Mechano-stimulated modifications in the chloroplast antioxidant system and proteome changes are associated with cold response in wheat

Background Mechanical wounding can cause morphological and developmental changes in plants, which may affect the responses to abiotic stresses. However, the mechano-stimulation triggered regulation network remains elusive. Here, the mechano-stimulation was applied at two different times during the growth period of wheat before exposing the plants to cold stress (5.6 °C lower temperature than the ambient temperature, viz., 5.0 °C) at the jointing stage. Results Results showed that mechano-stimulation at the Zadoks growth stage 26 activated the antioxidant system, and substantially, maintained the homeostasis of reactive oxygen species. In turn, the stimulation improved the electron transport and photosynthetic rate of wheat plants exposed to cold stress at the jointing stage. Proteomic and transcriptional analyses revealed that the oxidative stress defense, ATP synthesis, and photosynthesis-related proteins and genes were similarly modulated by mechano-stimulation and the cold stress. Conclusions It was concluded that mechano-stimulated modifications of the chloroplast antioxidant system and proteome changes are related to cold tolerance in wheat. The findings might provide deeper insights into roles of reactive oxygen species in mechano-stimulated cold tolerance of photosynthetic apparatus, and be helpful to explore novel approaches to mitigate the impacts of low temperature occurring at critical developmental stages. Electronic supplementary material The online version of this article (doi:10.1186/s12870-015-0610-6) contains supplementary material, which is available to authorized users.


Background
Chilling temperature significantly affects the early growth of winter wheat plants causing considerable reduction of grain yield and is one of the major factors limiting growth and productivity of crops [1]. Cold induced photosynthesis inhibition results in a complex array of reactive oxygen species (ROS) generation, especially in chloroplasts [2]. Over-accumulation of ROS may cause rigidification and leakage of the cell membrane, and destabilization of protein complexes [1]. Recent proteomic studies have revealed differential expression of proteins in wheat exposed to cold stress [3,4]. Among the down-regulated proteins due to cold stress, some key enzymes involved in Krebs cycle (isocitrate dehydrogenase, malate dehydrogenase) have been identified, together with many photosynthesisrelated proteins (e.g. oxygen-evolving complex proteins, ATP synthase subunits, ferredoxin NADPH oxidoreductase, and some Calvin cycle enzymes) [3]. Proteomic analysis of spring freezing stress responsive proteins in leaves revealed an increased accumulation of stress defense proteins, including LEA-related COR protein, Cu/Zn superoxide dismutase, and ascorbate peroxidases, which may play crucial roles in enhancing tolerance to spring freeze stress in bread wheat [4]. In addition, proteomic analysis of wheat in response to prolonged cold stress showed reinforcement in expressions of enzymes involving in ascorbate recycling (dehydroascorbate reductase, ascorbate peroxidase) and involving in tetrapyrrole resynthesis (glutamate semialdehyde aminomutase) [3].
Mechanical wounding can be caused by surrounding environmental factors, such as wind, rainstorms, and herbivores, and it has broad impacts on plants, including changes in morphogenetic characteristics, membrane potential [5], ROS, hormone signaling and gene expression [6]. Several alterations induced by mechanical wounding can allow plants to resist and acclimate to environmental stresses [6]. As previously observed in maize, bean, and rice, denser but smaller stomata in mechanically stimulated leaves could help plants to control transpirational water loss, thereby avoiding drought stress [7]. Mechano-stimulation was reported to increase cold tolerance in beans, tomato, and maize through maintenance of higher Photosystem II (PSII) efficiency and accumulation of higher levels of soluble sugars [8]. It was also suggested that similar defense mechanisms are operated in cold acclimation and mechanostimulation, resulting in similar morphological and developmental changes [8]. Recently, analysis of transcript profiles indicated various defense response genes were induced by mechano-stimulation, and were related to cold stress response, including general stress response (GSR), rapid stress response (RSR), and rapid wound response (RWR) [9]. In addition, it has been proposed that mechanical disruption of the cell wall may induce stress signaling [10]. Cold stress is perceived by the plant through detection of changes in membrane fluidity and protein conformation. Secondary messengers such as Ca 2+ and ROS are implicated in the initial signaling cascades in response to cold stress [1].
Many studies reported changes in ROS levels following mechano-stimulation [11,12]. For instance, mechanostimulation induced a significant increase in ROS levels in Mesembryanthemum crystallinum leaves [12]. Furthermore, proteomic studies have shown that plants transiently produce superoxide and H 2 O 2 , which might play critical roles in signal transduction during early wound response [13]. Mechano-stimulation induced increased expression of cytosolic H 2 O 2 -detoxifying enzyme, ascorbate peroxidase 2 (APX2) [14]. This increase in APX2 was independent of other mechanical wounding signals such as jasmonic acid (JA) or abscisic acid (ABA) [15]. It has also been suggested that NADPH-dependent H 2 O 2 signals contribute to the activation of specific mechano-stimulated signals which are not activated by the JA or ABA [15]. The cellular steady-state level of ROS is tightly regulated by a complex network involving Ca 2+ , protein phosphorylation, and ROS-scavenging/producing enzymes during wound response [15]. In addition, mechanical wounding has been found to induce a burst of superoxide and apoplastic peroxidase with both oxidative and peroxidative activities [15,16].
In this study, mechano-stimulation was applied to two contrasting winter wheat cultivars that differed in cold tolerance at different growth stages in order to investigate the effects of mechano-stimulation on the performance of the chloroplastic antioxidant system and changes of the proteome under late spring low temperature stress. The results obtained in this study may provide deeper insights into the roles of mechano-stimulated modifications within chloroplast antioxidant systems and proteome in cold tolerance in wheat. This information will be helpful for exploring novel approaches to mitigate the impacts of low temperatures which occur during critical developmental stages in wheat plants.

Plant materials
This experiment was carried out at Lianyungang Experimental Station of Nanjing Agricultural University (119°3 2′E, 34°30′N) during the wheat growing season in 2011-2012. The soil is a clay, contains 11.4 g kg −1 organic matter, 1.1 g kg −1 total N, 79.8 mg kg −1 available N, 32.4 mg kg −1 Olsen-P, and 132.4 mg kg −1 available K. Before sowing, 120 kg N ha −1 , 60 kg P 2 O 5 ha −1 and 120 kg K 2 O ha −1 were applied as basal fertilizer and a further 120 kg N ha −1 was used as a topdressing after jointing to avoid the potential impacts on stress treatments. Two winter wheat cultivars differing in cold tolerance but having close genetic backgrounds (Jimai 17 displays similar morphology and is related with Yannong 19 in pedigree), Yannong19 (YN19, cold tolerant) and Lianmai6 (LM6, cold susceptible, parents: YN19// Jimai 17/Zheng9023) were used in this experiment. The sowing date was 14 October 2011, with a seedling density of 160 m −2 and a row space of 0.25 m. The jointing date was confirmed through spike development checked with a Dino-Lite digital microscope (AM411 Version 1.4.1; Vidy Precision Equipment Co. Ltd, Wuxi, China).

Mechano-stimulation and cold treatments
To investigate the effects of mechano-stimulation applied at different stages on seedling performance under cold stress, four treatments were imposed: P 1 L, the early priming of mechano-stimulation for plants was applied at the Zadoks growth stage 26 (25 March 2012) and then subjected to a 4-day cold event at the Zadoks growth 31 (jointing stage, 8-12 April 2012); P 2 L, the later mechano-stimulation for plants was carried out 6 days before the cold event (2 April 2012); CL, the cold stress at jointing without early mechano-stimulation; CC, the normal temperature control. Mechano-stimulation was carried out using a cylinder roller with weight of 150 kg and diameter of 40 cm. The roller was rolled over the wheat plants with a pressure of 7000 N · m −2 at 9:00-9:30 am, which resulted in less than 20 % of the leaf area being damaged at jointing A 4-day cold stress was applied using four temperature control systems operated in the open top chamber condition. Air was cooled by a compressor, and then the cooled air was driven by an air blower to the field through ducting [16]. During cold treatment, plots were surrounded by 180-cm-high plastic film. All tubes were removed just after cooling treatment to avoid shading. Six temperature and humidity sensors were used to record the real-time data in each plot. The mean temperature in the cold treatment was 5.60°C lower than the normal temperature control. The mean temperature at night was −1.14°C, and the lowest temperature recorded during the cold treatment was −4.97°C (detailed temperature data are shown in Additional file 1: Figure S1). The experiment had a split-plot design with temperature treatment as the main plot and wheat cultivar as the subplot, with three replicates for each treatment. The size of each plot was 3 m × 4 m.

Chl a fluorescence transient
The fast chlorophyll a fluorescence induction curve was measured using a Plant Efficiency Analyzer (Pocket-PEA; Hansatech, Norfolk, UK) [17]. Before measuring, plants were dark adapted for 0.5 h. The collected data were processed by the program PEA Plus 1.04, and Biolyzer 3.0 software (Bioenergetics Lab., Geneva, Switzerland, http://www.fluoromatics.com/biolyzer_software-1.php) was used to calculate the fast chlorophyll a fluorescence induction (OJIP) test parameters.
Following the methods of Zheng et al. [18], H 2 O 2 concentration was measured by monitoring the absorbance of titanium peroxide complex at 410 nm, and the release rate of O 2 − was determined at an absorbance at 530 nm. APX (EC 1.11.1.11) activity was determined by monitoring the decrease in absorbance at 290 nm, the activity of SOD (EC 1.15.1.1) was measured by monitoring the inhibition of photochemical reduction of nitroblue tetrazolium (NBT), and GPX (EC 1.11.1.7) activity was calculated by monitoring the increase in absorbance at 470 nm due to the oxidation of guaiacol. GR (EC 1.6.4.2) activity was determined by the oxidation of NADPH at 340 nm, and CAT (EC 1.11.1.6) activity was measured following the method of Tan et al. [19]. DHAR (EC 1.8.5.1) was assayed by monitoring changes in absorbance at 265 nm after the addition of ascorbate oxidase as described by Miyake and Asada [20]. The activities of Ca 2+ -ATPase and Mg 2+ -ATPase in the chloroplasts suspension were measured following the method of Zheng et al. [18].

Rubisco activity
Leaf samples (0.2 g) were ground in 40 ml of extraction buffer (50 mM Tris-HCl, 1 mM EDTA, 1 mM MgCl 2 , 10 % PVP and 10 mM β-mercaptoethanol), and then centrifuged at 15 000 × g for 15 min. The supernatant was gently collected to measure Rubisco activity. The activity of Rubisco (EC 4.1.1.39) before (initial activity) and after (total activity) active site carbamylation was assayed using a spectrophotometric procedure coupled to NADH oxidation [21]. Rubisco activation was estimated as the percentage ratio of initial to total activities for each sample.

Protein extraction and 2-DE procedure
The extraction of protein in the latest fully expanded leave for 2 DE was performed following the trichloroacetic acid (TCA) acetone precipitation method described by Ding et al. [22].

Image analysis, protein identification, and functional annotation
The gels were scanned using a VersaDoc4000 image system (Bio-Rad) and the images were analysed with PDQUEST 8.0 software (Bio-Rad, USA). There were three biological replicates per treatment with at least three gels for each biological replicate. Only spots with a variation rate of ±0.5 in the three replicates were considered for further analysis. Stained protein spots were excised manually from the gels, in-gel digested with trypsin, and analysed using a MALDI-TOF/TOF mass spectrometer (ABI 4800). The MASCOT database search engine (http://matrixscience.com) was used to search for peptide mass lists from the obtained spectra against the NCBI database. The mass error tolerance was set to 80 ppm, and the score threshold was above or equal to 110.

Statistical analysis
All data were subjected to the two-way ANOVA using the SigmaSATA (Systat Software Inc., CA, USA). The Duncan's multiple range test was used to check the significance of difference between treatments. In 2-DE analysis, the difference of expression level at the given protein spots between treatments and the control (CC) for each cultivar was calculated and converted to a color scale by PageMan software (http://mapman.mpimpgolm.mpg.de/pageman/).

Chl a fluorescence transient
The increase in leaf fluorescence transients observed in CC treatment showed a typical OJIP shape in YN19 and LM6 (Fig. 1a, c). However, P 1 L (early mechano-stimulation + cold stress), P 2 L, (later mechano-stimulation + cold stress) and CL (non-mechano-stimulation + cold stress) showed repressed fluorescence transients in these two cultivars, particularly at step I (30 ms) and P. The main changes of fluorescence data were normalized between step I (30 ms) and P (300 ms) and presented as relative variable fluorescence W IP (Fig. 1b, d). Obvious changes in W IP during the fast rise period were observed under P 2 L and CL in YN19, while under P 1 L, P 2 L, and CL in LM6, compared with CC. W OP (Fig. 1e, g) and W OI (Fig. 1f, h) showed relatively variable fluorescence from O to step P (300 ms) and from O to I (30 ms). A significant decrease in W OP at step I was found in P 1 L, P 2 L, and CL in YN19, while W OP was increased remarkably by P 2 L at step I in LM6. Significant changes in W OI were found among P 1 L, P 2 L, and CL treatments in YN19, which were related to the reductions between PSI and reduced NADP + . However, with the exception of the P 2 L treatment, W OI was only slightly affected in LM6.

Rubisco activities and activation
Initial and total Rubisco activities and Rubisco activation in the latest fully expanded leaves were significantly decreased with CL, compared with CC in YN19 and LM6 (Fig. 2, P < 0.001). Both traits were slightly and marginally significantly increased by P 1 L (P = 0.077), whereas they were depressed by P 2 L compared with CL (P < 0.001). Rubisco activation in P 1 L was relatively higher than in CL, but was still lower than in CC for both cultivars (P < 0.001). In addition, no difference in Rubisco activation was found between P 2 L and CL.

ROS production, activities of antioxidant enzymes, and expressions of their encoding genes in chloroplasts
In YN19, CL increased the concentration of H 2 O 2 in chloroplasts in the latest fully expanded leaves by 63 % as compared with CC, whereas P 2 L showed a 34 % increment compared with CL (Fig. 3a). However, no difference was observed between CC and P 1 L. A similar pattern was observed in LM6. The highest rate of O 2 − release was found in P 2 L, followed by CL and P 1 L, whereas that in CC was lowest in both cultivars (Fig. 3b, P < 0.001). SOD activity in chloroplasts was increased by 16 % and 25 % with P 1 L in YN19 and LM6, respectively. Compared with CC, P 1 L increased while P 2 L and CL significantly decreased chloroplastic SOD activity in the two cultivars (Fig. 3c). In addition, in both cultivars, expression of Cu/Zn SOD was up-regulated by P 1 L compared with CL (Fig. 4a), whereas an up-regulation of Fe SOD due to P 1 L was only observed in YN19 (Fig. 4b). For both cultivars, CAT activity was lower in CL than in CC, whereas it was higher in P 1 L than in CL (Fig. 3d). In LM6, P 2 L decreased CAT activity by 20 % compared with CL, whereas no significant difference was found in YN19 (P = 0.112). An increase in the expression of CAT was found in P 1 L and P 2 L compared with CL in YN19; however, the difference was not statistically significant (Fig. 4c). In both cultivars, the combination of mechano-stimulation and low temperature (P 1 L and P 2 L) and CL enhanced APX activity compared with CC. In particular APX activity in P 2 L was significantly higher than in CL (Fig. 3e). Further, the same trend was found in thylakoid-bound APX (tAPX) expressions in YN19, whereas in LM6, a significant up-regulation of tAPX was only observed in P 1 L (Fig. 4d). In both cultivars, GPX activity was enhanced with P 1 L, but depressed with P 2 L and CL (no significant difference between P 2 L and CL) (Fig. 3f). In both cultivars, P 1 L and CL resulted in a significant increase in GPX activity, compared with CC (P < 0.001), whereas P 2 L slightly increased GPX activity (Fig. 3g, P = 0.105). Low temperature significantly enhanced DHAR activity in both cultivars (P < 0.001); however, P 1 L and P 2 L had opposite effects on DHAR activity in the two cultivars-namely, P 2 L decreased DHAR activity compared to CL in YN19, but increased activity in LM6; P 1 L increased DHAR activity in LM6, whereas no difference between P 1 L and CL was found in YN19 (Fig. 3h). Thus, P 1 L and P 2 L showed opposite patterns in the concentration of H 2 O 2 in chloroplasts, O 2 − release rate and most of the antioxidant enzyme activities, but the APX activity showed a similar trend in P 1 L and P 2 L.

ATPase activities in chloroplasts
In YN19, the Activities of both Mg 2+ -ATPase and Ca 2+ -ATPase were significantly decreased by CL as compared with CC ( Fig. 5, P < 0.001). However, both ATPase activities were increased by P 1 L, whereas they decreased by P 2 L. The activities of Mg 2+ -ATPase and Ca 2+ -ATPase in LM6 in response to different treatments were similar to those in YN19.

Proteomics
The reference 2-DE gel of proteins in wheat leaves affected by combination of mechano-stimulation and cold Fig. 2 Effects of mechano-stimulation on initial and total Rubisco activities and activation in the latest fully expanded leaves in winter wheat exposed to cold stress at jointing Fig. 3 Effects of mechano-stimulation on reactive oxygen species and antioxidant enzyme system in chloroplasts in the latest fully expanded leaves in winter wheat exposed to cold stress at jointing. a H 2 O 2 , hydrogen peroxide; b O 2 − , superoxide anion radical; c SOD, superoxide dismutase; d CAT, catalase; e APX, ascorbate peroxidase; f GPX, glutathione peroxidase; g GR, glutathione reductase; h DHAR, monodehydroascorbate reductase stress is shown in Fig. 6. More than 600 protein spots were detected in each gel. To demonstrate the proteomic response of the photosynthetic apparatus to mechano-stimulation and cold stress, variation in the expression of 12 protein spots related to photosynthesis, energy production, stress defense in chloroplasts is specifically shown in Fig. 7. The differentially expressed protein spots were identified by mass spectrometry (MS, Table 1). In the cluster related to photosynthesis, five protein spots, including enzymes involved in the Calvin cycle and Rubisco protein subunit-ferredoxin-NADP(H) oxidoreductase (spot 10), ribulose-1, 5-bisphosphate carboxylase activase (spot 11) and the Rubisco large subunitbinding protein subunit alpha (6)-were up-regulated by P 1 L in both cultivars; the exception being spot 10, which was missing in P 1 L in YN19. CL induced up-regulation in chloroplastic glutathione reductase (spot 5) and ascorbate peroxidase (spot 7) in both cultivars, whereas the expression of catalase-1 (spot 8) was down-regulated by CL. These proteins were, however, all up-regulated by P 1 L in both cultivars, except for catalase-1 in YN19. In addition, in both cultivars, the expression of ATP synthase β subunit (spot 9) was depressed by CL compared with CC, but was increased by P 1 L compared with CL. Interestingly, the chloroplastic fructose-bisphosphate aldolase (spot 2) was up-regulated by CL in the two cultivars, whereas under P 1 L, it was decreased in YN19 but increased in LM6. Proteomic analyses revealed that the oxidative stress defense, ATP synthesis, and photosynthesis-related proteins were similarly modulated by the mechanostimulation and the cold stress.

Discussion
It is well known that ROS production is a universal response to mechanical wounding in various plants [11]. The defense system can also be activated to alleviate ROS-induced oxidative stress and repair the damaged tissues [25]. Furthermore, many of the genes encoding enzymes involved in ROS metabolism are regulated by mechanical wounding [26]. Here, the leaf chloroplastic H 2 O 2 concentration in P 1 L plants was very close to that in CC plants. In contrast, P 2 L plants have a significantly higher H 2 O 2 concentration than CC plants (Fig. 3). This difference is related to the efficient ROS scavenging capacity of the antioxidant enzyme systems, particularly the water-water cycle in chloroplasts, which mainly includes SOD and APX [27]. The scavenging capacity of SOD and CAT activated by the mechano-stimulation in P 1 L Fig. 4 Effects of mechano-stimulation on relative transcript abundance of Cu/Zn SOD (a), Fe SOD (b), CAT (c) and thylakoid-bound APX (tAPX, d) in the latest fully expanded leaves in winter wheat exposed to cold stress at jointing had a significant inhibitory effect on the oxidative burst under low temperature stress. Further analysis revealed that the enhanced activities of SOD and CAT could be largely explained by the up-regulated expression of Cu/ Zn SOD, Fe SOD, and CAT in P 1 L (Fig. 7). Woundinduced activation of H 2 O 2 -detoxifying enzymes has previously been demonstrated using proteomic tools [26]. Our proteome analysis showed that the expression levels of ascorbate peroxidase and catalase-1 were paralleled by the activities of APX and CAT, respectively, under different treatments (Fig. 7). However, in YN19, the activity of APX in P 1 L was decreased compared to CL, whereas no significant difference was found in LM6. The qPCR analysis also showed that the tAPX expression was only slightly affected by P1L in YN19, but it was increased by P 1 L in LM6. APX activity under the combination of mechano-stimulation and cold stress was only partly consistent with that previously reported. It has been reported that expression of ascorbate peroxidase 2 (APX2) is involved in modulation of cellular H 2 O 2 levels in response to wounding [15,26,27]. The increase in APX activity in P 2 L and no increase in P 1 L suggested that APX may not play vitally important roles in the mechano-induced cold tolerance in wheat.
Increasing evidence supports the multi-signaling functions of H 2 O 2 in response to abiotic stresses in higher plants [11]. Here, under low temperature, for both texted cultivars the H 2 O 2 concentration in P 1 L was very close to the normal level in CC. However, the release rate of O 2 − in P 1 L was significantly higher than in CC. It was suggested that activated antioxidative enzymes, such as SOD and CAT induced by mechano-stimulation, modify the H 2 O 2 concentration to an appropriate level as a signal molecule, which prevents H 2 O 2 -induced damage to plant tissues [11]. In addition, modified GPX and GR activities have also been shown to be related to the down-regulation of H 2 O 2 levels in chloroplasts [28]. Here, the increased activities of GPX and GR did favour the relatively low level of H 2 O 2 in P 1 L (Fig. 3). Although the concentrations of AsA and GSH are only in the millimolar range in plant tissues, the AsA-GSH cycle plays a very important role in neutralizing H 2 O 2 released by As key members in the AsA-GSH cycle, the altered expression of glutathione reductase (GR)-and dehydroascorbate reductase (DHAR)related proteins were found via proteome analysis in the present study; chloroplastic GR was increased in P 1 L in both cultivars, whereas DHAR was enhanced in P 1 L only in LM6 compared to CL (Fig. 7). The changes in expression of these enzymes are in accordance with their activities in chloroplasts. Thus, we suggest that the AsA-GSH cycle is involved in mechano-stimulated cold tolerance in winter wheat.
Hardening with a previous abiotic stress endows plant with higher tolerance to recurring stresses [29]. For example, pre-anthesis heat hardening (or pre-treatment) can partially protect wheat plants from photosynthetic inhibition and oxidative damage under post-anthesis high-temperature stress, which is attributed to the modified expressions of photosynthesis-responsive and antioxidant enzyme-related genes [30]. Furthermore, many studies have shown that the mechanism underlying hardening includes the accumulation of soluble sugars, reduction of photosynthetic apparatus [30], scavenging of reactive oxygen species (ROS) [30], accumulation of osmoprotective proteins (dehydrins) [31], and other compatible solutes such as proline and betaína [31]. It is well known that cold acclimation reduces frost damage, and that this phenomenon involves a mechanism similar to that of drought acclimation [32]. Mechano-stimulation may induce many types of cold response proteins and genes [8]. In addition to the antioxidant system activated by mechano-stimulation, shown in the present study, many types of proteins related to photosynthesis, energy production, and C metabolism were modified by mechano-stimulation (Fig. 7). With respect to photosynthetic C assimilation, ribulose-1, 5-biophosphate carboxylase activase and its isoform 1 had a higher level of expression in P 1 L, but a relatively low level in CL, in the two tested cultivars. Ribulose-1, 5biophosphate carboxylase activase and Rubisco large subunit-binding protein have been shown to play a critical role in the activation of Rubisco [33]. Carbonic anhydrase enhances the CO 2 concentration in chloroplasts, which improves the carboxylation rate of Fig. 7 Relative expression ratio of altered proteins in wheat leaves affected by combination of mechano-stimulation and jointing cold stress. The difference of expression level at the given protein spots between CL and CC, P 1 L (or P 2 L) and CL was log-normalized and converted to a color scale. It was reorganized after analysis with the PageMan software (http://mapman.mpimp-golm.mpg.de/pageman/). Up-regulation and down-regulation were indicated in increasing red and blue, respectively. The missing proteins were indicated in white Differentially expressed protein spots in stress treatments (CL, P 1 L and P 2 L) compared with CC for each cultivar were indicated with arrows and listed in Table 1 Rubisco enzyme [21]. Our proteomic data also showed a higher abundance of Rubisco large subunit-binding protein subunit α in P 1 L than that in CL in both cultivars. This implies increases in the Rubisco activation state and carboxylation rate induced in the early mechanostimulated plants under low temperature stress (Fig. 2). The unaltered expression of ferredoxin-NADP(H) oxidoreductase and down-regulation of ferredoxin-NADP reductase, leaf isozyme, and chloroplastic-like protein in P 1 L observed in the present study, resulted in an increased level of NADPH-dependent H 2 O 2 as compared with CL. It has been suggested that increased NADPH-dependent H 2 O 2 is required for the activation of systemic wound responses [11]. The mechano-stimulation induced H 2 O 2 production is also involved in plant defence responses against invading pathogens [34]. It was reported that ROS can control Ca 2+ -permeable channel activity to regulate the intracellular Ca 2+ level [35]. The changes of ROS and Ca 2+ following mechano-stimuli were implicated in the induction of defense genes in response to fungal pathogens [34,36].
Photosynthetic electron transport generates energy (ATP) and reducing power (NADPH) to support carbon reduction and photorespiratory carbon oxidation in the dark reaction in photosynthesis and plays a key role in the maintenance of optimum photosynthetic rate and ensuring effective energy flow for growth [37]. To further investigate the interactive effects of mechanostimulation and cold stress on the process of photosynthetic electron transport, transient fluorescence kinetics were analysed (Fig. 1). The I-P phase of the transient fluorescence kinetics revealed changes in the electron flux from PQH 2 to the final electron acceptor and the size of the final electron acceptor pool of PS I [38]. The present study showed no significant effects of the combination of mechano-stimulation and cold stress on the electron flux from PQH 2 to the final electron acceptor. However, the O-I part of the kinetics was affected by cold stress and mechanostimulation, which reveals changes in the process involving exciton capture to PQ reduction [38]. In addition, the rise in fluorescence transient from O to P was faster in P 2 L in LM6, which indicates that the re-oxidation of Q A − was inhibited by the combination of cold and later mechano-stimulation [39]. However, it was not markedly affected by cold stress alone. The ATPases in chloroplasts, Mg 2+ -ATPase and Ca 2+ -ATPase, play a key role in ATP formation [18]. In this study, the activities of these two functional enzymes in chloroplasts were enhanced by mechano-stimulation in response to cold stress, which might favor the ATP formation. Consistently, the proteome data showed that, in both cultivars tested, there was a higher abundance of ATP synthase β subunit in P 1 L.

Conclusions
Early mechano-stimulation at the stage of at Zadoks growth stage 26 activated the antioxidant system and hence maintained the balance of reactive oxygen species, improved the electron transport and photosynthesis rate under cold stress applied at the jointing stage, whereas mechano-stimulation applied 6 days before the cold event induced an opposite effect, except for APX activity and ATPase activities in chloroplasts. Proteomic and transcriptional analysis revealed that the oxidative stress