Peroxiredoxins (Prxs) are a family of thiol-specific peroxidases that catalyze the reduction of peroxides, including H₂O₂ and organic peroxides. These antioxidant enzymes are found across all living organisms, and multiple Prxs are commonly found in one species (three in Escherichia coli, five in Saccharomyces cerevisiae, six in Homo sapiens, and ten in Arabidopsis thaliana) [1-5]. All Prxs contain an absolutely conserved peroxidatic cysteine residue named as the C_P, which is oxidized by different kinds of peroxides to cysteine sulfenic acid (C_P-SOH) [6]. In many Prxs, another cysteine exists, which is defined as the “resolving” Cys (C_R) for its action in resolving the C_P-SOH to a protein disulfide bond (C_P-C_R). This feature of the presence or absence of C_R has been used to distinguish the two subfamilies: 1-Cys and 2-Cys Prx families [6]. 2-Cys Prx contains a C_R within its molecule, whereas 1-Cys Prx lacks the intramolecular Cys, instead providing an extrinsic C_R by other proteins or small thiol molecules. There are two classes of 2-Cys Prxs, atypical and typical classes, depending on whether the C_P-C_R disulfide bond is formed between two adjacent subunits (typical) or within a single subunit (atypical). Prx isoenzymes are distributed in various subcellular locations and have specific physiological functions, such as detoxification or signaling [7-9]. Peroxide-oxidized Prxs are subsequently reduced by thiol-reducing equivalents, such as thioredoxin (Trx), glutathione, and glutathione transferase π [10-12]. The regenerated Prxs are then available for another catalytic cycle.

Reduction of the disulfide (C_P-C_R) formed in oxidized 2-Cys Prxs requires Trx systems consisting of Trx combined with Trx reductase (TrxR) [13, 14]. Trxs are small proteins with a molecular mass of 12-13 kDa, and ubiquitously distributed in most eukaryotes and prokaryotes. Trxs are characterized at the level of their sequence by the presence of two redox-active cysteines in the active motif (WCGPC). The two vicinal cysteines are key for reducing the disulfides in other proteins by cysteine thiol-disulfide exchange [15]. On the other hand, TrxR is a homodimeric protein containing an active disulfide and FAD and NADPH binding sites in each subunit. During the reduction of Prx, Trx directly reduces (C_P-C_R) in the oxidized 2-Cys Prx by disulfide exchange. Subsequently, TrxR catalyzes the reduction of oxidized Trx (Trx-S2) by using NADPH as the electron donor and its own redox-active cysteine pair with FAD as the cofactor [13, 16].

Some 2-Cys Prxs are readily inactivated by a high concentration of peroxide during catalysis. Such inactivation has been detected during in vitro characterizations of 2-Cys Prxs from eukaryotic cells [5, 17-19]. At high levels of peroxide, the C_P-SOH of the sensitive 2-Cys Prxs can be further oxidized by another peroxide molecule to form cysteine-sulfinic acid (Cys-SO₂H) in a process termed hyperoxidation, causing Prx inactivation [20-23]. The two sequence motifs “GGLG” and “YF” are well conserved in most eukaryotic typical 2-Cys Prxs, and these sequence features are common to sensitive 2-Cys Prxs [24]. The inactivation of 2-Cys Prxs has been suggested to be an evolutionarily adapted mechanism for eukaryotic cells, for the allowance of H₂O₂ accumulation to substantial levels for signaling purposes under certain circumstances [24].

Peroxide dose-dependent inactivation of 2-Cys Prxs is generally considered to be caused by irreversibly over-oxidizing the reduced C_P to cysteine-sulfinic acid (C_P-SO₂H) in Prxs. However, no concern was paid to the effects of the other two key components of the Prx systems, Trx and TrxR, on the inactivation during the oxidation process. In fact, at least TrxR has been reported to be inactivated by oxidants such as H₂O₂, which was proposed to give the enzyme the status of a cellular “redox sensor” [25]. Whether the activity of Trx is also sensitive to peroxide requires experimental investigation.

Many studies have clearly indicated that the substrates of Prx, peroxides, mediate a variety of physiological processes, such as life span, spore germination, cell survival, biomass production, and pathogen attack [26-30]. Therefore, systematic investigations on of the enzymatic properties of Prxs are necessary. To date, the biochemical properties and physiological functions of Prxs from most species, including bacterial, yeast, plant, and mammalian cells, have been investigated and described in detail, whereas the corresponding information on industrially and medically important filamentous fungi is very limited and fragmentary. Until now, only two Prxs from filamentous fungi, Aspergillus nidulans and A. fumigatus, had been investigated [16, 31]. A. fumigatus Prx, named Aspf3, was recently biochemically characterized in detail, and was confirmed to be a robust peroxidase. A. nidulans Prx, named An.PrxA [32], had also been confirmed to possess thioredoxin-dependent peroxidase activity. However, no additional information such as protein properties and catalysis characteristics is available for An.PrxA. Moreover, there are not any studies that provide insight into the H₂O₂ tolerability of these filamentous fungal Prxs.

A. nidulans is a representative of the important group of aspergilli with significance in the fields, including industry, medicine, and agriculture. These strains include the workhorse commercial fermentation fungus A. niger, the food processing fungus A. oryzae, the human pathogen A. fumigatus, and the toxigenic fungi A. flavus and A. parasiticus. Therefore, A. nidulans has become a classical model organism for developmental, genetic, cellular, and biochemical research [33], and it is one of the bestcharacterized of the filamentous fungi.

In the present study, we biochemically characterized A. nidulans recombinant Prx in detail and surprisingly observed its uniquely extraordinary ability to resist inactivation by high concentrations of H₂O₂. Furthermore, we found that filamentous fungal TrxR is the key element responsible for the extreme H₂O₂ tolerance of the Prx system.

Fungus, Media, and Culture Conditions

Aspergillus nidulans strain ABPU1 (biA1 pyrG89; wA3; argB2; pyroA4) was kindly provided by Dr. Hiroyuki Horiuchi (University of Tokyo, Japan). A. nidulans ABPU1 and its fungal variant were grown at 37°C in minimal medium (MM) (10 mM NaNO₃ ,1%glucose, 10 mM KH₂PO₄, 2 mM MgSO₄, 7 mM KCl, 2 ml/l Hunter’s trace metals) [33]. If needed amino acid was supplemented appropriately (arginine 0.2 g/l, uridine 0.6 g/l, uracil 0.55 g/l, biotin 0.4 mg/l, pyridoxine 0.4 mg/l). To analyze the tolerance of the strains against oxidative stress, three dilutions of mature A. nidulans conidia (10⁵, 10⁶, 10⁷) were spotted onto MM plates containing appropriate concentrations of oxidizing reagents. After incubation for 2 days at 37°C, the morphology and size of the fungal colonies were analyzed to determine their stress tolerance.

Gene Cloning and Expression Vector Construction

Trizol reagent (Invitrogen, USA) was used to extract total RNA from A. nidulans cells. The full-length cDNA of An.PrxA was obtained by 5’-3’ RACE using the GeneRacer Kit (Invitrogen, USA) as previously described [34]. The genes encoding An.PrxA, alkyl hydroperoxide reductase 1 (Sc.Ahp1), Sc.Trx, and Sc.TrxR were amplified from A. nidulans cDNA and S. cerevisiae genomic DNA by PCR using the specific primers listed in Table S1. The PCR products were digested with the corresponding restriction endonucleases and cloned into the expression vector pET28a. The resulting An.PrxA expression plasmid was used as the template DNA in subsequent PCR for site-directed mutagenesis. Two An.PrxA mutants, C31S and C61S, in which Cys 31 and Cys 61 were replaced individually by serine, were generated by following the protocol outlined in the Quick Change site-directed mutagenesis kit (Stratagene, USA) using complementary primers containing double-base mismatches that convert the codon for cysteine to a codon for serine. The correct introduction of mutations was verified by DNA sequencing. Plasmid constructions of the mutated versions were carried out as described for the wild-type enzyme.

Expression and Purification of the Recombinant Proteins

The above-mentioned expression plasmids and the previously constructed An.TrxA and An.TrxR expression plasmids [35] were introduced into E. coli BL21(DE3). Transformed cells were pre-cultured at 37°C in LB medium supplemented with 50 μg/ml kanamycin or ampicillin. Expression was induced by 0.5 mM isopropyl-β-D-thiogalactopyranoside at 20°C for 12 h. The cells were collected by centrifugation at 5,000 ×g for 5 min and resuspended in lysis buffer (20 mM sodium phosphate buffer, 500 mM NaCl, pH 7.4). The collected cells were disrupted with a homogenizer (JNBIO, China). The supernatant was obtained by centrifugation for 10 min at 10,000 ×g and then applied to HiTrap Chelating HP (GE Healthcare, USA). The target proteins were eluted with a linear gradient of imidazole from 0 to 500 mM. The fractions corresponding to the peak of target proteins were pooled for the following analyses.

Molecular Mass Determination

The purified protein was applied to a Superdex 200 increase 10/300 column (GE Healthcare) and then eluted at a flow rate of 0.5 ml/min with 20 mM sodium phosphate buffer containing 150 mM NaCl and 5 mM dithiothreitol (DTT). The molecular mass of the enzyme was determined on the basis of the mobilities of the standard proteins containing ribonuclease A (13.7 kDa), carbonic anhydrase (29 kDa), ovalbumin (44 kDa), conalbumin (75 kDa), aldolase (158 kDa), ferritin (440 kDa) and thyroglobulin (660 kDa).

Construction of AN8692 Disruption Strain

The primers used for AN8692 disruption are listed in Table S1. The AN8692 disruption cassette was constructed as a gene-replacement cassette comprising the argB gene of A. nidulans as the selection marker, flanked by 1 kb of 5’- and 3’- untranslated regions (UTR) of AN8692. PrimeSTAR HS DNA Polymerase (Takara Bio, Japan) was used for the PCR. The 1 kb of 5’- and 3’-UTR of AN8692 were amplified by PCR using gemomic DNA of A. nidulans ABPU1 and the respective primer pairs (ΔAn.PrxA-a and ΔAn.PrxA-c; ΔAn.PrxA-d and ΔAn.PrxA-f). The argB gene was amplified using pSSH1 (a plasmid harboring the A. nidulans argB gene) and argB-F and argB-R primers. The resulting three DNA fragments were mixed and amplified by fusion PCR using primers ΔAn.PrxA-b and ΔAn.PrxA-e to generate the An.PrxA disruption cassette. Finally, the resulting disruption cassette was introduced into the ABPU1 strain as previously described [36] to create an AN8692 disruption strain (An.PrxAΔ).

Southern Blot Analysis of AN8692 Disruption Strain

Genomic DNAs from the parent strain ABPU1 and the candidate disruption strain An.PrxA were digested using NdeI and HindIII, respectively, separated by gel electrophoresis, and then transferred to Hybond N+ membranes (Amersham Pharmacia Biotech, UK). Probes for hybridization were prepared by PCR using primers (Table S1) and then labeled using a digoxigenin (DIG) PCR labeling kit (Roche Diagnostics, Switzerland). Hybridization and signal detection were operated according to the manual supplied by the manufacturer.

Enzymatic Assays and Kinetic Parameter Determination

Peroxidase activity was measured by spectrophotometric assay monitoring the consumption of NADPH at 340 nm [13]. In the pretreatment operation, H₂O₂ and DTT were removed by the P6 column (Bio-Rad Laboratories, USA) prior to the activity assays. To determine the kinetic parameter, the reaction was carried out in a final volume of 1 ml of reaction mixture containing 50 mM phosphate buffer (pH 7.0), 200 μM NADPH, 1.13 μM Trx, 0.38 μM TrxR, 0.3 μM PrxA, and various concentrations of H₂O₂ or t-butyl hydroperoxide (t-BOOH). Absorbance spectra were all recorded with a U-5100 spectrophotometer (Hitachi, Japan). The initial reaction velocity was calculated on the basis of the linear portion of the curve. Apparent kinetic parameters of An.PrxA toward H₂O₂ and t-BOOH were determined at 0-16 μM H₂O₂ and 0-5 μM t-BOOH. The consumption of NADPH was monitored for 1 min, and the reaction velocity was calculated using the linear section of the plot.

Molecular Characterization of Recombinant An.PrxA

In 2007, Thon et al. [16] identified An.PrxA as the first filamentous fungal Prx. These authors isolated An.PrxA from A. nidulans cell extract using thioredoxin-affinity chromatography, and mass spectrometry analysis indicated that the protein is genetically encoded by AN8692 [16]. In an attempt to investigate the properties of the enzyme in detail, we cloned the full cDNA of this gene and gave a molecular characterization of the corresponding recombinant protein.

cDNA sequence analysis revealed that AN8692 is divided into three exons separated by two introns, which is completely in accordance with the predicted sequence information provided by AspGD (http://www.aspergillusgenome.org/). The deduced sequence of An.PrxA contains two cysteine residues at amino acid positions 31 and 61 (Fig. 1). The sequences surrounding Cys 61 show a universal active-site motif (PXXXS/TXXC) that is highly conserved in all Prx enzymes, which suggested Cys 61 is the site of oxidation by peroxides. Cys 31 of An.PrxA is well conserved through homologs from several species of bacteria and fungi, and the corresponding residue in S. cerevisiae Ahp1 has been experimentally assigned to C_R [13]. Arg 134 in the sequence distant to Cys 61 is also highly conserved and believed to be an essential active site in other Prxs [6, 37]. The two feature “GGLG” and “YF” motifs [24], which are conserved in peroxide-sensitive 2-Cys Prxs from eukaryotes, are not present in An.PrxA, suggesting An.PrxA is robust to peroxide. The C_R located preceding the C_P type of Prxs is found in some filamentous fungi, yeasts, and bacteria, but not in higher-order eukaryotes (Fig. 1).

Figure 1. Sequence alignment of deduced C_R ahead of C_P type of 2-Cys peroxiredoxin (Prx) family members. The deduced Prx sequences of Aspergillus fumigatus (XP_747849.1), Aspergillus niger (XP_001395908.1), Neurospora crassa (XP_964200.2), Penicillium nordicum (KOS40786.1), Saccharomyces cerevisiae (NP_013210.1), Candida albicans (XP_720512.1), Candida glabrata (XP_449850.1), Pichia pastoris (XP_002490288.1), Polynucleobacter wuianus (WP_068947794.1), Ralstonia syzygii (CCA854164.1), Burkholderia vietnamiensis (YP_001118388.1), and Herminiimonas arsenicoxydans (WP_011872175.1) were aligned using the programs Clustal and ESPrint. The conserved Cys 31, Cys 61, and Arg 134 of An.PrxA are indicated by the closed triangle, opened triangle, and closed circle, respectively. The putative PXXXS/TXXC motifs are indicated by the bar under the sequences.

The purified recombinant An.PrxA gave a single band with a molecular mass of approximately 20 kDa on SDS-PAGE under reducing conditions (Fig. 2A, lane 1). FPLC of An.PrxA on a gel filtration column in the presence of a buffer containing DTT yielded a peak at position corresponding to 40 kDa (Fig. 2B), suggesting that the native An.PrxA exists as a homodimer independent of the disulfide bond. When DTT was removed from the reduced mixture, the An.PrxA monomer band was oxidized gradually to a dimer band by oxygen in air under nonreducing conditions (Fig. 2A, lines 2 and 3). Under the same conditions, two major bands of the monomers with slightly different electrophoretic mobility were detected on SDS-PAGE (Fig. 2A, lines 2 and 3). The bands above and beneath might represent the originally reduced and the intramolecular disulfide bond formed forms of An.PrxA, respectively, because the ring-shaped protein should migrate slightly faster than the linear protein owing to the compactness of the protein conformation. To further confirm the redox state of two monomer forms, we treated the DTT-free An.PrxA with increasing concentrations of DTT. As a result, a gradual shift in the protein band from the position of higher mobility to that of lower mobility was observed (Fig. 2C), which supported our abovementioned supposition about the redox states of the two monomers. These observations suggested that oxidation of An.PrxA might result in the formation of both intramolecular and intermolecular linkages between Cys 31 and Cys 61, which seems to disturb the classification rule of An.PrxA between typical 2-Cys and atypical 2-Cys Prxs. However, it has been previously reported that in mammalian 1-Cys Prx, the C_P oxidation product, cysteine sulfenic acid (Cys-SOH), readily undergoes condensation with another noncatalytic cysteine to form disulfides in the denatured conformation of the protein during heating, which does not occur in its native state [38]. To investigate this possibility, we incubated the air-oxidized An.PrxA in the absence or presence of a thiol blocking reagent N-ethylmaleimide (NEM) to prevent the Cys-SOH group from forming disulfide after denaturation, before heating and followed by nonreducing SDS-PAGE analysis. As shown in Fig. 2D, the higher mobility monomer band was occluded by the thiol block reagent, while the dimer band still existed. These results suggested the higher mobility monomer contains a disulfide bond formed by heating during the denaturation process of SDS-PAGE, whereas the oxidized nondenatured An.PrxA contains a disulfide bond between intermolecular linkages but not intramolecular linkages. These structure features clearly indicate that An.PrxA belongs to the subfamily of typical 2-Cys Prxs.

Figure 2. SDS-PAGE and gel filtration analysis of recombinant Aspergillus nidulans peroxiredoxin (An.PrxA). (A) Nonreducing SDS-PAGE analysis (15% polyacrylamide gel) of An.PrxA. Line 1, An.PrxA (4 μg) with 10 mM dithiothreitol (DTT); line 2, An.PrxA (4 μg) first treated with 10 mM DTT followed by DTT removal by P6 column; line 3, DTT-free reduced An.PrxA (4 μg) exposed to air for 12h. (B) Estimation of the molecular mass of An.PrxA by gelfiltration. Inset: correlation between the molecular mass and elution volume of proteins. (C) Nonreducing SDS-PAGE analysis (15% polyacrylamide gel) of the DTT-free An.PrxA (4 μg) incubated with the indicated concentrations of DTT (lanes 1-7: 0, 1, 2, 4, 6, 8, and 10 mM DTT). (D) Effect of N-ethylmaleimide (NEM) treatment on the electrophoretic mobility (disulfide formation) of air-oxidized An.PrxA. The air-oxidized An.PrxA (4 μg) was incubated in the absence (Lane 1) or presence (Lane 2) of 10 mM NEM for 30 min before nonreducing SDS-PAGE analysis.

Catalytic Properties of An.PrxA

The oxidation of An.PrxA by peroxide was directly observed by non-reducing SDS-PAGE. Both H₂O₂ and t-BOOH triggered the oxidation of An.PrxA to form a major band as the disulfide-linked homodimer corresponding to 40 kDa (Fig. S1). The peroxidase activity of An.PrxA was monitored by following the decrease in A₃₄₀ attributable to the oxidation of NADPH in the reaction mixture containing NADPH, Trx, TrxR, and the substrates H₂O₂ (Fig. 3A) or t-BOOH (Fig. 3B). An.PrxA exhibited similar and robust Trx-dependent peroxidase activities toward both peroxides, which suggested that this enzyme acts as a peroxidase to remove both organic and inorganic peroxides. Enzymatic kinetics assays were performed to further analyze this bisubstrate reaction. The k_cat values measured with each peroxide substrate were almost equal (2.81 s^-1 and 2.87 s^-1 for H₂O₂ and t-BOOH, respectively). The K_m values for H₂O₂ and t-BOOH were both very low (2.80 μM for H₂O₂ and 1.05 μM for t-BOOH). The modest k_cat values and the low K_m values resulted in high catalytic efficiency (k_cat/K_m) for both H₂O₂ and t-BOOH (500 s^-1 mM^-1 and 1,370 s^-1 mM^-1 for H₂O₂ and t-BOOH, respectively), which are about 5- and 10-folds higher than that of Sc.Ahp1 (110 s^-1 mM^-1) [13], suggesting An.PrxA is a very efficient peroxidase towards both organic and inorganic peroxides in vitro. To confirm the catalytic role of the cysteine residues of An.PrxA, we replaced Cys 31 and Cys 61 individually with serine, thereby generating C31S and C61S mutant enzymes, respectively (Fig. S2). As shown in Fig. 3A and B, C31S and C61S mutants completely abolished the peroxidase activities towards both H₂O₂ and t-BOOH, whereas mixing of the two reduced mutants recovered half of their original activities. This clearly indicated that Cys 31 and Cys 61 are catalytic residues, and the disulfide bond was formed between the two cysteines from each subunit of the obligate homodimer.

Figure 3. Peroxidase activity analyses of Aspergillus nidulans peroxiredoxin (An.PrxA) variants. (A and B) NADPH oxidation was monitored as the decrease in A₃₄₀ in a 1 ml reaction mixture containing 50 mM phosphate buffer (pH 7.0), 1.2 μM An.TrxR, 10 μM An.TrxA, 150 μM NADPH, and 2 μM An.PrxA-C31S or An.PrxA-C61S or the mixture of 1 μM An.PrxA-C31S and 1 μM An.PrxAC61S, or 2 μM An.PrxA, with 0.5 mM H₂O₂ (A) or t-butyl hydroperoxide (t-BOOH) (B) as substrates, respectively. All the experiment were performed three times. (C) Growth of A. nidulans strains on agar plates containing 0 mM and 1 mM H₂O₂ or 0.5 mM t-BOOH. WT, wild type.

Protection of An.PrxA against H₂O₂ and t-BOOH

The in vitro catalytic properties strongly suggested that An.PrxA could provide protection against oxidative stress in vivo. To examine the effect of the An.PrxA gene on oxidative stress tolerance, we constructed a null mutant of the AnPrxA gene (Fig. S3). The unimpaired viability of the resulting AN8692 null strain under normal growth conditions demonstrated that this gene was not essential. On solid medium, the AnPrxA null mutant showed hypersensitivity towards either H₂O₂ or t-BOOH compared with its corresponding wild type (Fig. 3C), indicating An.PrxA represents a dual-function peroxidase to efficiently remove both organic and inorganic peroxides.

Extremely High Resistance of An.PrxA to H₂O₂

In other peroxidases, the sequence motifs “GGLG” and “YF” promote hyperoxidation by hindering the local unfolding necessary for disulfide formation with the resolving cysteine. By sequence analysis, we noticed that these motifs are not conserved in the An.PrxA sequence, thus suggesting An.PrxA may be a robust and resistant peroxidase to high oxidative stress. To prove this point, we investigated the resistance of An.PrxA to inactivation by high concentrations of H₂O₂. In comparison, we have also studied yeast Sc.Ahp1, another “GGLG” and “YF” absent Prx, which has been revealed to be more resilient to hyperoxidation than other eukaryotic Prxs [5, 39]. The time course of NADPH oxidation towards various concentrations of H₂O₂ showed that both peroxidase activities of the two Prx isoforms exhibited high resistance to high H₂O₂- mediated inactivation, but with different degrees (Figs. 4A and 4B). The A₃₄₀ decrease rates summarized in Fig. 4C indicated that even under high oxidative stress conditions, the peroxidase activity of An.PrxA increased or remained stable with the concentration of substrates increasing until to an extreme concentration of 600 mM. In comparison, 100 mM of H₂O₂ was the threshold concentration for Sc.Ahp1 to resist inactivation, while above 100 mM, the activities decreased in a dose-dependent manner. These results clearly indicated that both “GGLG” and “YF” motifabsent 2-Cys Prxs exhibited high abilities to resist inactivation by extremely high concentrations of H₂O₂, but An.PrxA system showed stronger tolerability.

Figure 4. Comparison of peroxidase activities of Aspergillus nidulans peroxiredoxin (An.PrxA) and Saccharomyces cerevisiae alkyl hydroperoxide reductase 1 (Sc.Ahp1) towards various concentrations of H₂O₂. (A and B) Peroxidase activities were measured in a 1 ml reaction mixture containing 5 μM An.TrxA, 0.6 μM An.TrxR, 150 μM NADPH, and 1 μM An.PrxA (A) or 5 μM Sc.Trx, 0.6 μM Sc.TrxR, 150 μM NADPH, and 1 μM Sc.Ahp1 (B) in the presence of 10-600 mM H₂O₂. (C) Quantitative comparison of peroxidase activities of An.PrxA and Sc.Ahp1 measured in (A) and (B). Peroxidase activities were reflected by A₃₄₀ decrease values at 1 min after the beginning of reaction. Results are the means ± SD of three independent experiments. *, p < 0.05; **, p < 0.01; ***, p < 0.001.

Molecular Basis for the Extremely High Resistance of An.PrxA to H₂O₂

An.PrxA and Sc.Ahp1 possess similar sequence characteristics, such as the absence of H₂O₂-sensitive feature motifs and the same order arrangements of C_P and C_R (Fig. 1). Therefore, we assumed that the great degree of different H₂O₂-dependent inactivation of the two peroxidases may be due to the Trx/TrxR systems rather than considering the H₂O₂-susceptibility distinction of the two Prxs. To determine the contribution of Trx/TrxR systems to the robust resistance, we set up a new assay by interchanging the Trx and TrxR pairs between the two Prx systems. The assay was carried out with 10 mM H₂O₂ and monitored by A₃₄₀ decrease. Unexpectedly, the catalytic activities of both recombination sets were even higher than those of their individual native forms, indicating that the Trx/TrxR combinations can functionally substitute in peroxidase assays of An.PrxA and Sc.Ahp1 (Fig. 5A). Interestingly, the enzyme recombination substantially altered the H₂O₂ tolerance of the two Prxs. As shown in Fig. 5B, equivalently changing An.TrxA/TrxR to Sc.Trx/TrxR critically promoted the susceptibility of An.PrxA to 600 mM H₂O₂, whereas An.TrxA/TrxR greatly increased the H₂O₂ tolerance of Sc.Ahp1, even slightly exceeding the level of the native An.PrxA combination, suggesting that the extremely high H₂O₂ tolerance of Prx requires robust An.TrxA/TrxR.

Figure 5. TrxR determined the H₂O₂-sensitivity of the peroxiredoxin (Prx) systems towards high conditions of H₂O₂. (A) An.PrxA combined with An.TrxA/TrxR or Sc.Trx/TrxR, and Sc.Ahp1 combined with Sc.Trx/TrxR or An.TrxA/TrxR were constituted to measure their peroxidase activities in the presence of 10 mM H₂O₂. (B) The peroxidase activities of the four Prx combination systems were measured in the presence of 600 mM H₂O₂. (C and D) An.TrxA/TrxR (C) and Sc.Trx/TrxR (D) were treated with indicated concentrations of H₂O₂ for 5 min before starting the reactions. (E) An.TrxA, An.TrxR, Sc.Trx, and Sc.TrxR was individually pretreated with 600 mM H₂O₂. (F) Sc.TrxR was pretreated with the indicated concentrations of H₂O₂. After pretreatment, the excess H₂O₂ was removed by P6 column and the pretreated enzymes were used for the peroxidase reaction systems in the presence of 0.5 mM H₂O₂. The reaction mixture contained the same amount of individual enzymes and NADPH as described in Figs. 4A and 4B. The results are representative of three independent experiments.

Further experiments were performed to compare the H₂O₂ tolerance of the two Trx/TrxR systems from the two separate sources. Individual Trx/TrxR systems were preexposed to a range of H₂O₂ concentrations for 5 min, and the activities were assayed by measuring NADPH oxidation coupled to 0.5 mM H₂O₂ reduction catalyzed by the two Prxs. As shown in Figs. 5C and 5D, for each pretreated concentration of H₂O₂, An.TrxA/TrxR displayed robust reduction activities, whereas concentrations higher than 200 mM H₂O₂ seriously inactivated Sc.Trx/TrxR in a dose-dependent manner, which is consistent with the data shown in Fig. 4. These results confirmed that it is Trx/TrxR dominating the H₂O₂ sensitivity of Prx systems under extreme conditions.

To further explore which element in Trx/TrxR is responsible for the extreme H₂O₂ tolerance of the An.PrxA system, we separately pretreated Trx and TrxR from the Prx assay set with 600 mM H₂O₂ and compared the inactivation levels reflected by the rate of NADPH oxidation catalyzed by Prx with 0.5 mM H₂O₂. As shown in Fig. 5E, pretreatment of Sc.TrxR seriously impaired the peroxidase activities of its corresponding Sc.Ahp1, whereas An.TrxR exhibited robust resistance to inactivation by 600 mM H₂O₂. It was easy to ascertain that neither of the pretreated Trxs determined the H₂O₂ sensitivity of their corresponding Prx systems because they showed identically high resistant abilities to extreme H₂O₂. To further confirm the H₂O₂ sensitivity of Sc.TrxR, we investigated the effect of the H₂O₂ concentration on the inactivation of Sc.TrxR using the pretreatment method. We found that the H₂O₂-induced decrease of H₂O₂ reduction activities was evident and that the inactivation rates were dose-dependent (Fig. 5F). To exclude the possibility that the high resistant abilities to H₂O₂ is Trx specific, we also tested the effects of the remaining Trx and TrxR combinations (An.Trx paired with Sc.TrxR and Sc.Trx paired with An.TrxR) on H₂O₂ sensitivity of the Prx systems (Figs. S5A and S5B). As was expected, possessing An.TrxR in the assay combinations would be beneficial to H₂O₂ tolerance, regardless of which Trx was combined. These data clearly indicated that it is the outstanding H₂O₂ tolerence of An.TrxR rather than Trx that ensures the extremely high resistance of An.PrxA to H₂O₂.

Activity Intolerance of the An.PrxA System to Excess t-BOOH

We subsequently investigated whether the same oxidative stress tolerance mechanism is present in the reduction process of t-BOOH, the organic substrate of An.PrxA. We measured and compared the activities of the two original fungal Prx systems reducing various concentrations of t- BOOH. To our surprise, a dose-dependent inhibition effect of t-BOOH on the peroxidase activities was observed in the assay systems of An.PrxA rather than Sc.Ahp1 (Figs. 6A and 6B). Unlike H₂O₂, 2 mM, a relatively lower concentration of t-BOOH, could trigger inactivation of the An.PrxA system. Interestingly, the interchange of Trx/TrxR combinations among the two fungal Prx assay sets did not alter the t-BOOH sensitivities of the Prx systems (Figs. 6C and 6D, Fig. S5C and S5D). It is notable that although the whole activities of the An.PrxA system were inhibited, the initial rates of NADPH oxidation within 10 sec kept as fast as that of the control with the appropriate substrate concentration of t-BOOH (1 mM), which was markedly different to the inactivation modes of Sc.Ahp1 by H₂O₂. These results suggested that the inactivation mechanism of Prx by excess H₂O₂ and t-BOOH, as well as the modes of action in detoxifying organic and inorganic peroxides by Prxs, may be distinct.

Figure 6. Comparison of t-butyl hydroperoxide (t-BOOH)-tolerant abilities of the Aspergillus nidulans peroxiredoxin (An.PrxA) and Saccharomyces cerevisiae alkyl hydroperoxide reductase 1 (Sc.Ahp1) systems. Peroxidase activity was measured in a 1 ml reaction mixture containing 150 μM NADPH, 1-8 mM t-BOOH, and the following different enzyme combinations: (A) 2 μM An.PrxA, 10 μM An.Trx, and 1.2 μM An.TrxR. (B) 2 μM Sc.Ahp1, 10 μM Sc.Trx, and 1.2 μM Sc.TrxR. (C) 2 μM An.PrxA, 10 μM Sc.Trx, and 1.2 μM Sc.TrxR. (D) 2 μM Sc.Ahp1, 10 μM An.Trx, and 1.2 μM An.TrxR. The results are representative of three independent experiments.

Filamentous fungi have industrial, pharmaceutical, and medical importance, and they frequently encounter high oxidative stress caused by host pathogen defense and fermentation conditions [40, 41]. Catalases, glutathione peroxidases, and Prxs have been considered as the central enzymes for protecting filamentous fungi from peroxide injury [30, 42-44]. The first two antioxidant enzymes have been intensively studied in filamentous fungi, but the information of filamentous fungal Prxs is very limited. Therefore, we molecularly characterized a recombinant filamentous fungal Prx cloned from A. nidulans, and discovered that its robust peroxidase activity to extremely high concentrations of H₂O₂ is dependent on the outstanding tolerance of filamentous fungal TrxR.

Although the bioinformatics analysis indicated that at least six Prx-like genes exist in the A. nidulans genome (http://www.aspergillusgenome.org/), we preferentially selected An.PrxA to study because of its strongly suggested antioxidative functions, and its special amino acid sequences property. “CCAAT” sequences were discovered in the promoter region of AN8692, and the CCAAT-binding factor (AnCF), which is an oxidative stress regulator, was found to fully repress the gene expression of AN8692 [32]. Additionally, AN8692 was greatly induced when strains were treated with H₂O₂ [32]. These information suggested An.PrxA should play a physiological function involving oxidative stress resistance in vivo. The predicted amino acids of An.PrxA show similar sequence characteristics to S. cerevisiae Sc.Ahp1 (Fig. 1), which was the first Prx characterized to have a C_R located at the N-terminal region preceding the C_P [13]. Sc.Ahp1 has been intensively investigated by several research groups, but some controversial conclusions in the determination of its catalytic residues, peroxide substrate specificities, subcellular locations, and substrate binding manners have been reached [5, 13, 39, 45, 46], which suggested that this type of Prx requires further characterization. The substrate preference of Sc.Ahp1 to t-BOOH is in agreement with its specialized cellular function as a t-BOOH scavenger [45]. In fact, almost all types of Prx isoenzymes exhibit different substrate preferences corresponding to their separated specialized functions [5]. Therefore, the efficient peroxidase activities to both H₂O₂ and t-BOOH suggested An.PrxA represents a dual-function peroxidase to efficiently remove both organic and inorganic peroxides. Our further physiological analysis verified the speculation that An.PrxA plays the oxidative protection function in cells without peroxide selectivity (Figs. 3C and 3D). Moreover, enzymatic assays clarified that Cys 61 is the C_P and Cys 31 is the C_R in An.PrxA (Figs. 3A and 3B), together with the information of the multiple-sequence alignment (Fig. 1), suggesting that the C_R located preceding the C_P type of Prxs is ubiquitously distributed in filamentous fungi, yeasts, and bacteria.

Our data indicated that the peroxidase tolerance to the extremely high concentrations of H₂O₂ is likely completely dependent on the oxidative stabilities of the reaction partner TrxR. In fact, TrxR sensitivity to peroxides has also been confirmed by other evidence. Incubation of mammalian TrxR (TR1) with H₂O₂ led to TR1 inactivation due to the oxidation of the enzyme in its reactive selenocysteine residue at the C-terminal end, which is required for catalytic activity [25]. The inactivation of TR1 will result in ROS accumulation, which will then affect cellular signaling components, including transcription factors, phosphatases, antioxidant enzymes, and kinases. Thus, the H₂O₂ sensitivity of TR1 was proposed to be involved in the redox-regulated cell signaling. Another possible reason for the sensitivity differences of the Prx systems to H₂O₂ is correlated to ROS-promoting protein denaturation or degradation. Many studies have shown that oxidation of proteins may lead to hydroxylation of aliphatic amino-acid side chains and aromatic groups, sulfoxidation of methionine residues, and conversion of some amino acids to carbonyl derivatives [47]. Such modifications directly lead to the loss of the enzymatic activities and were observed in some physiological processes such as aging, disease, and apoptosis. The modification efficiencies of the target proteins are dependent on their amino acid sequences and structural differences. Although the TrxR from A. nidulans shows high homology (68%) to the isoform of S. cerevisiae [16], the subtle distinction in amino acids may cause their different H₂O₂ sensitivity. However, the exact factors responsible for the resistance diversity of the two fungal TrxRs remain to be identified.

TrxRs from different organisms show a great diversity in their chemical mechanism of thioredoxin reduction [48]. However, the TrxR/Trx system-mediated Prx-dependent peroxide reduction processes are similar in most organisms. The Trx/TrxR system from E. coli has been shown to efficiently serve as the reduction system for yeast to support the peroxidase activity of Sc.Ahp1 [39]. In the present study, we also confirmed the functional substitutability between An.Trx/TrxR and S.c.Trx/TrxR in the Prx-dependent peroxide reduction reactions (Fig. 5A). These experiments demonstrated that oxidized 2-Cys Prxs are the nonselective substrates for Trx/TrxR systems from distinct organisms.

In spite of the outstanding resistance of An.PrxA to H₂O₂, the An.PrxA system was not resistant to excess t-BOOH, suggesting there are intrinsic differences in the propensity of the An.PrxA system to H₂O₂ or t-BOOH resistance ability (Fig. 6). To what kind of peroxide the Prx systems can be resistant and to what extent the Prx system is tolerable are dependent on the individual characters of the two constitute elements, Prx and Trx/TrxR. It seemed that the Prx variety determined the substrate preference and the resistance to relatively low peroxide, whereas Trx/TrxR is the key factor to decide the tolerance ability to extreme oxidative stress. It is noteworthy that the t-BOOHinactivated process of the An.PrxA system is different from that of the H₂O₂-inactivated Sc.Ahp1 system. In the presence of excess t-BOOH, the initial rate of NADPH oxidation in the An.PrxA system was independent of the t-BOOH concentrations. However, the rate decreased markedly with time after the initial phase (> 10 sec), and the higher the t-BOOH concentration, the faster the rate of decrease. Comparing the TrxR-determined H₂O₂-inactivated Sc.Ahp1 system, our speculation is that the inactivation manner of the An.PrxA system by t-BOOH could be attributable to the C_P hyperoxidation of An.PrxA, which could only occur in the catalytic cycle, as has been observed in human PrxI [17]. However, the exact mechanism of Prxs resistance to hyperoxidation (i.e., by organic and inorganic peroxides in different manners) needs to be further investigated.

This work was sponsored by the National Natural Science Foundation of China (21672065, 21636003, and 31471659), the Shanghai Pujiang Program (16PJ1402500), and the National Special Fund for State Key Laboratory of Bioreactor Engineering (Grant No. 2060204).