Introduction

The phytohormone salicylic acid (SA) plays an important role in regulating various physiological and biochemical processes, including plant defense, photosynthesis, cell expansion, respiration, thermotolerance, stomatal responses and nodulation1,2,3. Moreover, SA is an important signaling molecule for plant immune response and is necessary for the activation of SAR (systemic acquired resistance)4,5. However, the effect of SA on these processes may be indirect, because SA is heavily involved in crosstalk with other plant hormones6. These signaling networks and processes involve many protein to protein and protein to non-protein interactions1. Research on SA effects at the transcriptional and protein levels is abundant, but a more comprehensive picture of the change in phosphorylation induced by SA is needed.

Reversible phosphorylation is one of the most common post-translational protein modifications and is critical to the control of multiple cellular functions, including the stress/defense responses7. It is estimated that one-third of eukaryotic proteins are phosphorylated8. A wide range of signaling networks and processes, such as protein kinase activation, cell cycle progress, cellular differentiation and transformation, development and hormone responses are regulated by the state of protein phosphorylation9. Similarly, a variety of proteins has been reported to be phosphorylated in response to stimuli9. Hence, obtaining a complete systems-level analysis of molecular events requires direct measurements of proteins, including post-translational modifications that may affect their function10.

Our previous studies have identified several proteins that are responsive to SA2. In this work, based on isotope tags for relative and absolute quantification (iTRAQ) technology and TiO2 enrichment method combined with LC-MS/MS analysis, we monitored the response to SA in the phosphoproteome of maize leaves. Phosphoproteins with significantly changed phosphorylation levels were analysed to reveal their biological significance induced by SA. Our data provide new and valuable insights into the plant response induced by SA.

Results and Discussion

Phosphopeptide and phosphoprotein identification

About 1878 phosphopeptides were identified across the maize database (data not shown). Then, we removed the redundant and invalid peptides (PhosphoRS probabilities ≥75% and PhosphoRS scores ≥50). Thus, we identified 1495 unique phosphopeptides, collectively containing 2008 non-redundant phosphorylation sites (Original data for all the identified and quantified phosphopeptides were shown in Supporting Information Table S1). Among those phosphorylation sites, 1783 (88.79%) were found at serine, 208 (10.36%) at threonine and 17 (0.85%) at tyrosine residues (Fig. 1A). Among the 1495 unique phosphopeptides, 1034 were singly phosphorylated, 423 were twice phosphorylated, 40 were phosphorylated at three sites and only 2 were found phosphorylated at four sites (Fig. 1B). The contribution of phospho-Ser (pS) was consistent with results from rice (89.5%)11 and the contribution of phosphor-Thr (pT) was similar to that found in Arabidopsis (9.9%)11. Moreover, the distribution of tyrosine phosphorylation in maize was 0.85%, which was similar to that in cotton (0.8%)12 and was in agreement with the previous report from maize10,13.

Figure 1
figure 1

The distribution of phosphorylation sites.

(A) Distribution of phosphorylation on serine, threonine and tyrosine was assessed for all non-redundant localized phosphorylation sites. (B) Distribution of single- and multi-phosphorylated peptides showed that the majority of phosphopeptides have only one phosphorylation site.

Full size image

From the 1495 phosphopeptides, 858 phosphoproteins were identified. All the mass spectrometry data have been deposited to the ProteomeXchange Consortium via the PRIDE partner repository with the dataset identifier PXD002586. Passing through several steps in phosphopeptide identification, like protein digestion, phosphopeptide enrichment, fractionation and mass spectrometry analysis, it is possible that one or more phosphopeptides can be found in LC-MS analysis in one protein. In our study, among the 858 phosphoproteins, 520 were represented by a single phosphopeptide, 185 by two, 89 by three and 64 by more than three phosphopeptides (Fig. 2).

Figure 2
figure 2

The distribution of phosphoproteins based on identification by single or multi phosphopeptide.

Full size image

Quantitative analysis of phosphoproteins with phosphorylation levels significantly changed in response to SA

For quantitative analysis of phosphoproteins in response to SA, the change (treatment/control) of each phosphopeptide at 1, 4 and 9 h after SA treatment was calculated. An ANOVA with post hoc test was also performed which confirmed that there were significant differences among the three time points. Variance analysis results of these three periods were shown in supporting information Table S2. Of the analyzed samples, 291 phosphopeptides (244 phosphoproteins) showed a significant change (P <0.05, ratio ≥1.5, FDR <0.05) after SA treatment. Detailed information regarding phosphopeptides with statistically significant differences in maize after SA treatment was shown in Supporting Information Table S3. To graphically represent these t-test data, volcano plot –log10 (P value) vs. log2 (fold change) was constructed to graphically display the quantitative data (Supporting information Figure S2). The number of significantly changed phosphopeptides was 100, 150 and 160 after 1, 4 and 9 h, respectively. Some of the important phosphopeptides were shown in Table 1. In addition, the volcano plots indicate that some interesting peptides might be lost. Thus, it should be noted that this study cannot represent an exhaustive survey of all possible phosphorylation changes. A better approach would be to also accept lower abundance changes and confirm them in an independent experiment with higher number of biological replicates.

Table 1 Phosphopeptides changed significantly in maize after SA treatment, as identified by LC-MS/MS.
Full size table

An increasing trend was found for both up- and down-regulated phosphopeptides, of which there were 42 and 58 after 1 h, 60 and 90 after 4 h and 67 and 93 after 9 h, respectively (Fig. 3A). One reasonable explanation for this phenomenon is that following the SA treatment, an increasing number of phosphoproteins involved in signal transportation, but only a few phosphoproteins dephosphorylated during the analyzed time periods. In addition, Venn diagram analyses of the up- and down-regulated phosphopeptides were shown in Fig. 3B,C, respectively. There were 21 phosphopeptides whose levels were all up-regulated significantly at these three time points. On the other hand, there were seven phosphopeptides whose levels were all down-regulated significantly.

Figure 3
figure 3

Analysis of the phosphopeptides whose levels changed after exposure to SA.

(A) Numbers of phosphopeptides whose levels changed after exposure to SA for 1 h, 4 h and 9 h, respectively. N/a, not applicable, indicates that changes were not significant or were not detected in this group; (B) Venn diagram analyses of the up-regulated phosphopeptides in 1 h, 4 h and 9 h; (C) Venn diagram analyses of the down-regulated phosphopeptides in 1 h, 4 h and 9 h. Numbers in parentheses indicate the total number of phosphopeptides in 1 h, 4 h and 9 h, indicated by purple, yellow and green.

Full size image

Gene Ontology (GO) analysis

To reveal Gene Ontology (GO) distribution of phosphoproteins induced by SA in maize, functional enrichment analysis was used to assign biological relevance of these phosphoproteins using the Singular Enrichment Analysis (SEA) of AgriGO. The enriched GO annotations particularly concentrate on cellular component organization, cell part, cell, cellular process and binding (Fig. 4A, Table S4). By using the phosphoproteins whose phosphorylation levels were significantly regulated, GO analyses of up- and down- regulated phosphoproteins were performed. The 244 phosphoproteins can be classified into 13 functional categories based on their molecular function (Fig. 4B and Table S5). Compared to all phosphoproteins identified, the largest functional groups of the phosphoproteins with phosphorylation level up-regulated were unknown functional group (9.21%), nucleotide binding group (6.13%), metal ion binding group (5.19%) and a transporter activity group (2.71%). For the phosphoproteins with phosphorylation level down-regulated, the largest functional groups were catalytic activity (6.37%), unknown functional group (6.06%), nucleotide binding group (5.19%) and metal ion binding group (2.24%). The phosphoproteins could also be classified into 16 biological process groups (Fig. 4C and Table S5). The top four categories for the phosphoproteins with phosphorylation level up-regulated were metabolic processes (13.33%), unknown processes (8.28%), response to stimuli (7.43%) and regulation of biological processes (3.30%). For the phosphoproteins with phosphorylation level down-regulated, the top four categories were metabolic processes (9.55%), unknown processes (6.29%), regulation of biological processes (3.89%) and response to stimuli (3.42%).

Figure 4
figure 4

Phosphoproteins Gene Ontology (GO) distributions.

(A) GO term distribution of the parent proteins of the identified peptides to the total genome; X-axis represents the GO annotation. Y-axis represents the phosphoproteins percentage of GO annotation. Three categories of GO annotation are biological process, cellular component and molecular function. The references/background represents all protein in AgriGO database. (B) Distribution of the up and down regulated phosphoproteins based on molecular function to all phosphoproteins identified. (C) Distribution of the up and down regulated phosphoproteins based on biological process to all phosphoproteins identified.

Full size image

Otherwise, the cellular components were membrane (104, 42.62%), unknown (88, 36.07%), cytoplasm (51, 20.90%), nucleus (41, 16.80%), chloroplast (13, 5.33%), golgi (5, 2.05%), ribosome (5, 2.05%) and mitochondrion (5, 2.05%) (Fig. 5). These results indicate that diverse cellular processes and cellular components are involved in the response to SA. However, note that sometimes one protein could be assigned to more than one GO category. Therefore, the sum of percentages in each category is over 100%. These results indicate that diverse biological processes and cellular components are involved in the response to SA.

Figure 5
figure 5

Phosphoprotein distribution based on cellular components.

The percentage of differentially accumulated proteins was indicated.

Full size image

Phosphoproteins with catalytic activity

Many of the phosphoproteins that were significantly changed in phosphorylation levels in response to SA are involved in catalytic processes, including anthocyanidin, 5, 3-O-glucosyltransferase, cytokinin-O-glucosyltransferase, calcium-dependent protein kinase, glutamate decarboxylase and fructose-bisphosphate aldolase. Traditionally, we think that anthocyanins modify flower and fruit color by the production of blue anthocyanins. However, in recent years, many links between anthocyanins and plant resistance have been found14. 3-deoxyanthocyanidin has been proved to be a phytoalexin important for resistance to fungi in a wide range of plants, such as apple, pear and Sorghum bicolor15,16. 5-O-glucosyltransferase (5-UGT) is a key enzyme in anthocyanin synthesis and is also related to plant resistance. In potato cells, over-expression of 5-O-glucosyltransferase can greatly increase the content of anthocyanins and sucrose and confer significant resistance to bacterial infection17. In our study, after SA treatment, the phosphorylation level of anthocyanidin and 5, 3-O-glucosyltransferase showed a steady decline from 1 to 4 h and a sharp increase from 4 to 9 h, reaching a significant level at 9 h after SA treatment (Fig. 6A). These results indicate that the phosphorylation or dephosphorylation of anthocyanidin and 5, 3-O-glucosyltransferase may play an important role in the late response to SA treatment, maybe downstream in the signal pathway of SA.

Figure 6
figure 6

Change levels of the phosphopeptides significantly altered following SA treatment at 1 h, 4 h and 9 h.

n = 3; error bar indicates SE. (A) phosphopeptides involved in catalytic processes; (B) phosphopeptides involved in nucleotide and protein binding; (C) phosphopeptides involved in metal ion bonding.

Full size image

In contrast to anthocyanidin and 5, 3-O-glucosyltransferase, cytokinin-O-glucosyltransferase showed a steady increase from 1 to 4 h and a steady decline from 4 to 9 h, but the phosphorylation level of cytokinin-O-glucosyltransferase was lower than that in the control during the entire treatment time (Fig. 6A). This result is perhaps due to more storage forms of cytokinin-O-glucosides being converted to bioactive cytokinins after SA treatment. Cytokinins belong to a plant growth regulator group with important functions at all phases of plant development, from seed germination to senescence and also to stress response18. Exogenous cytokinins have been reported to increase tolerance to mild stress, speed up recovery, reduce the negative effects of water deficit, improve the recovery of stomatal conductance and net photosynthesis after rehydration19,20. Moreover, the transcription of many stress-induced genes can be stimulated by cytokinins21. Cytokinin-O-glucosyltransferase plays an important role in cytokinin-O-glucoside synthesis. Cytokinin-O-glucosides represent cytokinin storage forms, as they can be converted to bioactive cytokinins by the action of β-glucosidases in plants to confront adverse environmental conditions22. Over-expressed cytokinin-O-glucosyltransferase can increase the ABA and auxin content in Nicotiana tabacum and improve stress tolerance22. Over-expressed cis-cytokinin-O-glucosyltransferase in rice results in short-shoot phenotypes, delay of leaf senescence and a decrease in crown root number23. The significant regulation of phosphorylation of cytokinin-O-glucosyltransferase in this study might indicate that it’s phosphorylation or dephosphorylation may also have a physiological impact on the growth and development of maize.

Calcium-dependent protein kinase regulates the downstream components in calcium signaling pathways and is involved in crosstalk with other biotic and abiotic signaling pathways in plants24. Over-expression of an Oryza sativa calcium-dependent protein kinase, OsCPK12, promotes Oryza sativa’s tolerance to salt stress by reducing the accumulation of hydrogen peroxide (H2O2) in the leaves, but increases sensitivity to abscisic acid (ABA) and susceptibility to blast fungus. It showed that calcium-dependent protein kinase functions in multiple signaling pathways, positively regulating salt tolerance and negatively modulating blast resistance25. Recently, it is also reported that the phosphorylation level of calcium-dependent protein kinase changed significantly under drought stress in maize26. In this study, the change in phosphorylation level of calcium-dependent protein kinase is more intense than the two phosphoproteins discussed above, with a sharp increase from 1 to 4 h and a sharp decline from 4 to 9 h (Fig. 6A). These results suggest that the phosphorylation of calcium-dependent protein kinase may play a positive role at the beginning of SA treatment, but as time passes, the signal pathway collapses.

Glutamate decarboxylase (GAD) catalyzes the conversion of glutamate to γ-aminobutyric acid (GABA). Glutamate and GABA are linked with the tricarboxylic acid (TCA) cycle and are essential to the synthesis of other amino acids in plants. Therefore, the GAD enzyme may have a central role in primary metabolism and significant suppression of its expression could be lethal27. Moreover, GAD is involved in feedback controls of Ca2+-permeable channels to adjust intracellular GABA levels and thereby modulate pollen tube growth. Findings suggest that GAD activity linked with Ca2+-permeable channels relays an extracellular GABA signal and integrates multiple signal pathways to modulate tobacco pollen tube growth28. The change in phosphorylation level of GAD was similar to that of anthocyanidin 5, 3-O-glucosyltransferase, but the phosphorylation level of GAD was lower than that in the control (Fig. 6A). This result shows that SA treatment can damage the primary metabolism of maize via the phosphorylation or dephosphorylation of GAD.

Fructose 1, 6-biphosphate aldolase (FBA) is a key enzyme in plants, which is involved not only in glycolysis and gluconeogenesis in the cytoplasm, but also in the calvin cycle in plastids29. Research has shown that the over-expression of aldolase in plants increases photosynthetic rate, growth and biomass yields30. Other studies have shown that FBA plays important roles in plant growth, stress responses and development29. In addition, it is recently reported that phosphorylation level of this enzyme can be significantly increased under drought and heat combination stress26. In our study, the phosphorylation level of fructose 1, 6-biphosphate aldolase declined from 1 to 9 h, but was above control levels during the entire period of sampling (Fig. 6A). This indicates that SA treatment can damage the primary metabolism of maize and plants need more energy to repair the damage. Gluconeogenesis consumes plenty of energy thus up-regulates the phosphorylation level of the fructosebisphosphate aldolase which may inhibit gluconeogenesis for keeping energy in plants.

Nucleotide and protein binding

This study identified numerous phosphoproteins involved in nucleotide and protein binding, like the arginine serine-rich splicing factor, serine threonine-protein kinase, bZIP transcription factor and 40S ribosomal protein. Arginine serine-rich (SR) proteins constitute a conserved family of RNA-binding proteins with roles in both constitutive and alternative splicing31. The expression of SR protein under different environmental cues suggests they have an important role in plant defense and HR-like cell death32. It has been reported that the phosphorylation level of arginine SR splicing factor was significantly changed both under biotic and abiotic stresses26,33. In this study, the phosphorylation level of the arginine SR splicing factor was relatively stable from 1 to 4 h after SA treatment, but from 4 to 9 h, there was a sharp increase (Fig. 6B). These results suggest that transcription and translation may be major targets for regulatory phosphorylation during the response to SA in maize.

Threonine-protein kinase plays an important role in plant development and in the plant’s response to various unfavorable environmental conditions34. Over-expression of the threonine-protein kinase gene can enhance multistress tolerance in plants35. In wheat response to drought stress, phosphoproteome analysis reveals two threonine protein kinases36. Phosphorylation level changes of threonine protein kinases were also revealed in maize response to drought and heat combination stress26. Phosphorylattion level of serine threonine-protein kinase declined from 1 to 9 h after SA treatment (Fig. 6B). One reasonable explanation of this phenomenon was that serine threonine-protein kinase was upstream of the SA signal pathway.

The bZIP transcription factor is an important transcription factor regulating multiple biological processes including pathogen defense, responses to abiotic stresses, seed development and germination, senescence and responses to salicylic, jasmonic and abscisic acids in plants37. It has been shown that the phosphorylation of a bZIP transcription factor can act as positive regulators of ABA-responsive gene expression38. The expression of the bZIP transcription factor in rice can be rapidly induced by treating leaves with SA, suggesting that it plays a positive role in the SA-dependent signal transduction pathway for the defense of rice against pathogens39.

40S ribosomal protein is an important element of ribosomes in eukaryotes and plays important roles in protein synthesis. It is also responsible for rapid adjustments in plant growth patterns under environmental changes40. Phosphorylation levels of two ribosomal proteins were differentially regulated in maize after virus infection, with 40S ribosomal protein up-regulated and 60S ribosomal protein down-regulated33. In this study, the phosphorylation level of bZIP transcription factor showed a trend similar to that of the 40S ribosomal protein, increasing from 1 to 4 h and declining from 4 to 9 h after SA treatment. However, phosphorylation level of the bZIP transcription factor was up-regulated, whereas the 40S ribosomal protein was down-regulated, perhaps because the phosphorylation of bZIP transcription factor plays a positive role in the SA-dependent signal transduction pathway. SA can damage the protein synthesis system, so the phosphorylation of 40S ribosomal protein is down-regulated.

Metal ion bonding

In this study, we also found some phosphoproteins with phosphorylation level changed significantly involved in metal ion bonding, like ferritin, photosystem II phosphoprotein, chlorophyll a/b-binding (LHCB) proteins and calmodulin. Ferritins are molecules for iron storage and are present in most living things. In plants, ferritin is an essential iron homeostasis regulator and therefore plays a fundamental role in the control of iron induced by oxidative stress or by an excess of iron ions. Ferritin gene expression is modulated by various environmental factors, including the intensity of drought, cold, light and pathogenic attack41. It has also been reported that the phosphorylation level of ferritin was significantly changed under virus infection in maize, which indicated that the phosphorylation of ferredoxin might play a positive roles in maize for the resistance to pathogen infection33.

The primary reactions of plant and algal photosynthesis occur in the thylakoid membranes of chloroplasts. In the linear mode of electron transfer through the photosynthetic chain, photosystem II (PSII) and photosystem I (PSI) are connected in series through plastoquinone42. Phosphorylation of reaction-centre proteins affects their stability, possibly by slowing down their rate of degradation43. The light-harvesting chlorophyll a/b-binding (LHCB) proteins are the apoproteins of the light-harvesting complex of photosystem II. Down-regulation of any of the six LHCB genes results in abscisic acid (ABA)-insensitive phenotypes in seed germination and post-germination growth, demonstrating that LHCB proteins are positively involved in these developmental processes in response to ABA44. The activation of calmodulin (CaM) stimulates the DNA-binding activity of heat shock (HS) transcription factors, as well as the accumulation of HS proteins so as to confer thermotolerance. CaM can bind with target proteins to alter their function, acting as part of a calcium signal transduction pathway. In addition, the phosphorylation of CaM can lead to important physiological consequences for the cell as the diverse phosphocalmodulin species have differential actions as compared to nonphosphorylated CaM when acting on different CaM-dependent systems45.

The phosphorylation changes of the four phosphoproteins in this study were variable. Phosphorylation levels of ferritin and calmodulin were down-regulated and remained relatively stable from 1 to 9 h after SA treatment (Fig. 6C). This suggests that the phosphorylation of both ferritin and calmodulin may play a negative role in the SA signal pathyway. But the phosphorylation level of photosystem II phosphoprotein and chlorophyll a b-binding protein were up-regulated following SA treatment, showing a sharp increase from 1 to 4 h and a sharp decline from 4 to 9 h after SA treatment (Fig. 6C). This phenomenon may occur because the plant’s response to SA is an energy dissipation process and with the energy dissipation, the photosystem will collapse. Therefore, the phosphorylation level of photosystem II phosphoprotein and chlorophyll a b-binding protein declined during the late stages of the plant response to SA (Fig. 6C).

Other phosphorylated proteins

We also found that some phosphoproteins were involved in other processes. For example, the auxin efflux carrier and aquaporin PIP2-5 are involved in transporter activity and there are some phosphoproteins with unknown molecular function, like the small heat shock protein (sHSPs) 22. Division and growth of most types of plant cells require an external source of auxin. The auxin efflux carrier plays a negative role in controlling the level of auxin in plant cells and thus can affect cell division46. Aquaporin PIP2-5 is a plasma membrane intrinsic protein. Aquaporins are ubiquitously presented in living organisms including plants. It has been provided evidence for abiotic stress induced quantitative changes in aquaporin phosphorylation and its link with sub-cellular localization47. Our previous study showed that the phosphorylation level of aquaporin was up-regulated by virus infection, indicating that aquaporin phosphorylation can also be induced by biotic stress33.

Heat shock proteins are classified based on their molecular weight (HSP100, 90, 70 and 60 and sHSPs. sHSPs act as molecular chaperones and can bind thermally denatured proteins at their surface to maintain a folding-competent state48. The phosphorylation of sHSPs has been demonstrated in maize mitochondria49. In addition, recent study showed that phosphorylation levels of seven HSPs, including five small HSPs (sHSPs) and two HSP70s, changed significantly under drought and heat26. These results indicate that the phosphorylation of sHSPs appears to be important for the regulation of sHSP function in plant responses to stresses. This also suggests that the SA signal pathway in maize is a very complex biological process.

Comparison of the phosphoproteomic and proteomic changes

It will be interesting to compare the general proteomic results to this phosphoproteome. But it is a pity that we don’t have the proteomic results with the same materials and growth conditions in this study. So here we make a comparison of the phosphoproteomic and proteomic changes in response to sugarcane mosaic virus (SCMV) infection, which we have done separately before33,50. A total of seven proteins identified in the previous proteomic investigation were also detected with up- or down-phosphorylation in phosphoproteomic study. These proteins include phosphoenolpyruvate carboxykinase, serine/threonine-protein kinase, photosystem reaction center protein, eukaryotic translation initiation factor, chlorophyll a/b-binding protein, ATP synthase subunit alpha and 60S ribosomal protein. This comparison of phosphoproteomic and proteomic changes in response to SCMV infection suggests that most proteins with quantity change are not phosphoproteins or have no change in phosphorylation level. On the other hand, most differentially regulated phosphoproteins have modification level change instead of protein quantity change.

The consistency between protein expression levels and phosphorylation levels suggests that these proteins or phosphoproteins may be initially regulated at the protein expression level. Discrepancy between protein expression and protein phosphorylation levels has also been observed in several previous studies13,51,52. By way of explanation, it has been suggested that the abundance of a protein integrates phosphorylation processing, which modulates the quantity, temporal expression, localization and efficiency of the final product in the cell.

At last, one limitation of the present study should be noted. Although this study is meant to identify a number of potential protein targets of the phosphorylation/dephosphorylation machinery involved in the SA response, it cannot represent an exhaustive survey of all possible phosphorylation changes. The nature of an untargeted phosphoproteomics survey using a data dependent acquisition method on a mass spectrometer is inherently biased toward the identification of higher-abundance peptides in the sample. However, we believe that we have achieved an acceptable level of dynamic range, because we are able to identify high-abundance proteins, such as 60s ribosomal protein, as well as low-abundance transcription factors, such as bZIP. Furthermore, most of our reported phosphopeptide changes were identified in three biological samples and at more than one time point (Table 1 and Table S1). In addition, not all the phosphorylation changes in this study are biologically important. Promiscuous phosphorylation is a well-known phenomenon that occurs during in vitro assays since the concentration and location of reactants can be artificially high, thereby resulting in non-natural reactivity and phosphorylation patterns unreflective of true in plant chemistry53. Therefore, the forward or reverse genetics experiments are needed to be performed in the future to verify that these phosphorylation changes are directly involved in response to SA.

Conclusion

Protein phosphorylation has been found to play an important role in multiple cellular functions including stress/defense responses7. Phosphoproteomic studies on the molecular basis of regulatory mechanisms in plants will enhance our understanding of fundamental and complex biological processes and provide information that can be exploited for potential agricultural applications54. Our previous study identified several protein kinases responsive to SA2, suggesting that phosphorylation events play an important role in the plant response to SA. In the current study, we further characterized the phosphoproteomes of maize in response to SA treatment and provided a global analysis of protein phosphorylation regulated by SA. As we used intact plants instead of cell suspensions, we have obtained insight into some general physiological changes occurring in plants that mount a response to exogenous SA.

In summary, quantitative phosphoproteomic analysis of maize led to the identification of key phosphopeptides and phosphoproteins involved in the response to SA. We identified 858 phosphoproteins from 1495 phosphopeptides, among which 244 phosphoproteins were significantly differentially phosphorylated after SA treatment. While several of the phosphoproteins were well-known SA response phosphoproteins, many have not been reported previously. Our results also suggested that interaction with and inhibition of protein phosphatases may not be the only means of direct interaction of the SA receptor family with regulatory proteins and other currently unknown SA signaling pathways that result in decreased phosphorylation of downstream targets must be involved (Fig. 7). The functions of the differentially accumulated phosphoproteins in SA signaling can be further tested using reverse genetic approaches. It will also be very interesting to identify the kinases that phosphorylate these phosphoproteins in response to SA.

Figure 7
figure 7

Mechanistic model for SA-induced changes in protein phosphorylation.

Increased protein target phosphorylation is known to occur with activation of MAP kinases as well as from MAPK phosphatases inhibition or activation of RL kinases from an unknown phosphatases inhibition. Observed decreases in protein target phosphorylation could occur by the hypothetical activation of a phosphatase or inhibition of a kinase, which resulting from interaction with the SA receptor complex NPR1/NPR3/NPR4. Both phosphatase activation and kinase inhibition also may occur indirectly. NPR: Nonexpresser of pathogenesis-related genes; MAPK: Mitogen-activated protein kinases; RL kinases: Receptor-like protein kinases.

Full size image

Methods

Plant materials and hormonal treatments

Maize (Zea mays L.) seeds of inbred line B73 were grown in a greenhouse under a 16 h light/8 h dark cycle at 26 °C55. For hormonal treatments, the leaves of seedlings at the V4 development stage (when the fourth leaf has a visible “collar” at the base of the leaf) were sprayed with 10 ml of a solution containing 0.1% ethanol, 0.01% Tween-20 water solution and 200 μM SA (Sigma, USA). Each sample consisted of pooled leaves derived from six plants, with three biological replicates collected for each sample at 0 h, 1 h, 4 h and 9 h after treatment. Samples were stored at −80 °C for further extraction of protein or RNA.

To gain a more comprehensive understanding of the maize cellular processes in response to SA, the iTRAQ technique was performed. We quantitatively compared phosphopeptides isolated from maize leaves from SA treatment and control samples. The workflow of the analysis is shown in Supporting Information Figure S1. Briefly, after extraction from SA and control treatments using a TCA-acetone procedure, protein was digested according to the FASP procedure. Then, the resulting peptide mixture was labeled using the iTRAQ reagent according to the manufacturer’s instructions. Finally, phosphorylated peptides were enriched by TiO2 beads, followed by lyophilization and MS analysis.

Protein digestion

Total proteins were extracted from sample plants using a trichloroacetic acid/acetone procedure33. After quantification, protein digestion was performed according to the filter-aided sample preparation (FASP) procedure57. The FASP procedure is a method in which the sample is solubilized in 4% SDS, then retained and concentrated into microliter volumes in an ultrafiltration device. Notably, the presence of SDS efficiently inactivates detrimental enzymatic functions, such as protease and phosphatase activity57. The resulting peptide mixture was labeled using 4-plex iTRAQ reagent according to the manufacturer’s instructions (Applied Biosystems). Briefly, 200 μg of proteins from each sample were incorporated into 30 μl STD (4% SDS, 100 mM DTT and 150 mM Tris-HCl pH 8.0). The detergent, DTT and other low-molecular-weight components were removed using UA buffer (8 M urea and 150 mM tris-HCl, pH 8.0) by repeated ultrafiltration (Microcon units, 30 kDa). To block reduced cysteine residues, 100 μl iodoacetamide (0.05 M) in UA buffer was then added and the samples were incubated for 20 min in darkness. Filters were washed three times with 100 μl UA buffer and then twice with 100 μl DS buffer (50 mM triethylammonium bicarbonate at pH 8.5). Finally, the protein suspensions were digested with 2 μg trypsin (Promega) in 40 μl DS buffer overnight at 37 °C with the resulting peptides collected as a filtrate.

iTRAQ labeling

Peptide content was estimated by UV light spectral density at 280 nm using an extinction coefficient of 1.1 of 0.1% (g/l) solution that was calculated on the basis of the frequency of tryptophan and tyrosine in vertebrate proteins. For labeling, each iTRAQ reagent was dissolved in 70 μl ethanol and added to the respective peptide mixture. The samples, labeled as 114 (1 h), 115 (4 h), 116 (9 h) and 117 (0 h), were multiplexed and vacuum dried.

Enrichment of phosphorylated peptides using TiO2 beads

Labeled peptides were mixed, vacuum-concentrated and resuspended in 500 μl loading buffer [2% glutamic acid, 65% acetonitrile (ACN) and 2% TFA]. TiO2 beads (GL Sciences, Japan) were then added and the mixture was agitated for 40 min. Centrifugation was carried out for 1 min at 5000 g and the precipitated beads were set aside. The supernatant was mixed with another set of TiO2 beads and centrifugation and bead collection was carried out a second time. Beads from the two centrifugation rounds were combined; they were washed three times with 50 μl of washing buffer I (30% ACN and 3% TFA) and three times with 50 μl of washing buffer II (80% ACN and 0.3% TFA) to remove remaining non-adsorbed material. The phosphopeptides were then eluted with 50 μl elution buffer (40% ACN and 15% NH4OH), followed by lyophilization and MS analysis.

Mass spectrometry

Five microliters of the phosphopeptide solution were mixed with 15 μl of 0.1% (v/v) TFA; 10 μl of this mixture was analyzed for nanoLC-MS/MS using a Q Exactive mass spectrometer (Thermo Fisher Scientific) equipped with an Easy nLC HPLC (Proxeon Biosystems, now Thermo Fisher Scientific). The peptide mixture was loaded onto a C18-reversed phase column (15 cm long, 75 μm inner diameter, RP-C18 3 μm, packed in-house) in buffer A (0.1% formic acid) and separated with a linear gradient of buffer B (80% acetonitrile and 0.1% formic acid) at a flow rate of 250 nl min−1 controlled by IntelliFlow technology over 240 min. The peptides were eluted with a gradient of 0%–60% buffer B from 0 min to 200 min, 60% to 100% buffer B from 200 min to 216 min and 100% buffer B from 216 min to 240 min.

For MS analysis, peptides were analyzed in positive ion mode. MS spectra were acquired using a data-dependent top ten method dynamically choosing the most abundant precursor ions from the survey scan (300–1800 m/z) for HCD fragmentation. The range of charge is from +2 to +6. Determination of the target value was based on predictive automatic gain control. The dynamic exclusion duration was 40 s. Survey scans were acquired at a resolution of 70,000 at m/z 200 and resolution for HCD spectra was set to 17,500 at m/z 200. The normalized collision energy was 27 eV and the under fill ratio, which specifies the minimum percentage of the target value likely to be reached at maximum fill time, was defined as 0.1%. The instrument was run with peptide recognition mode enabled.

Data analysis

MS/MS spectra were searched using Mascot 2.2 (Matrix Science) embedded in Proteome Discoverer 1.4 against the UniProt_Poales, FASTA (426891 sequences) and decoy databases. For protein identification, the following options were used: peptide mass tolerance = 20 ppm, MS/MS tolerance = 0.1 Da, enzyme = trypsin and missed cleavage = 2; carbamidomethyl (C), iTRAQ4/4plex (K) and iTRAQ4/4plex (N-term) were set as fixed modifications; oxidation (M), phosphorylation (S/T/Y) and false discovery rate (FDR) ≤0.01 were set as variable modifications. Proteome Discoverer 1.4 was used to extract the peak intensity within 20 ppm of each expected iTRAQ reporter ion from each fragmentation spectrum. Only spectra in which all the expected iTRAQ reporter ions were detected were used for quantification. The phosphopeptide ratios were normalized by dividing by the average value of all peptides identified. Phosphorylated peptides were analyzed using Proteome Discoverer 1.4 (Thermo Electron, San Jose, CA, USA) with the score threshold for peptide identification set at a 1% FDR and with PhosphoRS site probability cutoffs of 0.75. Phosphopeptide quantification was carried out in three biological replicates. Student’s T test was used to evaluate the statistical significance and the false discovery rate (Benjamini-Hochberg) was calculated to correct for multiple comparisons. To state that a Phosphopeptide has a significant abundance changes, the following criteria have to be fulfilled: the abundance ratios has to be ≥1.5 and the P-value for student’s t test has to be less than 0.05.

Bioinformatics

Molecular functions of identified phosphoproteins were classified according to their gene ontology (GO) annotations combined with their biological function. Subcellular locations of unique phosphoproteins identified in this study were determined from the UniProt database (http://www.uniprot.org) or predicted using the publicly available program, WolfPsort (http://wolfpsort.org).

Additional Information

How to cite this article: Wu, L. et al. Quantitative analysis of changes in the phosphoproteome of maize induced by the plant hormone salicylic acid. Sci. Rep. 5, 18155; doi: 10.1038/srep18155 (2015).