, a high level of NADPH appeared to directly contribute to the production of H2O2 because a slight but significant increase of H2O2 accumulation was detected in NO3-induced nP.gsdA than that in WT under normal conditions (Fig. 3B). H2O2 exposure has further promoted intracellular H2O2 accumulation in both strains and enlarged the difference in H2O2 level between WT and nP.gsdA (Fig. 3B). Moreover, the elevation of H2O2 accumulation was prevented by the H2O2 scavenger N-acetyl-l-cysteine (NAC, 10 mM) (Fig. 3B), which also eliminated the H2O2-sensitivity difference between both strains (Additional file 1: Fig. S9). Therefore, it can be concluded that excess NADPH directly contributes to toxic level of H2O2 accumulation in fungal cells under oxidative stress conditions.

Fig. 3
figure 3

prxA suppression is responsible for the impairment of fungal resistance to H2O2 in G6PD-overexpression strains. All cultivations used NO3 as the sole nitrogen source. A, B Quantification analysis of intracellular O2·− and H2O2 in WT (WT_pyrG) and nP.gsdA. After precultivation, both strains were exposed to 0 or 1 mM H2O2 for 30 min followed by addition of ROS fluorescent probes. The ROS scavenger NAC (10 mM) was applied 1 h before the probe incubation. Fluorescence intensities of BES-So-AM and BES-H2O2-Ac were used to measure the level of intracellular superoxide and H2O2, respectively. All values were normalized by that in the unstressed WT (set to 100) (mean ± SD; n = 3, *P < 0.05, **P < 0.001, one-way ANOVA). C Relative expression levels of prxA in WT (WT_pyrG) and nP.gsdA. Strains were precultivated for 16 h, and then exposed to 1 mM H2O2 for 30 min. The level of prxA in WT without H2O2-treatment was set to 1 (mean ± SD; n = 3, *P < 0.05, one-way ANOVA). D Relative Prx-GFP levels in the WT (P_Gfp) and nP.gsdA (nP.gsdA/P_Gfp) strains. Inset, fluorescence spectra of Prx-GFPs from the corresponding cell lysates. After preculture, both strains were exposed to 0 and 1 mM H2O2 for 2 h. Cell lysates (1 mg/ml) were used for fluorescence analysis. E Effects of constitutive expression of prxA on fungal oxidative resistance. Conidia (1 × 105) of the strains were spotted and cultivated for 2 days on NO3-MM plates with or without 2 mM H2O2. Newly constructed strains are as follows: gP.prxA (replacing prxA promoter with gpdA promoter) and nP.gsdA/gP.prxA (replacing gsdA promoter with niaD promoter in gP.prxA)

Excess NADPH suppresses prxA transcription by downregulating NapA

Logically, intracellular H2O2 accumulation can be attributed to the inefficiencies of the key H2O2-decomposing enzymes. To explore whether excess NADPH impaired the antioxidant function of A. nidulans PrxA, we compared the transcriptional levels of prxA in WT and NO3-induced nP.gsdA. As expected, external H2O2 greatly increased PrxA transcriptional levels in WT (Fig. 3C), which was consistent with previous findings (Thon et al. 2010; Xia et al. 2018). H2O2-induced prxA expression was also observed in NO3-induced nP.gsdA strains; however, the induction strength was approximately 50% lower than that of WT (Fig. 3C). To investigate whether the transcriptional induction of prxA results of the corresponding changes of PrxA at protein level, we constructed GFP-tagged PrxA expression strains P_Gfp and nP.gsdA/P_Gfp, facilitating the quantification estimation of intracellular PrxA by fluorescence intensity measurements. The P_Gfp strain restored the oxidative resistance caused by prxA deletion (Additional file 1: Fig. S3), indicating the full function of Gfp-tagged PrxA. The same change tendency between gene transcription and protein expression was observed: the induction strength of PrxA in nP.gsdA was lower than in WT, which was indicated by the fluorescence intensity of PrxA-GFP in P_Gfp and nP.gsdA/P_Gfp under H2O2 treatment conditions (Fig. 3D). We hypothesized that the adverse induction of prxA expression caused by excess NADPH may account for the H2O2 accumulation and subsequent H2O2 defense defect in NO3-induced nP.gsdA. To verify this, we constructed two prxA-constitutively expressing strains (gP.prxA and nP.gsdA/gP.prxA) using WT and nP.gsdA as parent strains, respectively (Additional file 1: Fig. S2), and analyzed their antioxidant abilities. In both strains, constitutive expression of prxA was realized by replacing the prxA promoter with gpdA promoter. As expected, constitutive expression of prxA abrogated the distinct of H2O2-resistance between gP.prxA and nP.gsdA/gP.prxA, which was in sharp contrast to WT and nP.gsdA (Fig. 3E). Collectively, these data further illustrate that NADPH may determine the antioxidant ability of fungi via regulating the gene transcription of PrxA, the frontline defender against H2O2.

Repression of prxA transcription by accelerating intracellular NADPH production led us to infer that the function of NapA, the common transcriptional activator of fungal antioxidant genes, including prxA, is impaired under these conditions since NADPH should be the electron donor for NapA reduction and result in consequent deactivation of NapA (Thon et al. 2010). To validate this prediction, we first examined whether NapA can correctly localize in response to H2O2 exposure in the presence of excess intracellular NADPH. A GFP-tagged NapA was introduced to replace the original NapA in WT to construct N_Gfp (NapA-GFP) (Additional file 1: Fig. S3). The H2O2 resistance of N_Gfp was similar to that of WT (Additional file 1: Fig. S3), indicative of the functionality of this NapA::GFP fusion. The strain N_Gfp was further transformed with the pyrG-niaD.P-gsdA cassette to construct a new strain nP.gsdA/N_Gfp which can realize the NO3-inducible overexpression of gsdA in the fluorescent strain. Then, we characterized NapA::GFP localization and found that H2O2 exposure quickly resulted in NapA::GFP nuclear accumulation in both strains, indicating that activation of NapA was not interfered by excess intracellular NADPH (Fig. 4). Surprisingly, we found that nP.gsdA/N_Gfp showed significantly reduced fluorescence intensity compared with that of N_Gfp regardless with or without the presence of H2O2 (Fig. 4), indicating that excess intracellular NADPH impaired NapA production, which occurred prior to H2O2 exposure. Thus, we demonstrated that excess intracellular NADPH modulates fungal antioxidant activity by downregulating the amount of NapA rather than by affecting the redox state of NapA.

Fig. 4
figure 4

Overexpressing G6PD downregulated expression of NapA but did not interfere with its nuclear localization. Fresh conidia (1 × 105) of N_Gfp (NapA::GFP) and nP.gsdA/N_Gfp (replacing gsdA promoter with niaD promoter in N_Gfp) was incubated in NO3-MM for 10 h and was then exposed to 0 or 2 mM H2O2 for 20 min; nuclei were stained with Hoechst 33258 for 15 min. Scale bar, 20 μm. Images were captured using a laser confocal microscope

Excess NADPH obligatorily activates AnCF to repress napA expression

To elucidate the mechanism whereby NapA expression was downregulated by excess NADPH, we focused on the dynamics of the levels of AnCF, which is a key transcriptional repressor of NapA (Thon et al. 2010; Hortschansky et al. 2017). The HapB, HapC, and HapE subunits of AnCF are all necessary for DNA binding (Thon et al. 2010; Hortschansky et al. 2017). AnCF senses the redox status of the cell via oxidative modification of thiol groups within HapC; oxidized HapC is then unable to participate in AnCF assembly, but can be reduced by the thioredoxin system (TrxA and TrxR) for recycling in the AnCF assembly. Thus, we questioned if excess NADPH can over-reduce HapC, leading to the ROS-resistant defect of nP.gsdA. If this is the case, deletion of hapC should relieve fungal H2O2 sensitivity caused by excess NADPH. Thus, we have constructed a hapC deletion strain (∆hapC) (Additional file 1: Fig. S1) and overexpressed G6PD in this mutant (nP.gsdA/∆hapC) to understand the relationship among HapC, excess NADPH, and cell H2O2 resistance. Deletion of hapC resulted in a great growth defect under normal conditions (Additional file 1: Fig. S4), which is consistent with previous reports (Papagiannopoulos et al. 1996). Overexpression of gsdA using niaD.P under NO3 conditions also realized excess NADPH accumulation in nP.gsdA/∆hapC (Fig. 5A). Next, we compared conidia survival rates of WT, nP.gsdA, ∆hapC, and nP.gsdA/∆hapC strains in response to oxidative stress induced by H2O2 under NO3 induction (Fig. 5B). The ∆hapC strain showed significantly decreased survival rate in all H2O2 stress conditions compared with that of the WT strain, indicating that AnCF is indispensable to A. nidulans oxidative stress resistance. Notably, overexpressing gsdA in ∆hapC did not impair fungal oxidative stress resistance, and, in contrast, substantially rescued the survival rate of nP.gsdA/∆hapC strain (Fig. 5B), clearly indicating that, in the absence of AnCF, extra NADPH supply is advantageous for fungal ROS defense. That is to say, impairment of the antioxidant ability caused by excess NADPH in WT is mediated by the AnCF complex.

Fig. 5
figure 5

Intracellular NADPH levels determine the oxidative stress resistance via regulating AnCF complex assembly. A Overexpression of gsdA resulted NADPH accumulation in ΔhapC. WT (WT_pyrG), nP.gsdA, ∆hapC, and nP.gsdA/∆hapC were cultivated in NO3-MM liquid media for 16 h, then the lysates were used for the relative NADPH/NADP+ ratio calculation. NADPH/NADP+ value of WT was set to 1 (mean ± SD; n = 3, **P < 0.01, one-way ANOVA). B AnCF is involved in impairment of H2O2 resistance caused by excess NADPH. Fresh conidia (1 × 108) of four strains were spread on NO3-MM plates containing the indicated concentrations of H2O2. Colonies were counted after a 48-h incubation, and survival rates were expressed as percentages of the CFU for strains incubated without H2O2 (mean ± SD; n = 3). C Immunoblot quantification of AnCF complex levels (top) and HapC levels (bottom box) in WT and nP.gsdA during H2O2-treatment; H_Flag (WT expressing Flag-tagged HapC) and nP.gsdA/H_Flag (nP.gsdA expressing Flag-tagged HapC). Cell-free lysates (100 µg) from each sample were loaded to native- (top) and SDS-PAGE (bottom box). AnCF and HapC were detected using anti-Flag antibody; actin was used as a control and was detected using an anti-Actin antibody. D Quantitated graph for intracellular AnCF level normalized to actin (mean ± SD; n = 3, *P < 0.05, one-way ANOVA). E Time course analysis of NADPH/NADP+ level in H_Flag and nP.gsdA/H_Flag during 1 mM H2O2 treatment. Strains were precultivated in NO3-MM liquid media for 16 h and then exposed to 1 mM H2O2 for 30, 60, and 90 min (mean ± SD; n = 3, t-test). F Relative napA mRNA levels in these strains with or without treatment of H2O2. Strains were cultivated in NO3-MM liquid media for 16 h, and then treated by H2O2 for 30 min and 60 min (mean ± SD; n = 3, *P < 0.05, **P < 0.01, n.s., not significant, two-way ANOVA)

Next, we obtained insight into the relevance between the level of intracellular NADPH and AnCF complex assembly. The Flag-tagged HapC was introduced into both WT and nP.gsdA strains to replace native HapC and construct H_Flag and nP.gsdA/H_Flag strains, respectively (Additional file 1: Fig. S4), enabling the measurement of the cellular level of AnCF in both strains by western blotting. We first confirmed that HapC-Flag protein in both H_Flag and nP.gsdA/H_Flag strains complemented the growth delay caused by hapC deletions (Additional file 1: Fig. S4), indicating that the fusion protein was functional. No bands were detected in cells expressing untagged HapC in WT (data not shown). Considering that the assembly and dissociation of AnCF may affect the dynamics of the intracellular AnCF content, we measured the content of AnCF with time. The HapC bands on reducing SDS-PAGE showed that the total amount of HapC in WT cells was relatively stable across the 120-min observation period under H2O2-treatment conditions (Fig. 5C, bottom box, and Additional file 1: Fig. S10), while levels of AnCF complex in non-reducing native PAGE fluctuated with time (Fig. 5C, top and Fig. 5D). During the earlier 30 min of H2O2 exposure, the intracellular AnCF level declined to 2/3 the level of the pretreatment sample in H_Flag strain. Extending the H2O2 exposure time to 90 and 120 min has gradually recovered and stabilized the AnCF formation to the original level. Interestingly, we found that changes in the level of intracellular NADPH in H_Flag strain kept pace with the fluctuation of AnCF: a sudden drop in the first 30 min, which then returned to the original level within the next 60 min (Fig. 5E). Thus, we deduced that the NADPH intracellular contents may determine the level of the AnCF complex. This supposition was further supported by investigating the AnCF and intracellular NADPH profiles in NO3-induced nP.gsdA/H_Flag, which was proved to be very similar to those present in H_Flag (Fig. 5C–E); moreover, the NADPH level in nP.gsdA/H_Flag was found to be well above that in H_Flag at any time (Fig. 5E). In response to the elevated NADPH, AnCF content also keeps higher in nP.gsdA/H_Flag than that in H_Flag (Fig. 5D). These data, taken together, showed that the initial “down then up” fluctuations of NADPH levels and AnCF contents are the first response of the fungal cells to the H2O2 stimulus.

We further deduced that the “down then up” fluctuation of AnCF content would result in a reverse fluctuation of NapA levels and corresponding up- and downregulated expression of A. nidulans prxA. This was verified by the following transcriptional changes of napA in strains upon H2O2 treatment (Fig. 5F). Exposure to H2O2 for the first 30 min induced napA expression in WT and nP.gsdA, as opposed to the downregulation of intracellular AnCF level in both strains (Fig. 5F). Notably, deletion of hapC drastically elevated napA induction amplitude compared with that of WT during the first 30 min of H2O2 exposure, confirming the transcriptional repression effect of AnCF on napA. Extending H2O2 exposure from 30 to 60 min decreased napA transcription in WT and nP.gsdA (Fig. 5F), which contrasted with the changes in the levels of AnCF (Fig. 5C, top). Conversely, in ∆hapC and nP.gsdA/∆hapC, extending H2O2 exposure from 30 to 60 min did not lower napA transcription levels (Fig. 5F), further confirming the involvement of AnCF in the negative regulation of napA. Since NapA is the transcription activator of prxA, the “down then up” content fluctuation of NADPH should be ultimately used to trigger and subsequently break the induction of prxA to provide the on-demand cellular level of PrxA for oxidative stress defense in A. nidulans.

Reversible inhibition of G6PD may account for the NADPH fluctuation

Under oxidative stress conditions, a sudden “down” of intracellular NADPH level at the initial stage should be the result of NADPH consuming by PrxA for fungal antioxidant machinery. The following “up” of NADPH content suggests a quick activity acceleration of the NADPH-producing enzyme. A. nidulans G6PD may act as the key enzyme, because G6PDs from other sources have been reported to be reversibly inhibited by NADPH, which can be broken by rapid withdrawal of NADPH (Ramos-Martinez 2017). To verify that, we prepared recombinant G6PD of A. nidulans to test the NADPH-dependent inhibition and disinhibition of fungal G6PD in vitro. As shown in Additional file 1: Fig. S5, premixing G6PD with NADPH effectively inhibited fungal G6PD activity, which is indicative of self-braking of G6PD by its product NADPH. Next, we have investigated the disinhibition of G6PD by employing A. oryzae flavohemoglobin (a NADPH-dependent nitric oxide dioxygenase) (Zhou et al. 2011) as an NADPH scavenger. A rapid G6PD activation was achieved by the addition of flavohemoglobin and nitric oxide release reagent to the reaction buffer (Additional file 1: Fig. S5), indicating that regulation of A. nidulans G6PD activity is dependent on disinhibition. Taken together, these results supported our view that the most possible mechanism of NADPH fluctuation may be the result of the rapid NADPH consumption by PrxA upon oxidative exposure coupling the subsequent regeneration of NADPH via the disinhibition of G6PD. Moreover, the fungus may take advantage of the fluctuation of intracellular NADPH to regulate AnCF assembly in response to oxidative stress.


In A. nidulans, NADPH acts as a major reducing equivalent to reduce intracellular ROS, thus in this study, deficiency of NADPH, leads to low fungal viability under oxidative stress as expected. However, an increasing NADPH levels also impairs cellular tolerability to H2O2. Basing on our experiments, we reasoned for the unexpected phenotype as follows: excess NADPH promotes the assembly of AnCF, which in turn suppressed NapA expression, leading to low levels of PrxA and the eventual impairment of antioxidant ability. Moreover, the “down then up” fluctuation of the intracellular NADPH level, together with reversible inhibition activity of the fungal G6PD, lead us to deduce that the rhythm of NADPH may be an efficient survival strategy adapted by A. nidulans to defend against H2O2, as indicated in the proposed model (Fig. 6): the initial sudden decrease of NADPH resulting from PrxA-dependent H2O2-decomposition triggers the expression of prxA via the cascade regulation composed of AnCF and NapA. The subsequent recovery of NADPH allows this to shut down the further induction of prxA, which may be essential to maintain reasonable utilization of NADPH for other pathways under ROS stress conditions.

Fig. 6
figure 6

Schematic illustration of NADPH-mediated autoregulation mechanism for adaptation to oxidative stress in A. nidulans. AC Under normal conditions (A, Homeostasis stage) the NADPH pool provides an adequate supply of NADPH to reduce AnCF to its activated state, which will suppress NapA and subsequently lower the levels of PrxA. Due to the absence of high consumption of NADPH, the intracellular NADPH pool is maintained at a high level causing G6PD activity inhibition. Thus, the inflow of NADPH from G6PD to the NADPH pool is quite limited. Once fungi encounter H2O2 (B, Adaptive response stage), NADPH is suddenly needed to decompose H2O2 using PrxA, leading to an immediate drop within 30 min. Inevitably, the inflow of NADPH from the NADPH pool to reduce AnCF is declined. The accumulation of oxidized AnCF activates napA to supply more PrxA until the PrxA pool is saturated, accelerating NADPH consumption while resisting H2O2. Next, the depletion of NADPH reactivates G6PD, and NADPH eventually reaches to the initial level in the next 30 min (C, Cell survival stage). The increased NADPH level triggers the reactivation of AnCF, which consequently shuts down NapA induction, represses NapA activity via NADPH-dependent reduction to block PrxA induction, and ultimately keeps the PrxA pool at a moderate level (C, Cell survival stage)

In animals, NADPH also serves as an important participant in maintaining cellular redox homeostasis. However, the effects of intracellular NADPH accumulation on cell fates are not always the same case. In some cells, increased NADPH levels were found to be beneficial in protecting against oxidative stress. For example, overexpression of G6PD decreases ROS accumulation in response to exogenous and endogenous oxidant in vascular endothelial cells (Leopold et al. 2003) and in aldosterone-treated bovine aortic endothelial cells (Leopold et al. 2007). Additionally, overexpression of G6PD was reported to extend the life span of transgenic D. melanogaster (Legan et al. 2008). Furthermore, G6PD induction was reported to be instrumental in regenerating the intracellular GSH pool in human HeLa cells, which may facilitate the cellular protection against oxidant injuries (Salvemini et al. 1999) as GSH is a critical antioxidant and scavenges ROS directly or as cofactor of the glutathione and thioredoxin systems in animal cells (Stincone et al. 2015). The most notable difference in ROS-defense system between A. nidulans and animals is that fungal GSH is not essential in H2O2 defense (Sato et al. 2011), which may be one reason for the different phenotypes between A. nidulans and animals triggered by excess intracellular NADPH under oxidative stress conditions. On the other hand, the accumulation of intracellular NADPH by overexpression of G6PD can result in reductive stress and ultimately ROS production (Xiao et al. 2018, 2020). For example, overexpression of G6PD upregulated mRNA expression of NOX gp91phox and p22phox subunits, potentiating ROS production and oxidative damage in mouse pancreatic β cells and thymic lymphoma cells, which provided evidences for NOX-mediated oxidative stress by excess intracellular NADPH (Tome et al. 2006; Lee et al. 2011). Upregulated G6PD expression also increases oxidative stress in human tissue, such as failing human heart (Gupte et al. 2007). The adverse effects of NADPH on ROS damage protection have been interpreted as excess NAD(P)H being used by NOXs to produce ROS (Bedard et al. 2007; Sarsour et al. 2009; Handy et al. 2012). However, excess NADPH did not facilitate the intracellular O2·− accumulation in A. nidulans (Fig. 3A). Therefore, animals do not apparently share the same mechanism with fungi of a NOX-mediated increase of ROS triggered by excess NADPH. The homologs of AnCF and NapA are highly conserved in many filamentous fungi and yeasts, including most of the Aspergillus species (Brakhage et al. 1999; Zheng et al. 2015; Hortschansky et al. 2017; Mendoza-Martinez et al. 2017), S. cerevisiae (McNabb et al. 1995; Rodrigues-Pousada et al. 2019), S. pombe (McNabb et al. 1997; Boronat et al. 2014), Kluyveromyces lactis (Mulder et al. 1994; Imrichova et al. 2005) and Cryptococcus neoformans (Loussert et al. 2010; Pais et al. 2016). While systematic studies remain lacking, the striking mechanistic similarities in ROS defense between A. nidulans and these fungi indicate that utilizing NADPH fluctuation to assure timely activation and avoid overactivation of the key antioxidants as an oxidative adaptation strategy may be conserved among these fungi.

Is it necessary for A. nidulans to regulate the activity of PrxA in a feedback manner? If there was not a NADPH-mediated genetic regulation cascade between PrxA and AnCF as well NapA, the continued presence of H2O2 will result in substantial PrxA production. The accumulated PrxA will deplete the intracellular NADPH for its ROS decomposition reaction, depriving NADPH from other important NADPH-utilizing cell metabolisms such as the repair of oxidized proteins (Lu et al. 2014), fatty acid synthesis, reductive assimilation of inorganic sulfur (Thomas et al. 1991), and restoration of cellular pools of reduced glutathione and thioredoxin (Miller et al. 2018), which may lead to more cellular damage. Thus, as a double-edged sword in response to oxidative stress, prxA should be expressed on-demand. A similar case was observed in Schizosaccharomyces pombe (Day et al. 2012; Brown et al. 2013). At high levels of H2O2, which are acute stressful to the yeast, inactivating Tpx1 (a PrxA homolog) by hyperoxidation to a Trx1-resistant sulfinic (SOOH) derivative was essential to target reduced Trx1 toward other substrates, allowing the repair of oxidized proteins vital for cell survival under these conditions. In fact, inactivating Tpx1 to save reduced Trx1 is tantamount to saving NADPH since reduced Trx1 is derived from NADPH-dependent reduction in yeasts. Notably, the A. nidulans PrxA is a hyperoxidation-resistant peroxidase that can resist extremely high concentrations of H2O2 (Xia et al. 2018). Thus, PrxA cannot shift NADPH toward other substrates via self-inactivation by a high concentration of H2O2 as what occurs with yeast Tpx1. Alternatively, the balance of NADPH supply to H2O2-defense system and other metabolism in A. nidulans may be realized via the NADPH-mediated feedback mechanism.

NADPH-dependent metabolic reactions have been identified to be indispensable tools in biomanufacturing. A vast number of important targets including most natural products, amino acids, fatty acids, nucleotides, sterols and steroids are synthesized via NADPH-dependent pathways (Yu et al. 2018; Zhang et al. 2018; Gu et al. 2021). Furthermore, NADPH serves as the preferred electron donor for some of the largest and most versatile classes of enzymes such as cytochrome P450 enzymes and enoate reductases, which have broad applications in metabolic engineering. In these systems, the presence of NADPH-consuming reactions lowers the intracellular NADPH level and decreases the desired reaction rate, which may cause defective effects to cells; therefore, elevating G6PD enzymatic activity to enhance NADPH supply was a widely adopted strategy for metabolic engineering. Filamentous fungi are arguably the most industrially important group of microorganisms. Production processes involving these simple eukaryotes are often highly aerobic in nature, which implies that cultures are routinely subject to oxidative stress (Gibbs et al. 2000; Li et al. 2011). Thus, supply of more intracellular NADPH by overexpressing G6PD to improve fermentation products may be challenging because of the possible oxidative damage caused by the excess intracellular NADPH. However, the coincidental overexpression of G6PD and PrxA can partially alleviate oxidative damage (Fig. 3E), leading us to hypothesize that further the elevating cellular PrxA level by a genetic method may cover the defective effects produced by G6PD overexpression on ROS resistance and ultimately contribute to NADPH-demanding production synthesis.


Increasing intracellular NADPH promotes the assembly of the CCAAT-binding factor AnCF, which in turn suppressed NapA, a transcriptional activator of the key ROS scavenger PrxA, leading to low levels of PrxA and the eventual impairment of antioxidant ability in Aspergillus nidulans.

Availability of data and materials

The data and the materials are all available in this article as well as the Additional file 1.



Reactive oxygen species


Glucose-6-phosphate dehydrogenase


Nicotinamide adenine dinucleotide phosphate






Thioredoxin reductase


6-Phosphogluconate dehydrogenase


Glutathione reductase


NADPH oxidase


Download references


Not applicable.


This study was supported by the International S&T Innovation Cooperation Key Project (2017YFE0129600), the National Natural Science Foundation of China (21672065, 22077032 and 21636003), the National Major Science and Technology Projects of China (2019ZX09739001), the Fundamental Research Funds for the Central Universities (22221818014), and the 111 Project (B18022).

Author information




SZ and PW conceived of the study, JL performed the experiments and developed the manuscript with SZ. YS and FL contributed to the gene deletion experiment and helped analyzed the data. YZ and YY provided key technical support. SZ, ZZ and PW revised manuscript.

Corresponding authors

Correspondence to Ping Wang or Shengmin Zhou.

Ethics declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Additional file 1: Table S1.

A. nidulans strains used in this study. Table S2. Primers used in this study. Fig. S1. CRISPR/Cas9-mediated disruptions of hapC gene in A. nidulans. A Schematic diagram of gene disruption by in vitro assembled Cas9/sgRNA and donor DNAs. B Gene replacement of the target gene loci via the homology-directed repair pathway. Left, Schematic diagram of locus changes of target genes before and after gene replacement. Right, Confirmation of gene disruptions by PCR using the indicated primer pairs. hapC disruptant was verified with primer pairs AN4034-uF/pyrG-check-5′-R (lane 1), AN4034-dR/pyrG-check-3′-R (lane 2), respectively; primer pair hapC-check-F/hapC-check-R was used to amplify AN4034 ORF in the parent (lane 3) and disruptant (lane 4) strains. M, marker. Primers are listed in Table S2. Fig. S2. Constructions of promoter substitution strains of A. nidulans. AD Conditional promoter replacement strategy for homologous recombinant strains nP.gsdA (A), gP.gsdA (B), gP.gndA (C), gP.prxA (D) and validation of the corresponding recombinations by PCR with the indicated primers (right). The isolated nP.gsdA transformant was verified with primer pairs AN2981-uF/pyrG-check-5′-R (lane 1), AN2981-dR/pyrG-check-3′-R (lane 2), respectively; Primer pair AN2981-uF/AN2981-dR was used to amplify corresponding regions in the transformant (lane 3) and parent (lane 4) strains. M, marker. The similar methods were performed to validate the resultant transformants of gP.gsdA and gP.gndA. For verifying the indicated recombination in gP.prxA, primer pair M13-R/pyroA-check-3′-F was used to amplify corresponding regions in the transformant (lane 3) and parent (lane 4) strains. Primers are listed in Table S2. Fig. S3. Construction and phenotype analysis of GFP-tagged NapA strain and GFP-tagged PrxA strain. A and C Schematic diagram of locus changes of napA gene (A) and prxA gene (C) before and after the gene replacements (left) and the corresponding confirmation using PCR (right). Primer pair napA-5′-F/napA-3′ UTR-R was used to amplify the corresponding region in the isolated transformant (lane 1) and the parent (lane 2) strains. M, marker. Primers are listed in Table S2. B GFP-tagged NapA strain (N_Gfp) shows similar H2O2 resistance to the control strain (WT_argB). Conidia (1 × 105) from both strains were spotted on MM plates with or without 2 mM H2O2 and incubated at 37 °C for 2 days. D GFP-tagged GFP strain (P_Gfp) recovered the H2O2 resistance and similar to the control strain (WT_argB). Conidia (1 × 105) from both strains were spotted on MM plates with or without 1 mM H2O2 and incubated at 37 °C for 2 days. Fig. S4. Construction of Flag-tagged HapC strain and phenotype analysis of variants of hapC mutant under unstressed conditions. A The recombinant plasmid pUC19-pyroA-hapC-Flag was used to transform to ∆hapC to construct H_Flag strain. The indicated recombinant transformant was isolated and further confirmed by PCR with the primer pair M13-R/pyroA-check-3′-F (right, lane 1). Lane 2 showed the corresponding result of the control strain. Primer are listed in Table S2. B Conidia (1 × 105) from the control (WT_argB), ∆hapC, nP.gsdA/∆hapC and HapC::Flag fusion protein-expressing strains H_Flag, and nP.gsdA/H_Flag were spotted on MM plate without H2O2 and incubated at 37 °C for 2 days. Fig. S5. Reversible inhibition of NADPH to G6PD activity. A SDS-PAGE (12%) analysis of purified recombinant AnG6PD expressed by E. coli. M, maker. Lane 1, purified AnG6PD. B Activity of recombinant G6PD samples was estimated by generation of NADPH using a UV–Vis spectrophotometer at 340 nm. G6PD activity was measured in a 1 ml reaction mixture containing 5 mM glucose-6-phosphate, 0.3 mM NADP+, and 5 µM G6PD; 30 µM NADPH was added to the reaction mixture to evaluate the inhibition effects on G6PD activity. Further addition of 20 µM flavohemoglobin (Fhb1), and 50 µM NO donor MAHMA NONOate was used to consume NADPH for inhibition relief of G6PD activity. Fig. S6. Downregulation of A. nidulans gsdA is detrimental to fungal growth and oxidative stress resistance. A Survival rates of WT (WT_pyrG) and nP.gsdA on MM plates using ammonium tartrate as nitrogen source under oxidative stress conditions. Fresh conidia (1 × 108) of both strains were spread on MM plates containing the indicated concentrations of H2O2. Colonies were counted after a 48-h incubation, and survival rate are expressed as percentages of the CFU for strains incubated without H2O2. BC G6PD activities and the relative NADPH/NADP+ ratio in WT and NH4+-repressed nP.gsdA strains before and after treatment of H2O2. Both strains were cultivated in MM liquid media using ammonium tartrate as the nitrogen source for 16 h, and then treated with the indicated concentrations of H2O2 for 30 min. (mean ± SD; n = 3, *P < 0.05, **P < 0.01, ***P < 0.001; n.s., not significant, one-way ANOVA.) Fig. S7. Replacement of gsdA native promoter with gpdA promoter also perturbed NADPH rhythm and impaired fungal resistance to H2O2. AB Relative expression levels of gsdA (A) and NADPH/NADP+ ratios (B) in WT (WT_pyrG), nP.gsdA and gP.gsdA (replacing gsdA promoter with gpdA promoter in WT) strains. All strains were precultivated in liquid NO3-MM for 16 h and then exposed to 1 mM H2O2 for 30 min. The WT level of gsdA was set to 1, and the levels of gsdA in other strains were normalized to this. C Comparison of the H2O2 resistance of WT (WT_pyrG), nP.gsdA and gP.gsdAstrains. (mean ± SD; n = 3, *P < 0.05, **P < 0.001, t-test.) Fig. S8. Enhancing NADPH by overexpression of gndA impairs fungal resistance to H2O2. AB Relative expression levels of gndA (A) and NADPH/NADP+ ratios (B) in WT (WT_pyrG) and gP.gndA (replacing gndA promoter with gpdA promoter in WT) strains. All strains were precultivated in liquid NO3-MM for 16 h and then exposed to 1 mM H2O2 for 30 min. The WT level of gndA was set to 1. (C) Comparison of the H2O2 resistance of WT (WT_pyrG) and gP.gsdA strains. (mean ± SD; n = 3, *P < 0.05, **P < 0.001, t-test.) Fig. S9. Intracellular H2O2 accumulation accounts for the growth retardation of nP.gsdA strain under oxidative stress conditions. Growth comparison of WT (WT_pyrG) and nP.gsdA strains under oxidative stresses or unstressed conditions. Conidia (1 × 105) from both strains were spotted and cultivated for 2 days on NO3-MM plates supplied with or without 10 mM NAC and 1 mM H2O2 as indicated by “+” and “−”, respectively. Fig. S10. Quantitative analysis of relative levels of intracellular HapC. Total contents of intracellular HapC levels of WT and nP.gsdA strains at different periods are shown by the intensity of bands on denaturing and reducing PAGE. HapC levels were normalized to actin contents calculated by the same method for the further quantitative comparison. Each value represents the mean ± SD of triplicate determinations (mean ± SD; *P < 0.05, **P < 0.01, one-way ANOVA).

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Li, J., Sun, Y., Liu, F. et al. Increasing NADPH impairs fungal H2O2 resistance by perturbing transcriptional regulation of peroxiredoxin. Bioresour. Bioprocess. 9, 1 (2022).

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI:


  • Peroxiredoxin
  • Oxidative stress
  • Aspergillus
  • Glucose-6-phosphate dehydrogenase
  • AnCF