Differential molecular response of monodehydroascorbate reductase and glutathione reductase by nitration and S-nitrosylation

Highlight Nitration and S-nitrosylation, two post-translational modifications (PTMs) mediated by nitric oxide, differentially regulate MDAR and GR activities which are key components of the ascorbate–glutathione cycle.


Introduction
The ascorbate-glutathione cycle is composed of monodehydroascorbate reductase (MDAR), glutathione reductase (GR), ascorbate peroxidase (APX), and dehydroascorbate reductase (DHAR), as well as the antioxidant metabolites ascorbate and glutathione and the reductive coenzyme NADPH.This metabolic pathway is essential for the detoxification and regulation of the cellular level of hydrogen peroxide (H 2 O 2 ) in plant cells (Hossain and Asada, 1985;Asada, 1992;Noctor and Foyer, 1998;Shigeoka et al., 2002).Thus, H 2 O 2 is reduced to water by APX using ascorbate as the electron donor.The oxidized ascorbate (monodehydroascorbate or dehydroascorbate) is then regenerated by MDAR and DHAR using reduced glutathione (GSH) with the concomitant generation of oxidized glutathione (GSSG).Finally, GSSG is reduced by GR with the aid of NADPH as the electron donor.At the subcellular level, these enzymes have been demonstrated to be located in different cellular compartments including the cytosol, chloroplasts, peroxisomes, and mitochondria (Foyer and Halliwell, 1976;Groden and Beck, 1979;Jiménez et al., 1998;Asada, 2006;Palma et al., 2006;Reumann and Corpas, 2010).
Nitric oxide (NO) belongs to a family of related molecules known as reactive nitrogen species (RNS).Although the specific source of NO in plants is still under debate (see Corpas et al., 2009), there is no doubt that plants have an endogenous NO source(s).S-Nitrosoglutathione (GSNO) is formed by the S-nitrosylation reaction of NO with GSH, and it has a significant physiological relevance because GSNO is considered to function as a mobile reservoir of NO bioactivity.On the other hand, the reaction of NO with superoxide radicals (O 2 • -) generates a powerful oxidant, designated peroxynitrite (ONOO -).Furthermore, these NO-derived molecules can mediate several post-translational modifications (PTMs) such as nitration and S-nitrosylation.Protein tyrosine nitration involves the addition of a nitro group (-NO 2 ) to one of the two equivalent ortho-carbons of the tyrosine residue aromatic ring.Protein nitration is affected by various features such as the protein quaternary structure, the environment in which the protein is located, and the nitration mechanism.Consequently, these covalent changes may result in effects such as loss or gain of protein function or no change in function (Souza et al., 2008;Radi, 2013).On the other hand, S-nitrosylation involves the binding of an NO group to a protein cysteine residue and is also able to change the function of many proteins (Astier et al., 2012).Proteomic studies have identified potential plant protein candidates for nitration and S-nitrosylation which belong to diverse functional categories such as redox-related, stress-related, metabolic, and signalling/regulating proteins (Lindermayr et al., 2006;Corpas et al., 2015).These analyses have identified some enzymatic components of the ascorbate-glutathione cycle as potential PTM targets mediated by NO-derived molecules.However, little is known about the specific impact of these NO-related PTMs on the activity and structure of particular proteins involved in antioxidative systems (Holtgrefe et al., 2008;Chaki et al., 2011;Astier et al., 2012;Begara-Morales et al., 2013a, 2014).In previous studies, in order to understand the fine-tuned regulation of this key antioxidant system by NO, the effect of NO-related PTMs on the activity of cytosolic APX was analysed.Thus, APX was identified as a target of S-nitrosylation in Arabidopsis plants (Fares et al., 2011;Keyster et al., 2011), and pea cytosolic APX has very recently been demonstrated to have a dual mechanism of regulation mediated by NO-derived molecules; while the nitration of Tyr235 provoked a deactivation of APX activity, S-nitrosylation at Cys32 activated this activity (Begara-Morales et al., 2014).On the other hand, a nitroproteomic study of sunflower hypocotyls identified GR as a target for tyrosine nitration (Chaki et al., 2009) and for S-nitrosylation in rice (Lin et al., 2012).Additionally, MDAR has also been identified as a potential candidate for both S-nitrosylation and nitration in citrus plants (Tanou et al., 2012) and for S-nitrosylation in rice (Lin et al., 2012) and Arabidopsis (Hu et al., 2015).However, to the authors' knowledge, no information is available on the effect of any PTMs mediated by NO-derived molecules on the molecular function of MDAR or GR.
Thus, the aim of the present study is to gain a more indepth understanding of the regulation of the antioxidative ascorbate-glutathione cycle by NO-PTMs.Pea (Pisum sativum) was selected as the model plant for several reasons: (i) previous studies have allowed significant information on NO metabolism to be obtained (Barroso et al., 2006;Corpas et al., 2008Corpas et al., , 2013b)); (ii) the isolation and characterization of a full-length genomic clone encoding the pea MDAR has been reported; and (iii) a previous study on pea APX showed dual regulation by S-nitrosylation and nitration (Begara-Morales et al., 2014).On the other hand, both chloroplastic and cytosolic isoforms of GR (Edwards et al., 1990;Stevens et al., 1997;Gill et al., 2013;Wu et al., 2013) were used to obtain the most complete information about this enzyme.Consequently, using in vitro approaches, the potential effect of NO-derived molecules such as ONOO -and GSNO, which trigger nitration and S-nitrosylation, respectively, on the molecular function of two enzymes (peroxisomal MDAR and cytosolic and chloroplastic GR) in this cycle was analysed.
As a result, peroxisomal MDAR was deactivated by both nitration and S-nitrosylation, which compromise the cycle's antioxidant capacity.However, cytosolic and chloroplastic GR were unaffected by any of these NO-PTMs in an attempt to maintain the levels of GSH and the cellular redox state.

Plant material and growth conditions
Pea (Pisum sativum L., cv.Lincoln) seeds were obtained from Royal Sluis (Enkhuizen, The Netherlands).Seeds were surface sterilized with 3% (v/v) commercial bleach solution for 3 min, and then were washed with distilled water, and germinated in vermiculite for 3-4 d under the following growth chamber conditions: 24 ºC/18 °C (day/night), 80% relative humidity, a 16 h photoperiod, and a light intensity of 190 μE m -2 s -1 .Healthy and vigorous seedlings were selected and grown in nutrient solutions (Corpas et al., 1993).After 14 d, plants were transplanted to similar media supplemented with 150 mM NaCl and were grown for 4 d (Begara-Morales et al., 2014).
Recombinant proteins carrying an N-terminal choline-binding domain were produced using Escherichia coli strain BIVU0811, which were routinely cultured overnight at 37 ºC in LB with kanamycin (25 mg l -1 ) and ampicillin (100 mg l -1 ).Gene expression was induced by the addition of 1 mM salicylate and 10 mM 3-methyl benzoate in a 250 ml culture grown at 20 ºC overnight in order to produce a higher proportion of soluble protein.Cells were harvested by centrifugation and re-suspended in 20 ml of phosphate-buffered saline (PBS; pH 7.0) containing 25 U ml -1 DNAse I, 10 mM MgCl 2 , and commercial protease inhibitor (Complete, Roche).Cells were lysed with a Niro Soavi NS1001L Panda High-Pressure homogenizer at a pressure of 800-900 bar.The cell lysate was then centrifuged at 10 000 g at 4 °C for 15 min, and the supernatant was used for the purification of recombinant proteins with the aid of a 1 ml LYTRAP column (Biomedal).The column was washed with 20 ml of 20 mM K phosphate buffer (pH 7.0) containing 300 mM NaCl and 5 mM choline.The protein was eluted in 1 ml fractions using a discontinuous gradient of choline prepared in the same buffer with 100 mM NaCl and 20 mM choline (fraction E1), 50 mM choline (E2), 75 mM choline (E3), 100 mM choline (E4), 150 mM choline (E5), 200 mM choline (E6), 250 mM choline (E7), and 500 mM choline (E8).The samples were analysed by 10% SDS-PAGE and stained with Coomassie blue dye.Supplementaryy Fig. S1 available at JXB online shows the SDS-PAGE analysis of the purification of recombinant peroxisomal MDAR (Supplementary Fig. S1A) and chloroplastic and cytosolic GR (Supplementary Fig. S1B, C).
MDAR and GR activity assays: treatment with a peroxynitrite donor (SIN-1) and a nitric oxide donor (GSNO) MDAR (EC 1.6.5.4) activity was determined spectrophotometrically by measuring the reduction of absorbance at 340 nm according to the technique described by Hossain et al. (1984) with some modifications.The 1.0 ml assay mixtures contained 50 mM TRIS-HCl (pH 7.8), 0.2 mM NADH, 1 mM ascorbate, and sample.The reaction was initiated by adding 0.2 U of ascorbate oxidase (EC 1.10.3.3 from Cucurbita; Sigma-Aldrich, St. Louis, MO.USA), and the decrease in A 340 due to NADH oxidation was monitored.One milliunit of activity was defined as the amount of enzyme required to oxidize 1 nmol NADH min -1 at 25 ºC.GR (EC 1.6.4.2) activity was assayed by monitoring NADPH oxidation coupled with the reduction in GSH (Edwards et al., 1990).The reaction rate was corrected for the slight non-enzymatic oxidation of NADPH by glutathione disulphide (GSSG).
The molecule SIN-1 (3-morpholinosydnonimine) has been shown to generate peroxynitrite (ONOO -), a protein-nitrating compound (Daiber et al., 2004).Recombinant proteins were therefore incubated at 37 °C for 1 h with increasing concentrations (0-5 mM) of SIN-1 (Calbiochem) freshly made up before use (Begara-Morales et al., 2014).For treatments with GSNO (NO donor), recombinant proteins were incubated at room temperature for 30 min with 0.5 mM and 2 mM GSNO (Begara-Morales et al., 2014).As a control, proteins were also incubated with 0.5 mM and 2 mM GSH.The protein concentration was determined with the aid of the Bio-Rad protein assay using bovine serum albumin (BSA) as standard.
Identification of nitrated tyrosine residues in recombinant pea MDAR using mass spectrometric techniques Purified recombinant pea MDAR was processed according to a protocol involving reduction with dithiothreitol (DTT), derivatization with iodoacetamide (IAA), and enzymatic digestion with trypsin (37 ºC, 8 h).The sample was purified using solid-phase extraction cartridges to eliminate choline interference.The resulting peptide mixture was analysed using a MALDI-TOF/TOF (matrix-assisted laser desorption ionization-time of flight/time of flight) mass spectrometer (4800, AB Sciex) to evaluate the quality of the sample.MALDI-TOF spectra were interpreted using a peptide mass fingerprinting (PMF) database search (Protein Prospector program).The database used for identification was UniProt (release 2011_02).The sample was then analysed by liquid chromatography-tandem mass spectometry (LC-MS/MS) using a Velos-LTQ mass spectrometer equipped with a micro-ESI ion source (ThermoFisher, San Jose, CA, USA).The sample was evaporated to dryness and diluted up to 40 μl with water containing 5% methanol and 1% formic acid.The sample was then loaded in a chromatographic system consisting of a C18 pre-concentration cartridge (Agilent Technologies, Santa Clara, CA, USA) connected to a 10 cm long, 150 μm id Vydac C18 column (Vydac, IL, USA).Separation was carried out at 1 μl min -1 with a 3-40% acetonitrile gradient for 30 min (solvent A, 0.1% formic acid; solvent B, acetonitrile with 0.1% formic acid).The highperformance liquid chromatography (HPLC) system contained an Agilent 1200 capillary pump, a binary pump, a thermostated microinjector, and a micro switch valve.
The Velos-LTQ instrument was operated in positive ion mode with a spray voltage of 2 kV.The scan range of each full MS scan was m/z 400-2000.After each MS scan, a collection of targeted MS/ MS spectra was obtained in order to identify both the unmodified and nitrated form of the predicted tyrosine-containing peptides.The parent mass list of the targeted scan was selected to ensure maximum coverage of the tyrosine-containing tryptic peptides for MDAR.The list of targeted m/z values was obtained after in silico digestion of the proteins using nitrated tyrosine as a dynamic modification.The resulting list of predicted peptides (in both nitrated and unmodified form) was filtered to exclude all peptides not containing tyrosine residues.
The MS/MS spectra were searched using Proteome Discoverer software (ThermoFisher) on the basis of the following parameters: peptide mass tolerance 2 Da, fragment tolerance 0.8 Da, enzyme set as trypsin, and no missed cleavages.The dynamic modifications were cysteine carbamidomethylation (+57 Da), methionine oxidation (+16 Da), and tyrosine nitration (+45).The searches were carried out using a database containing all the peptides listed in Table 1.Identifications were filtered with XCorr >3, P(pep) <15%.The MS/ MS spectra of the nitrated tyrosines were manually validated by comparing the spectra obtained for the unmodified peptide and the nitrated peptide.

Site-directed mutagenesis
Conversion of the tyrosine codon (TAT) to phenylalanine (TTT) in the pea cDNA of peroxisomal MDAR (accession no, AY662655) was accomplished by oligonucleotide-directed mutagenesis.The template for PCR mutagenesis was also the pALEXb expression vector carrying the entire P. sativum pMDAR I gene obtained as described above.The site-directed mutant of MDAR Tyr345Phe was obtained by using the following primers: 5′-CTTTGATCTTTTCCAATCCAC-3′ (MDAR Tyr345Phe forward) and 5′-GTGGATTGGAAAAGATCAAAG-3′ (MDAR Tyr1345Phe reverse) where the mutated nucleotides are underlined.
The mutation was introduced using the QuikChange kit following the manufacturer's protocol.The mutant plasmid encoding nucleotide sequence was confirmed by DNA sequencing.The plasmids containing the mutation were transformed into XL1-Blue supercompetent cells and stored at -80 °C in 85% glycerol.The expression and purification of the recombinant mutant MDAR protein were obtained as previously described for the wild-type MDAR protein.

In silico analysis of MDAR
The three-dimensional structure of pea peroxisomal MDAR was modelled at the Geno3D server (Combet et al., 2002) using as template the structures of the putidaredoxin reductase from Pseudomonas putida (PDB code access 1q1w and 1q1r) (Sevrioukova et al., 2004) and the ferredoxin reductase from Rhodopseudomonas palustris (PDB code access 3fg2) (Xu et al., 2009) with an identity of 26.8% and 30.1%, respectively.The analysis of the quality of the models was carried out at the Structural Analysis and Verification Server (SAVES) in terms of atomic non-local environment assessment (ANOLEA) (Melo and Feytmans, 1998), three-dimensional profiles (Verify3D) (Eisenberg et al., 1997), and Procheck (Laskowski et al., 1993).The co-ordinates of FAD and NAD were calculated by superposition of the model on those of the X-ray structure 2YVG, which shares 30% identity with the primary structure of the pea MDAR and 1.55 Å rms (backbone atoms) with the pea MDAR model.
Docking of the model of pea MDAR with GSH was carried out at the SwissDock sever (Grosdidier et al., 2011) in accurate mode and without defining the region of interest (blind docking) but allowing flexibility for the side chains within 5 Å of any atom of the ligand in its reference binding mode.Analysis of the results was carried out with the help of Swiss PdbViewer (Guex and Peitsch, 1997) and UCSF Chimera (Pettersen et al., 2004) Molecular evolution studies were carried out at the Evolutionary Trace server (Mihalek et al., 2004) using the model of the tertiary structure as input.The phylogenetic significance and evolutionary conservation were explored by BLASTP searches (Altschul et al., 1997) on the subsection Viridiplantae of UniProtKB release 2013_05 (UniProt Consortium, 2011).The phosphorylation score was computed at the NetPhos 2.0 Server (Blom et al., 1999), and the solvent-accessible area by DSSP (Kabsch and Sande, 1983).
The estimation of the pKa and the analysis of the interactions was carried out with Propka 3.1 (Olsson et al., 2011).The algorithm was originally described by Li et al. (2005) as a very fast empirical method for structure-based pKa prediction that relies on the estimation of desolvation effects and intraprotein interactions to account for the variation in the standard pKa of ionizable groups.

Biotin switch method
For in vitro S-nitrosylation, peroxisomal MDAR and chloroplastic and cytosolic GR were incubated with GSNO as the NO donor for 30 min at room temperature.S-Nitrosylated recombinant proteins were subjected to the biotin switch method as described by Begara-Morales et al. (2014).The non-nitrosylated free cysteine residue was blocked by incubation with 30 mM methyl methane thionsulphonate and 2.5% SDS at 50 °C for 20 min with frequent vortexing.Residual methyl methane thionsulphonate was removed by precipitation with 2 vols of -20 °C acetone, and the proteins were resuspended in 0.1 ml of HENS buffer (25 mM HEPES pH 7.7 buffer containing 1 mM EDTA, 0.1 mM neocuproine, and 1% SDS) per milligram of protein.Biotinylation was obtained by adding 1 mM N- [6-(biotinamido) hexyl]-3′-(2′-pyridyldithio) propionamide (biotin-HPDP) and 0.1 mM ascorbate, and incubating at room temperature for 1 h.The proteins were then precipitated with 2 vols of -20 ºC acetone.Biotinlabelled proteins were separated by non-reducing 10% SDS-PAGE and then transferred onto polyvinylidene difluoride (PVDF) membranes (Immobilon P, Millipore, Bedford, MA, USA) using a semidry transfer system (Bio-Rad Laboratories).PVDF membranes were blocked using TRIS-buffered saline (TBS)+1% BSA.The blots were incubated with anti-biotin antibody at a dilution of 1:20 000 for 1 h, and the immunoreactive bands were detected using a photographic film (Hyperfilm, Amersham Pharmacia Biotech) with an enhanced chemiluminescence kit (ECL-PLUS, Amersham Pharmacia Biotech).

Purification of biotinylated proteins and immunodetection of MDAR
Purification of biotinylated proteins from control and NaCl-treated pea plant leaves was carried out as described by Begara-Morales et al. (2014).Briefly, biotinylated proteins and 30 μl of neutravidin Some peptides detected do not contain tyrosines.These peptides were not included in the targeted MS/MS detection.They were detected and identified as their molecular weight coincides with that of predicted peptides.
agarose 50% (w/v) slurry (high capacity neutravidin agarose resin, Thermo Scientific) per milligram of protein were equilibrated with a neutralization buffer [20 mM HEPES pH 7.7 containing 100 mM NaCl, 1 mM EDTA, and 0.5% (v/v) Triton X-100].Proteins were added to the neutravidin agarose matrix and were incubated 1 h at room temperature with gentle shaking.The matrix with bound proteins was washed several times with washing buffer [20 mM HEPES pH 7.7 containing 600 mM NaCl, 1 mM EDTA, and 0.5% (v/v) Triton X-100] and was transferred to an empty column.Finally, biotinylated proteins were eluted after incubation for 30 min with elution buffer (20 mM HEPES pH 7.7 containing 0.1 M NaCl, 1 mM EDTA, and 100 mM β-mercaptoethanol) at room temperature.Purified biotinylated proteins were separated by 12% SDS-PAGE and transferred to PVDF membranes as described above.

Expression and purification of recombinant pea MDAR and GR proteins: effect of peroxynitrite
In order to increase the knowledge of the regulatory mechanism of the ascorbate-glutathione cycle involved in the decomposition of H 2 O 2 , the recombinant proteins of two of its components, the MDAR and GR isoforms, were obtained by sequencing the pea clones and overexpression in E. coli (see the Materials and methods).Supplementary Fig. S1 at JXB online shows the electrophoretic analysis of the different fractions obtained after LYTRAP affinity column chromatography of recombinant MDAR and GR proteins.Recombinant MDAR (Supplementary Fig. 1A) showed a molecular mass of 68.6 kDa, which is within the range of the theoretical value predicted for the peroxisomal MDAR protein (47.3 kDa) with the Ly-tag (21.3 kDa).The fractions E3-E5 with an MDAR activity of 1200 nmol NADH min -1 mg -1 protein showed an adequate grade of purity for this protein which was used for subsequent experiments.On the other hand, the recombinants chloroplastic and cytosolic GR (Supplementary Fig. 1B, C) showed a molecular mass of ~81 kDa and 75 kDa, respectively, which is within the range of the theoretical value predicted for both GR proteins (59.7 kDa with the Ly-tag 21.3 kDa for the chloroplastic isoform, and 53.7 kDa with the Ly-tag 21.3 kDa for the cytosolic isoform).The fractions E2-E4 with GR activities of 18 μmol NADPH min -1 mg -1 protein were used for subsequent experiments.
In order to evaluate the potential action of different NO-derived molecules, an in vitro assay was carried out in the presence of ONOO -using SIN-1 as the peroxynitrite donor (Chaki et al., 2009;Begara-Morales et al., 2014).Figure 1A depicts the inhibitory effect of ONOO -on MDAR activity in a dose-dependent manner that ranges from 30% with 0.1 mM SIN-1 to 67% with 1 mM and 5 mM SIN-1, respectively.On the other hand, Fig. 1B shows that chloroplastic GR activity was not affected by any assayed concentration of ONOO -.Similar behaviour was observed for cytosolic GR (results not shown).
The consistency of nitration by SIN-1 was confirmed by immunoblot analysis of the recombinant proteins using an antibody against 3-nitrotyrosine.Figure 1C and D show that the degree of nitration of both MDAR and GR isoforms increases as a function of the SIN-1 concentration.

Effect of S-nitrosylation of recombinant pea MDAR and chloroplastic and cytosolic GR
In order to gain additional insight into the regulation of pea MDAR and GR proteins, the effect of increasing concentrations of GSNO, a well-known NO donor (Begara-Morales et al., 2014), on the enzymatic activities was evaluated.As shown in Fig. 1E, 0.5 mM and 2.0 mM GSNO considerably inhibited MDAR activity by between 58% and 65%, respectively.In contrast, chloroplastic GR activity is not significantly affected (Fig. 1F).When both activities were assayed in the presence of 0.5 mM and 2 mM GSH to determine whether this effect was due to the release and binding of NO to the protein, GSH was found not to affect either MDAR or GR activities (Fig. 1G and H, respectively).Similar behaviour was observed for cytosolic GR (results not shown).This indicates that none of these enzymes is affected by S-glutathionylation.Furthermore, to show that GSNO treatment of recombinant MDAR and GR isoforms causes S-nitrosylation, the biotin switch assay method (Begara-Morales et al., 2014) was specifically used to detect S-nitrosylated proteins.Figure 1I and J shows that MDAR and chloroplastic and cytosolic GR are S-nitrosylated after treatment with 2 mM GSNO (lane 3), whereas treatment with GSH does not produce any signal in the biotin switch assay (lane 2).As expected, S-nitrosylation of these proteins is reversible and can be down-regulated by adding a reducing agent such as DTT to the S-nitrosylated proteins (lane 4) or in the absence of ascorbate (lane 5) which is used as an SNO-specific reducing agent, further demonstrating S-nitrosylation.

Mapping tyrosine nitration sites and spectral characterization of nitrated pea peroxisomal MDAR
Given that GR is not affected by any of the PTMs mediated for NO-derived molecules, further study was conducted on MDAR.With the aim of identifying which of the 22 tyrosines present in the pea MDAR is(are) target(s) of this PTM, peroxynitrite-treated recombinant MDAR was subjected to trypsin digestion followed by MALDI-TOF/TOF mass spectrometry examination.Table 1 shows the list of peptides scanned and those identified by LC-MS/MS.Among the peptides identified, only three contained a nitrated tyrosine.Figure 2 shows the comparison of the nitrated (top) and unmodified (bottom) MS/MS spectra of these identified peptides from the pea and chloroplastic and cytosolic GR (D), treated with different concentrations of SIN-1 and detected with an antibody against 3-nitrotyrosine (dilution 1:2500).A 5 μg aliquot of protein was used per line.Effect of S-nitrosoglutathione (GSNO) on recombinant MDAR (E) and chloroplastic GR (F).Effect of glutathione (GSH) on recombinant MDAR (G) and chloroplastic GR (H).Purified MDAR and GR proteins were incubated at different concentrations of SIN-1 at 37 ºC for 60 min, GSNO at 25 ºC for 30 min, or GSH at 25 ºC for 30 min.The specific activity of the recombinant MDAR was 1200 nmol NADH min -1 mg -1 and for GR proteins it was 18 μmol NADPH min -1 mg -1 .S-Nitrosylation of recombinant MDAR (I) and chloroplastic and cytosolic GR (J).A 5 μg aliquot of purified recombinants proteins was treated with 2 mM GSH and 2 mM GSNO and was subjected to the biotin switch method.Control treatments were carried out with water (lane 1) and 2 mM GSH (lane 2).Additionally, recombinants proteins were S-nitrosylated with 2 mM GSNO (lane 3) and reduced again with 50 mM DTT (lane 4).Furthermore, GSNO-treated recombinant proteins underwent the biotin switch method without ascorbate (lane 5).Proteins were separated under non-reducing conditions by SDS-PAGE and blotted onto a PVDF membrane.Biotinylated proteins were detected using an anti-biotin antibody.Data are means ±SEM of at least three replicates.*Differences from control values were significant at P<0.05.
MDAR.The nitrated peptide LFTSEIAAFYEGY*YANK (Z=2) has a total of 17 amino acids and a mass of 2032 Da (1987 Da plus 45 Da) which is compatible with the acquisition of a nitro group in Tyr213 (Fig. 2A).The nitrated peptide TSVPDVY*AVGDVATFPLK (Z=2) has a total of 18 amino acids and a mass of 1924 Da (1879 Da plus 45 Da) which is also compatible with the acquisition of a nitro group in Tyr292 (Fig. 2B).The nitrated peptide SVEEYDYLPY*FYSR (Z=2) has a total of 14 amino acids and a mass of 1876 Da (1831 Da plus 45 Da) which is also compatible with the acquisition of a nitro group in Tyr345 (Fig. 2C).

Modelling of pea MDAR and identification of potential residues affected by peroxynitrite and GSNO
The low identity of pea MDAR with the structures available from the PDB made homology modelling a difficult task.The best co-ordinates were obtained from Geno3D server (Combet et al., 2002) using as template the PDB entries 1q1w, 1q1r (26.8% identity), and 3fg2 (30.1% identity) (Sevrioukova et al., 2004;Xu et al., 2009).The model was refined to -18 225.1 kcal mol -1 and comprises from residue 6 to residue 433 (98.8% coverage).The analysis of the quality of the model yields an Errat overall quality factor of 88.496, 81.82% of the residues with averaged 3D-1D score larger than 0.2, and 64.6% at the most favoured regions in the Ramachandran plot, 1.9% (i.e.seven residues) being at disallowed regions.
The location on the model of the pea MDAR of the three tyrosine residues identified as nitrated by LC-MS/MS and the two cysteine residues that are potential targets for GSNO did not provide any conclusive information regarding the modulation of the enzyme by tyrosine nitration and/or for S-nitrosylation (Fig. 3A).However, since MDAR plays an important role in homeostasis, it is reasonable to assume that its regulation has been preserved through evolution.On the other hand, the fact that the modulation takes place via the modification of only a few residues points to a particular reactivity of those residues.Bearing in mind both ideas, residues identified as potential targets of the PTM were further analysed.The results are summarized in Supplementary Table S1 at JXB online and reveal that Tyr292 is absolutely preserved and shows the lowest estimated pKa, Tyr213 is prone to phosphorylation, and Tyr345 is the most accessible.Cys197 is buried and less preserved that Cys68.Molecular docking with GSNO failed due to problems with the parameterization of the S-O bond, but when the calculations were carried out with GSH an area close to Cys68 was spotted where GSH fits with an estimated K d of 6.5 μM (Fig. 3B).

Effect of SIN-1 on recombinant mutant pea MDAR (Tyr345Phe) obtained by site-directed mutagenesis analysis
Of the nitrated tyrosines, in silico analysis outputs Tyr345 as the most likely candidate responsible for the observed inhibition by ONOO -since this residue is located at the active site and closely interacts with the cofactor.To corroborate this hypothesis, the Tyr345 residue in MDAR was replaced with phenylalanine by site-directed mutagenesis to obtain mutant MDAR Tyr345Phe.The overexpression and purification of the mutant were obtained as the recombinant wild type which yielded a functional enzyme resistant to inhibition by 0.5 mM and 5.0 mM ONOO -(Fig.4A).Additionally, the nitration by SIN-1 of the mutant MDAR Tyr345Phe was confirmed by immunoblot analysis using an antibody against 3-nitrotyrosine (Fig. 4B) which shows that the degree of nitration increases as a function of the SIN-1 concentration.This allows the confirmation that the nitration of Tyr345 is the residue responsible for the inhibition of MDAR activity.

Analysis of protein and gene expression of MDAR under salt-induced oxidative stress
In order to gain additional insight into the physiological relevance of MDAR under an oxidative stress situation, it was analysed in leaves of pea plants grown in the presence of 150 mM NaCl as was previously reported (Begara-Morales et al., 2014).Figure 5A shows by immunoblot analysis the MDAR protein expression which was found to increase clearly under salinity conditions.MDAR gene expression was also analysed and a similar behaviour to the protein expression was observed, with an increase of 3.5-fold under salt stress (Fig. 5B).

Purification of total S-nitrosylated proteins under salinity stress and detection of S-nitrosylated MDAR
To evaluate if MDAR under salinity stress conditions undergoes a process of S-nitrosylation, total S-nitrosylated proteins were purified from leaves of pea plants grown under control and salinity stress conditions and then the presence of MDAR protein among these S-nitrosylated protein was evaluated by immunoblotting.Figure 5D depicts the electrophoretic analysis of total S-nitrosylated proteins.Thus, under salinity stress, the pattern of S-nitrosylated proteins showed an increase in the number and in the intensity of some specific bands.Figure 5D shows the immunoblot analysis of the total S-nitrosylated proteins probed with an antibody against cucumber MDAR where an increase under salinity stress was also observed.Taken together, the results indicate that MDAR is S-nitrosylated in vivo and this process is increased under salinity conditions, which supports the data observed in in vitro conditions (Fig. 1I).

Discussion
PTMs such as nitration and S-nitrosylation mediated by NO-derived molecules are now considered to be crucial elements in the fine-tuned regulation of the function of their protein targets.In higher plants, proteomic analyses of nitration/S-nitrosylation have shown that a certain number of proteins are targets of PTMs mediated by NO-derived molecules (Lindermayr et al., 2005;Chaki et al., 2009;Álvarez et al. 2011;Lozano-Juste et al., 2011;Tanou et al., 2012;Begara-Morales et al., 2013a, b, 2014;Corpas et al., 2007Corpas et al., , 2015)).However, information on the specific impact of NO-PTMs on key proteins involved in antioxidative systems and the consequences for their functionality and protein structure is scarce (Yu et al., 2014).Very recently, in vitro assays of several recombinant Arabidopsis superoxide dismutases (MnSOD1, FeSOD3, and CuZnSOD3) have shown that these SOD activities were not altered upon GSNO treatment but were inhibited to different degrees by ONOO - (Holzmeister et al., 2015).In this context, the ascorbate-glutathione pathway is one of the key antioxidant systems involved in the regulation of H 2 O 2 levels in plant development and under abiotic stress conditions.In this respect, it has very recently been demonstrated that pea cytosolic APX, a key enzyme system in the antioxidant ascorbate-glutathione cycle, presents dual regulation by both tyrosine nitration and S-nitrosylation (Begara-Morales et al., 2014).Furthermore, several proteomic studies have identified GR and MDAR, other important components in the ascorbate-glutathione cycle, as targets of tyrosine nitration/Snitrosylation processes (Chaki et al., 2009;Lin et al., 2012;Tanou et al., 2012).To elucidate the molecular mechanism and physiological relevance of these NO-PTMs to GR and MDAR functionality and their effect on the potential operation of the cycle, the present study analyses the regulation of MDAR and GR activities at the molecular level by using NO-derived molecules such as peroxynitrite and S-nitrosoglutathione that have the capacity to mediate tyrosine nitration and S-nitrosylation, respectively.
GR is an important enzyme in the antioxidative defense system that converts oxidized glutathione (GSSG) to reduced glutathione (GSH) using NADPH as cofactor.This reaction enables the GSH/GSSG ratio to be maintained at a high level, which is very important given that GSH is considered to be the most abundant soluble antioxidant in plants.This enzyme system has different isoenzymes located in a diverse range of cell compartments (Edwards et al., 1990;Romero-Puertas et al., 2006;Wu et al., 2013) and plays an important physiological role in maintaining and regenerating GSH in response to biotic and abiotic stresses in plants (Creissen et al., 1992;Foyer and Noctor, 2011;Leterrier et al., 2012;Gill et al., 2013;Signorelli et al., 2013).Under the present experimental conditions, the activities of either isoform tested was unaffected by any NO-PTM assayed and mediated by ONOO -and GSNO, suggesting that this could be a mechanism for maintaining GSH regeneration in order to sustain antioxidant capacity of the ascorbate-glutathione cycle against nitro-oxidative cell conditions.Moreover, it must be pointed out that, in the case of tyrosine nitration mediated by peroxynitrite, this behaviour is unusual as, to the authors' knowledge, the pea GR is the first case of a nitrated protein found to be unaffected by this NO-PTM in higher plants as, up to now, most analyses have shown that nitration causes loss of function in all proteins identified in higher plants (Astier and Lindermayr, 2012;Corpas et al., 2013b).
Given that pea GR was unaffected by NO-PTMs either for peroxynitrite or for GSNO, this study was focused on MDAR which is the enzyme in the ascorbate-glutathione cycle involved in the regeneration of reduced ascorbate.The pea (P.sativum) plant has only one gene encoding MDAR (Murthy and Zilinskas, 1994) whose corresponding protein has been immunolocalized in the different subcellular compartments including chloroplasts, peroxisomes, mitochondria, and the cytosol (Leterrier et al., 2005).As part of the ascorbate-glutathione cycle, MDAR also plays an important role under environmental stresses in which nitro-oxidative stress could be a significant component.In pea, although MDAR activity increased under high light intensity and cadmium, it was reduced by the herbicide 2,4-D (Leterrier et al., 2005).However, during the natural senescence of pea leaves, a simultaneous decrease in MDAR and APX activities has been reported (Jiménez et al., 1998).In tomato, MDAR activity is also increased by salinity (Mittova et al., 2003) and high light intensity (Gechev et al., 2003), in rice by low temperature (Oidaira et al., 2000), and in Arabidopsis by UV-B radiation (Kubo et al., 1999).However, in Arabidopsis, stresses such as high temperature (30 ºC), enhanced light intensity (200 μE m -2 s -1 ), water deficiency (water deprivation for 2 d), and low temperature (5 ºC) did not affect MDAR activity (Kubo et al., 1999).As mentioned above, several proteomic studies have identified MDAR as a potential target for both S-nitrosylation and nitration (Lin et al., 2012;Tanou et al., 2012).However, the specific effects of these NO-PTMs on MDAR functions are not known.Thus, the present data show that both processes cause loss of MDAR function.Three putative candidates have been identified for tyrosine nitration and two cysteine residues are present in pea MDAR, but the fact that none of them is located in a relevant position from a functional point of view makes it difficult to understand the mechanism of inhibition and whether all of them are equally relevant.However since MDAR plays an important role in homeostasis and only a few residues are modified, it is reasonable to assume that those residues have been preserved through evolution and that they show a particular reactivity (Sano et al., 2004).For the particular case of tyrosine nitration, it is accepted that it proceeds though a radical mechanism where tyrosine cannot react directly with peroxynitrite but rather with carbon dioxide or a metal centre to yield a secondary oxidizing species that reacts with tyrosine to form the tyrosyl radical (Alvarez and Radi, 2003).Thus, from both evolutionary and chemical points of view, Tyr292 should be the strongest candidate for nitration, Tyr345 the weakest, and Tyr213 may be discarded and considered as a good candidate to undergo phophorylation.However, an analysis of the interaction network of the protein with its cofactors shows that only Tyr345 is involved (Table 2).Thus, the second interaction sphere of Tyr345 includes residues that directly interact with the atoms N3 (His313) and N10 (Glu45 and Asp296) from FAD (Table 2).Interestingly the prediction of the pKa values output by PropKa 3.1 for the ionizable groups present in FAD and NAD included in the model of pea MDAR reveals that atoms N3 and N10 from FAD show extreme anomalous values, the former being -1.72 (expected 5) and the latter 4.35 (expected 10).This analysis led to the hypothesis that nitration of Tyr345 should influence the functionality of MDAR.This hypothesis was confirmed by site-directed mutagenesis and, as expected, the functionality of the mutant MDAR Tyr345Phe was not affected by 5.0 mM ONOO -.
The enzymatic activity of MDAR is also regulated by GSNO, and a link between MDAR and NO metabolism throughout iron metabolism has been reported.Specifically, plant oxyhaemoglobin (Hb) can act as an NO scavenger with the concomitant production of nitrate and Fe 3+ -Hb which is Table 2. Analysis of the first and second interaction spheres (in italics) of the three tyrosine residues identified as nitrated The contribution of hydrogen bonding (involving side chains and backbone), coulombic interactions, and desolvation effects (regular, which is calculated according to Coulomb's law, and RE, which includes all interactions between the ionizable residue and the remaining protein, apart from the Coulomb energy, that affects the deprotonation energy of the residue).Residues of the second interaction sphere that interact with any of the cofactors are shown in bold.reduced to Fe 2+ -Hb in the presence of ascorbate and NADH, and MDAR activity is a key element in this process because it facilitates the regeneration of ascorbate (Igamberdiev et al., 2006;Wang and Hargrove, 2013).The sequence of MDAR comprises two cysteine residues and, according to Sano et al. (1995), one cysteine is more reactive towards DNTBA and it seems to participate in the reduction of the enzyme by NADH.

Residue
As depicted in Supplementary Table S1 at JXB online, Cys68 seems to be the best candidate for S-nitrosylation because it is better preserved and more accessible.The identification was approached taking into account that S-nitrosylation by GSNO is a transnitrosylation reaction, an affinity between GSNO and the target proteins is expected, and hypothesizing that it may be detected in docking experiments.In fact it has been demonstrated that such an affinity is strong enough to be exploited to isolate targets proteins (Begara-Morales et al., 2013a).Blind docking experiments with GSH found a region close to Cys68 where GSH fits with an estimated K d of 6.5 μM (Fig. 3B).This result is significant because amino acid sequence alignment reveals that Cys68 is equivalent to Cys117 of chloroplastic MDAR that has been reported to be involved in the activity and structural stability of chloroplastic MDAR (Li et al., 2010) Additionally, both ONOO -and GSNO showed concentration-dependent inhibitory effects on MDAR activity, which suggests that these NO-derived molecules can have a finetuned regulatory effect depending on their cellular production and physiological and stress conditions.It is well known that peroxisomes have a remarkable oxidative metabolism.However, it must also be pointed out that peroxisomes are subcellular compartments where the presence of l-argininedependent nitric oxide synthase activity (Barroso et al., 1999), NO generation, and other NO-related species including ONOO -and GSNO has been demonstrated (Barroso et al., 2013a;Corpas and Barroso, 2013).Therefore, peroxisomal MDAR could be modulated through loss of function when these molecules are overproduced under adverse conditions and could consequently contribute to a nitro-oxidative stress.In order to assess the in vivo relevance of the present study, salt stress (150 mM NaCl) was applied to plants as an inducer of nitro-oxidative stress (Begara-Morales et al., 2014).It was found that MDAR expression (mRNA, protein, and activity) increased, which may reflect a mechanism to compensate the inhibitory effect of S-nitrosylation and nitration upon the enzyme in salt-stressed pea leaves.To date, other peroxisomal enzymes such as catalase (Clark et al., 2000) and NADH-dependent hydroxypyruvate reductase (Corpas et al., 2013a) have been demonstrated to be targets of NO, which confirms the importance of NO in the peroxisomal metabolism.
In summary, the present results provide new insights into the molecular mechanism involved in regulating MDAR and GR through PTMs mediated by NO-derived molecules and confirm the close involvement of NO and ROS metabolism in the antioxidant defence against nitro-oxidative stress situations in plants.These data, together with previous findings on the dual regulation of APX by S-nitrosylation/ nitration (Begara-Morales et al., 2014), are summarized in Fig. 6.It shows the modulation of the antioxidative response of key enzymes in the ascorbate-glutathione cycle by NO-PTMs, where MDAR was deactivated by nitration and S-nitrosylation, which could compromise the cycle's antioxidant capacity.However, GR was not affected by any of these NO-PTMs in an attempt to maintain the levels of GSH and the cellular redox state, suggesting that this could be a crucial mechanism to sustain the antioxidant capacity of the ascorbate-glutathione cycle against nitro-oxidative cell conditions.Figure S2.Multiple alignment of the deduced amino acid sequences of MDAR in different plant species.
Table S1.Characterization in terms of evolutionary conservation, likelihood of tyrosine phosphorylation, solvent-accessible surface area, and estimated pKa of the three tyrosine and two cysteine residues susceptible to being responsible for the modulation of the enzymatic activity of pea MDAR by peroxynitrite and GSNO.

Fig. 1 .
Fig. 1.Effect of nitration and S-nitrosylation on recombinant monodehydroascorbate reductase (MDAR) and glutathione reductase (GR).Effect of SIN-1 (peroxynitrite donor) on recombinant MDAR (A) and GR (B) activities.Representative immunoblot showing the grade of tyrosine nitration of MDAR (C)and chloroplastic and cytosolic GR (D), treated with different concentrations of SIN-1 and detected with an antibody against 3-nitrotyrosine (dilution 1:2500).A 5 μg aliquot of protein was used per line.Effect of S-nitrosoglutathione (GSNO) on recombinant MDAR (E) and chloroplastic GR (F).Effect of glutathione (GSH) on recombinant MDAR (G) and chloroplastic GR (H).Purified MDAR and GR proteins were incubated at different concentrations of SIN-1 at 37 ºC for 60 min, GSNO at 25 ºC for 30 min, or GSH at 25 ºC for 30 min.The specific activity of the recombinant MDAR was 1200 nmol NADH min -1 mg -1 and for GR proteins it was 18 μmol NADPH min -1 mg -1 .S-Nitrosylation of recombinant MDAR (I) and chloroplastic and cytosolic GR (J).A 5 μg aliquot of purified recombinants proteins was treated with 2 mM GSH and 2 mM GSNO and was subjected to the biotin switch method.Control treatments were carried out with water (lane 1) and 2 mM GSH (lane 2).Additionally, recombinants proteins were S-nitrosylated with 2 mM GSNO (lane 3) and reduced again with 50 mM DTT (lane 4).Furthermore, GSNO-treated recombinant proteins underwent the biotin switch method without ascorbate (lane 5).Proteins were separated under non-reducing conditions by SDS-PAGE and blotted onto a PVDF membrane.Biotinylated proteins were detected using an anti-biotin antibody.Data are means ±SEM of at least three replicates.*Differences from control values were significant at P<0.05.

Fig. 2 .
Fig. 2. Comparison of the nitrated (top) and unmodified (bottom) MS/MS spectra of the peptides identified from the pea peroxisomal MDAR in the corresponding panels: (A) LFTSEIAAFYEGY*YANK, (B) TSVPDVY*AVGDV ATFPLK, and (C) SVEEYDY*LPYFYSR.Peptide fragment ions are indicated by 'b' if the charge is retained on the N-terminus and by 'y' if the charge is maintained on the C-terminus.The subscript indicates the number of amino acid residues in the fragment studied from either the N-terminus or the C-terminus.The superscript indicates the charge (1 + or 2 + ) of the backbone fragmentation.(This figure is available in colour at JXB online.)

Fig. 5 .
Fig. 5. Protein and gene expression of MDAR and analysis of S-nitrosylated MDAR in leaves of pea plants under salinity (150 mM NaCl) stress conditions.(A) Immunoblotting analysis of MDAR protein expression using an antibody against cucumber MDAR (dilution 1:3000).A 10 μg aliquot of protein was used per lane.(B) Real-time quantitative RT-PCR transcript analysis (arbitrary units) of the MDAR gene.Data are means ±SEM of at least four independent RNA samples.*Differences from control values were significant at P>0.05.(C) Detection of total S-nitrosylated proteins separated under non-reducing conditions by 12% SDS-PAGE and blotted onto a PVDF membrane.Biotinylated proteins were detected using anti-biotin antibodies as described in the Materials and methods.(D) Immunoblot of total S-nitrosylated proteins probed with an antibody against cucumber MDAR (dilution 1:3000).A 5 μg aliquot of protein was used per lane.Downloaded from https://academic.oup.com/jxb/article/66/19/5983/697589 by guest on 08 May 2024

Figure
Figure S1.SDS-PAGE analysis of the purification of the recombinant proteins.FigureS2.Multiple alignment of the deduced amino acid sequences of MDAR in different plant species.TableS1.Characterization in terms of evolutionary conservation, likelihood of tyrosine phosphorylation, solvent-accessible surface area, and estimated pKa of the three tyrosine and two cysteine residues susceptible to being responsible for the modulation of the enzymatic activity of pea MDAR by peroxynitrite and GSNO.

Table 1 .
List of pea MDAR peptides scanned and identified by LC-MS/MS