Introduction

During the evolutionary arms race, plants have developed many mechanisms to counteract the pathogen ingress and limit the severity of infection. Defense activation in plants is based on a complex signaling network orchestrated by reactive nitrogen (RNS) and oxygen species (ROS). Among them, nitric oxide (NO) as a free radical gas is a key player involved in diverse processes in plants, including response to multiple biotic and abiotic stress factors. A crucial role of NO in plant immune responses has been reported in different pathosystems including Solanaceae plants and oomycete pathogens (Floryszak-Wieczorek et al. 2012; Janus et al. 2013). Noritake et al. (1996) reported for the first time the involvement of NO in potato innate immunity and demonstrated that pre-treatment of potato tubers with an NO donor (NOC-18) induced the accumulation of rishitin, an important phytoalexin in an establishment of constitutive resistance.

Moreover, the synchronized overproduction of NO and superoxide (O2 •-) in plants may lead to the formation of peroxynitrite (ONOO−)—considered as an NO downstream signal and a potent modulator of the redox regulation in various cell signal transduction pathways. In contrast to animal system, where ONOO− executes cell death, in plants ONOO− is not indispensable in the activation of hypersensitive response (HR) (Delledonne et al. 2001). Hence, it is the co-operation of NO and active oxygen species, such as hydrogen peroxide (H2O2), induces HR in Arabidopsis after inoculation with Pseudomonas syringae pv. maculicola (Psm) or tomato (Pst) (Delledonne et al. 1998).

Efficient immune responses in plants are related to the early generation of highly reactive molecules with a high oxidizing potential, causing an imbalance in delicate redox homeostasis (Noctor et al. 2012). Therefore, restoration of cell balance by redox re-adjustment processes may somehow determine future adequate response (Groß et al. 2013). Maintaining the redox state by enzymatic and non-enzymatic apparatus in plants under stress conditions may eventually determine the efficiency of redox signaling in response to avr and vr pathogens, respectively.

Post-infection generation of NO may modify the total pool of S-nitrosothiols (SNOs) and convert NO-exported bioactivity into altered metabolism and PR gene expression (Astier et al. 2011). Interestingly, the NO-coded message may be stored in the post-translational modification of proteins. These, in turn, may be controlled by thioredoxins system and by trans-nitrosylation reactions with low-molecular mass thiols, governed mainly by GSNO reductase activity (Chaki and Lindermayr 2014).

Protein S-nitrosylation, based on NO equivalent transfer to cysteine thiol, is a reversible and redox-dependent modification regulating the activity of an increasing repertoire of proteins. The plant S-nitrosoproteome has been explored since 2005 and so far more than two hundred protein S-nitrosylation targets have been proposed (e.g. Lindermayr et al. 2005; Kato et al. 2013; Vanzo et al. 2014). Among them many await functional validation under control and stress conditions (Fares et al. 2014). In potato more than 80 proteins were found to undergo S-nitrosylation after exposure of potato leaves and tubers to S-nitrosoglutathione (Kato et al. 2013). In the presented paper, we used a modified biotin-switch technique to detect S-nitrosylation of proteins in mock- and P. infestans-inoculated potato leaves. Accumulating data suggest that storage of inactive signaling proteins and transcription factors may promote rapid immune response after de-nitrosylation and subsequent activation (Malik et al. 2011). Thus, deciphering the mechanism of NO-based signal storage and sensing on PR gene expression are pivotal in potato resistance to P. infestans.

In this study, we investigated how potato leaves exposed to avirulent and virulent P. infestans could activate post-stress NO signaling, moving towards or compromising resistance. For this purpose early and late interactions between reactive nitrogen and oxygen species were compared in the incompatible and compatible response. Thus, we provided evidence that NO-mediated changes re-written on protein S-nitrosylation were linked to redox homeostasis re-adjustment leading to PR protein accumulation and late blight disease control.

Materials and methods

Plant material and pathogen culture

A resistant potato line, Solanum tuberosum L. cv. ‘Bzura’ was derived from in vitro tissue culture and kept in sterile soil in a phytochamber (16 h/8 h : day/night; 180 μmol m−2 s−1) at 18 ± 1 °C and 60 % relative humidity up to the stage of eight leaves.

Pathogen culture and inoculation with P. infestans

Phytophthora infestans (Mont.) de Bary avirulent 1.3.4.7.10.11 (MP946) and virulent race 1.2.3.4.6.7.10 (MP977) were obtained from the Plant Breeding and Acclimatization Institute, Research Division at Młochów, Poland. Isolate MP946 in response to the used potato line triggered hypersensitive pointed cell death (HR) identified by us earlier as TUNEL-positive (Floryszak-Wieczorek et al. 2013). Potato plants were inoculated by spraying leaves with 5 ml of an oomycete zoospore suspension at a concentration of 1.0 × 105 per 1 ml of water and they were kept overnight at 100 % relative humidity and 18 °C and afterwards they were transferred to a growth chamber.

Assessment of disease index

The area under disease progress was assessed on potato leaves 7 days after inoculation with P. infestans and was based on a scale from I to IV (James 1971), which represented the percentage of leaf area covered by late blight symptoms (I = 1 to 9 %; II = 10 to 24 %; III = 25 to 49 %; IV = 50 to 100 %). Disease symptoms were also determined using trypan blue staining of P. infestans mycelium according to the assay proposed by Wilson and Coffey (1980), i.e. potato leaf discs (ø = 2 cm) were inoculated with a zoospore suspension (40 μl). Imaging was performed by scanning leaf discs with a LIDE 210 Scanner (Canon). The blue stain corresponded to the area covered by P. infestans mycelium and was analyzed using the ImageJ 1.47v open source software (Wayne Raspand National Institutes of Health, USA).

Nitric oxide generation

The FL-NO fluorescence in extracts of potato leaves after inoculation was assayed spectrofluorimetrically using a selective nitric oxide sensor (CuFL) (Lim et al. 2006). The copper-complex of FL (2-{2-Chloro-6-hydroxy-5-[2-methylquinolin-8-ylamino)methyl]-3-oxo-3H-xanthen-9-l}benzoic acid) was prepared as 1 mM water stock solution according to the manufacturer’s instructions (Strem Chemicals). Leaf tissue (500 mg of fresh weight) was homogenized in 2 ml of 10 mM potassium-phosphate buffer (pH 6.0). The extract was centrifuged at 21,000 × g for 30 min at 4 °C. Then, 100 μl of supernatant were immediately used for NO assay by adding CuFL to the final concentration of 2 μM. Fluorescence intensity was determined with the Fluorescence Spectrometer Perkin Elmer LS 50B (UK) using 488 and 516 nm for excitation and emission, respectively. Each value was expressed as relative fluorescence intensity [Int × g−1 FW].

Superoxide radical production

Superoxide accumulation was determined by monitoring the reduction of NBT to diformazan in the presence of O2 •− according to Doke (1983). The amount of reduced NBT was measured at a wavelength of 580 nm. The reference sample consisted of the incubation mixture lacking plant material. The O2 •− level was expressed as ΔA580 [h−1 × g−1 FW].

Peroxynitrite formation

The level of peroxynitrite was assayed according to Huang et al. (2007) using folic acid as the peroxynitrite scavenger, giving a high fluorescent emission product. Fluorescence intensity of the solution was recorded at 460 nm with the excitation wavelength set at 380 nm. The standard curve was prepared for SIN-1 (Sigma-Aldrich) as a donor of peroxynitrite.

Hydrogen peroxide accumulation

The concentration of H2O2 was assayed spectrophotometrically using a titanium (Ti4+) method described by Becana et al. (1986). Fresh leaves (0.25 g) were homogenized in 3 ml of 0.1 M potassium phosphate buffer (pH 7.8). After centrifugation (15,000 × g for 30 min), the supernatant was used for further assays. The reaction mixture (1.5 ml) contained 0.1 M potassium-phosphate buffer (pH 7.8), enzymatic extract (400 μl) and titanium reagent. Titanium reagent was prepared on the day of assay by mixing 0.6 mM solution of 4-(2-pyridylazo) resorcinol and 0.6 mM potassium titanium tartrate at a 1:1 ratio. The concentration of H2O2 was determined by measuring absorbance at a wavelength of 508 nm against a calibration curve and expressed as μmol H2O2 × g−1 FW.

S-nitrosothiol total pool quantification

Total SNO content was determined by chemiluminescence using a Sievers® Nitric Oxide Analyzer NOA 280i (GE Analytical Instruments, USA) according to the procedure proposed by Chaki et al. (2009). The detection of SNOs was based on reductive decomposition of nitroso compounds by an iodine/triiodide mixture in the presence of copper and conducted under red safety light. Fresh leaves (0.250 g) were homogenized in Tris–HCl 0.1 M buffer pH 7.5 (1:4, w/v) containing 100 μM DTPA, 1 mM EDTA, 1 mM EGTA, 1 mM PMSF, 0.1 mM neocuproine, 3.5 % (w/v) PVPP, 0.25 % (v/v) Triton X-100 and centrifuged at 3,000 × g for 10 min. The supernatants were incubated with 10 mM NEM (N-ethylmaleimide) for 15 min at 4 °C and subsequently two aliquots were prepared for each sample. To remove nitrite one aliquot was incubated for 15 min with 10 mM sulphanilamide at 4 °C. To eliminate nitrite and decompose SNOs the next aliquot was treated with 10 mM sulphanilamide and 7.3 mM HgCl2 for 15 min at 4 °C. The difference between detected signals obtained from these aliquots demonstrated the total SNO content.

Total sulfhydryl group content

The status of -SH groups was assayed spectrophotometrically according to Rice-Evans et al. (1991). Fresh leaves (250 mg) were homogenized in 2 ml of 0.1 m citrate buffer (pH 3.0). After centrifugation at 15,000 × g for 15 min, 300 μl of supernatant were collected and 10 % sodium dodecyl sulfate (SDS) and 10 mM sodium-phosphate buffer (pH 8.0) were added. Initial absorbance (A0) was measured after mixing at a wavelength of 412 nm. Next, 5,5′-dithiobis-(2-nitrobenzoic acid) (DTNB) was added and the mixture was incubated for 1 h at 37 °C. Simultaneously, the control sample was prepared with 10 mM sodium-phosphate buffer (pH 8.0) instead of DTNB. Final absorbance (A1) was measured at a wavelength of 412 nm. The difference in absorbance A1-A0 (after subtracting an analogous value obtained for the control sample) was a measure of –SH group content in the sample. Sulfhydryl group content was expressed as glutathione equivalents [μmol GSH × g−1 FW].

Ascorbate level

For the determination of ascorbate the spectrophotometric assay was used as described by Mukherjee and Choudhuri (1983). Fresh leaves (250 mg) were homogenized in 5 % triacetic acid (TCA) (1:8; w/v) and then centrifuged at 15,000 × g for 20 min at 4 °C. To TCA-diluted supernatant 10 μl of 10 % thiourea and 500 μl of 0.28 % dinitrophenyl hydrazine in 1 M HCl (w/v) were added. After 20 min incubation at 100 °C and cooling on ice bath, 1.25 ml 80 % H2SO4 was added to stop the reaction. Absorbance was measured at 530 nm and obtained results were calculated according to the standard curve in the range 1–12 μg ml−1 ascorbate.

Total antioxidantive capacity

Total antioxidative capacity was based on leaf extract capacity to reduce the 2,2′-azinobis-(3-ethylbenzothiazoline-6-sulfonic acid) (ABTS) cation radical according to the method proposed by Re et al. (1999). Initial ABTS+ solution was diluted with 0.1 M potassium-phosphate buffer (pH 7.4) to set absorbance at a wavelength of 414 nm on 1.0. Fresh leaves (250 mg) were homogenized in 2 ml of 5 % TCA and centrifuged at 15,000 × g for 15 min. The volume of 980 μl diluted ABTS+ was pipetted to a cuvette and absorbance (A0) was measured at a wavelength of 414 nm. Next, 20 μl of the extract were added and absorbance was measured again after 10 s (A2) and 30 min (A1), respectively. Fast antioxidants were calculated as ΔAfast = A1-A0. Slow antioxidants were calculated as ΔAslow = (A2-A1)-(A2 ’-A1 ’). The calibration curve was prepared by successively adding 5 μl portions of 0.01 mM Trolox® to ABTS+ and measuring a decrease of absorbance. The final result of total antioxidative capacity was expressed in mM Trolox × g−1 FW.

Enzyme activities

NADPH oxidase [EC 1.6.3.1]

The NADPH dependent O2 •− generating activity was determined by a modified assay based on a reduction of XTT by O2 •− anions according to the method of Able et al. (1998). Fresh leaves (0.5 g) were homogenized in 50 mM potassium-phosphate buffer, pH 7.0 (1:4; w/v), containing 0.1 % Triton X-100 (v/v), 1 % PVP, 0.04 % Na 2O5, 1 mM EDTA and centrifuged at 18,000 × g for 20 min. Supernatants were passed through Sephadex G-25 gel filtration columns (Illustra NAP-10, GE Healthcare) and served as the enzyme extract. The volume of 1 ml assay reaction mixture contained 0.5 mM XTT, 0.1 mM NADPH and 30 μl enzyme extract in 50 mM Tris–HCl buffer, pH 7.5. XTT reduction was determined at 470 nm and rates of O2 •− generation were calculated using an extinction coefficient for XTT of 2.16 × 104 M−1 cm−1 and the enzyme activity was expressed as μmol O2 •− × min−1 × mg−1 protein.

Superoxide dismutase [EC 1.15.1.1.]

SOD activity was assayed by measuring SOD ability to inhibit the photochemical reduction of NBT using the method of Beauchamp and Fridovich (1971). Fresh leaves (250 mg) were homogenized in 0.05 M potassium-phosphate buffer, pH 7.0 (1:12; w/v), containing 1 % PVPP, 1 mM EDTA, 0.01 M NaCl and centrifuged at 20,000 × g for 30 min. The assay mixture contained 0.05 M sodium phosphate buffer (pH 7.8), 13 mM methionine, 75 μM NBT, enzymatic extract and 2 μM of riboflavin. The reaction was initiated by UV radiation (15 W) and was run for 15 min. The absorbance was measured at a wavelength of 560 nm. The amount of enzyme that caused inhibition of the NTB reduction reaction by 50 % was assumed as a unit of SOD activity (U × mg−1 protein).

S-nitrosoglutathione reductase [EC 1.2.1.46]

The GSNOR activity was determined according to the procedure proposed by Barroso et al. (2006) with minor modifications. Fresh leaves (500 mg) were homogenized in 0.1 M Tris–HCl buffer, pH 7.5 (1:4 w/v) containing 0.2 % Triton X-100 (v/v), 10 % glycerol (v/v), 0.1 mM EDTA, 2 mM DTT at 4 °C and centrifuged at 27,000 × g for 25 min. The supernatant was passed through Sephadex G-25 gel filtration columns (Illustra NAP-10, GE Healthcare), then immediately through Amicon Ultra 3 K Filters (Millipore) and served as the enzyme extract. The assay reaction mixture of 1 ml contained 0.5 mM EDTA, 0.2 mM NADH, 0.4 mM GSNO and 30 μl enzyme extract in 25 mM Tris–HCl buffer, pH 8.0. The reaction was held at 25 °C and initiated with an addition of GSNO (Sigma Aldrich). NADH oxidation was determined at 340 nm and rates of NADH consumed at min−1 were calculated using an extinction coefficient of 6220 M−1 × cm−1.

β-1,3-glucanase [EC 3.2.1.6]

The β-1,3-glucanse (PR-2) activity was determined according to the procedure proposed by Abeles and Forrence (1970) in a colorimetric assay utilizing laminarin as a substrate. Fresh leaves (250 mg) were homogenized in 0.05 M potassium-acetate buffer, pH 5.0 (1:16; w/v), containing 0.125 g Polyclar AT and then centrifuged at 15,000 × g for 25 min. The supernatant (0.5 ml) was added to 0.5 ml 2 % (w/v) laminarin aqueous solution and was incubated for 2 h at 50 °C. After stopping the reaction and dilution (1 : 10), optical density was read at 500 nm. The β-1,3-glucanase activity was determined as the level of reducing sugars produced and served as glucose equivalents [μmol glucose × min−1 × mg−1 protein]; finally PR-2 activity was expressed as % of control.

Chitinase [EC 3.2.1.14]

The chitinase (PR-3) activity was determined in a colorimetric assay utilizing CM–Chitin–RVB (LOEWE Biochemica) as a substrate (Pauly et al. 1999). Fresh leaves (250 mg) were homogenized in 0.05 M potassium-acetate buffer, pH 5.0 (1:16; w/v), and then centrifuged at 10,000 × g for 10 min. The volume of 0.2 ml of the enzymatic fraction was added to 0.2 ml CM–Chitin–RVB (2 mg × ml−1) and 0.4 ml of homogenization buffer. Then the reaction mixture was incubated for 1 h at 37 °C and stopped by adding 0.2 ml 2 M HCl. The reaction mixture was centrifuged at 10,000 × g for 5 min and the supernatant was collected to measure optical density at 550 nm. Chitinase activity was calculated according to the standard curve prepared with the use of recombinant Streptomyces griseus chitinase at 0.005–0.130 mU (Sigma-Aldrich). Enzyme activity was expressed as % of control.

Gene expression measurement

The RNA was isolated from 150 mg of frozen leaf using TriReagent® (Sigma) according to the method of Chomczynski and Sacchi (1987). The obtained RNA was purified with the use of a Deoxyribonuclease I Kit (Sigma). For the reverse transcription 1 μl of RNA from every experimental variant was processed with a RevertAid™ Reverse Transcriptase Kit (Thermo Scientific) according to the manufacturer’s instructions. Real-time PCR was performed on a Rotor Gene 6000 Thermocycler (Corbett Life Sciences). The reaction mixture contained 0.1 μM of each primer, 1 μl of 5× diluted cDNA, 10 μl of the Power SYBR® Green PCR Master mix (Applied Biosystems) and DEPC-treated water to the total volume of 20 μl. The real-time PCR reaction conditions included an initial 5-min denaturation at 95 °C, followed by 55 cycles consisting of 10 s at 95 °C, 20 s at 53 °C and 30 s at 72 °C. The reaction was finalized by denaturation at a temperature rising from 72 to 95 °C by one degree at every 5 s. Reaction specificity was confirmed by the occurrence of one peak in the melting curve analysis. PR-1 primers used in real-time detection were as follows:

F: CCGCGTTGAGCTGGGGGAAA, R: GAGCTGGGGACTGCAGGATGC (Tm = 53 °C). The data were normalized to the reference gene encoding elongation factor (ef1α, AB061263; F: ATTGGAAACGGATATGCTCCA, R: TCCTTACCTGAACGCCTGTCA, Tm = 53 °C). All used primers were designed using Primer-BLAST (Ye et al. 2012). The Ct values were determined with the use of a Real-time PCR Miner (Zhao and Fernald 2005) and the relative gene expression was calculated with the use of efficiency corrected calculation models presented by Pfaffl (2001).

Detection of endogenously S-nitrosylated proteins in potato leaves by the modified biotin-switch technique

The in vivo S-nitrosylated proteins were detected with the use of the modified biotin-switch technique (Lindermayr et al. 2005; Vanzo et al. 2014). Briefly, 250 mg of frozen leaf powder was mixed with 1 ml of HENT buffer containing 100 mM HEPES-NaOH pH 7.4, 10 mM EDTA, 0.1 mM neocuproine and 1 % (v/v) Triton X-100. After 15 min incubation the mixture was centrifuged at 4 °C for 10 min at 12,000 × g. Protein concentration was measured using Bradford reagent and was adjusted to 1 μg × μl of HEN buffer, containing cOmplete™ protease inhibitor cocktail tablets (Roche). The blocking step was performed with the use of four volumes of freshly prepared HENS buffer (225 mM HEPES-NaOH pH 7.2, 0.9 mM EDTA, 0.1 mM Neocuproine, 2.5 % (w/v) SDS) and 30 mM NEM. Then the extracts were incubated at 37 °C for 30 min. After incubation excess blocking agent was removed by ice-cold acetone precipitation. The pellet was re-suspended in 0.1 ml HENS buffer (without NEM) per 1 mg of protein. Reduction of –SNOs and simultaneous biotinylation were performed using 1 mM biotin-HPDP (EZ-LinkTM Biotin-HPDP; Thermo scientific) and 3 mM reductant sinapinic acid (Sigma-Aldrich) for 1 h at 25 °C in darkness. Next, the precipitation using ice-cold acetone was conducted and the obtained pellets were re-suspended in non-reducing sample buffer and subjected to electrophoresis and Western blot analysis. Series of controls were performed including the positive signal control (omitting the blocking step) and the false-positive signal (omitting the biotinylation step). Electrophoresis was performed in non-reducing SDS-PAGE on gradient 4-20 % polyacrylamide gels (Mini-PROTEAN® Precast Gels, Biorad) followed by electrotransfer to the PFDF membrane (Millipore). The membrane was blocked using 1 % bovine serum albumin for 1 h, and the blots were then incubated with the anti-biotin mouse monoclonal antibody conjugated with alkaline-phosphatase (dilution 1:25000; Sigma-Aldrich) for 24 h. Further, the protein bands were visualized using 5-bromo-4-chloro-3-indolyl phosphate and nitro blue tetrazolium (SigmaFast BCIP, Sigma-Aldrich) according to the manufacturer’s instructions.

Identification of proteins using mass spectrometry

Identification and analyses of proteins were performed with the use of a liquid chromatograph coupled with a mass spectrometer at the Mass Spectrometry Laboratory, the Institute of Biochemistry and Biophysics, the Polish Academy of Sciences. Protein bands were excised from the PVDF membrane and placed in 100 mM ammonium carbonate (pH 8.0) and directly subjected to the digestion procedure (overnight with 10 ng × μl−1 trypsin). After simultaneous reduction with 10 mM DTT (30 min at 56 °C) and alkylation with iodoacetamide (in darkness for 45 min at room temperature) the resulting peptide mixtures were concentrated and desalted on a RP-C18 pre-column (Waters). Further peptide separation was run on a nano- Ultra Performance Liquid Chromatography (UPLC) RP-C18 column (Waters, BEH130 C18 column, 75 μm i.d., 250 mm long) of a nanoACQUITY UPLC system, using a 45-min linear acetonitrile gradient. The column outlet was directly coupled to the Electrospray ionization (ESI) ion source of the Orbitrap Velos type mass spectrometer (Thermo), working in the regime of data dependent on the MS to MS/MS switch with HCD type peptide fragmentation. An electrospray voltage of 1.5 kV was used. Raw data files were pre-processed with the Mascot Distiller software (version 2.4.2.0, MatrixScience). The obtained peptide masses and fragmentation spectra were matched to the National Center Biotechnology Information (NCBI) non-redundant database (37425594 sequences; 13257553858 residues) with a Viridiplantae filter (1760563 sequences) using the Mascot search engine (Mascot Daemon v. 2.4.0, Mascot Server v. 2.4.1, MatrixScience). The following search parameters were applied: enzyme specificity was set to semiTrypsin, peptide mass tolerance to ± 30 ppm and fragment mass tolerance to ± 0.1 Da. The protein mass was left unrestricted, while mass values were set as monoisotopic with two missed cleavages being allowed. Alkylation of cysteine by carbamidomethylation, oxidation of methionine and carboxymethylation on lysine were set as a variable modification.

Statistical analysis

All results were based on at least three independent experiments, each with at least three biological replicates. Analysis of variance was conducted and the least significant differences (LSD) between means were determined using Tukey’s test at a level of significance P = 0.05. SigmaPlot 11.0 (Systat) was used to perform statistical tests. Randomization was performed during collection of samples in histochemical assay of trypan blue staining.

Results

Generation of reactive nitrogen and oxygen species in potato leaves exposed to P. infestans

Semi-quantitative measurement of NO generation in potato leaves after inoculation was performed using the NO-selective fluorescent probe copper-complex of FL (Cu-FL). The avirulent race triggered an enhanced NO generation in potato leaves early after inoculation, with the maximum overproduction at 3 hpi (Fig. 1a). In contrast, virulent P. infestans provoked minor changes in the NO status similar to that of the mock-inoculated leaves. Augmented NO synthesis in avr P. infestans-inoculated potato was accompanied by enhanced O2 •- generation (Fig. 1b). Both races of the pathogen caused a sustained increase of the O2 •- level with the similar kinetics. Synchronized generations of NO and O2 •- in the avr P. infestans-potato system increased the formation of peroxynitrite (ONOO−) at 3 hpi (Fig. 1c). In turn, in the compatible interaction the kinetics of ONOO− formation showed a late accumulation at 24 hpi (Fig. 1c). Boosted generation of O2 •- could be a result of controlled O2 •- synthesis by NADPH oxidase. Hence, in the incompatible interaction the activity of NADPH oxidase was strongly induced from 1 hpi, reaching a two-fold increase in relation to uninoculated leaves (Fig. 1d). In the compatible interaction an up-regulation of NADPH oxidase was recorded later from 3 hpi (Fig. 1d). It was also found that avr P. infestans caused a considerably enhanced H2O2 overproduction in each tested time-point after inoculation (Fig. 1e), whereas in the compatible interaction only moderate changes were recorded. Moreover, in the incompatible interaction the above-mentioned changes in H2O2 content correlated in the time-dependent manner with the activity of superoxide dismutase (SOD). SOD was significantly up-regulated throughout the whole analyzed 24-h-period, reaching a steady two-times higher level in relation to healthy leaves (Fig. 1f).

Fig. 1
figure 1

The effect of P. infestans inoculation on potato leaves cv. ‘Bzura’: a nitric oxide (NO) and b superoxide radical (O2 •-) generation, c formation of peroxynitrite (ONOO−), d activity of NADPH oxidase (NOX), e accumulation of hydrogen peroxide (H2O2), f activity of superoxide dismutase (SOD), * significantly different from mock-inoculated potato leaves, P < 0.05. Values represent the average of data ± SD of three independent experiments

Full size image

Metabolic status of NO in potato leaves after inoculation

To gain further insight into NO-dependent down-stream signaling in potato, the S-nitrosothiols (SNOs) total pool was analyzed using precise chemiluminescence assay. An increased storage of the NO message in the total SNOs content was observed in avr P. infestans-treated potato from 1 hpi, attaining the highest level at 24 hpi (Fig. 2a). In contrast, the SNO pool remained unchanged in potato inoculated with the virulent race of the pathogen. The SNO level is prone to be regulated by GSNO reductase, capable of GSNO turn-over and affecting trans-nitrosylation processes. GSNOR activity was only moderately down-regulated in avr P. infestans-inoculated leaves (Fig. 2b). However, it must be stated that a lack of strong time-dependent correlation was found between GSNOR activity and the SNO pool. Furthermore, the complex management of NO-initiated pathway involves changes in S-nitrosoproteome structure. By performing the modified biotin-switch assay coupled to quantitative LC-MS/MS analysis, 104 protein targets of S-nitrosylation in potato were proposed. Supplementary Table 1 presents a list of in vivo S-nitrosylated proteins in mock- and P. infestans-treated potato leaves. Proteins typed for S-nitrosylation were clustered into 5 main functional groups (Fig. 6d Appendix). The most abundant cluster of ‘metabolic enzymes’ involved 44 % of the identified S-nitrosoproteome. ‘Photosynthesis-involved proteins’ comprising proteins orchestrating photosynthetic light reactions and the Calvin-Benson cycle were the second most intensely represented (24 %). The following less frequent groups included ‘redox-related proteins’ and ‘defense-related proteins’ each representing 6 % of the total number of modified proteins, respectively. Approx. 60 % of identified S-nitrosylation candidates have not been presented so far as potential members of the S-nitrosoproteome in plants and only 17 of them were reported earlier as S-nitrosylation targets in potato leaves and tubers supplied with GSNO (Kato et al. 2013). Since S-nitrosylation is under strong control of redox status of the cell, antioxidative enzymatic and non-enzymatic changes have been recorded upon P. infestans inoculation.

Fig. 2
figure 2

The effect of P. infestans inoculation on potato leaves cv. ‘Bzura’: a activity of GSNO reductase (GSNOR), b total content of S-nitrosothiols (SNOs) determined by chemiluminescence assay using Sievers® Nitric Oxide Analyzer NOA 280i, * significantly different from mock-inoculated potato leaves, P < 0.05. Values represent the average of data ± SD of three independent experiments

Full size image

Redox balance in inoculated leaves

Pathogens could trigger an imbalance in redox homeostasis in host-plant cells to maximize the opportunity of colonization. Therefore, selected non-enzymatic and enzymatic components of the redox maintenance apparatus were analyzed in potato leaves. In the incompatible interaction with P. infestans the total content of free sulfhydryl groups (-SH) (Fig. 3a) and ascorbate (Fig. 3b) increased notably at the early phase (1-3 hpi). It reached an approximately 2-fold and 3-fold higher concentration in relation to mock-inoculated and vr P. infestans-treated leaves, respectively. The early shift into the reductive potential was also confirmed by the expanded total antioxidative capacity referred to both the slow (Fig. 3c) and the fast (Fig. 3d) antioxidants. Thus, non-enzymatic components of the redox apparatus had a significant impact on the redox state re-adjustment towards the reductive potential in potato after ROS and RNS generation (Fig. 1a, b, c).

Fig. 3
figure 3

The effect of P. infestans inoculation on redox state re-adjustments in potato leaves cv. ‘Bzura’: a total sulfhydryl (–SH) group content, b ascorbate concentration, c slow and d fast antioxidants referred to as total antioxidant capacity, * significantly different from mock-inoculated potato leaves, P < 0.05. Values represent the average of data ± SD of three independent experiments

Full size image

Pathogenesis-related protein up-regulation and disease progress

Late blight disease spots developed mainly after colonization with the virulent race of P. infestans. Severe foliar damage was observed in more than 75 % of leaves at 7 dpi (Fig. 4d) and the oomycete mycelium covered more than 20 % of leaf discs at 5 dpi (Fig. 5 Appendix). In avr P. infestans-treated leaves disease spots were smaller than 3 % of total area. The reduced disease symptoms corresponded to the enhanced defense responses in avr P. infestans-inoculated potato. Semi-quantitative real time RT-PCR revealed that PR-1 gene expression was rapidly activated upon inoculation with avr P. infestans at the early phase (10-fold increase) and the mRNA level remained multiplied till 24 hpi (Fig. 4a). In the compatible interaction the robust PR-1 gene transcript accumulation was recorded solely at the late phase at 24 hpi. However, the PR-1 mRNA level raised gradually since 3 hpi. The indication of PR-1 gene was accompanied by an effectively up-regulated β-1,3-glucanase activity (PR-2) (Fig. 4b). A less intense activation of PR-2 was found in the compatible interaction after pathogen ingress. In turn, chitinase (PR-3) activity was induced with a distinct magnitude at 3 hpi in both interactions (Fig. 4c), showing slight contributions to the effective defense against oomycete pathogens. The comparison of spatiotemporal PR-1 and PR-2 analyses revealed stronger induction of the aforementioned PRs in response to the avirulent rather than the virulent race of P. infestans.

Fig. 4
figure 4

The effect of P. infestans inoculation on potato leaves cv. ‘Bzura’: a time-course of PR-1 gene expression upon inoculation using real-time RT-PCR semi-quantitative analysis, b relative activity of PR-2 protein (β-1,3-glucanase), c relative activity of PR-3 protein (chitinase), d index of disease development at 7 dpi, * significantly different from mock-inoculated potato leaves, P < 0.05. Values represent the average of data ± SD of at least three independent experiments

Full size image

Discussion

In order to gain further insight into the NO-mediated multilevel interaction leading to potato immune response, two P. infestans races were used to induce the incompatible and compatible interaction, respectively. The activated defense signaling network was potentially NO-initiated and ranged from the S-nitrosylation of proteins to the execution of PR proteins storage. To date, NO functions in plants are far from being fully described. However, much progress has been achieved in the understanding of NO pleiotropic effects in plant system, considering the last two decades of intensive research. Since NO biosynthesis remains unknown, a myriad of its down-stream effects was found to be affected by the cellular redox state. Therefore, the maintenance of the redox state in inoculated tissue appeared to have a pivotal role in the immune response establishment in potato.

P. infestans triggered NO burst in potato

Since NO burst is one of the first events observed after pathogen ingress, it is crucial to execute NO-dependent message to immune response in plants. At first, in the presented study we showed that potato leaves have had generated NO following inoculation with the avirulent race of P. infestans. In turn, the virulent race failed to initiate an early NO production. Primary NO over-production was shown to be specific for the incompatibility of potato and the attacker. Many reports showed that NO action is dependent on its concentration and spatial production patterns (Besson-Bard et al. 2008; Kovacs and Lindermayr 2013). Others represented the standpoint that an early spatiotemporal NO burst is restricted to the incompatible interaction (Bennett et al. 2005; Mur et al. 2005). On this basis, we attempt to record NO-driven down-stream effects in both interactions. The attention was given to peroxynitrite and SNOs. A rapid formation of ONOO− as a result of the equimolar reaction of NO and O2 •- in potato leaves treated with the avirulent race of pathogen indirectly indicated the activation of signaling towards resistance (Arasimowicz-Jelonek and Floryszak-Wieczorek 2011). However, Delledonne et al. (2001) documented that ONOO− is not a direct executor of HR in plants unlike in the animal system. It is worth noting that our previous findings revealed that in potato-P. infestans system oxidative burst has occurred since first hpi in both interactions. The difference between these responses has been observed afterwards, between 6 and 10 hpi (Floryszak-Wieczorek et al. 2011). The superoxide radical as the main component of oxidative burst may be produced in a controlled manner by plasma membrane NADPH oxidase (Noirot et al. 2014). In the presented study total NADPH oxidase activity was shown to be up-regulated since 1 hpi in avr P. infestans-treated leaves, what comprehensively correlated with the boosted O2 •- accumulation. In turn, an increase in NADPH oxidase activity was weaker and occurred later in vr P. infestans-inoculated leaves. Additionally, it needs to be emphasized that H2O2 accumulation governed by SOD activity increased greatly only in the incompatible interaction.

Obtained data are in line with previous findings that HR in soybean cells was activated after NO cooperation with H2O2 generated from O2 •- by superoxide dismutase (De Stefano et al. 2006). These results revealed that in tobacco NO and H2O2 can act independently or in synergy and can share common components in regulating gene expression during defense responses in and outside the cell death program.

Collectively, these data highlighted the involvement of NO burst in the induction of redox changes in avr P. infestans-treated leaves. Further, the up-regulation of NADPH oxidase and the controlled O2 •- production were tuned with SOD-mediated dismutation of O2 •- to H2O2 triggering, in combination with NO, the cell death in potato inoculated leaves.

Redox re-adjustments determined resistance

Since an efficient immune response of potato leaves to P. infestans necessitates rapid changes in reactive nitrogen and oxygen species, an imbalance in redox homeostasis may occur as a result of these changes. To counteract oxidative stress, a spatiotemporally synchronized accumulation of non-enzymatic and enzymatic antioxidants must be employed (Groß et al. 2013). The cytoprotective effect of NO was reported to be linked to the NO-dependent regulation of the redox state and the controlling of ROS generation (Arasimowicz-Jelonek et al. 2014a, b). During the incompatible response of potato leaves an enhanced accumulation of ascorbate and total thiol content was observed early after pathogen recognition (1-6 hpi). This effect was triggered only in avr P. infestans-inoculated leaves, indicating fast and precisely controlled redox state re-adjustment. Markedly increased total antioxidative capacity recorded after inoculation with avr P. infestans was a symptom of the shifted redox potential to more reductive conditions. Interestingly, Arabidopsis pad2-1 mutants, with an insufficient GSH supply, were unable to activate defense against the biotrophic Phytophthora brassicae or bacterial pathogen Pseudomonas syringae (Dubreuil-Maurizi et al. 2011). In turn, 6-phosphogluconolactonase 3 knock-down mutants of Arabidopsis (pgl3) with an enhanced PRs gene expression and a decreased redox potential possessed the constitutive activation of immune responses against P. syringae pv. maculicola and Hyaloperonospora arabidopsidis Noco2 (Xiong et al. 2009).

Summarizing, the ability of potato plants to control the redox state precisely under the pathogen pressure may modify the level of susceptibility/resistance to P. infestans. The implementation of redox re-adjustment processes was highly relevant to the type of interplay between potato and an oomycete pathogen. In fact, the antioxidative counteractions to redox imbalance impinged on the efficiency of defense expression in potato.

Nitric oxide-mediated S-nitrosylation

Considering NO bioactivity in plants, emphasis must be given to the storage of NO-coded message into the total SNO pool and S-nitrosylation of targeted proteins. The presence of NO-sensitive reactive Cys residues in low- and high-molecular mass molecules may serve as the molecular switch of NO functions in plants (Spadaro et al. 2010). Since particular Cys residues are sensitive to both the oxidative status of the cell and NO generation, they may combine two pathways that orchestrate early signaling, i.e. NO- and ROS-dependent (Yu et al. 2014). An augmented NO production in potato inoculated with the avirulent pathogen was tuned in time with a significant rise in the SNO pool (from 1 hpi). By contrast, no significant changes were noted in the compatible response. In proteins Cys may be a direct target of NO itself albeit NO may act indirectly via trans-nitrosylation reactions (Yu et al. 2014). Overall, the covalent attachment of NO to –SH groups of cysteine residues in protein and non-protein thiols may serve as a reversible redox-based modification, responsible for the execution of the NO-converted signal to a physiological response in plants (Spadaro et al. 2010). S-nitrosylation may modulate the activity of enzymes, the localization of proteins and the likelihood of interactions with other high-molecular mass molecules. Emerging evidence from particular studies indicate that an elevated SNO formation promoted plant resistance to a pathogen (Rustérucci et al. 2007). However, the accumulation of SNOs above a certain threshold may affect the SA-signaling pathway by a negative feedback loop, promoting NPR1 oligomerization in the cytoplasm and the suppression of defense (Malik et al. 2011). Therefore, SNO turnover controlled by an array of processes i.e. trans-nitrosylation reactions and enzymatic decomposition may be critical in the defense execution. Often, the total SNO pool reflects the change of GSNO level – a mobile reservoir of NO bioactivity in plants. Discrepancies in GSNO content found in diverse pathosystem and GSNO localization must be kept in mind and may be a result of GSNOR distribution and different activity in plant cell compartments, tissues and organs (Corpas et al. 2013). For example, the tendency towards GSNO allocation in epidermal cells of sunflower plants was observed after inoculation with avr Plasmopara halstedii (Chaki et al. 2009). Emphasizing the role of denitrosylation processes, it must be stated that GSNOR is mainly involved in the breakdown of GSNO and currently limited information concerns other cellular specific SNO reductases or SNO lyases (Malik et al. 2011). Nevertheless, the excessive SNOs promoted the susceptibility of plants (Feechan et al. 2005). An emerging aspect of SNOs turn-over is the existence of specific forms of de-nitrosylation and the identification of cellular machinery to fine-tune SNOs pool.

It is generally accepted that changes in SNO contents, mediated by GSNO reductase, together with ROS governed by NADPH oxidase facilitated the immune promoting activity (Yun et al. 2011). However, in the presented study the postinfection GSNOR activity did not interfere entirely with the kinetics of SNO formation. The importance of GSNO reductase as a potato immune response controller was also highlighted in our previous reports concerning systemic acquired resistance and cross-tolerance of potato (Floryszak-Wieczorek et al. 2012; Janus et al. 2013; Arasimowicz-Jelonek et al. 2014a, b). As noted before, NO exported bioactivity was exhibited as an enhanced SNO storage at a relatively low threshold, governed temporarily by GSNOR activity.

One of the basic requirements of signal generation and transduction must be its transient character, enabling the reverse of signaling when it is required. Therefore, Cys residues may serve as an ideal substrate for targeted signaling since their NO-based modifications are mostly reversible and, as mentioned previously, exclusively redox-dependent (Spadaro et al. 2010).

Nitric oxide-mediated SNO formation influences rearrangements in the S-nitrosoproteome structure in physiological timelines and stress conditions (Fares et al. 2014). By performing the biotin-switch technique coupled with LC-MS/MS analysis, 104 proteins typed for S-nitrosylation were found in potato leaves at a decisive time-point after pathogen ingress (3 hpi). The presented list of potato S-nitrosylation targets (Table 1 Appendix) contains only in vivo S-nitrosylated proteins identified in mock and P. infestans challenge inoculated potato leaves. So far, a limited repertoire of functionally described S-nitrosylated proteins was uncovered in plants. Proteins subjected to S-nitrosylation presented in this study orchestrate an ample scope of cellular processes, including primary and secondary metabolism, redox maintenance and defense-related changes. Since available methods of SNOs identification in proteins are imperfect, we presented the supplemental list of potential S-nitrosylation targets in potato. Many among identified proteins are abundant in plant system and have been indicated as NO targets in different plant species. In the presented study, both CA and StSABP2 isoforms were found to be S-nitrosylation targets in potato challenged with P. infestans. It is worth noting that the attachment of NO to the cysteine residue of carbonic anhydrase was earlier recorded in GSNO-treated potato (Kato et al. 2013). Additionally, the role of CA in plant immunity is possibly linked to the lipid-base signaling and JA-responsive gene down-regulation (Hoang and Chapman 2002). These changes are believed to be profound in the relevant disease resistance development in plants. As documented before, the transfer of NO moiety to a specific Cys280 of AtSABP3 led to the inhibition of its carbonic anhydrase activity as well as blocked the ability of the protein to bind SA in an NO dose-dependent manner (Wang et al. 2008b). Both functions of AtSABP3 are essential in the establishment of Arabidopsis defense against virulent bacterial pathogens PstDC3000, so that their inhibition by S-nitrosylation potentially contributed to the negative feedback loop, modulating SA-dependent plant immune responses. We would like to indicate that in potato exposed to P. infestans, thioredoxin M4 and 2-Cys peroxiredoxin – important players in the thioredoxin/thioredoxin reductase system, also undergo S-nitrosylation (Lindermayr et al. 2005; Romero-Puertas et al. 2007; Kato et al. 2013). In turn, superoxide dismutase as the following redox-related enzyme reported herein as S-nitrosylation target confirmed the findings of Kato et al. (2013). SOD activity has been believed to raise upon S-nitrosylation (Sehrawat et al. 2013) and that could potentially explain the up-regulation of SOD in the incompatible interaction of potato and P. infestans. So far, S-nitrosylation of SOD was presented in Arabidopsis (Lindermayr et al. 2005), Kalanchoe pinnata (Abat et al. 2008), citrus plants (Tanou et al. 2010) and potato (Kato et al. 2013). As well, the functional validation of SOD S-nitrosylation was performed only in Brassica juncea (Sehrawat et al. 2013). Intriguingly, recently SOD has been reported to undergo inhibition by nitration of Tyr63 (Holzmeister et al. 2015).

From 104 presented S-nitrosylation candidates in the potato-P. infestans system, 17 were identified by us as nitration targets (unpublished data). The above mentioned list included e.g, two isoforms of carbonic anhydrase, chitinase, subtilisin-like protease-like, 2-Cys peroxiredoxin A, Rubisco small and large subunits, aldolase and chloroplast manganese stabilizing protein. This finding underpins the importance of dual protein regulation by Cys S-nitrosylation and Tyr residue nitration. It may serve as a basis of NO-based re-programming of the cell physiological status.

Hallmarks of potato defense

In our experimental approach, treatment of potato with avr P. infestans has activated PR-1 and PR-2, the key indicators of plant defense against pathogens at the level of transcript accumulation and protein activity up-regulation, respectively. These findings are in line with previous reports concerning the compatible and incompatible interaction of plants and hemibiotrophic pathogens (Vleeshouwers et al. 2000; Wang et al. 2008a). Early up-regulation of the PR-1 gene and β-1,3-glucanase were restricted only to the incompatible interaction. Activation of the PR-1 gene depends on the signaling cascade often initiated by the NO burst and is orchestrated by salicylic acid bioactivity and redox state re-adjustments (Tada et al. 2008). As it was reported previously PR-2 activation at the transcript accumulation and protein activity level is likewise a part of potato defense against P. infestans (Arasimowicz-Jelonek et al. 2014a, b). However, the precise mode of action accompanying NO-mediated PR gene expression in potato is far from being resolved and needs further research. Nevertheless, PRs became fully operational at the early phase after inoculation. The delayed timing of PR-1 expression (starting at 24 hpi) indicated insufficient defense activation against the virulent race of the pathogen. Wang et al. (2008a) demonstrated that a less virulent US-1 genotype of P. infestans caused the accumulation of the PR-1gene transcript earlier, from 8 hpi, in relation to the new more aggressive US-8 linage. Aggressive race induced the late gene up-regulation (at 48 hpi). In turn, only a slight increase of PR-3 activity in potato leaves was observed at 3 hpi, suggesting a less evident role of this enzyme in response to pathogens lacking chitin in cell walls. Consequently, in the light of presented findings, it may be concluded that mainly PR-1 and PR-2 effectively participated in the limiting of late blight disease.

Concluding remarks

The cross-talk of ROS- and RNS-based signaling is largely linked with redox-dependent modification on targeted cysteines in proteins, which may be considered as redox-switches of immune responses in potato leaves. So far, many protein targets were found to be subjected to S-nitrosylation, but still the precise mechanism is not fully understood and needs further research. Moreover, special emphasis must be given to perform a functional analysis of S-nitrosylated proteins engaged in non-model plant resistance, i.e. potato challenged by a pathogen. In consequence, the apprehension of NO fate is essential to warrant full understanding of NO influence on defense-orientated signaling in plants.

It is also true that currently we have no information as to whether and to what extent NO mediates P. infestans pathogenicity. Generally it is known that P. infestans is thought to accomplish colonization by molecular re-programing of the host defense strategy specifically by introducing an array of effectors (Vleeshouwers and Oliver 2014). According to Arasimowicz-Jelonek and Floryszak-Wieczorek (2014), the pathogen metabolic equipment to reset the NO signal and counteract nitrosative stress not only plays a role as a modulator of the host immune response, but might also be implicated in virulence. It is time to enrich our knowledge on P. infestans NO sensing and signaling, which may complete our understanding of pathogen-triggered re-programing of plant metabolism, moving elementarily the host-organism towards resistance or susceptibility.