Transcriptomic and phosphoproteomic profiling and metabolite analyses reveal the mechanism of NaHCO3-induced organic acid secretion in grapevine roots

Background Organic acid secretion is a widespread physiological response of plants to alkalinity. However, the characteristics and underlying mechanism of the alkali-induced secretion of organic acids are poorly understood. Results Oxalate was the main organic acid synthesized and secreted in grapevine (a hybrid of Vitis amurensis, V. berlandieri and V. riparia) roots, while acetate synthesis and malate secretion were also promoted under NaHCO3 stress. NaHCO3 stress enhanced the H+ efflux rate of grapevine roots, which is related to the plasma membrane H+-ATPase activity. Transcriptomic profiling revealed that carbohydrate metabolism was the most significantly altered biological process under NaHCO3 stress; a total of seven genes related to organic acid metabolism were significantly altered, including two phosphoenolpyruvate carboxylases and phosphoenolpyruvate carboxylase kinases. Additionally, the expression levels of five ATP-binding cassette transporters, particularly ATP-binding cassette B19, and two Al-activated malate transporter 2 s were substantially upregulated by NaHCO3 stress. Phosphoproteomic profiling demonstrated that the altered phosphoproteins were primarily related to binding, catalytic activity and transporter activity in the context of their molecular functions. The phosphorylation levels of phosphoenolpyruvate carboxylase 3, two plasma membrane H+-ATPases 4 and ATP-binding cassette B19 and pleiotropic drug resistance 12 were significantly increased. Additionally, the inhibition of ethylene synthesis and perception completely blocked NaHCO3-induced organic acid secretion, while the inhibition of indoleacetic acid synthesis reduced NaHCO3-induced organic acid secretion. Conclusions Our results demonstrated that oxalate was the main organic acid produced under alkali stress and revealed the necessity of ethylene in mediating organic acid secretion. Additionally, we further identified several candidate genes and phosphoproteins responsible for organic acid metabolism and secretion. Electronic supplementary material The online version of this article (10.1186/s12870-019-1990-9) contains supplementary material, which is available to authorized users.


Background
Soil alkalinity is an important environmental problem, and alkali stress primarily caused by NaHCO 3 and Na 2 CO 3 severely affects crop growth and development in more than 434 million hectares of land worldwide [1]. Compared with neutral salt stress, alkali stress not only causes osmotic stress and ion injury but also leads to high pH injury [2,3]; therefore, alkali stress is more destructive than salt stress [4]. However, in contrast to the extensive studies on plant salt tolerance, much less attention has been focused on exploring the mechanisms underlying alkali stress tolerance.
Under alkali stress, plants must regulate intracellular pH and the pH outside roots to maintain root functions. Organic acids play a key role in regulating the cell and rhizosphere pH levels. It was reported that organic acid metabolism is closely correlated with alkali stress tolerance [4]. Puccinellia tenuiflora roots accumulate and secrete citric acid into the rhizosphere in response to alkali stress [5]. HCO 3 − induces the production of malate, succinate, and citrate in rice roots [6] and malate in maize roots [7]. Additionally, organic acids have been reported to function in plant adaptation to other abiotic stresses. For example, cotton and alfalfa accumulate more citric acid under drought and salt stresses, respectively [8,9]. Plants accumulate organic acids in cells and secrete them into rhizospheres when facing Al, P and Cd toxicity [10][11][12]. With regard to the mechanisms of alkali-induced organic acid synthesis and secretion, some key enzymes involved in organic acid metabolism, such as citrate synthase, malate synthase and isocitrate lyase, have been suggested to determine alkali stress tolerance [13]. Plasma membrane H + -ATPases, such as Arabidopsis AHA2 and AHA7, are necessary for proton secretion for plant tolerance to alkali stress [11,14]. Additionally, some studies have demonstrated that ethylene and auxin have roles in the regulation of root H + secretion and alkali stress by regulating H + -ATPase [15,16].
To date, the pathways of organic acid metabolism in the cell and extrusion out of the cell under alkali stress remain largely unknown in grapevine. Genome-scale analysis of gene expression profiles is a powerful method to reveal plant abiotic stress tolerance mechanisms [17,18]. For example, transcriptome profiling reveals the genetic basis of alkalinity tolerance in wheat [18] and gene networks responsive to NaHCO 3 stress in Tamarix hispida [17]. Quantitative and comprehensive insights into the mRNA transcriptome will contribute to unraveling the key mechanism of organic acid secretion in response to alkali treatment. On the other hand, phosphorylation represents one of the most important posttranslational modification functions in diverse biological pathways. The activities of several proteins related to organic acid synthesis and extrusion, such as phosphoenolpyruvate carboxylase (PEPC) and H + -ATPase [19,20], have been shown to be regulated by phosphorylation. An analysis of the comprehensive phosphorylation modifications induced by alkali stress is needed to better understand the process of organic acid secretion under alkali stress.
Grapevines are an economically important fruit crop worldwide that grow as deep-rooted perennial plants, and their growth and fruit quality are largely influenced by soil alkalinity and salinity. The majority of commercial rootstocks and varieties used in viticulture are moderately sensitive to alkali stress. However, the candidate rootstock cultivar A15 possesses strong alkali tolerance and is a suitable material for identifying alkali toleranceassociated genes. Additionally, A15 was identified to have an higher capacity to secrete organic acids than other grapevine rootstocks with moderate or weak tolerance to alkali stress [21]. The present study determined the accumulation and secretion of organic acids and H + efflux at different time points after NaHCO 3 treatment. Thereafter, the key genes and phosphoproteins involved in the above processes were identified using transcriptomic and phosphoproteomic analyses. Additionally, the roles of ethylene and indoleacetic acid (IAA) in mediating organic acid secretion under alkali conditions were evaluated. The results provide valuable information for dissecting the metabolism and regulatory pathways of alkali stress tolerance in grapevine roots.

Results
Effects of NaHCO 3 on the synthesis and secretion of organic acids in grapevine roots A total of six organic acids were identified, oxalate, malate, tartrate, succinate, citrate and acetate. Oxalate was the main organic acid and accounted for 69.5-71.3% of the sum of the six organic acids in the control roots. In contrast, the five other organic acids, particularly citrate, showed relatively low concentrations ( Fig. 1a-f). Similar differences in organic acid concentrations were also found in the culture solution ( Fig. 1g-l); therefore, oxalate was the main organic acid synthesized and secreted by grapevine roots. The content of oxalate in the NaHCO 3 -treated roots was significantly enhanced from 3 to 24 h after treatment (HAT) and reached a maximum increase of 87.1% at 12 HAT compared to that in the control roots (Fig. 1a). The increased synthesis of oxalate promoted oxalate secretion and led to continuous accumulation in the culture solution; the oxalate content in the NaHCO 3 culture solution reached 1.56 times that of the control solution at 24 HAT (Fig. 1g). Additionally, NaHCO 3 treatment continuously significantly enhanced acetate synthesis but did not promote its secretion (Fig. 1f, l). In contrast, the other organic acids were generally not affected by NaHCO 3 treatment. However, notably, the content of malate in the culture solution containing NaHCO 3 was substantially increased at all time points (Fig. 1h). On the other hand, NaCl treatment as a positive control produced different effects on organic acid synthesis and secretion, e.g., NaCl imparted a reduced and stronger influence on oxalate and malate secretion, respectively, than NaHCO 3 (Fig. 1g, h). Therefore, the changes in organic acid synthesis and secretion under NaHCO 3 were attributed to both Na + and -HCO 3 .

NaHCO 3 treatment promotes H + secretion of grapevine roots
To detect root H + secretion, the vine roots with different treatments were placed in solid medium with pH-sensitive bromocresol purple for 10 h (Fig. 2). Color changes occurred around the roots to indicate changes in pH. For the control roots, only less acidic substances were secreted, as indicated by the weak yellow color, at 0.5 and 3 days after treatment (DAT), and the acidic substances were gradually secreted and accumulated, generating a clear yellow color in the small area around the roots at 5 DAT (Fig. 2a). In contrast, the roots treated with NaHCO 3 began to produce a clear yellow color at 0.5 DAT and produced large areas of yellow color at 3 and 5 DAT (Fig. 2c), indicating that large amounts of acidic substances had accumulated. Additionally, the NaCl treatment as a positive control led to a clear yellow color, but the area and intensity were smaller than those under NaHCO 3 treatment (Fig. 2b, c); therefore, the production of acidic substances was attributed to both Na + and − HCO 3 . Moreover, the H + -ATPase inhibitor Na 3 VO 4 almost completely inhibited the production of yellow color (Fig. 2d), suggesting that acid secretion is related to the PM H + -ATPase activity. Notably, the production of yellow color was accompanied by the occurrence of newly grown roots (Fig. 2a-c), and the production of yellow color and new roots was almost completely inhibited by the simultaneous treatment of NaHCO 3 plus Na 3 VO 4 (Fig. 2d). Further experimentation using solid medium, half of which contained Na 3 VO 4 , indicated that root growth and H + secretion were substantially affected by H + -ATPase activity (Fig. 2e). Therefore, NaHCO 3 treatment induced H + secretion, which was accompanied by new root growth.
On the other hand, H + flux was determined using noninvasive microtest technology to further determine H + secretion under NaHCO 3 treatment (Fig. 3a). High negative values of H + flux prior to treatment indicated the influx of H + into root cells. In contrast, the H + flux gradually changed to efflux, as indicated by the positive Fig. 1 The organic acid content in grapevine roots (a-f) and their culture solution (g-l) under nonstress, salt-stress and alkali-stress conditions. The organic acid content in the root exudate solution was expressed using the ratio of the content of organic acids in the root exudate solution to the root weight to reflect root secretion capacity. Values represent the means ± SD of three replicates. * Significant difference, P < 0.05; ** highly significant difference, P < 0.01 values in the control and NaHCO 3 -treated roots from 1 HAT. Compared to the low H + efflux rate of the control roots, the H + efflux rate of the roots treated with NaHCO 3 was very high at 6 and 12 HAT. Additionally, the highest H + efflux rate in the treated roots occurred at 12 HAT. On the other hand, the NaHCO 3 -treated roots possessed higher H + -ATPase activity than the control roots, and the highest activity of the treated roots was found at 12 HAT, which was 0.46-fold higher than that in the control (Fig. 3b).
Identification of the changes in the transcriptome profile of grapevine roots in response to NaHCO 3 To explore the mechanism underlying the alkali stressinduced oxalate secretion, RNA-Seq analysis of the control and NaHCO 3 -treated vine roots was conducted to quantify gene changes. A total of 3232 and 1714 genes were up-and downregulated by at least one-fold in the NaHCO 3 -treated vine roots, respectively (Fig. 2), demonstrating that NaHCO 3 treatment caused massive transcriptional reprogramming in the vine roots. All of the annotated differentially expressed genes (DEGs) were associated with 20 biological processes. The process of carbohydrate metabolism contained the most DEGs; additionally, biosynthesis of other secondary metabolites, environmental adaptation, amino acid metabolism and signal transduction were also clearly changed biological processes (Fig. 4a).
In the carbohydrate metabolism process, the gene expression of seven genes involved in oxalate and malate metabolism was significantly altered by NaHCO 3 treatment ( Fig. 2 and Fig. 4b). Two PEPCs and their kinases Fig. 2 Effects of NaHCO 3 , NaCl and vanadate on rhizosphere acidification. The roots of five-week-old in vitro shoot cultures were treated with water (pH 7.0) (a), 75 mM NaCl (pH 7.0) (b), 75 mM NaHCO 3 (pH 8.7) (c), and 75 mM NaHCO 3 plus 0.1 mM Na 3 VO 3 (d, e) for 6 h. After the roots were washed with deionized water, the roots were carefully spread on the bottom of a culture dish and then covered with an agar sheet containing 0.006% bromocresol purple and 0.8% agar. In addition to bromocresol purple and agar, Na 3 VO 4 was added into the whole agar sheet in panel d and into the right side of the agar sheet in panel e (PEPCKs) were significantly upregulated, positively contributing to the biosynthesis of oxaloacetate and thereby providing the substrate for oxalate and malate biosynthesis. An oxalate--CoA ligase gene (OCL) and two NADP-malic enzyme (ME) genes were significantly downregulated, which decreased the degradation of oxalate and malate, respectively. Additionally, the expression of five ATP-binding cassette (ABC) transporters, belonging to the process of membrane transport, was substantially increased, and in particular, the expression of ATP-binding cassette B19 (ABCB19) increased 9.66-fold. Two aluminum-activated malate transporters (ALMTs) exhibited more than 5.0-fold higher expression in the NaHCO 3 -treated roots than in the control roots. Plasma membrane (PM) H + -ATPases PMA2 and AHA11 were only detected in the NaHCO 3 -treated roots but showed very low values of reads per kilobase of transcript per million mapped reads (RPKM), suggesting that they might not be the primary proton pump in grapevine roots.
With respect to the signal transduction process, the most significantly altered pathway was the ethylene signaling pathway, for which 51 genes related to ethylene biosynthesis and signaling were transcriptionally modified. Six 1-aminocyclopropane-1-carboxylic acid (ACC) synthases (ACSs) were substantially upregulated by NaHCO 3 treatment; in contrast, two ACC oxidases (ACOs) were transcriptionally changed, and ACO1 was downregulated 2.97-fold (Additional file 1: Table S1). Additionally, IAA metabolism and signaling were also significantly changed. The expression of four genes related to IAA metabolism and signaling was upregulated more than 5-fold, including indole-3-acetic acid-amido synthetase (GH3) and three IAA responsive factors (IAA12, SAUR32 and SAUR40). Moreover, six protein kinases located in the plasma membrane were substantially upregulated by NaHCO 3 treatment, suggesting that plasma membrane proteins were regulated by phosphorylation.
On the other hand, 34 differentially expressed genes, which contained genes involved in organic metabolism and transport and hormone biosynthesis and signaling, were detected by qRT-PCR at different times after NaHCO 3 treatment. Similar expression changes in the DEG tag profiles were found, which not only validated the reliability of the DEGs but also demonstrated their expression patterns under alkali stress (Fig. 4c).
Quantitative analysis of phosphoproteins with phosphorylation levels significantly changed in response to NaHCO 3 We identified 2669 unique phosphoproteins, collectively containing 6312 nonredundant phosphorylation sites. Among those phosphorylation sites, 5404 (85.6%) were found at serine, 877 (13.9%) were found at threonine and 31 (0.49%) were found at tyrosine residues. Of the 2669 phosphoproteins, 2141 phosphoproteins were identified, which contained different quantities of phosphorylation sites (from 1 to 25); a total of 1822 and 608 phosphoproteins contained one and two phosphorylation sites, respectively (Additional file 2: Table S2).
When comparing the phosphorylation levels between the NaHCO 3 -treated and control samples, a total of 197 phosphoproteins (270 phosphorylation sites) showed a significant change (ratio ≥ 1.5, P < 0.05), with 107 upregulated and 163 downregulated (Additional file 2: Table  S2). The 197 phosphoproteins were annotated using Blast2GO according to the biological process, cellular component and molecular function (Fig. 5a). The phosphoproteins were classified into 11 biological processes, with metabolic process, cellular process and localization as the top three categories. For the molecular function, phosphoproteins were classified into 9 categories, and the top 3 categories were binding, catalytic activity and transporter activity. For the cellular components, cell part, cell and membrane possessed the highest number of phosphoproteins (Fig. 5a). The phosphoproteins modified by NaHCO 3 were primarily localized in the nucleus, cytoplasm, chloroplast and plasma (Fig. 5b). Additionally, Motif-X analysis identified 16 significantly enriched motifs (Fig. 5c).
Aconitate hydratase (ACH) and PEPC3-1, which are involved in oxalate and malate synthesis, were significantly phosphorylated in the NaHCO 3 treatment compared to the control ( Fig. 4b; Table 1). The phosphorylation levels of two plasma membrane-localized ATPases (ATPase4-1 and particularly ATPase4-2) were substantially enhanced, which are responsible for H + efflux and provide energy for the transport of oxalate and malate across the plasma In panel b, the schematic was constructed using the genes and proteins related to organic acid metabolism and extrusion. The normalized values of gene expression are shown on a color scale, and a P highlighted with a yellow background color denotes a phosphoprotein. The gray color indicates the possible transporters of oxalate. In panel c, the heatmap was constructed using the genes related to organic acid metabolism and transport, metabolism and signaling of ethylene and IAA, and receptor kinases. The normalized values of gene expression are shown on a color scale. Fold changes of qRT-PCR were calculated by comparing the relative expression values of the selected genes in the NaHCO 3 -treated and control vine roots membrane. Additionally, two plasma membrane ABC transporters ABCB19-2 and pleiotropic drug resistance 12 (PDR12), potential transporters of oxalate, were also significantly phosphorylated. Moreover, the phosphorylation levels of two plasma membrane-localized serine/threonineprotein kinases (STY46 isoform X1 and CDL1), ACO4 and ACO11 for ethylene synthesis and EIN2 for ethylene signaling were significantly changed ( Fig. 4b; Table 1).

Organic acid secretion by vine roots involves ethylene and IAA
Transcriptomic and/or phosphoproteomic profiling indicated that the metabolism and signaling of IAA and ethylene were significantly altered by NaHCO 3 treatment. To reveal the role of ethylene and IAA in mediating NaHCO 3 -induced organic acid secretion, the changes in ethylene and IAA under NaHCO 3 treatment and the effects of the inhibition of ethylene and IAA biosynthesis and/or signaling on organic acid secretion were determined. Compared with the control, NaHCO 3 treatment significantly increased ethylene production at 3 and 6 HAT but substantially decreased ethylene production at 12 and 24 DAT. In contrast, the IAA content was substantially reduced by NaHCO 3 at 6 DAT, and the decrease in the IAA content reached 57.6% at 24 HAT (Fig. 6a, b). Additionally, the inhibition of ethylene biosynthesis and perception via aminoethoxyvinylglycine (AVG) and 1methylcyclopropene (1-MCP) completely blocked NaHCO 3 -induced organic acid secretion, and the inhibition of IAA biosynthesis via 1-N-naphthylphthalamic acid Fig. 5 The distribution of differentially phosphorylated proteins (a, b) and significantly enriched phosphorylated sites (c). The 197 phosphorylated proteins were classified according to their biological process, cellular component, molecular function and subcellular location (a, b). The Gene Ontology (GO) annotation proteome was derived from the UniProt-GOA database (www. http://www.ebi.ac.uk/GOA/). Soft motif-x was used to analyze the model of sequences constituted with amino acids in specific positions of modify-21-mers (10 amino acids upstream and downstream of the site, but phosphorylation with modified-13-mers was 6 amino acids upstream and downstream of the site) in all protein sequences. All of the database protein sequences were used as background database parameters and other parameters with default (NPA) significantly reduced NaHCO 3 -induced organic acid secretion (Fig. 6c, d). Therefore, ethylene signaling is necessary to regulate organic acid secretion under NaHCO 3 , and IAA also participates in the process; additionally, their roles are largely affected by their concentrations in plants.

Discussion
Accumulation and secretion of oxalate are important physiological responses of grapevine roots to alkali stress Plant roots are in direct contact with the soil and can adjust to an adverse rhizosphere environment by secreting a large amount of compounds, which is the most direct and obvious response of roots to environmental stresses [5]. Under alkali stress, the accumulation of organic acids is observed in many plants.
Oxalate was highly accumulated and secreted by grapevine roots under alkali stress (Fig. 1a, g). A large amount of oxalate was also found in the roots of Kochia sieversiana and Suaeda glauca under alkali stress [22,23]. In contrast, Chloris virgata was found to predominantly accumulate citrate, and wheat roots secrete large amounts of lactic, acetic and formic acids under alkali stress [24,25]. The accumulation and secretion of organic acids, including oxalate, citrate and malate, play important roles in osmoregulation, pH adjustment, and ionic balance maintenance by providing a negative charge under alkali and salt stresses [5,22,23]. This study demonstrated that both NaHCO 3 and NaCl stimulated the secretion of organic acids, but NaHCO 3 created a more acidic rhizosphere environment (Fig. 2b, c). Additionally, NaHCO 3 and NaCl imparted different effects on oxalate, acetate and malate (Fig. 1). Similarly, previous reports have also demonstrated that alkali and salt stress have different effects on organic acid metabolism and secretion. For example, the concentration of citrate, the most dominant organic acid, is substantially increased by alkali stress but decreased by salt stress in Puccinellia tenuiflora roots; additionally, alkali stress causes the secretion of citric acid into the rhizosphere, while salt stress does not [5]. Different changes in citrate and malate were also found in Chloris virgata under alkali and salt stresses [25]. Therefore, organic acid secretion is a widespread physiological response of plants, but oxalate is the primary organic acid synthesized and secreted by vine roots. Additionally, − HCO 3 plays a key role in inducing oxalate synthesis and secretion in grapevine roots. In panels a-c, values represent the means ± SD of three replicates. ** highly significant difference, P < 0.01; values indicated by different capital letters are significant at P < 0.01 at the same treatment time The possible pathway of organic acid synthesis and secretion based on the key genes and phosphoproteins under alkali stress Three pathways for oxalate biosynthesis have been proposed in plants, i.e., the glycolate/glyoxylate pathway, the ascorbate pathway and the oxaloacetate (OAA) pathway [26,27]. Under alkali stress, substantial amounts of − HCO 3 enter the root and provide the substrate for PEPCs, which catalyze the β-carboxylation of PEP to OAA using − HCO 3 in the cytosol in an irreversible process [28]; thereafter, OAA can be converted to oxalate, acetate and malate (Fig. 4b). In Arabidopsis, the pepc3 mutant abolished the salt stress-induced increase in malate, suggesting the role of AtPEPC3 in regulating organic acid synthesis [29]. The significant upregulation of the expression of two PEPC3 genes under alkali stress might enhance OAA synthesis and the subsequent conversion of OAA to oxalate and acetate (Fig. 4b, c). In addition to transcriptional regulation, PEPC activity is also regulated by phosphorylation catalyzed by phosphoenolpyruvate carboxylase kinase (PEPCK) [20]. PEPC phosphorylation is abolished in the pepc3 mutant of Arabidopsis under salt stress [20], suggesting that PEPC3 might be the target protein of PEPCKs. The significant increases in the expression levels of two PEPCK1 genes and the phosphorylation level of PEPC3-1 also suggested the phosphorylation regulation of PEPC3-1 by PEPCK1s (Fig. 4b, c). Collectively, the above results suggest that the OAA pathway is the main pathway for oxalate and acetate biosynthesis under alkali stress. However, notably, the glycolate/glyoxylate pathway might also be modified in grapevine roots under alkali stress, as indicated by the significant increase in the phosphorylation level of aconitate hydratase (aconitase), a key protein in the glycolate/glyoxylate pathway [30].
H + extrusion and the transport of organic acid anions across the plasma membrane are controlled by PM H + -ATPases, providing energy by creating an electrochemical proton gradient for transporters [31]. The PM H + -ATPase is important for the root proton-secretion adaptation to alkaline stress [15]; high PM H + -ATPase activity and proton secretion have been shown to enhance plant tolerance to alkaline stress [32], while lower PM H + -ATPase activity and proton secretion result in sensitity to alkaline stress [33]. PM H + -ATPase activity is tightly regulated by phosphorylation at several N-and C-terminal residues, especially upon exposure to various environmental stimuli [34]. The significant increase in the phosphorylation levels showed that the two ATPases 4 are the key proton pumps in grapevine roots under alkali stress (Fig. 4b, Table 1). The ATPases 4 from Arabidopsis (AHA4) and tobacco (PMA4) plants are also reported to be involved in salt tolerance [35,36]. Therefore, ATPase 4 might function in plant tolerance to alkali and salt stress.
In contrast, the specific transporter of oxalate remains unclear. Nevertheless, some experiments have indicated the role of ABC transporters in transporting organic acids in plants. Four ABC transporters (ABCG11, ABCG21, ABCA2, and ABCB21) were considered to be potentially involved in oxalate and/or citrate secretion under Al stress in Grain amaranth roots [10]. AtPDR6, belonging to the ABCG subfamily of the ABC transporter family, is involved in the root extrusion of organic acids, including succinate, fumarate and malate [37]. In this study, the expression levels of five ABC transporters, particularly ABCB19-1, were substantially increased by NaHCO 3 treatment (Fig. 4b, Additional file 1: Table S1), suggesting their association with organic acid secretion. In particular, the expression of ABCB19-1 was well correlated with the oxalate content in roots (Fig. 4c, Fig. 1a), suggesting its role in transporting oxalate. Additionally, two ABC transporters, ABCB19-2 and PDR12, were significantly phosphorylated (Table 1), and PDR12 has been reported to mediate the extrusion of water-soluble carboxylate anions in yeast [38]; therefore, ABCB19-2 and PDR12 might participate in the transport of organic acids. However, the detailed function of the above transporters must be characterized in future studies.
ALMTs are found throughout plant genomes and are involved in a range of distinct functions in different cell types. ALMT1 functions in mediating organic acid secretion in the Al-tolerance response in many plant species [39]. In contrast, the ALMT2 transporter mediates an Al-independent electrogenic transport of organic anions, such as malate and citrate, across the plasma membrane in wheat [40]. The upregulation of two ALMT2s (Fig. 4b, c) and the increased content of malate implied that ALMT2 transporters mediate the salt-and alkali-induced malate extrusion in vine roots. Notably, a large increase in malate secretion was not accompanied by its accumulation under salt and alkali stress (Fig. 1b, h). A similar phenomenon was also found in other plants under Al stress [22,41]. Therefore, it is suggested that malate metabolism is not a limiting factor but rather that transporters are more important for its secretion under alkali stress.

The signaling pathway mediating alkali stress-induced organic acid secretion
Ethylene is an important signaling molecule mediating numerous important biological processes, including responses to abiotic stresses [42]. In this study, the application of inhibitors related to ethylene biosynthesis and perception indicated the necessity of ethylene in organic acid secretion under alkaline stress (Fig. 6c, d). However, the exogenous application or endogenous overproduction of ethylene substantially inhibited H + -ATPase activity and H + efflux in rice roots under alkali stress [17]; additionally, ethylene production was reduced with NaHCO 3 treatment (Fig. 6a). Similarly, transgenic tobacco plants with poor ethylene biosynthesis exhibited elevated salt tolerance, while rice plants treated with ethylene exhibited salt hypersensitivity [43,44]. Therefore, the role of ethylene might be concentration-dependent [42,45], and low-concentration ethylene might be necessary to mediate NaHCO 3 -induced organic acid secretion. Ethylene biosynthesis is mainly regulated by ACS and ACO at the transcriptional and posttranslational levels [46,47]. When grapevines were subjected to alkali stress, the decline in ethylene production was accompanied (Fig. 6a) by the decreased expression of ACO1 (Fig. 4c) and phosphorylation levels of ACO4 and ACO11 (Table 1) but the increased expression of ACS3 and ACS7 (Fig. 4c), suggesting the key role of ACOs in regulating ethylene synthesis under alkali stress. Additionally, EIN2, a central regulator of ethylene signaling [48], controls the transduction of the ethylene signal from the ER membrane to the nucleus in Arabidopsis [48], and its phosphorylation inhibited ethylene signaling (Table 1; 48). After the signal cascade mediated by EIN5, EIN6, EIN3 and others, ethylene signals are delivered to ethylene responsive factors (ERFs), the last downstream components of the ethylene signaling pathway, which lead to the regulation of ethylene controlled gene expression [43]. Here, the changes in the expression of a large amount of ERFs may correspond to the different biological processes regulated by ethylene under alkali stress.
Auxin has been reported to control root apoplastic acidification, to enhance the Al-induced exudation of citrate and to promote the phosphorylation of PM H + -ATPases [13,45,49]. Additionally, PIN2 (an auxin efflux transporter) activates plasma membrane H + -ATPases to release protons, which is necessary for the adaptation of Arabidopsis to alkali stress [15]. Therefore, auxin plays a key role in regulating organic acid secretion under alkali stress. A recent study found that endogenous auxin controls apoplastic acidification; however, an endogenous increase in the overexpression of auxin biosynthesis gene or exogenous increase in the auxin level induces a transient alkalinization [45], which is similar to the role of ethylene discussed above. Here, NaHCO 3 treatment induced a decrease in IAA, but NPA treatment reduced organic acid secretion (Fig. 6b-d), suggesting that the fine-tuning of IAA biosynthesis may be essential for the regulation of organic acid secretion. Additionally, the genes of IAA12 and GH3 as IAA repressors [45,50] were also substantially upregulated (Fig. 4c), probably reducing IAA synthesis and signaling. Small auxin-up RNA (SAUR) genes represent the largest family of auxin-responsive genes and participate in auxin-mediated PM H + -ATPase activation [45,51]. The large increase in the expression of SAUR32 and SAUR40 suggests that they likely regulate organic acid secretion in response to auxin in vine roots. Particularly, it has been demonstrated that auxin mediates ethylene signaling to control root growth [52]. Therefore, we inferred that ethylene most likely regulates organic acid secretion through auxin signaling in vine roots under alkali stress.
On the other hand, the cell membrane harbors hundreds of different receptor kinases that receive environmental signals at the receptor domain on the extracellular side of the membrane and convert these signals into cellular responses via an intracellular protein kinase domain [34]. The PM H + -ATPase was reported to interact with multiple such receptor kinases [53]. The receptor kinase-mediated control of primary active proton pumping at the plasma membrane, e.g., PSY1R, increases proton efflux from roots by interacting with and phosphorylating AHA2/AHA1 [54]. In this study, the high-level expression and/or phosphorylation of the 11 plasma membrane-located receptor kinases might participate in the regulation of ATPases. Although the exact roles of candidate genes remain to be examined, our data provide a platform for further functional analyses of these genes.

Conclusion
Oxalate was the primary organic acid synthesized and secreted by vine roots under NaHCO 3 stress. The OAA pathway, including two PEPC3s and PEPCK1s, plays a key role in oxalate synthesis. The secretion of organic acids and H + were controlled by PM H + -ATPases, and two phosphorylated PM ATPases 4 were the main proton pumps under NaHCO 3 stress. Additionally, ABCB19-2 and PDR12 might participate in the transport of oxalate and other organic acids. Low-concentration ethylene mediates NaHCO 3 -induced organic acid secretion, and IAA also participates in this process.

Determination of organic acids in grapevine roots and root exudate solutions
Healthy apical growth tips of A15 vines were removed in early summer to establish grapevine in vitro shoot cultures. The shoot cultures were subcultured on Murashige and Skoog medium containing 3% (w/v) sucrose, 7 g. L − 1 agar and 0.2 mg. L − 1 IBA. Five-week-old grapevine in vitro shoot cultures were transferred to glass bottles with a 10-cm height and 5-cm diameter. The vines were treated with 50 ml water (pH 7.0) as a control, 75 mM NaCl (pH 7.0) and NaHCO 3 (pH 8.7). Each glass bottle was provided sufficient oxygen with an oxygen machine (SenSen Group, China). The vines were grown in a controlled-environment growth cabinet at 25°C, a 16-h photoperiod and a light intensity of 600 μmol/m 2 /s. At different time points after different treatments, the roots were collected and immediately frozen in liquid nitrogen for organic acid extraction. The treatment solution was collected and evaporated to dryness with a rotary evaporator, and the residue was dissolved in 1 ml double distilled water. The filtrate, which was passed through a 0.45-μm filter, was used for organic acid determination. The extraction of organic acids from the roots and their determination for root extract and treatment solutions were performed using a capillary electrophoresis system (Beckman P/ACE, Palo Alto, CA) as described in our previous study [55].

H + secretion test
H + secretion was detected according to a previously described method [56]. Five-week-old grapevine A15 in vitro shoot cultures were treated with water (pH 7.0), 75 mM NaCl (pH 7.0), 75 mM NaHCO 3 (pH 8.7), and 75 mM NaHCO 3 plus 0.1 mM Na 3 VO 3 for 6 h. Then, the roots were rinsed, carefully spread in Petri dishes and covered by solid medium (pH 5.8) which consisted of 0.006% bromocresol purple (pH indicator, discoloration range of 5.2-6.8), 1 mmol l − 1 CaSO 4 , 2.5 mmol l − 1 K 2 SO 4 and 0.8% agar. The vines were grown in a controlled-environment growth cabinet at 25°C and continuous illumination at 400 μmol/m 2 /s light intensity.

Measurement of H + -ATPase activity and H + flux in roots
Plant materials were the same as those described in the organic acid determination Section. The extraction of plasma membrane (PM) protein and activity determinations of PM H + -ATPase of the root tips were conducted according to the method of Yan et al. (2002) [56]. The activity of H + -ATPase was determined by the Pi amount after 30 min of reaction in 0.5 ml of 30 mM BTP/MES buffer (pH 6.5), 5 mM MgSO 4 , 50 mM KCl, 1 mM Na 2 MoO 4 , 1 mM NaN 3 , 5 mM ATP, and Brij 58 (0.02% w/v). The reaction was initiated by adding 5 mg of membrane protein and stopped with 1 ml of stopping reagent (2% H 2 SO 4 , 5% SDS, and 0.7% (NH 4 ) 2 MoO 4 ) followed immediately by the addition of 50 μl ascorbic acid (10% w/v). The color development was completed after 30 min. The absorbance at 820 nm was measured using a spectrophotometer.
Root H + fluxes were measured using the scanning ionselective electrode technique with a Noninvasive Microtest Technology System (NMT, Xuyue Beijing Sci. & Tech. Co., Ltd., Beijing, China). The 10-mm long roots from the root apex were equilibrated in the measurement solution for 20 min and then immobilized in the measurement chamber in 5 ml of fresh measuring solution.

RNA-Seq and quantitative real-time PCR (qRT-PCR)
Total RNA of 10-mm-long grapevine roots from the apex was extracted using TRIzol Reagent (Invitrogen, Carlsbad, CA, USA), and mRNA was isolated using poly-T oligo attached magnetic beads. Transcriptome sequencing and analysis were conducted by OE Biotech Co., Ltd. (Shanghai, China). Sequencing libraries, constructed using NEBNext® Ultra™ RNA Library Prep Kit for Illumina® (7530 L, NEB, United States), were sequenced on an Illumina HiSeq 4000 platform, and 150bp paired-end reads were generated. Clean reads were assembled into transcripts using Cufflinks referencing the grape genome (http://genomes.cribi.unipd.it/grape/). The unigene expression levels were quantified using RPKM. Unigenes that were differentially expressed between two samples were screened using a false discovery rate < 0.05 and absolute log 2 (fold change) ≥1 as the threshold. Three biological replicates were generated for the control and NaHCO 3 treatment.
qRT-PCR was performed using SYBR Green Master-Mix (SYBR Premix EX Taq TM, Dalian, China) on a Bio-Rad iQ5 (Hercules, CA, United States) instrument, and the primers are listed in Additional file 3: Table S3.

Phosphopeptide and phosphoprotein identification and quantification
Proteins from grapevines were extracted and quantified with the BCA kit (Beyotime, Beijing, China). After trypsin digestion, the peptide mixture was desalted on a Strata X C 18 SPE column (Phenomenex, Torrance, CA, USA). Each peptide was vacuum-dried and reconstituted in 0.5 M TEAB (Sigma, USA). The peptide mixture was labeled using TMT kit (Thermo Fisher Scientific, Torrance, CA, USA) according to the manufacturer's introduction. The TMT labeled peptides were fractionated by high pH reverse-phase HPLC and concentrated by vacuum centrifugation. The phosphopeptides were enriched using IMAC microspheres, and they were then eluted with elution buffer, followed by lyophilization and LC-MS/MS analysis. The phosphopeptides were dissolved in 0.1% formic acid (solvent A) and loaded onto a C 18 -reversed phase column (15-cm length, 75 μm i.d., packed in-house) and separated with a linear gradient of solvent B (0.1% formic acid in 98% acetonitrile) at a constant flow rate of 0.4 μL/min on an EASY-nLC 1000 UPLC system (Thermo). The peptides were eluted with a gradient of 6 to 23% solvent B for 26 min, 23 to 35% solvent B for 8 min, and 80% solvent B for 6 min. The peptides were detected and identified by tandem mass spectrometry (MS/MS) in Q ExactiveTM Plus (Thermo) coupled online to the UPLC, which was supported by Jingjie PTM BioLabs (Hangzhou, China). MS/MS data were searched using a Maxquant search engine (v1.5.2.8) against the Vitis vinifera proteome concatenated with reverse decoy database. The parameters in Maxquant searches were as follows: max missing cleavage of Trypsin/P, 2; peptide mass tolerance, 20 ppm in the first search and 5 ppm in the main search; MS/MS tolerance, 0.02 Da; fixed modification, carbamidomethyl on