Comparative Proteomic Analysis of Differentially Expressed Proteins Induced by Hydrogen Sulfide in Spinacia oleracea Leaves

Hydrogen sulfide (H2S), as a potential gaseous messenger molecule, has been suggested to play important roles in a wide range of physiological processes in plants. The aim of present study was to investigate which set of proteins is involved in H2S-regulated metabolism or signaling pathways. Spinacia oleracea seedlings were treated with 100 µM NaHS, a donor of H2S. Changes in protein expression profiles were analyzed by 2-D gel electrophoresis coupled with MALDI-TOF MS. Over 1000 protein spots were reproducibly resolved, of which the abundance of 92 spots was changed by at least 2-fold (sixty-five were up-regulated, whereas 27 were down-regulated). These proteins were functionally divided into 9 groups, including energy production and photosynthesis, cell rescue, development and cell defense, substance metabolism, protein synthesis and folding, cellular signal transduction. Further, we found that these proteins were mainly localized in cell wall, plasma membrane, chloroplast, mitochondria, nucleus, peroxisome and cytosol. Our results demonstrate that H2S is involved in various cellular and physiological activities and has a distinct influence on photosynthesis, cell defense and cellular signal transduction in S. oleracea leaves. These findings provide new insights into proteomic responses in plants under physiological levels of H2S.


Introduction
Hydrogen sulfide (H 2 S) has been emerging as a potential messenger molecule, strikingly similar to nitric oxide (NO) and carbon monoxide (CO), involved in the modulation of a wide range of physiological processes in animals and plants [1][2][3][4][5][6]. Since the 1970s, the phenomenon of H 2 S emission from plants has been demonstrated by many researchers [7][8][9]. Moreover, the production of H 2 S can be altered under biotic or abiotic stresses [10]. In addition, H 2 S has a dual function, either as a cytotoxin or a cytoprotectant, which depends on the concentration of H 2 S and the status of the environment. At low concentration, H 2 S has an obvious signaling regulatory function in plants. For instance, H 2 S could promote seed germination of wheat and ameliorate copperinduced damage of plasma membrane integrity in root tips [11]. H 2 S has also been reported to counteract chlorophyll loss and reduce oxidative damage due to osmotic stress in sweet potato seedling leaves [12]. Furthermore, boron toxicity, chromium toxicity and cadmium toxicity in plants could be alleviated by H 2 S through enhancing the activities of antioxidant enzymes and decreasing the accumulation of toxic ions [13][14][15][16]. In addition, a low H 2 S concentration has been shown to promote the embryonic root length of Pisum sativum [17]. Similarly, our prior study has shown that H 2 S could enhance photosynthesis through promoting chloroplast biogenesis, photosynthetic enzyme expression and thiol redox modification in Spinacia oleracea seedlings [18]. Besides, our results have indicated that H 2 S plays an ameliorative role in protecting barley seedlings against aluminum toxicity by inducing the activities of antioxidant enzymes, increasing citrate secretion and the gene expression of citrate transporter, and enhancing the protein expression of PM H + -ATPase [19]. Interestingly, some evidences have recently demonstrated that H 2 S may delay senescence of cut flowers and prolong flower vase life in a wide spectrum of botanical species, including herbaceous and woody plants. In addition, H 2 S also could prolong the postharvest shelf life of strawberries and play an antioxidative role in fruits [20,21]. However, at high concentrations, H 2 S may interfere with plant's normal growth and metabolism. For instance, high concentrations of H 2 S may impair photosynthetic electron transport and depress plant growth [22,23].
Previous studies on H 2 S mainly focused on morphological, physiological and biochemical processes in plants. However, the detailed molecular mechanisms underlying plant response to H 2 S signal remain largely unknown. Moreover, none of the studies mentioned above have provided information on the changes of protein expression induced by physiological levels of H 2 S. Recently, proteomic approaches have emerged as a powerful tool for gaining insight into physiological changes at the cellular and biochemical level, allowing the function and regulation of a specific signaling molecule to be explored in detail. For instance, by using proteomic approaches, Bai et al. [24], found that Gprotein coupled signaling is an early event that works upstream of NO biogenesis. Similarly, Lum et al. [25], investigated the downstream signaling pathways of NO in mung bean using a proteomic approach, suggesting that exogenous sodium nitroprusside (SNP), a donor of NO, could affect the expression level of photosynthetic enzymes and glucose metabolism. Therefore, comparative proteomic studies have been successfully applied to systematically investigate protein expression changes in several plant species to elucidate the roles of specific signaling molecules.
In the present study, we used a quantitative proteomic approach to identify global protein expression changes of S. oleracea seedlings under NaHS treatment, an exogenous H 2 S donor. Using this powerful tool, we observed extensive changes of protein expression relating to energy production and photosynthesis, cell rescue, development and defense and so on. The present results would provide some new insights into H 2 S-mediated metabolic and physiological changes in plants and also would accelerate the study of H 2 S signaling function in plants

Plant materials and growth conditions
Seeds of Spinacia oleracea were first sterilized in 75% ethanol for 3 min, then in 10% sodium hypochlorite solution for an additional 10 min followed by washing with distilled water and germinated in a soil/vermiculite (1:1) mixture. Two-week-old seedlings were transferred to 1/2 Hoagland's solution (pH 6.0) in a controlled growth chamber with a light/dark regime of 15/9 h, relative humidity of 80%, temperature of 21/27uC and a photosynthetically active radiation (PAR) of 190 mmol m 22 s 21 . NaHS was purchased from Sigma and used as an exogenous H 2 S donor as described by Hosoki et al., [26]. The seedlings were treated with 100 mM NaHS for 30 d and the solution were changed every 3 d. The S. oleracea leaves were collected and immediately frozen in liquid N 2 and stored at 280uC for subsequent experiments. Each experiment was repeated at least three times.
Leaf area calculation, dry weight and relative water content analysis Twenty leaves with the same leaf position were collected from control and NaHS treated seedlings, respectively, and then flatted on clean coordinate paper. Leaves were photographed with a digital camera at the same image resolution. Leaf area was calculated with pixels as described previously by Xiao et al. [27], using the Adobe Photoshop 7.0 software (Adobe Systems Inc., San Jose, CA). Each treatment comprised three biological replicates. Thus, the average leaf area was calculated from three replicates.
Leaves were removed from the same leaf position and immediately weighed to obtain the leaf fresh weight (FW). Leaves were subsequently placed into vials filled with distilled water for 24 h, then blotted to remove excess water and re-weighed to determine the leaf turgid weight (TW). Leaves were dried to a constant weight at 65uC and re-weighed to obtain the leaf dry weight (DW). Leaf relative water content (RWC) was calculated as (FW-DW)/(TW-DW)6100 [28].

Pigment analysis, gas exchange and stomatal aperture measurements
Chlorophyll content was measured according to Lichtenthaler [29] with some modifications. After extraction using 10 ml of 80% (v/v) aqueous acetone, the content of total chlorophyll was calculated from the absorbance of leaf chlorophyll extracts at 470, 646 and 663 nm.
The net photosynthetic rate (P n ) was measured using a portable photosynthesis system (Li-6400, Li-Cor, Lincoln, NE, USA) on the third fully developed leaf of each seedling. Air temperature, light intensity, CO 2 concentration and air relative humidity were maintained at 25uC, 800 mmol m 22 s 21 , 380 ml l 21 , and 90%, respectively. P n was expressed on a leaf area basis.
Measurements of stomatal apertures were performed as described by Desikin et al., [30]. Abaxial epidermal strips from similar rosette leaves were floated in 10 mM 2-(N-morpholino)ethanesulfonic acid (MES) buffer (pH 6.15) containing 50 mM KCl and 50 mM CaCl 2 for 2 h under light conditions to open the stomata before the addition of NaHS. Next, 100 mM NaHS was added to the buffer solution and incubated for another 2 h to assay stomatal aperture. Finally, stomatal aperture was calculated as the ratio of width to length using Sigma Scan Pro 5 software.

Measurement of amino acid content
Amino acid analysis was carried out by ion exchange chromatography as described by Oliveria et al., [31]. For each treatment, samples were ground in 50 ml of a methanol, chloroform, and water mixture (60:25:15 v/v/v) for 1 min. The ground samples were centrifuged, and then the clear supernatant was decanted into 100 ml beaker and allowed to partially evaporate overnight in a hood to remove the methanol and chloroform. The samples were taken to dryness in a vacuum desiccator and the dried extracts were suspended in 5 ml of citrate buffer (pH 2.2). The measurement of amino acid content was performed by an amino acid analyzer (Model L-8800, Hitachi Co. Ltd., Tokyo, Japan) with a column packed with Hitachi custom ion-exchange resin, which temperature were controlled from 30 to 70uC. The lithium citrate buffer and ninhydrin flow rates were 0.35 and 0.30 ml/min, respectively.

Protein extraction and 2-DE analysis
Total proteins were extracted by the phenol procedure [32]. Briefly, one to two grams of the fresh leaves were ground in liquid nitrogen and total soluble proteins were extracted at 4uC for 1 h in 2 ml of 20 mM Tris-HCl buffer (pH 7.5) containing 250 mM sucrose, 10 mM ethylene diamine tetraacetic acid (EDTA), 1 mM phenylmethyl-sulfonyl fluoride (PMSF), 1% (w/v) Triton X-100, 5% b-mercaptoethanol and 1% (w/v) polyethylene polyvinyl pyrrolidone (PVPP). The homogenates were subjected to centrifugation at 12,000 g, 4uC for 15 min, after which the supernatants were added to two volumes of Tris-saturated phenol (pH 8.0) and the mixture was further vortexed for 30 min. Proteins were precipitated by adding five volumes of ammonium sulfatesaturated methanol and incubated at 220uC for at least 4 h. After centrifugation as described above, the protein pellets were re-suspended and rinsed with ice-cold methanol, followed by washing with ice-cold acetone twice, and spun down at 15,000 g, 4uC for 10 min after each washing. The final washed pellets were air-dried and dissolved in lysis buffer containing 8 M urea, 2 M thiourea, 4% (w/v) 3-[(3-Cholamidopropyl)dimethylammonio]-1propanesulfonate (CHAPS), 1% (w/v) DL-dithiothreitol (DTT) and 1% (v/v) IPG buffer (pH 4-7). Protein concentrations were determined by the Bradford assay [33]. Two-dimensional electrophoresis (2-DE) was carried out according to Bjellqvist et al., [34]. Samples containing 1.2 mg protein were loaded onto an IPG strip holder fitted with dry IPG strips (length 18 cm, pH 4-7) and rehydrated for 16 h at room temperature. Isoelectric focusing was carried out with an Ettan IPGphor system (GE Healthcare Amersham Bioscience, Little Chalfont, U.K.) using the following voltage program: 300 V for 1 h, 600 V for 1 h, 1000 V for 1 h, a gradient to 8000 V for 2 h, and then maintaining a voltage of 8000 V for 64000 V?h. Focused strips were then equilibrated by soaking in an equilibration solution (6 M urea, 30% glycerol, 2% SDS, 50 mM Tris-HCl, pH 8.8, and 1% DTT) for 15 min, followed by the same equilibration solution but with 2.5% iodoacetamide instead of DTT for another 15 min. Separation of proteins in the second dimension was performed on SDS-12.5% polyacrylamide gels. Each separation was repeated three times to ensure the protein pattern reproducibility.
Gel staining, imaging and data analysis SDS-PAGE gels were stained with Coomassie Brilliant Blue (CBB) R-250 and then scanned at 600 dots per inch (dpi) resolution using a scanner (Uniscan M3600, China). Gels were analyzed using PDQuest software (Version 7.0, Bio-Rad). For each gel, a set of three images was generated, corresponding to the original 2-D scan, the filtered image and the Gaussian image. The Gaussian image, containing three-dimensional Gaussian spots, was used for the quantification analysis. After normalization and background subtraction, a matchset was created by comparing the control gels. The intensity of each spot was determined using the spot quantification tool and expressed using the ratio of pixel intensity of a single spot to the pixel intensity of all spots on the gels. The intensity of each spot was normalized by the local regression model of the software to compensate for gel-to-gel variation [35]. The expression of a protein in a spot was defined as the relative pixel volume of that spot. Meanwhile, protein spots that changed by more than 2-fold and passed the Student's t test (P,0.05) were selected and identified by MALDI-TOF MS. To compensate for subtle differences in sample loading or gel staining/destaining during individual repeat experiments, We normalized spot volumes based on total intensity of valid spots were calculated for each 2-DE gel and used for statistical calculations of protein abundance. For some missing values, we manually checked, and a consistent reproducibility between normalized spot volumes was found in the three replicates (Table  S1). Each experiment was repeated three times.

In-gel protein digestion, protein identification and protein classification
Protein spots showing significant changes in abundance during the treatments were excised manually from colloidal CBB stained 2-DE gels and protein digestion with trypsin was performed as follows. Briefly, gel slices were first destained with a 1:1 (v/v) solution of methanol and 50 mM NH 4 HCO 3 for at least three times until the color of CBB was removed, then washed several times with water and completely dried in a vacuum centrifuge. Depending on protein amount, 2-3 ml of 0.1 mg ml 21 modified trypsin (Promega, sequencing grade) in 25 mM NH 4 HCO 3 was added to the dehydrated gel spots. After 30 min incubation, 7 ml of 25 mM NH 4 HCO 3 was added to submerge the gel spots and left at 37uC overnight. After digestion, the gel slices were washed with 0.1% trifluoroacetic acid (TFA) in 50% v/v acetonitrile (ACN) three times to acquire the peptides. Matrix was prepared by dissolving a-cyano-4-hydroxycinnamic acid (CHCA) in 50% ACN and 0.1% TFA. Ten microliters of matrix solution was added into the dry peptides, and vortexed for 30 min. 0.5 ml of peptide was mixed with 0.5 ml of matrix solution. A total of 1 ml of reconstituted in-gel digest sample was spotted onto an Anchorchip target plate. The dried sample on the target plate was washed twice with 1 ml of 0.1% TFA, and left for 30 s before solvent removal. MALDI-TOF MS analysis (ReFlexTMIII, Bruker) was used to acquire the peptide mass fingerprint (PMF). A standard peptide mixture was spotted adjacent to all samples for external calibration, followed by internal mass correction using peptide ions generated by trypsin autoprotolysis (m/z 842.5 and m/z 2211.10). Spectra were analyzed using the flexAnalysis software (Version 3.2, Bruker-Daltonics). Then, the measured tryptic peptide masses were transferred through the MS BioTool program (Bruker-Daltonics) as inputs to search against the taxonomy of green plants in the NCBI (NCBInr, downloaded on September 9, 2011) database. The parameters of PMF were as follows: 100 ppm tolerance as the maximum mass error, MH + monoisotopic mass values, allowance of oxidation (M) modification, allowance for one missed cleavage, and fixed modification of cysteine by carboxymethyl (carbamidomethylation, C). Potential matches were identified by considering the Mascot score, the putative functions and differential expression patterns on 2-DE gels. Several criteria were used to assign a positive match with a known protein. These were as follows: (i) Protein identifications were validated manually, ensuring that at least 4 peptides matched. (ii) The coverage of protein sequences by the matching peptides had to reach a minimum of 10%. (iii) The score that was obtained from the analysis with the Mascot software indicated the probability of a true positive identification and had to be at least 70. Positive matches were BLAST searched against the UniPort (http://www. uniprot.org) and/or NCBI protein (http://www.ncbi.nlm.nih.gov) databases for updated annotation and identification of homologous proteins. The identified proteins were searched with the UniPort and TAIR databases to find out if their functions were known, then they were further classified using Functional Catalogue software (http://mips.gsf.de/projects/funcat).

Statistical analysis
Values in figures were expressed as means 6 SE. The statistical significance of the data was analyzed using an univariate analysis of variance (P,0.05) (one-way ANOVA; SPSS for Windows, Version 13.0). For proteomic experiment, protein samples for 2-DE gel image analysis were extracted from three independent seedlings grown in three different pots in the same growth chamber. Thus, for Control and H 2 S treatment, three independent biological replicates were performed in 2-DE gel image analysis. The ratio of H 2 S and CK in the table 1 was the average of three replicates. Statistic analysis for 92 protein spots on 2-DE gels was performed using Student's t-test (P,0.05) provided by PDQuest software as mentioned earlier.

Results
Effects of H 2 S on growth and photosynthesis of S. oleracea H 2 S could significantly affect growth and photosynthesis in S. oleracea seedlings. As shown in Fig. 1A, leaf area in S. oleracea seedlings treated with NaHS was profoundly increased (P,0.01) by 40% compared to control plants. Similarly, dry weight of S. oleracea seedlings under NaHS treatment was significantly increased (P,0.01) from 0.32 g per seedling to 0.47 g per seedling  (Fig. 1B). The relative water content (RWC) of seedlings treated with NaHS was also increased (P,0.05) (Fig. 1C). In addition, chlorophyll content showed a remarkable increase (P,0.01) in seedlings treated with NaHS (Fig. 1D). Consistently, NaHS treatment obviously enhanced the photosynthetic rate of S. oleracea seedlings (P,0.01) (Fig. 1E). However, the stomatal aperture was reduced by around 20% in S. oleracea seedlings treated with NaHS (Fig. 1F). Changes in amino acid content in S. oleracea under NaHS treatment To investigate whether a low concentration of H 2 S affected amino acid metabolism, we measured the contents of 17 kinds of amino acid in S. oleracea seedlings after NaHS treatment. As shown in Fig. 2, the contents of 10 amino acids, including Arg, Tyr, Thr, Val, Cys, Met, Ile, Phe, His and Pro, increased to some extent, whereas the contents of other seven amino acids, including Asp, Ser, Glu, Gly, Ala, Leu and Lys, decreased in S. oleracea seedlings under NaHS treatment.

Differentially expressed proteins in S. oleracea under NaHS treatment
To elucidate the possible mechanisms underlying H 2 S-induced increased in plant growth, we performed 2-DE to identify differences in the whole protein profiles of S. oleracea seedlings under 100 mM NaHS treatment compared with control plants. Representative images are presented in Fig. 3. The proteome was evaluated over an isoelectric point (pI) ranging from 4 to 7 and molecular weight (MW) ranging from 12 to 110 KDa (Fig. 3A). More than 1000 proteins were reproducibly resolved from the 2-D gels. Inspection of the gel patterns revealed that the MW and/or pI values of the spots differed from the theoretical values. Alternatively, some proteins were present in multiple spots, possibly due to translation from alternatively spliced mRNAs. Close-up views of several protein spots are shown in Figure 3B. A total of 92 proteins were positively identified by MALDI-TOF MS and listed in Table 1. Besides, the detail peptide information of identified proteins was listed in Table S2. Among these proteins, 65 protein spots were up-regulated and 27 were down-regulated in S. oleracea seedlings after H 2 S treatment.

Reproducibility and variation in proteomics data
It was important to test the variation between biological replicates for both control and NaHS treatment sets. We carried out three biological replicates for our experiments (Table S1). Meanwhile, we also calculated for each protein the Pearson's linear correlation for protein abundant values across three biological replicates ( Table 2). The reproducibility between the each two biological replicates of accumulation profiles of proteins showed a very high correlation (0.921 for CK-R1 and R2, 0.916 for CK-R2 and R3, 0.945 for CK-R1 and R3, 0.948 for H 2 S-R1 and R2, 0.931 for H 2 S-R2 and R3, 0.941 for H 2 S-R1 and R3). The reproducibility decreased to still significant levels with decreasing protein abundance ( Table 2). In this study, we found that over 1000 protein spots were reproducibly resolved, of which the abundance of 92 spots was changed by at least 2-fold (sixty-five were up-regulated, whereas 27 were down-regulated) ( Table 1). Besides, to further check the reproducibility of the replicates, we also analyzed the correlation of 92 differentially expressed proteins among three biological replicates ( Figure S1), suggesting the well correlation and reproducibility among different biological replicates.

Functional classification of differentially expressed proteins
Among the 92 identified proteins, 14 proteins had unknown functions or were hypothetical proteins. However, 78 had assigned functions and could be classified into 9 groups based on their biochemical function (Table 1, Fig. 4). The majority of the protein profile corresponded to energy production and photosynthesis associated proteins (16.30%), followed by cell rescue, development and defense related proteins (15.22%). In addition, transcription, protein synthesis, folding and modification related proteins (13.04%) and proteins with binding functions, cofactor requirements and cellular components (11.97%) took a large part of the identified proteins. Cellular signal transduction (8.70%) and metabolism related proteins (amino acid, nitrogen and sulfurrelated protein were 3.26%, C-compound related protein was 6.52% and phosphate-related protein was 3.26%) were also found in our present study (Fig. 4, Table 1).

Discussion
H 2 S has recently been discovered to be an important signaling molecule involved in many different plant physiological processes, including seed germination [11], root organogenesis [17], abiotic stress tolerance [13,14,36,37], photosynthesis [18] and guard cell movement [38,39], senescence of cut flowers [20] and postharvest shelf life of fruits [21]. Although the functional roles of H 2 S have been studied in these physiological processes, the exact mechanisms by which H 2 S transmits the signal are still unclear. Therefore, our present study aimed to investigate H 2 S responsive differentially expressed proteins in plants using a proteomic approach.

H 2 S affected energy production and photosynthesisassociated protein expression
Carbon dioxide fixation is an essential process of photosynthesis, and this pathway involves in many enzymes that catalyze and regulate energy generation [40]. In this study, we identified several photosynthesis-associated proteins and their expression levels were significantly up-regulated after NaHS treatment, e.g., the ribulose-1,5-bisphosphate carboxylase/oxygenase large subunit (rubisco LSU, spots 3, 7, 8 and 53) and rubisco activase precursor (spots 1 and 34). Rubisco, the CO 2 fixing enzyme in Calvin cycle, is the primary limiting factor of net photosynthesis. H 2 S could obviously increase the protein expression of rubisco LSU. Similarly, we have previously reported that the activity of rubisco and protein and gene expression of rubisco LSU were significantly enhanced after 100 mM NaHS treatment [18]. Here, proteomic results confirmed that the increased expression of rubisco LSU and Rubisco activase precursor by H 2 S may enhance Calvin cycle activity, promoting the increment of photosynthetic CO 2 assimilation and plant growth. Besides, we also detected that phosphoenolpyruvate carboxylase (PEPc, spot 55), which plays a key role in photosynthetic CO 2 assimilation and plant growth, was up-regulated after H 2 S treatment. This result is consistent with protein expression data obtained using the western blotting method (data not shown). Similarly, these results were supported by physiological measurements (Fig. 1E).
It's well known that large amounts of ATP are needed by plants to provide sufficient energy for growth, development and photosynthesis [41,42]. Therefore, ATP synthase and ATPase are key enzymes in energy production and conversion. In this study, we identified ATP synthase (spot 47) and found that its expression was up-regulated by H 2 S, suggesting that H 2 S treatment could increase energy production which could be utilized for the growth and development of plants. We also identified another important protein, chlorophyll a/b binding protein CP26 (LHCB, spot 11), whose expression was significantly increased under NaHS treatment. The LHCB proteins are the apoproteins of the light-harvesting complex of photosystem II (PSII), which are normally complexed with chlorophyll and xanthophylls and serve as the antenna complex [43]. Therefore, high expression of LHCB may increase PSII activity and electron transfer efficiency. Similarly, our previous results have shown that the maximal photochemical efficiency of PSII (F v /F m ) is reached in the presence of H 2 S [18]. In other words, H 2 S appears to have a positive effect on the process of light capture in photosynthesis.

H 2 S affected cell rescue, development and defenserelated protein expression
The gaseous hormone ethylene plays multiple roles in regulating plant growth and development [44]. Ethylene is produced biologically from S-adenosylmethionine (SAM) via the following pathway: SAMR1-aminocyclopropane-1-carboxylic acid (AC-C)Rethylene [44]. Ethylene-forming enzyme (EFE) is responsible for oxidation of ACC to ethylene [44,45]. Here, we identified an ethylene-forming enzyme-like protein (spot 29) whose abundance was obviously increased after H 2 S treatment, indicating that ethylene biosynthesis may also be associated with H 2 S signaling pathway. Interestingly, Liu et al. [46], have reported that exogenous application of ethylene could significantly increase endogenous H 2 S content in Arabidopsis seedlings. Meanwhile, they also found that the generation of H 2 S induced by NO might mediate ethylene-induced stomatal closure in Arabidopsis [46]. Similarly, it's well known that jasmonic acid (JA) is involved in a wide range of stress, defense and development processes in plants [47]. In our study, we identified allene oxide synthase (AOS, spot 16), which is involved in the biosynthesis of JA. Moreover, AOS abundance significantly increased under H 2 S treatment, suggesting that H 2 S could enhance plant defense by increasing the level of JA. Interestingly, Hou et al. [48], have reported that H 2 S may function downstream of H 2 O 2 in JA induced-stomatal closure of Vicia faba. In addition, JA could enhance the generation of endogenous H 2 S and L-cysteine desulfhydrase activity in guard cells of Vicia faba leaves. Therefore, we hypothesized that the signaling pathway of JA may also be associated with H 2 S signaling networks. Lipoxygenases (LOXs) are non-heme iron-containing dioxygenases widely distributed in plants and animals [49]. Moreover, LOXs may be involved in a number of diverse aspects of plant physiology including growth and development, pest resistance, and senescence or responses to wounding [49]. Besides, LOXs are required for the wound-induced biosynthesis of JA in leaves [49]. Here, we detected LOX (spot 14) and its protein expression level was obviously increased after H 2 S treatment. Therefore, we speculated that H 2 S signaling function may be related with JA pathway to some extent. Glutaredoxins (Grxs) are small oxidoreductases of the thioredoxin family of proteins regulating the thiol redox state of several proteins [50]. Thereby, Grxs play key roles in different aspects of plant development and defense through regulating and maintaining thiol redox homeostasis. In our study, we identified glutaredoxin family protein (spot 6) and found that H 2 S treatment obviously up-regulated this protein expression, indicating that H 2 S may be involved in thiol redox modification and cell redox homeostasis, in line with our previously published paper [18].
In this study, several proteins involved in plant response to biotic and abiotic stresses were identified under H 2 S treatment. For instance, cc-nbs-lrr resistance protein (spot 19), putative NBS domain resistance protein (spot 59), Rp1-like protein (spot 26) and Table 2. Pearson's linear correlation for protein expression abundance values of Spinacia oleracea across three replicates. RPM1-interacting protein 4 (spot 30). These proteins are essential regulators of plant defenses and play central roles in resistance against infection by pathogens [51]. Our results suggest that H 2 S could protect plants from pathogen infection by increasing these protein abundances. Further, glycine-rich proteins (GRPs), containing .60% glycine, have been found in the cell walls of many higher plants and form a group of structural protein components of the wall in addition to extensins and proline-rich proteins [52]. GRPs play very important roles in the post-transcriptional regulation of gene expression in plants under various stress conditions, in most cases, they are accumulated in the vascular tissues and their synthesis is a part of the plant's defense mechanism [53]. Interestingly, we also identified this important protein, glycine-rich protein GRP17 (spot 45), whose abundance obviously increased under H 2 S treatment. Pollination and elongation of the pollen tube are important processes for the normal growth and development of flower plants [54]. Pollen coat protein (PCP) plays a multiplicity of roles in the pollination process [55]. In our study, we identified a pollen coatlike protein (spot 2) and Phl p 3 allergen (spot 10), both of which were up-regulated by H 2 S. Therefore, we hypothesized that H 2 S might be involved in the pollination process and elongation of the pollen tube and this physiological function of H 2 S is the same as that of NO signaling molecule in flower plants [54]. H 2 S affected substance metabolism, lipid, fatty acid and isoprenoid metabolism related protein expression Substance metabolism is the basic life activity and is vulnerable to environmental stresses in plants. As shown in Fig. 1A and 1B, leaf area and dry weight were obviously increased in S. oleracea seedlings treated with 100 mM NaHS, a donor of H 2 S. These results showed that H 2 S affected plant growth and the accumulation of organic compounds. In addition, it's well known that aminotransferase and methyltransferase are key links between carbon and nitrogen metabolism [56,57]. Here, we found that glutathione synthetase (spot 42), class I glutamine aminotranferase (spot 58) and caffeoyl CoA O-methyltransferase (spot 27) were all up-regulated under NaHS treatment. Among them, class I glutamine aminotranferase is known to be involved in Lmethionine (Met) biosynthesis, suggesting that high expression of this protein may increase the formation of Met, which was completely consistent with Met content (Fig. 2F). Besides, the upregulated of caffeoyl CoA O-methyltransferase and down-regulated of cysteine desulfurase (spot 75) may increase the cysteine (Cys) content, as supported by our previously published data on Cys and GSH content [18].
Carbon compounds in the cell wall serve as an important physical barrier, having a significant role in cell defense against various external stresses [58]. Conventionally, accumulation of carbon compounds, such as callose and lignin, has been considered to be one of the most important defense mechanisms. In the present study, we identified cellulose synthase CesA10 (spot 13) and secondary wall-associated glycosyltransferase family 8D (spot 64) and found that their expressions significantly increased after NaHS treatment. Both CesA10 and secondary wallassociated glycosyltransferase family 8D are known to be important for cellulose synthesis and lignin biosynthesis [59]. Therefore, we speculated that H 2 S may play a crucial role in plant defense by increasing the biosynthesis of cell wall related compounds.
We also identified several lipid, fatty acid and isoprenoid metabolism related proteins, including three up-regulated proteins, i.e., serine palmitoyltransferase 2 (spot 24), adenylate isopentenyltransferase (spot 33) and putative acyl-CoA dehydrogenase (spot 65), and three down-regulated proteins, i.e., putative plastid 1deoxy-D-xylulose 5-phosphate reductoisomerase precursor (spot 68), lipase class 3 family protein (spot 69) and cyclopropane-fattyacyl-phospholipid synthase (spot 84). These results indicate that H 2 S may function as a signaling molecule in lipid, fatty acid and isoprenoid metabolic pathways. As we known, in plants, acyl-CoA dehydrogenase is involved in the b-oxidation process of fatty acid [60]. Our results suggested that H 2 S may be involved in regulating the b-oxidation process of fatty acid by changing the protein expression of acyl-CoA dehydrogenase. Regulation of gene expression is achieved at several levels, i.e., transcriptional, post-transcriptional, translational, and post-translational. Several proteins implicated in transcription, protein synthesis and modification were identified in the present study, including three ribosomal proteins (spots 17,18,35), PRP-like protein (spot 63), putative Skp1 (spot 70), hypersensitive-induced response (HIR) protein (spot 76), two-component response regulator ARR 15 (spot 92). Among them, three ribosomal proteins, whose abundances were increased following H 2 S treatment, indicated that H 2 S signaling molecule could accelerate the biosynthesis process of protein in plants. PRP-like protein, which is involved in mRNA processing and regulation of the timing of the transition from a vegetative to reproductive phase in plants, was also found to be obviously increased under H 2 S treatment. Besides, Skp1 is a core component of the Skp1-Cullin-F-box (SCF) family of E3 ubiquitin ligases and serves to tether the rest of the complex to an F-box protein, which provides specificity for binding ubiquitin ligase substrate proteins and plays a role during embryogenesis and early postembryonic development, especially during cell elongation and division [61]. In our experiment, the protein expression of Skp1 was decreased following H 2 S treatment, indicating that H 2 S could inhibit protein degradation and delay embryogenesis in plant. In addition, HIR proteins are a group of proteins involved in hypersensitive reaction (HR). They belong to the PID (proliferation, ion and death) superfamily, whose members function in cell proliferation, ion channel regulation and cell death [62]. HIR protein expression in maize and barley is associated with localized host cell death and disease resistance responses [62,63]. Here, we detected a HIR protein and found that expression of this protein was clearly decreased after H 2 S treatment, indicating that H 2 S may act as a signaling molecule that regulates the plant immunity by changing the expression of HIR protein. Finally, the two-component response regulator ARR 15 is transcriptional activator that binds specifically to the DNA sequence 59-[AG]GATT-39 and functions as a response regulator involved in the His-to-Asp phosphorelay signal transduction system for cytokinin and meristem stem cell maintenance [64]. We found that the protein expression of ARR 15 was obviously changed after H 2 S treatment, suggesting that H 2 S may be involved in His-to-Asp phosphorelay signal transduction system. H 2 S affected cellular transport, transport facilities, transport routes and cellular signal transduction related protein expression It's well known that WRKY transcription factors (WRKY TFs) are a large family of regulatory proteins involved in various plant processes but most notably in coping with diverse biotic and abiotic stresses [65,66]. In our study, we identified WRKY TF 23 (spot 5) and found that its abundance was obviously increased after H 2 S treatment, indicating that H 2 S-enhanced plant immunity may be associated with the regulation of WRKY TFs. Besides, a previous study in animal research has shown that H 2 S, as an endogenous gaseous signal molecule, could induce the opening of K ATP channels [3]. However, it is not clear whether H 2 S could affect potassium channels in plants. Interestingly, in our experiment, we found that H 2 S increased the protein expression of potassium channels (spot 31) in S. oleracea seedlings. Moreover, our unpublished data also has shown that H 2 S could increase K content and promote K + influx in barley seedlings roots under high salt stress by changing the expression of K channel related genes. In addition, García-Mata and Lamattina have reported that H 2 S could induce stomatal closure and participate in the ABAdependent signaling pathway, possibly through the regulation of ABC transporter proteins in guard cells [38]. This result is surprisingly consistent with our proteomic data, because we also identified an ABC transporter protein (spot 43) and found that its expression was obviously increased under H 2 S treatment. It's well known that phytochromes are a family of photoreceptors that modulate the expression of a large number of light-responsive genes and control plant growth and development [67]. Recently, phytochrome has been found to regulate various biotic and abiotic stresses, such as salinity, drought, cold and herbivory [68]. Crosstalk between phytochrome-mediated light signals and some other signaling pathways has been reported in diverse plants [69]. Thus it is possible that phytochrome is involved in H 2 S signaling pathway. In our study, we identified phytochrome C (spot 37) and found that its expression was up-regulated under H 2 S treatment, suggesting that H 2 S could promote plant growth and development and these processes were associated with the signal pathway of phytochrome. Further research need to widen our understanding of the relationship between phytochrome and H 2 S signal in plants.

H 2 S affected proteins with binding function or cofactor requirements and cellular components related protein expression
In the present study, we identified several proteins with binding function or cofactor requirement and cellular components, including beta tubulin (spot 15), lipid binding protein (spot 38), protein binding protein (spot 41), 14-3-3e protein (spot 60), 14-3-3 protein Lil 1433-3 (spot 62), cytosolic phosphoglucose isomerase (spot 85) and actin (spot 89). Among them, actin and tubulin dynamics have important functions in cellular homeostasis. H 2 S treatment affected actin and tubulin expressions, suggesting that H 2 S might be involved in cellular homeostasis. 14-3-3 proteins are known to bind certain phosphorylated proteins to complete phosphoregulation events. In plants, 14-3-3 protein binding activity includes the regulation of key metabolic enzymes, such as nitrate reductase and sucrose synthase [70,71], and the activation of plasma membrane H + -ATPase [72]. In addition, 14-3-3 proteins also participate in a wide array of signal transduction regulatory events [73]. In our experiment, we identified two 14-3-3 proteins and found that their expression levels were obviously increased after H 2 S treatment, indicating that H 2 S may function in the signal transduction regulatory events involving 14-3-3 protein. However, the detailed mechanism of cross-talk between H 2 S and 14-3-3 protein in signal transduction regulatory events is still unclear.
In summary, by using a comparative proteomic strategy, we compiled an overview of the systematic mechanism by which S. oleracea seedlings respond to the H 2 S signaling molecule. Quantitative analysis of more than 1000 highly reproducible proteins on 2-DE profiles identified 92 proteins which expressions were significantly changed in response to H 2 S. These proteins were classified into 9 functional groups, the main one being energy production and photosynthesis associated proteins, followed by cell rescue, development and defense related proteins. In addition, transcription, protein synthesis, folding, modification related proteins and proteins with binding function or cofactor requirement and cellular components formed a large part of the identified proteins. Cellular signal transduction and metabolism related proteins were also found. Taken together, the above-mentioned results indicated that H 2 S played important roles in a set of proteins associated with energy production, photosynthesis, metabolism, cell rescue, cell defense and protein synthesis, folding and signal transduction, etc. On the basis of above results, we also studied another important question regarding H 2 S signaling in plant cells, the localization of its sub-cellular target, and proposed a schematic model of systematic response mechanism of S. oleracea seedlings to H 2 S (Fig. 5). We found that several subcellular organelles were predominantly affected by H 2 S or H 2 S signaling in plants, including mitochondria, chloroplast, nucleus, and peroxisome. This research provides valuable information about the response of plants to H 2 S as a signaling molecule.

Acknowledgments
We are grateful to Bing-Bo Li, Wei-Zhi Lu and Fei-Hua Wu for assistance with the experiments and Sees-editing Ltd for critically editing the manuscript.