Elevated CO2 concentration promotes photosynthesis of grape (Vitis vinifera L. cv. ‘Pinot noir’) plantlet in vitro by regulating RbcS and Rca revealed by proteomic and transcriptomic profiles

Background Plant photosynthesis can be improved by elevated CO2 concentration (eCO2). In vitro growth under CO2 enriched environment can lead to greater biomass accumulation than the conventional in micropropagation. However, little is know about how eCO2 promotes transformation of grape plantlets in vitro from heterotrophic to autotrophic. In addition, how photosynthesis-related genes and their proteins are expressed under eCO2 and the mechanisms of how eCO2 regulates RbcS, Rca and their proteins have not been reported. Results Grape (Vitis vinifera L. cv. ‘Pinot Noir’) plantlets in vitro were cultured with 2% sucrose designated as control (CK), with eCO2 (1000 μmol·mol− 1) as C0, with both 2% sucrose and eCO2 as Cs. Here, transcriptomic and proteomic profiles associated with photosynthesis and growth in leaves of V. vinifera at different CO2 concentration were analyzed. A total of 1814 genes (465 up-regulated and 1349 down-regulated) and 172 proteins (80 up-regulated and 97 down-regulated) were significantly differentially expressed in eCO2 compared to CK. Photosynthesis-antenna, photosynthesis and metabolism pathways were enriched based on GO and KEGG. Simultaneously, 9, 6 and 48 proteins were involved in the three pathways, respectively. The leaf area, plantlet height, qP, ΦPSII and ETR increased under eCO2, whereas Fv/Fm and NPQ decreased. Changes of these physiological indexes are related to the function of DEPs. After combined analysis of proteomic and transcriptomic, the results make clear that eCO2 have different effects on gene transcription and translation. RbcS was not correlated with its mRNA level, suggesting that the change in the amount of RbcS is regulated at their transcript levels by eCO2. However, Rca was negatively correlated with its mRNA level, it is suggested that the change in the amount of its corresponding protein is regulated at their translation levels by eCO2. Conclusions Transcriptomic, proteomic and physiological analysis were used to evaluate eCO2 effects on photosynthesis. The eCO2 triggered the RbcS and Rca up-regulated, thus promoting photosynthesis and then advancing transformation of grape plantlets from heterotrophic to autotrophic. This research will helpful to understand the influence of eCO2 on plant growth and promote reveal the mechanism of plant transformation from heterotrophic to autotrophic. Electronic supplementary material The online version of this article (10.1186/s12870-019-1644-y) contains supplementary material, which is available to authorized users.


Background
Increasing atmospheric CO 2 concentration influences plant growth [1,2]. Photosynthesis, respiration and water relations are the three primary physiological processes influenced by elevated CO 2 concentration (eCO 2 ) in plants [3]. CO 2 concentration inside the culture vessels decreased when plantlets grown in vitro, which limits the photosynthetic rate of the plants [4,5]. The biomass accumulation of the in vitro cultured plants increased under photoautotrophic and CO 2 enrichment conditions, also affected nutrient absorption and secondary metabolism [6,7].
Plantlet grown vigorously under CO 2 enriched photoautotrophic and photomixotrophic conditions, with high photosynthetic photon flux density [8]. Photosynthetic response to light and CO 2 increased with Rubisco activities and proteins of plantlets grown in vitro [7]. Rubisco, the main catalytic enzyme determines photosynthetic rate [8], would respond to eCO 2 [9] and increase carboxylation efficiency under eCO 2 [10]. Succinctly, the synthesis of the Rubisco holoenzyme is mainly affected by ribulose bisphosphate carboxylase small chain (RbcS) [11]. The activity of Rubisco is related to Rubisco activase (Rca) and other proteins [12,13].
In some species, it is reported that the transcript levels of RbcS are differentially regulated by red and blue light or growth temperature [14]. The abundance of the RbcS multigene family transcript has been researched in many plants [15]. RbcS regulates Rubisco through coordinated expression of RbcL and RbcS in plants [11]. In addition to the folded RbcL subunits assemble [16], RbcS could combine more CO 2 than the RbcL in all Rubiscos [17]. The detailed mechanism of RbcS mediated assembly of RbcL under different environment and how the expression of RbcS and its protein responds to eCO 2 remains to be investigated. The Rca could gain energy from ATP hydrolysis to remodel Rubisco inhibitors and activate Rubisco [18]. Inhibit expression of Rca in some plants results in severe photoautotrophic growth defects [19]. Rca proteins belong to a subgroup of the ATPases associated with various cellular activities (AAA) called AAA + [20]. There are two Rca forms both can activate Rubisco [21]. Rca is regulated by the intracellular ATP/ ADP ratio [22] or the C-terminal extension of the α-isoform of Rca in some plants [18]. Some research indicated that Rca could reduce the effects of abiotic stresses on plants, such as high temperature, drought, salt [23][24][25] and heavy metal [26]. The expression of Rca is regulated by trans-acting factors in soybean [27]. The actual change mechanism of Rca expression and whether Rca related to other proteins under eCO 2 is less studied.
'Pinot Noir' is a wine grape variety widely planted in worldwide and its growth influenced by various environmental factors [28]. The increasing CO 2 concentration could promote plant growth. Although, many studies have focused on the effects of CO 2 on grape ripening [29] and postharvest [30]. It is unclear the mechanism of how eCO 2 affects the plant growth and photosynthesis. Additionally, there are a few reports on the analysis of transcriptome combined with proteome to study the effects of eCO 2 on grape growth and development. In light of this situation, the experiment was conducted based on the hypothesis that eCO 2 will enhance photosynthesis by regulating the expression of related genes and proteins in grape plantlets. Therefore, grape plantlets grown in vitro cultured with eCO 2 were used in this study based on transcriptome, proteome and photosynthetic physiology analysis.

Effects of eCO 2 on growth and chlorophyll fluorescence
Grape plantlets were cultured for 25 days at 1000 μmol· mol − 1 of CO 2 and compare with control conditions. The results showed that the leaf area, plantlet height and shoot fresh weight increased significantly in Cs and C0 compared with CK (Additional file 1: Table S1). In addition, the number of adventitious roots in tubers also was increased in Cs and C0 (Fig. 1a).
In compared to CK, the Fv/Fm decreased in Cs and C0, and significantly lower in C0 than that of CK (Fig. 1b, c). The qP and ETR rised in Cs and C0 (Fig. 1d, e). The ETR of C0 was significantly higher than the Cs and CK. The qP of C0 was significantly higher than Cs and CK (Fig. 1d). The NPQ was in the following order: Cs < C0 < CK (Fig. 1d). The decrease of NPQ indicated that eCO 2 enhanced the efficiency of PSII and reduced the damage caused by biotic and abiotic stress. The ΦPSII of Cs and C0 were significantly higher than CK (Fig. 1d). These results suggested that eCO 2 improved photosynthesis and reflected by chlorophyll fluorescence parameters, including Fv/Fm, ETR, qP, NPQ and ΦPSII.

Transcriptome and proteome differences expression in eCO 2
In the transcriptome project, three RNA-Seq groups with three replications were sequenced, 29.5Gb clean bases were generated from the 9 libraries. After data processing, 46.49-47.46 million high-quality reads were obtained (Table 1). Through transcriptome analysis, a total of 1814 DEGs were observed by comparing with CK, of which 116 up-regulated and 632 down-regulated DEGs were identified in Cs versus CK, 349 up-regulated and 717 down-regulated DEGs were identified in C0 versus CK (Fig. 2a). According to SDS-PAGE analysis, protein sample could be tested in the next step (Additional file 2: Figure S1). After analysis of proteomic profiling, a total of 177 DEPs were observed from the pooled data for above two comparison groups. Among them, 48 up-regulated DEPs and 67 down-regulated DEPs were identified in Cs versus CK, 32 up-regulated DEPs and 30 down-regulated DEPs were identified in C0 versus CK (Fig. 2b).

GO analysis of DEGs and DEPs
Of the 25,679 genes identified in the transcriptome analysis, 17,750 genes (69.12%) were annotated via GO analysis. Compared with CK, 748 DEGs identified in Cs were enriched in the biological process (BP), cellular component (CC), and molecular function (MF) categories. In the cellular components category, most of DEGs were involved in integral component of membrane (99 genes) and cytoplasm (94 genes). In the biological process category, most of DEGs were involved in defense response (70 genes) and transcription, DNA-templated (59 genes). In the molecular function category, most of DEGs were involved in transcription factor activity, sequence-specific DNA binding (54 genes) and ATP binding (48 genes) (Additional file 3: Table S2 A).
The 1066 DEGs of C0 versus CK were detected. Most DEGs mainly enriched in cytoplasm (98 genes) and integral component (97 genes) of cellular components. In the biological process category, most of DEGs were involved in defense response (53 genes) and transcription, DNA-templated (48 genes). In the molecular function category, most of DEGs were involved in metal ion binding (52 genes) and transcription factor activity, sequence-specific DNA binding (45 genes) (Additional file 3: Table S2 B).

KEGG pathway analysis for DEPs
To further investigate the plant reaction to eCO 2 , DEPs were identified by searching the KEGG database. The 115 DEPs of Cs were assigned to 52 KEGG pathways, and the top 5 pathways with the highest rich factor were photosynthesis-antenna proteins, metabolic pathways, biosynthesis of secondary metabolites, carbon metabolism, biosynthesis of amino acids (Additional file 5: Table  S4 A). The 62 DEPs of C0 were assigned to 20 KEGG pathways, and the top 5 pathways with the highest rich factor were photosynthesis-antenna proteins, photosynthesis, metabolic pathways, phenylpropanoid biosynthesis, phenylalanine, tyrosine and tryptophan biosynthesis (Additional file 5: Table S4 B). The common pathways with the highest rich factor of Cs versus CK and C0 versus CK were photosynthesis-antenna proteins, photosynthesis and metabolic pathways. Simultaneously, 9, 6 and 48 proteins were involved in the three pathways, respectively. Moreover, 12 proteins involved in metabolic pathway were overlaps with photosynthesis (Table 4).

Combined analysis of transcriptome and proteome data
To reveal eCO 2 regulates photosynthesis gene via transcript and protein levels, the transcript data were used to analyze 18 DEPs associated with photosynthesis and metabolic pathways.   (Figs. 3, 4). The RbcS (XP_002276967.1) and its corresponding gene were up-regulated in Cs and C0. ATP synthase delta chain (XP_002274963.1) and Rca (XP_0022 82979.1) were up-regulated but their corresponding genes were down-regulated in Cs and C0 (Figs. 3, 4).
The results make clear that eCO 2 have different effects on gene transcription and translation. RbcS was not correlated with its mRNA level, suggesting that the change in the amount of RbcS is regulated at their transcript levels by eCO 2 . However, Rca was negatively correlated with its mRNA level, it is suggested that the change in the amount of its corresponding protein is regulated at their translation levels by eCO 2 .

Confirmation of qRT-PCR
In order to evaluate our transcriptome-sequencing data, 18 genes in the photosynthesis and metabolic pathway were selected for qRT-PCR. The results analyse indicated that 15 genes (83.33%) showed similar trends in the relative expression levels, which suggested that the gene expression changes detected by transcriptome-sequencing analysis were reliable. But 3 genes (16.67%) analyzed by qRT-PCR, i.e., PsbE (4025054), PsbD (4025083) and LHCB3 (LOC100252004) were not consistent with our RNA-seq data (Fig. 5).

Discussion
Proteins involved in photosynthesis were regulated by eCO 2 Light-harvesting complexes (LHC) of photosynthetic plant bind pigments essential for augmenting light capture and photoprotection [31]. LHCI and LHCII belong to photosystem I (PSI) and photosystem II (PSII), respectively. LHCII is a trimeric light-harvesting complex (Lhc) composed of a combination of the Lhcb gene products and others [32]. Plants can develop strategies of acclimating varies light conditions during seasons and can rapidly adjust photosynthesis antenna sizes in case of excess light, avoiding over excitation and formation of harmful by products [33]. CO 2 concentration could affect the primary light reaction of photosynthesis in soybean leaves [34]. In our research, 8 proteins of LHCII were up-regulated in eCO 2 (Table 4), this change indicated that eCO 2 could induce more light-harvesting proteins (Fig. 3), and cause an increase in the size of the PSI and PSII antenna. The light-harvesting complex II (LHCII) could convert most photons to biochemical energy and biomass [35]. With the increase of LHCII in eCO 2 , more light energy can be absorbed and converted into photosystem. The increase of qP and decrease of NPQ confirming that leaves could absorb more light energy under eCO 2 (Fig. 1d). In present study, the number of up-regulated light-harvesting proteins of PSII was more than that of PSI, which showed that the eCO 2 had a great influence on PSII. Additionally, the LHCII conditions would migration from PSII to PSI under deficient CO 2 environment [36]. CP24 was up-regulated in eCO 2 , it was essential for connecting LHCII to the PSII complex [37,38]. Lack of the light-harvesting complex CP24 affects the structure and function of the grana membranes of higher plant chloroplasts [37]. Overall, these proteins, which were up-regulated under eCO 2 , could absorb and convert more light energy into the photosystem.
In photosynthesis pathway, the expression of PetE (XP_002285904.1) and Chlorophyll a-b binding protein (XP_003633024.1) were descend in Cs. Interestingly, the expression of PetE (LOC100248911) and LHCB3 (LOC100252004) were ascend in Cs and C0. These results indicated that most of the DEPs and their corresponding genes expression were inconsistent. The eCO 2 may cause various modifications of related proteins after translation and needs to be study for further.

eCO 2 regulates metabolic protein expression
There were 48 DEPs involved in metabolic pathway, while 12 of them were overlaps with photosynthesis. This might indicate that eCO 2 would affect other metabolic through adjusting photosynthesis. Our results indicated many of down-regulated DEPs were enriched in metabolic pathway in eCO 2 , which were related to biosynthesis of secondary metabolites (Table 4). This change is suggesting that eCO 2 probably decreased biosynthesis of secondary metabolites [39]. Therefore, plant could accumulation more primary metabolism products to encourage growth.
The eCO 2 could ameliorate the effects caused by drought [40], high temperatures [41], and maintaining higher photosynthetic rates. This may be linked to the reduction in stomatal conductance [42]. Moreover, increasing photosystem antenna size must inevitably cause structural changes needed to ensure high efficiency of its functioning [43]. There were 4 DEPs (PsbQ, PsbE, PsbD and PsaN; Fig. 3) up-regulated in eCO 2 . Those proteins could maintain the stability of the photosystem reaction center [44,45]. By analyzing changes of those proteins in eCO 2 , we can conclude that eCO 2 could trigger some proteins to maintain the stability of the photosynthesis system. Therefore, eCO 2 could ameliorate the adverse effect under abiotic stress. PsbQ can increase PSII activity and stability of oxygen release complexes (OECs) [45]. It is also the water decomposition subunit [46]. The PsbQ was up-regulated in eCO 2 , this means eCO 2 could promote water decomposition and maintain stability in OECs by regulating PsbQ. The other 3 proteins (PsbE, PsbD and PsaN) related to photosynthetic electron transport and accumulation of photosynthetic substances [44]. Those proteins (LHCs, PsbQ, PetE, PsbD, PsaN) increased in eCO 2 (Fig. 3), resulting in absorbing more light energy and promoting more photosynthetic electron transport. This is causing the advance of qP and ETR, and the reduction of NPQ (Fig. 1d, e).
ATP synthase delta chain is CF 1 subunit (δ) belongs to the F-type ATPase, which utilizes the energy of a transmembrane electrochemical gradient to generate ATP by rotary catalysis [47]. F-type ATPase products would provide energy for photosynthesis carbon fixation [48]. ATP synthesis in the hydrophilic α 3 β 3 head (CF 1 ) is powered by the CF 0 rotary motor in the membrane [49]. Previous studies have shown that the ATP synthase delta chain is mainly related to the component linkage of the F-type ATPase sector [50,51]. In our study, ATP synthase delta chain protein was up-regulated in eCO 2 (Fig. 3), this indicated that eCO 2 can affect leaf redox pathways by changing the F-type ATPase subunit accumulation. Our results confirmed that ATP synthase delta chain act as a stator to prevent unproductive rotation of CF 1 with CF 0 , this is consistent with previous study [49].

eCO 2 promotes up-regulation of RbcS and Rca
Rubisco is L 8 S 8 hexadecamer complex [52] and inefficient [53]. RbcS regulates Rubisco through coordinated expression of RbcL and RbcS in plants [11]. RbcS is linked to the folded RbcL subunits assemble [54] and as a 'reservoir' for CO 2 storage [17]. In our results, RbcS was up-regulated in eCO 2 (Fig. 4), this indicated that RbcS not only has high affinity with CO 2 , but also responds to eCO 2 in the environment. It has reported that RbcS mRNA levels and RbcS synthesis simultaneously increased in RbcS-sense plants [11]. The RbcS transcript was found to be inhibited in source of sugar (sucrose or glucose) in the media of photoautotrophic Chenopodium callus and some plants, but over-expression of RbcS was found in low CO 2 [55]. Interestingly, RbcS mRNA level was up-regulated in C0, and down-regulated in Cs and CK, which indicate that the medium with sugar inhibits the expression of RbcS, this is consistent with previous studies. The amount of RbcS synthesize was tightly correlated with RbcL mRNA level [11]. In our research, large amounts of RbcS accumulated under eCO 2 but there was no significant change in RbcL mRNA level. This result showed that RbcS accumulated not only associated with RbcL mRNA level, but also related to CO 2 concentration. It has reported that long-term growth of Arabidopsis at high CO 2 (1000 μmol·mol − 1 ) resulted in nonstructural carbohydrates increased and an even greater decline in mRNA of RbcS [56]. Nevertheless, the mechanism of eCO 2 regulates RbcS accumulated would research in future.
Sugar phosphate inhibits Rubisco activity [57], such as RuBP, CATP and Xu5P [12]. Rca catalyzes the remodeling of inactive Rubisco, releases it's bound sugar phosphate and activate Rubisco [20]. Heat [23], drought [24] and salt [25] could increase Rca. In our results, Rca was up-regulated under eCO 2 . Through the previous analysis, LHCII, PsbQ, PsbE, PsbD, PsaN and ATP synthase delta chain were up-regulated, indicating that these proteins would absorb more energy and produce more ATP, which could change ATP/ADP ratio. Rca uses the hydrolysis of ATP to facilitate the dissociation of RuBP bound as an inhibitor at the active site of uncarbamylated and inactive Rubisco [58]. Therefore, the activity of Rca was affected by ADP/ATP ratio [22]. We speculated that eCO 2 affect Rca activity by up-regulating the expression of light-harvesting proteins and F-type ATPase, and all of those changes ultimately affect the activation of Rubisco. Galmés et al. [59] reported that Rubisco content reduced was the primary driver in the regulation of Rubisco activity to eCO 2 . At normal conditions, Rca negatively affects the Rubisco content [60]. However, Rca level is a major limiting factor of non-steady-state photosynthesis [61]. Therefore, Rca up-regulated to adjust the non-steady-state photosynthesis caused by eCO 2 . Overproduction of Rubisco does not enhance photorespiration as well as CO 2 assimilation probably due to partial deactivation of Rubisco [62]. Rca was negatively correlated with mRNA levels, it is suggested that changes in the expression of these proteins are regulated at their translation levels by eCO 2 .

Conclusions
The detailed analysis of transcriptome and proteome of grape (V. vinifera L. cv. 'Pinot Noir') plantlets in vitro under differential concentration of CO 2 revealed crucial molecular mechanism difference in transformation from heterotrophic to autotrophic. The results indicated that eCO 2 triggers the RbcS and Rca up-regulated, then promoting photosynthesis and then advancing transformation of grape plantlets from heterotrophic to autotrophic. The study provided deep refinements into the existing knowledge of plantlets in vitro response to eCO 2 , and the molecular mechanism was revealed through identification and comparative analysis of genes and proteins from photosynthesis-antenna, photosynthesis and metabolism pathways. The expression level of RbcS was not related to protein expression and the expression of Rca was highly inverse correlated with protein expression. Consequently, these datas provide clues as to the fundamental regulatory network targeted by eCO 2 , and will lead to future functional analyses that may be valuable for both agronomic improvement and our understanding of the means by which new phenotypes may arise.

Plant materials
'Pinot Noir' (V. vinifera L.) grape plantlets, which was kept in the Fruit Tree Physiology and Biotechnology Laboratory, College of Horticulture, Gansu Agricultural University, were used as test materials in an in vitro experiment. The grape plantlets were propagated in advance and were vigorous in growth without contamination. Each nodal segment (approximately 2.0 cm long) with two bud was cultured on modified B5 solid medium + IAA (0.1 mg·L − 1 ) (50 mL of medium was taken in 150 mL Erlenmeyer flasks). Plantlets were grown in controlled climate chamber (PQX-430D) at a day/night regime of 16 h/8 h (light/dark), an irradiance of 120 μmol·m − 2 ·s − 1 , temperatures of 26°C day and night. One climate chamber (PQX-430D-CO 2 ) have TC-5000 (T) intelligent CO 2 controller to regulate CO 2 concentration. The CO 2 concentration treatments were as follows: environmental atmospheric CO 2 concentrations (380 ± 40 μmol·mol − 1 ); and elevated CO 2 concentrations (1000 μmol·mol − 1 ). The grape plantlets were cultured with 2% sucrose designated as control (CK), with eCO 2 while without sucrose as C0, with both 2% sucrose and eCO 2 as Cs. Each treatment had three biological replicates with 15 plantlets per replicate. Plantlet leaves were harvested at 25 days after inoculation.
The leaf samples were transferred immediately to liquid nitrogen and stored at − 80°C for subsequent analysis. Different treatments were simultaneously sampled from three comparable plants used as three biological replications.

RNA isolation and library preparation for transcriptome analysis
Total RNA samples were extracted using the mirVana miRNA Isolation Kit (Ambion). The RNA samples were evaluated using the Agilent 2100 Bioanalyzer (Agilent Technologies, Santa Clara, CA, USA), with RNA Integrity Number (RIN) ≥ 7 were subjected to the subsequent analysis. The libraries were constructed using TruSeq Stranded mRNA LTSample Prep Kit (Illumina, San Diego, CA, USA) according to the manufacturer's instructions. Then these libraries were sequenced on the Illumina sequencing platform (HiSeqTM 2500 or Illumina HiSeq X Ten) and 125 bp/150 bp paired-end reads were generated.

Analysis of RNA-sequencing data
Raw data (raw reads) were filtered into clean reads using NGS QC Toolkit. The reads containing ploy-N and the low quality reads were removed to obtain the clean reads. Then the clean reads were mapped to reference genome sequence (http://www.genoscope.cns.fr/externe/ GenomeBrowser/Vitis/) using HISTA 2. Briefly, the number of mapped reads for each transcript was normalized into a reads per kb per million reads value (RPKM) to calculate level of differential expression for each transcript. In analysis, a criterion of P value < 0.05 and fold change > 2 or fold change < 0.5 was used to identify DEG. Functional gene classification was performed using UniProtKB/Swiss-Prot database. GO enrichment and KEGG pathway enrichment analysis of DEGs were performed using the R programming language based on the hypergeometric distribution, respectively.

qRT-PCR analysis
One micrograms total RNA was subjected to reverse transcription using SYBR Green PCR Master Mix (TaKaRa) Kit with gDNA Eraser (Perfect for Real Time). Real-time PCR was carried out by using SYBRs Premix Ex Taq II (TaKaRa) in ABI StepOne™ Plus Real-Time PCR System (Roche, Switzerland). All primers used for qRT-PCR were listed in Additional file 6: Table S5.

Protein extraction
Fresh leaves (0.5 g) from each biological replicate were ground into power in liquid nitrogen and dissolved (vortex blending) with 500 μL extraction buffer (0.7 M sucrose, 0.1 M NaCl, 0.5 M Tris-HCl (pH 7.5), 50 mM EDTA and 0.2% DTT). The samples were grinded at the power of 60 Hz for 2 min. Then supplemented with extraction buffer for 1 mL and mixed and added with Tris-phenol buffer and mixed for 30 min at 4°C. The mixtures were centrifuged at 7100 g for 10 min at 4°C. Collect phenol supernatants and added for 5 volumes of 0.1 M cold ammonium acetate-methanol buffer and precipitated at − 20°C overnight. The samples were centrifuged at 12,000 g for 10 min to collect precipitations. The precipitations were dried and dissolved in lysis buffer (1% DTT, 2% SDS, 10% glycerinum, 50 mM Tris-HCl (pH 6.8) for 3 h. The samples were centrifuged at 12000 g for 10 min to collect supernatants. The supernatants were centrifuged again to remove precipitations completely. The protein concentration was quantified by BCA method [64] and theprotein purity was detected by SDS-PAGE [65], 15μg proteins of each sample were separated on 12% SDS-PAGE gel.

Protein digestion and iTRAQ labeling
Protein digestion was performed according to the FASP procedure [66]. Brifely, protein sample (100 μg) was subjected with 120 μL reducing buffer (10 mM DTT, 8 M Urea, 100 mM TEAB, pH 8.0) on 10 K ultrafiltration tube and the solution was incubated at 60°C for 1 h. IAA was added to the solution with the final concentration of 50 mM in the dark at room temperature for 40 min. The solutions were centrifuged on the filters at 12,000 g for 20 min at 4°C. Remove the supernatant and add TEAB (100 μL, 100 mM) to the solutions and centrifuged at 12,000 g for 20 min. Collection the filter units into new tubes, add TEAB (100 μL, 100 mM) and followed with 2 μL sequencing-grade trypsin (1 μg·μL − 1 ), incubated for digestion at 37°C for 12 h. The collections of digested peptides were centrifuge at 12,000 g for 20 min. The solutions were collected and lyophilized. The lyophilized samples were resuspended in TEAB (100 μL, 50 mM) and 40 μL of each sample was transferred into new tubes for labeling. Each sample add iTRAQ label reagent (iTRAQ® Reagents-8plex kit, Sigma) following the manufacturer's protocol (Applied Biosystems, Foster City, CA, USA). All labeled peptides were pooled together.

Protein identification and function annotation
Raw data of iTRAQ-labeled proteins by was search against V. vinifera (Grape) genome protein database in National Center for Biotechnology Information (NCBI) using the Proteome DiscovererTM 2.2 (Thermo, USA). Database searches were performed with trypsin digestion specificity, and the cysteine alkylation was considered as parameters in the database searching. For protein quantification method, iTRAQ8-plex was selected. For protein identification, a decoy database search approach