Introduction

During growth and development, reactive oxygen species (ROS) are produced continuously in the course of various metabolic processes, and their levels are greatly influenced by the environmental conditions (Hasanuzzaman et al. 2020). Their main sources are the photosynthetic and mitochondrial electron transport chains. Excess of ROS damages the various macromolecules and the membrane lipids. There are many ROS-processing pathways in plants, which control their amount and consequently the redox environment in the plant cells (Kocsy et al. 2013; Considine and Foyer 2014). A major regulator of the cellular redox state is the H2O2-degrading ascorbate–glutathione (ASA-GSH) cycle for which NADPH ensures the reducing power (Sahoo et al. 2020). ASA and GSH are the most abundant low-molecular-weight non-enzymatic antioxidants in plant cells. The size and redox state of their pools greatly determine the capacity of cells to adjust the control of redox environment to the growth conditions during various developmental stages of plants (Bartoli et al. 2017; Asadi karam et al. 2017). The stress-induced adjustment of the appropriate ASA and GSH levels was observed in Brassica napus treated with Cd (Jung et al. 2020). The amount and ratio of the reduced form of glutathione and its precursors in the total non-protein thiol pools increased in wheat subjected to osmotic stress compared to the control plants (Kocsy et al. 2004). An increase in γ-glutamylcysteine (GSH precursor) content, γ-glutamylcysteine synthetase activity and transcript level was observed after chilling and ozone treatments in maize and Arabidopsis, respectively (Gómez et al. 2004; Sasaki-Sekimoto et al. 2005). Besides the stress response, the cellular redox environment also has an important role in the regulation of physiological processes under optimal growth conditions as shown for the ascorbate-dependent transient oxidation of the root apical meristem in Arabidopsis during cell cycle (de Simone et al. 2017).

The central metabolic pathways, such as carbon fixation and the synthesis of starch, lipids, amino acids and chlorophylls are under redox control but they also affect the redox status (Geigenberger and Fernie 2014). Based on the observations of several studies in different plant species, oxidative stress modified the levels of many metabolites (Savchenko and Tikhonov 2021). The amounts of several sugars (sucrose, fructose, glucose, raffinose, stachyose, verbascose) and most free amino acids increased after oxidative stress. However, the asparagine (Asp), glutamic acid (Glu) and methionine (Met) contents and the levels of certain tricarboxylic acid cycle compounds (malate, succinate, 2-oxoglutarate, isocitrate) became lower. The oxidative stress-induced metabolic changes indicate the rerouting of glycolytic carbon flow into the oxidative pentose phosphate pathway (OPPP) in order to provide NADPH for the antioxidant system (Obata and Fernie 2012). In Arabidopsis, the oxidative stress raised by methyl viologen led to the reprogramming of carbon metabolism as shown by the repression of the anabolic pathways (e.g., Calvin cycle) and the induction of catabolic ones (e.g., OPPP) (Scarpeci and Valle 2008). The importance of the appropriate adjustment of the metabolic pathways during treatment with this herbicide was also demonstrated by the comparison of wild-type Arabidopsis with Rcd1 (radical-induced cell death1) mutant being tolerant to methyl viologen (Sipari et al. 2020). In wild-type plants the level of glycolytic and tricarboxylic acid cycle intermediates declined and the photosynthesis and respiration were inhibited, while in the mutants these processes and the energy salvaging pathways were sustained. In catalase-deficient Arabidopsis mutants with increased H2O2 level, the amounts of many amino and organic acids were greater compared to wild-type plants (Noctor et al. 2015). In rice cell suspension culture, menadione-induced oxidative stress also increased the amount of many amino acids and decreased the level of several sugars including glucose 6-phosphate (glucose 6-P) and fructose 6-phosphate (fructose 6-P) (Ishikawa et al. 2010). The effect of oxidative stress on metabolism of amino acids and carbohydrate was also shown at protein and transcript levels in H2O2-treated rice and wheat, respectively (Wan and Liu 2008; Li et al. 2011a).

The effect of oxidants on various physiological and biochemical processes was extensively studied (Demidchik 2015; Noctor et al. 2015; Savchenko and Tikhonov 2021), whereas there are only a few publications about the influence of reductants in plants (Hancock and Veal 2021). Treatment of Arabidopsis with the reductant dithiothreitol increased organic and amino acid synthesis and decreased sucrose formation (Kolbe et al. 2006). Additionally, foliar application of ASA stimulated the accumulation of free amino acids and soluble carbohydrates in flax cultivars (El-Bassiouny and Sadak 2014). However, the possible negative influence of reductants used in high concentration or for longer duration was not studied in plants. Such effect can be supposed based on results in animal systems where widespread metabolic influences of reductants were observed (Xiao and Loscalzo 2020). Thus, excess of antioxidants resulted in negative physiological effects, such as mitochondrial dysfunction in 3T3-L1 mouse cell lines (Peris et al. 2019). On the basis of our hypothesis, ASA and H2O2 may differently effect metabolism in shoots where the major redox-responsive metabolic processes occur. Accordingly, the aim of the present studies was the comparison of the metabolic effects of ASA and H2O2 in the shoots. The induced redox changes were monitored through the investigation of the components of ASA-GSH cycle including enzyme activities, amounts of metabolites and the related transcripts.

Materials and Methods

Plant Material, ASA and H2O2 Treatments

Seeds of the wheat (Triticum aestivum L.) cultivar Chinese Spring were cultivated between wet filter papers at 25 °C for 1 day. Afterwards, they were moved to 4 °C for 1 day and then moved back to 25 °C for 2 days. Wheat seedlings were transferred to half-strength modified Hoagland solution and were grown for 10 days at 20/17 °C day/night temperature and 75% relative humidity, with 16 h illumination at 250 µmol m−2 s−1 in a growth cabinet (Poleco, Poznan, Poland). Then 0, 5 and 20 mM ASA or H2O2 was added to the nutrient solution. These concentrations of ASA and H2O2 were determined based on growth and gene expression data in preliminary experiments, in which the two compounds were applied in 0, 0.5, 1, 5 10 and 20 mM concentrations Sampling of the whole shoots for biochemical, physiological and molecular measurements was done after 0-, 3- and 7-day treatments in the middle of the photoperiod. The experiments were repeated three times and in each experiment three parallel samples were collected for the various analyses. The control values of the studied biochemical and molecular biological parameters did not change significantly during the experiment therefore they are not shown after 3 days and 7 days.

Determination of Membrane Injury, Pigment Composition and Photosynthetic Parameters

The membrane injury of the wheat leaves was determined by measuring the electrolyte leakage of 1 cm long leaf segments with a conductometer after their shaking in 4 mL deionised water for 2 h. The conductivity was measured again after destroying the cell membranes by incubating the samples at 100 °C for 30 min. The relative electrolyte leakage was then calculated as the ratio of the first and second values (Gulyás et al. 2014).For the measurement of chlorophyll (a and b) and carotenoid contents, leaf tissues of 50 mg were homogenized in 1 mL of 80% acetone, which was followed by centrifugation at 12,000×g for 10 min. These steps were repeated until the pellet was completely white. All of the supernatants were collected and diluted to 12 ml. The absorbance of the solutions was then determined spectrophotometrically (Varian, Middelburg, The Netherlands) at 470, 646, 664 and 750 nm. The chlorophyll a, b and carotenoid contents were calculated using Arnon's equations (Lichtenthaler and Buschmann 2001).

The Fv/Fm chlorophyll fluorescence parameter was determined on fully developed, detached leaves after 20-min dark adaptation with the use of a pulse amplitude modulated fluorometer (PAM) (Imaging-PAM MSeries, Walz, Effeltrich, Germany). The Fv/Fm reflects the maximal quantum efficiency of PSII.

The photosynthetic activity of leaves was determined by the use of a Ciras 3 Portable Photosynthesis System (PP Systems, Amesbury, MA, USA) using a narrow (1.7 cm2) leaf chamber. The measurements were performed on five fully developed attached leaves after 0, 3 and 7 days of treatments. The net photosynthetic rate (Pn), stomatal conductance (gs), transpiration rate (E) and internal CO2 concentration (Ci) were determined at the steady state level of photosynthesis using a CO2 level of 390 μL L−1 and light intensity of 250 μmol m−2 s−1.

Measurement of H2O2 Content and Lipid Peroxidation

The H2O2 concentration of the shoots was analysed by the FOX1 method in a colorimetric reaction using a spectrophotometer as described earlier (Kellős et al. 2008). Shoot samples of 200 mg were homogenized in 1 mL 10% H3PO4. The reaction is based on the oxidation of ferrous ion to ferric ion by H2O2 and the latter one was detected by xylenol orange (Jiang et al. 1990).

The lipid peroxidation was determined by the analysis of malondialdehyde (MDA) levels. Shoot samples of 200 mg were ground in 600 µl of 0.1% (w/v) trichloroacetic acid. After centrifugation at 12,000×g for 10 min, 2 mL of 0.5% (w/v) thiobarbituric acid dissolved in 20% (w/v) trichloroacetic acid was added to 300 µl supernatant. Then, the samples were incubated at 90 °C for 30 min, which step was followed by centrifugation at 12,000 × g for 10 min. The MDA levels were determined spectrophotometrically at 532 nm with the subtraction of the non-specific absorption at 600 nm. The concentration of lipid peroxides, together with that of the oxidatively modified proteins were thus calculated in terms of MDA level using an extinction coefficient of 155 mM−1 cm−1 (De Paula et al. 1996).

Enzyme Activity Analyses

Shoot samples of 200 mg were crushed in a mortar with liquid nitrogen, then 1 mL of 50 mM MES/KOH (pH 6.0) buffer containing 2 mM CaCl2, 40 mM KCl and 1 mM ASA was added to the powder (Murshed et al. 2008). The ascorbate peroxidase (APX) enzyme activity was measured in a reaction mixture which contained 50 mM potassium phosphate buffer (pH 7.0), 5 mM H2O2, 0.25 mM ASA and 50 µl extract in a total volume of 1 mL (Murshed et al. 2008). The activities of the other three enzymes of ASA-GSH cycle were determined using 50 mM Hepes buffer (pH 7) and 50 µl plant extract in a total volume of 1 mL (Murshed et al. 2008). Besides these components, the reaction mixture contained 2.5 mM GSH, 0.1 mM EDTA and 0.2 mM dehydroascorbate for the dehydroascorbate reductase (DHAR); 0.25 mM NADPH, 2.5 mM ASA and 1 U ascorbate oxidase for the monodehyroascorbate reductase (MDHAR) and 0.25 mM NADPH, 0.5 mM EDTA and 0.5 mM GSSG for the glutathione reductase (GR). Catalase (CAT) and glutathione S-transferase (GT) activities were determined as described earlier (Beers and Sizer 1952; Habig et al. 1974). The activities of the enzymes are displayed on a protein basis which was determined according to the method of a previous study (Bradford 1976). A Cary 100 UV–visible spectrophotometer (Varian, Middelburg, The Netherlands) was used for the measurements.

Determination of Thiols

Shoots of 200 mg fresh weight were pulverized with liquid nitrogen, then 1 mL of 0.1 M HCl was added to the powder. The levels of non-protein thiols and their disulphide forms were determined according to an earlier report (Gulyás et al. 2017). The total thiol content was determined after reduction with dithiothreitol and derivatisation with monobromobimane. For the measurement of oxidised thiols, the reduced thiols were blocked by N-ethylmaleimide and the surplus was removed by toluene. Then the amounts of the oxidised thiols were determined similarly to the analysis of total thiols. Cysteine (GSH precursor), hydroxymethylglutathione (hmGSH, a homologue of GSH in Poaceae), and GSH were separated by reverse-phase HPLC (Waters, Milford, MA, USA) and detected by a W474 scanning fluorescence detector (Waters, Milford, MA, USA). The reduced thiol levels were calculated as the difference between the amounts of total and oxidised thiols.

Analysis of Ascorbate and Dehydroascorbate

Shoot samples of 500 mg fresh weight were crushed in a mortar with liquid nitrogen and extracted with 3 ml of 1.5% meta-phosphoric acid (Szalai et al. 2014). After centrifugation, reduced (ASA) and total ascorbic acid (the latter after reduction by dithiothreitol) contents were measured in the supernatants by HPLC using an Alliance 2690 system equipped with a W996 photodiode array detector (Waters, Milford, MA, USA) (Szalai et al. 2014). The concentration of dehydroascorbate (DHA), the oxidized form of ASA was calculated by subtracting the amount of ASA from the total ascorbic acid content.

Metabolite Profiling

Shoot samples of 100 mg were extracted twice with 0.5 mL 60 v/v % MeOH, then twice with 0.5 mL 90 v/v % MeOH after the internal standard (30 µL 1 mg mL−1 ribitol solution) was added as described earlier (Canellas et al. 2019). The whole extraction solution was approximately 2 mL 75 v/v % MeOH. The extraction was carried out with vortex for 30 s, then with ultrasonic bath for 5 min at room temperature, and subsequently the samples were centrifuged at 10,000 g for 5 min at 4 °C. Supernatants were collected, and aliquots (150 µL) were dried under vacuum. The derivatization for GC analyses was carried out with methoxyamine hydrochloride (20 mg mL−1 in pyridine) at 37 °C for 90 min. Following this step, N-Methyl-N-trimethylsilyl-trifluoroacetamide was added and the samples were incubated for 30 min at 37 °C. They were injected into the LECO Pegasus 4D GCxGC TOFMS (LECO, Benton Harbour, MI, USA) equipped with a 30 m column (Rxi-5MS phase) and a 1.5 m column (Rxi-17Sil MS phase). One µL of the sample was injected onto the column in split mode at 230 °C. The carrier gas was He and constant flow rate (1 mL/minute) was used. In addition, the transfer line and the ion source were kept at 250 °C. The thermal program was started at 70 °C and the temperature was on hold for 3 min, then increased to 320 °C at 7 °C/minute rate. The high temperature was on hold for 5 min with 3.25 s modulation period in the 2D GC mode. For the identification of the metabolites, standards and the Kovats retention index were used. Both the GC analyses and the data evaluation and normalization for each metabolite were carried out by the LECO ChromaTOF program. During the evaluation, the ChromaTOF 4.72 was used with Finn and Nist databases. The analysis was calibrated for 55 compounds but only 31 of them could be detected at least after one treatment.

Measurement of Free Amino Acids

Shoot samples of 300 mg fresh weight were extracted with 2 mL cold 10% trichloroacetic acid during 1 h agitation on a shaker at room temperature (Kovács et al. 2012). After filtration through a 0.2 µm pore membrane filter, the samples were examined by an automatic amino acid analyser (Ingos Ltd., Praha, Czech Republic) equipped with an Ionex Ostion LCP5020 cation exchange column. The stepwise separation of free amino acids was done by a Li+-citric buffer system (Ingos Ltd., Praha, Czeh Republic).

Gene Expression Studies

Shoot samples of 50–100 mg fresh weight were pulverized in liquid nitrogen. Afterwards, 600 µl TRIzol reagent was added to the samples for extraction. Total RNA was extracted from the shoot samples by Direct‐zolTM RNA Miniprep Kit (Zymo Research), following the instructions of the supplier. The reverse transcription (RT) was performed using M‐MLV reverse transcriptase and oligo (dT) 15 primer (Promega) as described by the supplier. For the synthesis of cDNA, the final volume of the template was 11 µl containing 1 µg RNA in 1 µl H2O, 9 µl DEPC-treated water, and 1 µl oligo-dT primer (0.5 µg/µl). For the master-mix (25 µl) preparation, we used 5 µl M-MLV RT 5xbuffer, 1.25 µl dNTP mix (100 mM), 0.5 M-MLV RT (200 U/µl), 7.25 µl DEPC-treated water, and 11 µl template. The following conditions were used for reverse transcription: 25 °C for 5 min, 42 °C for 60 min, 70 °C for 15 min and then the samples were kept at 4 °C. The final volume of the cDNA solution was 100 µl after the addition of 75 µl sterilized MQ water. The gene expression levels were determined by qRT-PCR using a CFX96 TouchTM Real‐Time PCR Detection System (Bio‐Rad, Hercules, CA, USA) with specific primers (Table S1) (Paolacci et al. 2009; Sečenji et al. 2010; Gulyás et al. 2014; Gyugos et al. 2019; Toldi et al. 2019; Tian et al. 2021). The qPCR reaction mixture contained 5 µl qPCRBIO SyGreen blue mix, 0.4 µl each of the primers, 3.2 µl MQ water and 1 µl cDNA. The following conditions were set for the qPCR: 95 °C for 3 min, 95 °C for 0.05 min, 60 °C for 0.30 min, GOTO 2 for 39 times, 65 °C for 0.05 min, and 95 °C for 0.5 min. The samples were analysed in triplicates. The relative gene expressions [2−ΔΔCq, where ΔCq = Cq(ref) − Cq(target)] were calculated using the Ta30797 gene as a reference (Livak and Schmittgen 2001; Paolacci et al. 2009).

Statistics

The significant differences were calculated by analysis of variance (ANOVA) using SPSS statistics (16.0) software. The least significance difference (LSD) test was used to compare means at 5% probability level. By the correlation analysis the magnitude of the significant correlations was interpreted according to previously described method (Guilford 1950).

Results

ASA- and H2O2-Dependent Changes in Root and Shoot Growth

The shoot length of wheat seedlings was significantly greater on the 7th day of 5 mM and 20 mM ASA and 5 mM H2O2 treatments compared to its value before the addition of the two compounds (Fig. S1A). However, no significant changes were exhibited in root length (Fig. S1B). The shoot fresh weight increased during most treatments except for the 3-day application of 5 mM and 20 mM ASA and 20 mM H2O2 (Fig. S1C). For root weight, the only significant difference was seen following 7 days of the 5 mM H2O2 application (Fig. S1D).

Influence of ASA- and H2O2 on Photosynthetic Parameters

The chlorophyll a content was decreased by 30–40% following the treatments except for 3-day 20 mM ASA and 5 mM H2O2, while the chlorophyll b content was not affected by the treatments except for the application of 20 mM H2O2 for 3 days (Table 1). The carotenoid content was reduced by ASA, and it was not influenced by H2O2 except for the 7-day 5 mM H2O2 treatment (Table 1).

Table 1 Effects of ascorbate (ASA) and hydrogen peroxide (H2O2) on photosynthetic pigments, chlorophyll a fluorescence and CO2 assimilation parameters
Full size table

Longer application of ASA treatment caused small, but significant decrease in Fv/Fm regardless to ASA concentration (Table 1). H2O2 reduced Fv/Fm only after a longer time at the higher concentration.

The gas exchange parameters were measured before and 3 and 7 days after the treatments. The control data (without ASA or H2O2 treatments) showed similar values along the experiments, thus only the first day data (before the treatments) were presented. Compared to the control, both ASA and H2O2 treatments decreased the CO2 assimilation rate (Pn) after 3 and 7 days of treatments (Table 1). The decrease showed both concentration and time dependence: e.g.. higher decrease was observed when the concentration of ASA or H2O2 increased, and also when longer period was applied. Significant difference between the ASA and H2O2 treatments was found only when the chemicals were applied at 5 mM concentration and longer (7 days) period. In this case, the ASA treatment caused higher decrease in Pn than the H2O2 treatment. Stomatal conductance (gs) and transpiration rate (E) were also decreased by both treatments as compared to control values, however significant differences between either the concentrations or the duration of the treatment were rarely found (Table 1). Intercellular CO2 level decreased both after ASA and H2O2 treatments, which was more intensive at 3 days of treatments than at 7 days.

ASA- and H2O2-Induced Changes in H2O2 Content, Lipid Peroxidation and Electrolyte Leakage

The amount of endogenous H2O2 was decreased by all treatments, especially after 7 days, except for the 20 mM ASA (Fig. 1A). Following this treatment, the H2O2 content was greater after 3 days compared to the control, while after 7 days it was similar to it.

Fig. 1
figure 1

Effects of ascorbate (ASA) and hydrogen peroxide (H2O2) on the endogenous hydrogen peroxide (H2O2) level (A), lipid peroxidation (B) and electrolyte leakage (C). Values marked with different letters are significantly different from each other at p ≤ 0.05 level (ANOVA followed by least significant difference test, three independent experiments with 3 parallels each). MDA: malondialdehyde

Full size image

The lipid peroxidation, as indicated by the MDA levels, was increased by all treatments (Fig. 1B). These changes were especially large after 7-day 5 mM and 20 mM ASA addition and 3-day 20 mM H2O2 treatment.

While the applied treatments did not affect the electrolyte leakage after 3 days, their application for 7 days greatly increased it (Fig. 1C). This increase ranged from 1.8-fold to 2.9-fold.

Effects of ASA and H2O2 Treatments on ASA and Non-protein Thiols

The ASA and DHA contents were greatly increased by 7-day 20 mM ASA treatment (Fig. 2A, B). A great decrease in the DHA level was observed after the 7-day 5 mM ASA and 20 mM H2O2 treatments (Fig. 2B). The lowest DHA/ASA ratios were detected after 3-day 20 mM ASA and 7-day 20 mM H2O2 treatments (Fig. 2 C). The EDHA/ASA value was increased by 7-day 5 mM and 20 mM ASA treatments, while it was reduced by the other treatments except for 7-day 5 mM H2O2 application (Fig. S2A).

Fig. 2
figure 2

Effects of ascorbate (ASA) and hydrogen peroxide (H2O2) on the endogenous levels of ascorbate. A ascorbic acid contents (ASA), B dehydroascorbic acid contents (DHA), C DHA/ASA: ratio of the two forms. Values marked with different letters are significantly different from each other at p ≤ 0.05 level (ANOVA followed by least significant difference test, three independent experiments with 3 parallels each)

Full size image

The amount of the Cys (GSH precursor) was significantly reduced only by 20 mM ASA after 7 days compared to the control plants (Fig. 3A). Applying ASA in 5 mM concentration, the cystine (CySS) content, CySS/Cys ratio and the ECySS/Cys value were greater after 3 days and lower after 7 days compared to the controls, while 5 mM H2O2 did not affect these parameters (Figs. 3B, C, S2B). Adding the two compounds in 20 mM concentration, ASA increased and H2O2 decreased these parameters during the whole experiment.

Fig. 3
figure 3

Effects of ascorbate (ASA) and hydrogen peroxide (H2O2) on the amount and redox state of cysteine and glutathione. A Cys: cysteine B CySS: cystine, C ratio of the two forms, D GSH: glutathione, E GSSG: glutathione disulphide, F GSSG/GSH: ratio of the two forms. Values marked with different letters are significantly different from each other at p ≤ 0.05 levels (ANOVA followed by least significant difference test, three independent experiments with 3 parallels each)

Full size image

The GSH contents were significantly reduced by the treatments except for the 7-day treatment with H2O2 (Fig. 3D). The trends of changes in GSSG content, GSSG/GSH ratio and EGSSG/GSH value were similar to those of the corresponding parameters of Cys/CySS redox couple (Figs. 3E, F, S2C). The large difference between the effects of the 20 mM ASA and H2O2 should be emphasized here, too.

A significant reduction of the hmGSH content was only observed after 7-day 20 mM ASA treatment (Fig. S3A). However, it was increased by 20 mM H2O2 both after 3 and 7 days. The hmGSSG content was greater following 5 mM and 20 mM ASA treatments for 3 and 7 days, respectively (Fig S3B). Interestingly, longer treatment (7 days) with 5 mM ASA or shorter application (3 days) of 20 mM ASA substantially decreased its level. The hmGSSG content was also significantly lower after 20 mM H2O2 treatment. The hmGSSG/hmGSH ratio was much greater after 3-day 5 mM ASA treatment, while 20 mM ASA applied for 7 days) and treatments with H2O2 decreased it except for 7-day 20 mM H2O2 application (Fig. S3C). The EhmGSSG/hmGSH value was increased by 3-day 5 mM ASA and by 7-day 20 mM ASA, and it was decreased by the other treatments except for 5 mM H2O2 (3d, 7 days) compared to the control plants (Fig. S2D).

Modification of the Antioxidant Enzyme Activities by ASA and H2O2

The activity of the examined antioxidant enzymes was increased by the most treatments except for MDHAR and GR (Fig. 4). In the case of APX, this increase was significantly greater after the addition of 20 mM ASA compared to the 20 mM H2O2 treatment, while after the addition of 5 mM H2O2 only a transient change was observed after 3 days (Fig. 4A). In contrast to the other enzymes, the MDHAR activity was only greater after a 3-day 5 mM ASA treatment compared to the control (Fig. 4B). The DHAR activity increased in turn after all treatments except for the 7-day treatment with 20 mM ASA (Fig. 4C). Similarly to APX, the activity of GR was greater after 7 days following the addition of ASA than that of H2O2 (Fig. 4D). This difference derives from the twofold activity increase by ASA and the unchanged GR activity after H2O2 treatment. The activities of CAT and GT were increased by both chemicals except for CAT after 7-day 5 mM H2O2 treatment and for GT following 3-day 20 mM H2O2 application (Fig. 4E, F). The greatest change in CAT and GT activity was observed after 7-day 20 mM and 5 mM H2O2 addition, respectively.

Fig. 4
figure 4

Effects of ascorbate (ASA) and hydrogen peroxide (H2O2) on the activity of the antioxidant enzymes. A APX: ascorbate peroxidase, B MDHAR: monodehydroascorbate reductase, C DHAR: dehydroascorbate reductase, D GR: glutathione reductase, E CAT: catalase, F GT: glutathione transferase. Values marked with different letters are significantly different from each other at p ≤ 0.05 level (ANOVA followed by least significant difference test, three independent experiments with 3 parallels each)

Full size image

ASA- and H2O2-Dependent Changes in the Metabolic Profile of Wheat Seedlings

In total, 31 metabolites were identified in the samples and the amount most of them was influenced by ASA and H2O2 treatments (Fig. 5; Table S2). Based on hierarchical clustering, the metabolite profiles after all treatments were separated from that of the control plants. The samples from ASA- and H2O2-treated plants formed separate clusters, except for the 7-day 5 mM ASA and H2O2 treatments. The greatest difference was observed between the 7-day 20 mM ASA and H2O2 treatments, which were positioned on the two margins of the heatmap.

Fig. 5
figure 5

Effects of ascorbate (ASA) and hydrogen peroxide (H2O2) on the metabolite levels. The C, A5-3, A5-7, A20-3, A20-7, H5-3, H5-7, H20-3 and H20-7 treatments represent the control, 5 mM ASA for 3 days, 5 mM ASA for 7 days, 20 mM ASA for 3 days, 20 mM ASA for 7 days, 5 mM H2O2 for 3 days, 5 mM H2O2 for 7 days, 20 mM H2O2 for 3 days and 20 mM H2O2 for 7 days, respectively. The ASA and H2O2 refer to ascorbate and hydrogen peroxide, respectively. The data and their statistical analysis are shown in Table S2 (Color figure online)

Full size image

Based on the similarities of the redox compounds-induced changes in the amounts of metabolites, they were grouped into three major clusters. The members of the first cluster (mainly organic acids and sugars) showed no or slight alterations in their levels after the various treatments compared to the control samples. The glucose and glucopyranose were present in high amounts after all treatments. In the second cluster (mainly organic acids, sugars and amino acids), the treatments resulted in large differences compared to each other and the control samples. Both compounds induced a decrease in the Glu, quinic acid and phosphoric acid contents after several treatment duration or concentration. In the third cluster (mainly amino acids and sugars) there were no differences in the sucrose and galactose contents between the samples. Interestingly, several compounds were not present in detectable amounts after certain treatments. Mannose, sucrose and galactose were present in higher amounts in all samples.

Effect of ASA and H2O2 on Free Amino Acid Levels

The total protein content exhibited similar decrease after all treatments (Fig. S4A), however, the total free amino acid level was differently affected by the various treatments (Fig. S4B). It was greatly increased by 20 mM H2O2 and significantly reduced by 20 mM ASA after 7 days, which led to a great difference between these two treatments.

Analysing the effects of the various treatments on the amounts of the individual amino acids by hierarchical clustering, the control and 5 mM H2O2-treated (3 and 7 days) plants were grouped together, but those ones treated with 20 mM H2O2 or both concentrations of ASA formed separate clusters, indicating the clear effect of these treatments (Fig. 6; Table S3). Within the latter cluster, amino acid levels after 7-day 20 mM ASA addition were separated from those of all other treatments. The distance was especially large from the cluster of 20 mM H2O2 treatment. Similarly to the total free amino acid content, the amounts of the individual amino acids were also smaller in general after 20 mM ASA treatment then after the addition of 20 mM H2O2 for 7 days.

Fig. 6
figure 6

Effects of ascorbate (ASA) and hydrogen peroxide (H2O2) on the amino acid levels. The C, A5-3, A5-7, A20-3, A20-7, H5-3, H5-7, H20-3 and H20-7 treatments represent the control, 5 mM ASA for 3 days, 5 mM ASA for 7 days, 20 mM ASA for 3 days, 20 mM ASA for 7 days, 5 mM H2O2 for 3 days, 5 mM H2O2 for 7 days, 20 mM H2O2 for 3 days and 20 mM H2O2 for 7 days, respectively. The ASA and H2O2 refer to ascorbate and hydrogen peroxide, respectively. The data and their statistical analysis are shown in Table S3 (Color figure online)

Full size image

According to the clustering of the redox treatment-induced changes, the free amino acids formed two large groups. In the first cluster (16 amino acids), the levels of amino acids were similar after most of the treatments except for the 7-day 5 and 20 mM ASA treatments. In the second cluster (containing Asn, 1mHis, Pro, Cysta, His, 2mHis, Val, Tyr, Phe, and Arg), large differences were found in the amino acid contents of the individual groups; the most considerable contrast was observed between the 3-day 5 mM H2O2 and 7-day 20 mM ASA treatments. Interestingly, members of the same amino acid family were grouped in different clusters based on the effect of the various treatments on their levels as shown for glutamate and aspartate family.

Transcriptional Regulation of Glutathione and Amino Acid Metabolism by ASA and H2O2

The separate cluster of the transcript levels in the control plants indicates a pronounced effect of all treatments on this parameter (Fig. 7; Table S4). This influence differed after ASA and H2O2 treatments as shown by the different clusters, except for 7-day 5 mM ASA and 20 mM H2O2. The greatest effect on the gene expression was induced by the addition of 5 mM H2O2 for 3 days and by the treatment with 20 mM ASA for 7 days as shown by the large decrease and increase in the transcript levels, respectively.

Fig. 7
figure 7

Effects of ascorbate (ASA) and hydrogen peroxide (H2O2) on the gene expression. The C, A5-3, A5-7, A20-3, A20-7, H5-3, H5-7, H20-3 and H20-7 treatments represent the control, 5 mM ASA for 3 days, 5 mM ASA for 7 days, 20 mM ASA for 3 days, 20 mM ASA for 7 days, 5 mM H2O2 for 3 days, 5 mM H2O2 for 7 days, 20 mM H2O2 for 3 days and 20 mM H2O2 for 7 days, respectively. The ASA and H2O2 refer to ascorbate and hydrogen peroxide, respectively. The data and their statistical analysis are shown in Table S4 (Color figure online)

Full size image

Based on the similarities in the expression changes of the individual genes, there were two main clusters (Fig. 7). In the first one, the genes were repressed or they were not affected by most of the treatments except for OrnATF. This cluster contained nearly all genes associated with the amino acid metabolism and those ones encoding orgenellar enzymes of antioxidant system or OPPP. In the second cluster the gene expressions were increased in general. To this cluster belonged several genes encoding the cytoplasmic enzymes of antioxidant system.

Discussion

Alterations in Photosynthesis and Membrane Stability After ASA and H2O2 Treatments

ASA and H2O2 decreased the CO2 assimilation rate (Pn) of plants in both concentration and time dependent manner. The decrease of Pn is mainly associated with the stomatal closure at the 3 days of treatments as indicated by the decrease of gs. Furthermore, the stomatal closure also reduced the transpiration rate in plants. The increase of Ci level at 7 days of treatments (as compared to 3 days of treatments) suggested that some non-stomatal limitation can also contribute to the decrease of CO2 assimilation rate. The chlorophyll a content also decreased at 7 days of ASA and H2O2 treatments, which manifested in a slight, but significant changes in Fv/Fm fluorescence parameters. Altogether the, photosynthetic parameters, together with pigment contents reveal a mild stress state of the leaves, which is about to get worse if the treatments (especially ASA and 20 mM H2O2) would have been maintained for a longer period. The disturbed photosynthesis can also contribute to/ be responsible for the increased lipid peroxidation resulting in greater electrolyte leakage (decreased membrane stability) after one-week ASA and H2O2 treatments. In contrast to lipid peroxidation, the H2O2 levels decreased after the most treatment which change may result from the efficient activation of the enzymes of ASA-GSH cycle for its degradation. As observed in wheat, ASA treatment (8 mM for 5 or 10 days) also increased the lipid peroxidation in Arabidopsis (Qian et al. 2014). Its similar effect on lipid peroxidation was found after its use in smaller concentration during a shorter period (0.5 mM, 2 days) in tomato leaves (Elkelish et al. 2020). In contrast to wheat, ASA increased chlorophyll content and decreased electrolyte leakage in tomato. These latter results indicate the different effects of ASA on photosynthetic pigments and membrane stability in the various plant species which also depend on the applied concentration and duration of the treatments. Similarly to ASA, the influence of H2O2 on certain photosynthesis-related parameters also differed in wheat and other plant species as shown by the unchanged chlorophyll level and net photosynthetic rate in H2O2-treated cucumber (Sun et al. 2016). In addition, chlorophyll content and chlorophyll fluorescence did not change in leaves of Ficus deltoidae after 3-week treatment with 8 mM H2O2 (Nik Muhammad Nasir et al. 2021).

Different Modification of the Redox Environment by ASA and H2O2

The present experiments demonstrated that both the excess of antioxidants (reductants) and oxidants could disturb the photosynthesis and redox homeostasis, which is important for the maintenance of the metabolism in plants. Similar to wheat, treating Arabidopsis with reductants (ASA, GSH) and an oxidant (H2O2) increased and decreased the GSSG content and the GSSG/GSH ratio, respectively (Gulyás et al. 2017). Interestingly, after one-week addition of ASA and H2O2 in the higher concentration, an overcompensation of the induced redox changes in the opposite direction was observed. Namely, after ASA treatment the cysteine (GSH precursor), glutathione and ascorbate pools became more oxidised, and following H2O2 application they were more reduced compared to control plants as shown by the amounts of the oxidised forms, their ratios to the reduced ones and the half-cell reduction potential values. This phenomenon is also indicated by the greater endogenous H2O2 content in ASA-treated plants and by the lower one following H2O2 application (20 mM, 7 days). In animal system, the formation of ROS was also observed during treatment with reductant because of the lack of sufficient electron acceptor in the mitochondrial electron transport chain (Xiao and Loscalzo 2020). As a consequence of the more oxidising redox environment after 20 mM ASA treatment for a week in wheat, a greater activity of the H2O2-removing enzymes, APX and GR was observed compared to the 7-day addition of 20 mM H2O2. A transcriptional regulation of the GSH metabolism-related enzymes was indicated by the similar differences at gene expression level. The observed different redox changes after the 20 mM ASA and H2O2 treatments were accompanied by the great reduction (shoot) or stop (root) of growth in the case of both chemicals shown by the fresh weight data.

The present observations indicate the coordinated adjustment of the components of ASA-GSH cycle, which was also shown in poplar through the increase in ASA content after the overexpression of GR (Foyer et al. 1995). Consistent with our results, ASA treatment also activated the antioxidant defence system including superoxide dismutase (SOD), peroxidase (POD) and CAT in wheat and other plant species in earlier studies (Athar et al. 2009; Xu et al. 2015; Wang et al. 2018). The effect of ASA on antioxidants was also proved in an ascorbate-deficient vtc-1 Arabidopsis mutant, in which the activities of MDAR and DHAR enzymes decreased (Huang et al. 2005). As observed in the present experiments, exogenous application of H2O2 reduced the endogenous ROS levels (MDA and superoxide radical) in wheat plants (Li et al. 2011b; Ashfaque 2014). Treatment with H2O2 also modified the GSH levels in mung bean (Yu et al. 2003). These results corroborate that ASA and H2O2 affect several compounds of the redox system, however, their influence is different from what was observed in the present study.

Redox-Dependent Differences in the Adjustment of Metabolite Levels by ASA and H2O2

The overcompensation of the redox changes during treatment with ASA and H2O2, respectively, can derive from the effect of ASA and H2O2 on the NAD(P)H producing/consuming metabolic pathways, such as the glycolysis, tricarboxylate cycle and saccharopine pathway (Arruda and Neshich 2012; Geigenberger and Fernie 2014). After the addition of ASA, the glucose-6-P level did not change, which was a sign of its use for the glycolysis and/or the oxidative pentose phosphate pathway like under control conditions (Fig. 8). In addition, the saccharopine pathway was repressed by ASA as indicated by the decreased Lys and alpha-aminoadipate levels. Thus, the NAD(P)H production in these pathways probably remained constant and became lower, respectively. Its increase was not necessary, since exogenous ASA might ensure enough reducing power and less NADPH was also sufficient for the reduction of its oxidised form. Consequently, the amount of the available NADP+ was not enough to accept all electrons from photosynthesis, and ROS were formed in greater amount as discussed above. After the addition of H2O2, the decrease in glucose-6-P content was an indicator of the increased glycolysis and/or OPPP [NAD(P)H production], which was observed in several plant species during oxidative stress (Scarpeci and Valle 2008; Savchenko and Tikhonov 2021). This change resulted in increased NAD(P)H levels, which could ensure the sufficient speed of the H2O2 degradation in the ASA-GSH cycle, during which enough NADP+ was produced for accepting the electrons from photosynthesis, leading to reduced ROS production.

Fig. 8
figure 8

An overview of metabolic changes involved in primary pathways of wheat seedlings exposed to a reductant (ASA) and oxidant (H2O2). The red and blue arrows and horizontal lines refer to 7 days 20 mM ASA and H2O2 treatments, respectively. Their relative changes compared to the control are indicated as follows: downregulation – small and large downwards arrows for 0.7- to 0.35-fold and < 0.35-fold changes, respectively; no effect – horizontal lines for 0.7- to 1.4-fold changes; upregulation – small and large upwards arrows for 1.4- to 2.8-fold and > 2.8-fold changes, respectively. The maximum ratio was around 7, but for proline the increase was much greater as indicated besides the arrows. The continuous lines show the direct metabolic connection between two metabolites, while the dashed lines show several intermediate products between them. Besides the three-letter abbreviations of the proteinogenic amino acids, the following ones were used: AAA alfa-aminoadipic acid. Cit citrulline, GABA gamma-aminobutyric acid, γGluCys gamma-glutamylcysteine, γGluCysGly gamma-glutamylcysteinylglycine, GSH glutathione, Orn ornithine, OPPP oxidative pentose phosphate pathway, TCA tricarboxylic acid (Color figure online)

Full size image

The observed difference in the effect of ASA and H2O2 on the main metabolic pathways influenced not only the redox environment but also the levels of many metabolites connected to these pathways. Thus, the amount of most amino acids (exceptions: Phe, Tyr, Cys, Val, Asn, Pro, Arg, citrulline and oxoproline), the formation of which is linked to the glycolysis and tricarboxylate cycle, decreased after ASA treatment and increased or remained constant after H2O2 treatment (Fig. 8). This phenomenon can be explained by the unchanged and increased use of glucose-6-P for glycolysis after ASA and H2O2 addition, respectively. Similar to wheat, the amount of certain metabolites of glycolysis and tricarboxylic acid cycle also decreased in Arabidopsis after oxidative stress induced by methyl viologen (Sipari et al. 2020). Interestingly, Pro exhibited a much greater increase, 123-fold and 27-fold compared to other amino acids (maximum sevenfold increase), after the addition of ASA and H2O2, respectively (Fig. 8), which indicates its special importance in the response to reductants and oxidants. Similar to wheat, ASA treatment also increased its endogenous level in flax cultivars (El-Bassiouny and Sadak 2014). In fact, its synthesis is an NADPH-consuming process, while during its degradation NADPH will be produced; therefore, it contributes to the adjustment of the cellular redox environment (Szabados and Savouré 2010). Glu as a precursor of both Pro and GSH plays an important role in the interconnection of GSH- and Pro-dependent redox adjustment. The greater increase in Pro content after ASA treatment was accompanied by decreased Glu and GSH content, while its smaller increase following H2O2 addition was associated with unchanged Glu and GSH contents (Fig. 8).

The redox sensitivity of carbohydrate metabolism was indicated by the different effects of ASA and H2O2 on the amounts of certain carbohydrates (sucrose, fructose, glucose-6-P and mannose; Fig. 8). Thus, the amount of mannose, being a precursor of ASA, was decreased by exogenous ASA and increased by H2O2. Although organic acids of the citrate cycle (malate and citrate valves) play an important role in the integration of redox metabolism (Igamberdiev and Bykova 2018; Sipari et al. 2020), their levels remained constant after both treatments except for the similar reduction in alpha-ketoglutarate content. This decrease could derive from its greater use for the amino acid synthesis of the glutamate family (Sipari et al. 2020). Following the ASA treatment, the simultaneously great decrease in the alpha-ketoglutarate and Glu contents can be explained by the very large Pro accumulation (123-fold increase). In the case of H2O2 addition, besides the smaller increase in Pro content (27-fold), the amounts of Gln and Arg also increased.

The different effect of ASA and H2O2 treatments on the ASA-GSH cycle-associated redox parameters and their relationships with the amount of the metabolites was also corroborated in a correlation analysis (Table S5). Following H2O2 addition, the coordinated adjustment of the studied parameters of the ASA-GSH cycle was shown by their high or very high positive correlations except for the ASA content. However, such adjustment was not observed after ASA treatment, since among the 36 comparisons only 10 were high or very high. Regarding the metabolites shown on Fig. 8, many of them were differently influenced by ASA and H2O2 as indicated by the directions and level of their correlations with the redox parameters (Table S5). After addition of ASA, the amount of fructose, aconitic acid, α-ketoglutaric acid, shikimate, quinic acid, Ser, Gly, Ala, Leu, Ile, Thr, Met, Lys, Glu, GABA, and following H2O2 treatment the sucrose, glucose, glucopyranose, tagatofuranose, ribose, malate, His, Phe, Tyr, Asn and Pro content had (very) high negative correlation with the level of ASA-GSH cycle components, respectively.

ASA and H2O2 Differentially Affect the Transcripts Associated with the Studied Metabolites

Similar to the metabolite profile, the transcription of the studied genes related to the antioxidant system, amino acid metabolism and OPPP was also differently affected by the shift of the redox environment of the leaf tissue to more oxidizing and more reducing directions after ASA and H2O2 treatments, respectively. Interestingly, 5 mM H2O2 greatly reduced the expression of most genes after 3 days, while 20 mM ASA increased their expressions after 7 days compared to the control plants, which indicate that the redox control of metabolism also exists at transcriptional level. In the case of ASA treatment, the activation of the genes encoding the enzymes related to OPPP and the antioxidant system may contribute to the recovery of the redox environment in the leaf tissue. Regarding the OPPP, several glucose-6-P dehydrogenase genes were released from their known reductive feedback inhibition (Foyer and Hodges, 2011). An opposite effect was expected after 7-day 20 mM H2O2 addition for the readjustment of the redox environment by the repression of genes removing ROS. Interestingly, such effect was only found after 3-day 5 mM H2O2 treatment. As observed in wheat, many genes related to the redox control of metabolic processes were also induced by H2O2 in Arabidopsis (Desikan et al. 2001) and by methyl viologen in pepper (Lee et al. 2010). A coordinated control of the expression of the studied genes associated with amino acid metabolism, antioxidant system and OPPP was indicated by the hierarchical clustering of the transcript level, since the organellar redox system and OPPP-related genes were not affected or were repressed by most treatments, while the cytoplasmic genes were in general induced by ASA and H2O2.

Conclusions

The applied reductant (7 days 20 mM ASA) and oxidant (7 days 20 mM H2O2) shifted the redox environment of the leaf tissues to the opposite directions, since in the former case it became more oxidized and in the latter one more reduced based on the redox state of the ascorbate and glutathione pools, respectively. These redox changes resulted in different modifications of the primary metabolic pathways as indicated by the levels of organic acids, carbohydrates and free amino acids. Simultaneously, such alterations occurred at gene expression levels, which may contribute to the recovery of the redox conditions existing in the control conditions.