Research ArticlesBotany

Inactivation of PYR/PYL/RCAR ABA receptors by tyrosine nitration may enable rapid inhibition of ABA signaling by nitric oxide in plants

See allHide authors and affiliations

Science Signaling  01 Sep 2015:
Vol. 8, Issue 392, pp. ra89
DOI: 10.1126/scisignal.aaa7981

NO more ABA activity

Abscisic acid (ABA) is a critical plant hormone, controlling developmental processes and immune responses. Long-term regulation of ABA signaling involves changes in gene expression that reduce ABA synthesis and enhance ABA metabolism. Castillo et al. found that various members of the ABA receptor PYR/PYL/RCAR family were modified posttranslationally by tyrosine nitration or S-nitrosylation at cysteine residues, two covalent modifications that can result from increased nitric oxide (NO). These NO-mediated modifications and polyubiquitylation, which target proteins for degradation, occurred in a complex, potentially interconnected, and receptor-specific pattern in plants overexpressing individual receptors. Tyrosine nitration, but not S-nitrosylation, inhibited ABA-induced activity in vitro, suggesting that tyrosine nitration may be a mechanism to rapidly tune the cellular responsiveness to ABA.


Abscisic acid (ABA) is a phytohormone that inhibits growth and enhances adaptation to stress in plants. ABA perception and signaling rely on its binding to receptors of the pyrabactin resistance1/PYR1-like/regulatory components of ABA receptors (PYR/PYL/RCAR) family, the subsequent inhibition of clade A type 2C protein phosphatases (PP2Cs), and the phosphorylation of ion channels and transcription factors by protein kinases of the SnRK2 family. Nitric oxide (NO) may inhibit ABA signaling because NO-deficient plants are hypersensitive to ABA. Regulation by NO often involves posttranslational modification of proteins. Mass spectrometry analysis of ABA receptors expressed in plants and recombinant receptors modified in vitro revealed that the receptors were nitrated at tyrosine residues and S-nitrosylated at cysteine residues. In an in vitro ABA-induced, PP2C inhibition assay, tyrosine nitration reduced receptor activity, whereas S-nitrosylated receptors were fully capable of ABA-induced inhibition of the phosphatase. PYR/PYL/RCAR proteins with nitrated tyrosine, which is an irreversible covalent modification, were polyubiquitylated and underwent proteasome-mediated degradation. We propose that tyrosine nitration, which requires NO and superoxide anions, is a rapid mechanism by which NO limits ABA signaling under conditions in which NO and reactive oxygen species are both produced.


The phytohormone abscisic acid (ABA) inhibits key developmental processes, such as seed germination, and enhances stress-related plant processes, including stomatal guard cell closure, and multiple defense responses against abiotic and biotic factors (1). In many of these ABA-regulated processes, nitric oxide (NO) is a downstream mediator of ABA signaling (2). Increased ABA content correlates with enhanced NO production (3, 4). However, NO may antagonize ABA function as demonstrated by the enhanced effects of ABA on development and responses to stress observed in the NO-deficient Arabidopsis thaliana triple mutant nia1 nia2 noa1-2 (4). This complex scenario in the functional interaction between ABA and NO likely reflects both positive and negative effects exerted by NO on ABA signaling, depending on the nature of the process and the site and timing of production and action of both regulators (5).

ABA signaling is orchestrated in plants through a module constituting soluble receptors of the PYR/PYL/RCAR family, protein phosphatases of the PP2C family, three ABA-activated kinases of the sucrose nonfermenting1-related subfamily 2–related protein kinase 2 (SnRK2) family, and ABA-responsive transcription factors (68). The core ABA signaling pathway also exerts nontranscriptional effects on the activity of ion transporters localized in the plasma membrane of guard cells and root cells (9, 10). In the absence of ABA, the pathway is inhibited by the action of PP2Cs (11), which mediate the dephosphorylation and inhibit the activation of SnRK2.2, SnRK2.3, and SnRK2.6 (8), and subsequently inhibit the activity of several transcription factors from different families (12). ABA perception and signaling involve ligand binding by PYR/PYL/RCAR and interaction of the ligand and receptor complex with PP2C (13, 14). The formation of the ternary complex inhibits phosphatase activity (13), enabling SnRK2s to remain phosphorylated and active to mediate phosphorylation of ABA-regulated transcription factors (12).

Because NO-deficient plants display ABA hypersensitivity in multiple plant processes, including seed germination, seedling establishment, and stomata closure, as well as enhanced tolerance to dehydration (4), NO can exert a negative effect on ABA signaling, which could occur at the level of hormone perception (15). The application of ABA, which stimulates NO production, or the application of NO to wild-type, unstressed Arabidopsis leaves results in stomatal closure. However, stomata of abi1-1 or abi2-1, which are PP2C-mutant plants, do not close in response to either exogenous ABA or NO (16), suggesting that NO functions upstream of PP2Cs in this ABA-regulated process.

Most of the direct regulatory effects exerted by NO are mediated by posttranslational modification (PTM) of target proteins through either tyrosine (Tyr) nitration or cysteine (Cys) S-nitrosylation (1720). S-nitrosylation is reversible (21, 22). In contrast, tyrosine nitration is not reversible and would require degradation and resynthesis of the modified proteins. Tyrosine nitration also requires the production of reactive oxygen species (ROS) in addition to NO, because the reactive intermediate is the strong nitrating compound peroxynitrite, which results in vivo from the reaction of NO with superoxide anion (23). Thus, tyrosine nitration may represent a rapid mechanism for limiting ABA signaling in conditions during which NO and ROS both increase.

Here, we assessed whether NO could modify ABA receptors of the PYR/PYL/RCAR family, and we analyzed whether these NO-mediated modifications affected receptor function. We performed both in vitro and in planta analyses and detected specific nitration and S-nitrosylation patterns on different PYR/PYL/RCAR proteins. We found that several members of this family were inactivated through NO-mediated nitration of key Tyr residues, but not upon S-nitrosylation of Cys residues. These types of regulation represent rapid mechanisms to modulate ABA perception when transcription- and translation-based mechanisms are too slow.


ABA receptors are targets of nitration and S-nitrosylation

NO-mediated PTMs are often associated to altered function of the target protein. To assess whether ABA receptors of the PYR/PYL/RCAR family are susceptible to Tyr nitration, we purified His-tagged proteins and exposed the recombinant proteins to the peroxynitrite donor 3-morpholinosydnonimine (SIN-1) and detected Tyr-nitrated proteins by Western blotting. On the basis of reported functions, information regarding monomeric and dimeric forms of the proteins, and those with three-dimensional (3D) structural information, we selected 6 of the 14 members of the Arabidopsis PYR/PYL/RCAR family to evaluate in detail. We found that all six were nitrated in vitro with SIN-1: PYR1, PYL1, PYL4, PYL5, PYL6, and less efficiently PYL8 (Fig. 1A). Nitration was proportional to the amount of nitrating SIN-1 reagent and depended on peroxynitrite, as demonstrated by competition with the peroxynitrite scavenger epicatechin (fig. S1).

Fig. 1 Receptor function of Tyr-nitrated and S-nitrosylated PYR/PYL/RCARs.

(A) Tyr nitration of recombinant ABA receptors. The indicated recombinant purified proteins were treated (+) or not (−) with 0.5 mM SIN-1 in 0.1xTBS or 0.1xTBS alone. Tyr-nitrated proteins were detected by Western blot with the antibody against 3nitroY (α-3NY). Total protein was detected with an antibody against the His tag (α-His). Data are representative of three experiments. (B) Effect of Tyr nitration on ABA receptor activity in ABA-stimulated phosphatase inhibition assay. SIN-1 samples are nitrated. Others are buffer-treated control (0.1xTBS) or untreated receptors. Values shown are the mean percentage of HAB1 activity with maximal recombinant HAB1 activity set to 100% (n = 3 independent experiments) ± SE. The 100% HAB1 phosphatase activity was 34.94 ± 1.67 nmol min−1 mg−1 protein. SIN-1–treated values were significantly different than untreated values for each tested receptor with P < 0.0005 in Student’s t test (**). The 0.1xTBS buffer–treated values were also significant with P < 0.005 for PYL4 (**) and P < 0.05 for PYL6 and PYL8 (*) in Student’s t test. (C) The indicated recombinant purified proteins were treated (+) or not (−) with 0.25 mM GSNO in HEN buffer or HEN buffer alone, and S-nitrosylation was detected by Western blot with an antibody against the tandem mass tag (α-TMT). Total protein was detected with an antibody against the His tag (α-His). Data are representative of three experiments. (D) Effect of S-nitrosylation on ABA receptor activity in ABA-stimulated phosphatase inhibition assay. GSNO samples are S-nitrosylated. Others are buffer-treated control (HEN) and untreated receptor. Values shown are the mean percentage of HAB1 activity with maximal recombinant HAB1 activity set to 100%. The 100% HAB1 activity was 46.78 ± 4.12 nmol min−1 mg−1 protein (D).

To determine whether Tyr nitration affected PYR/PYL/RCAR function, we measured the ABA-induced activation of the receptors in a phosphatase-coupled assay using the phosphatase HAB1 (24). Reactions containing ABA and recombinant ΔNHAB1, which is an N-terminal truncation mutant of HAB1 that has greater activity and stability than the full-length protein and that we refer to as HAB1 hereafter, and individual PYR/PYL/RCAR proteins showed that HAB1 phosphatase activity was inhibited >80% in the presence of control purified receptors or receptors exposed to buffer, but that receptors that had been preincubated with SIN-1 and were nitrated were significantly less able to inhibit HAB1 activity (Fig. 1B). The sensitivity to Tyr nitration–mediated inactivation differed among the receptors, with PYR1 and PYL1 becoming the most inactivated (Fig. 1B). PYL6, PYL8, and, to a larger extent, PYL4 were already partially inactivated by incubation with buffer in the absence of SIN-1 (Fig. 1B), thus likely indicating that these receptors are more thermolabile than the others.

We also tested whether these six ABA receptors were S-nitrosylated upon exposure to the NO donor nitrosoglutathione (GSNO) in vitro. PYR1, PYL1, PYL4, PYL5, PYL6, and PYL8 were nitrosylated (Fig. 1C). S-nitrosylated PYR/PYL/RCARs inhibited HAB1 phosphatase activity in the presence of ABA to a similar extent to that observed for the unmodified proteins (Fig. 1D). Moreover, the nitrosylated receptors maintained activity in this ABA-induced phosphatase inhibition assay even when exposed to a high concentration of GSNO (1 mM; fig. S2).

Thus, our findings suggested that the negative regulatory effect exerted by NO on ABA signaling involved Tyr nitration, which would require the simultaneous production of superoxide and NO to produce peroxynitrite. Under this condition, nitration and the subsequent inactivation of the receptor could occur.

Different Tyr residues contribute to nitration-triggered inactivation of PYR1

To assess how ABA receptors were inactivated upon Tyr nitration, we focused PYR1, which has four Tyr residues per molecule. Tyr58, Tyr120, and Tyr143 are conserved within the family in Arabidopsis (fig. S3A) and within PYR1 homologs in other plant species (fig. S3B). Tyr23 is located far from the ABA-binding pocket (fig. S4A), but Tyr58, Tyr120, and Tyr143 residues are oriented toward and close to the ABA-binding site in the 3D structure (fig. S4B). Factors that increase the probability of nitration of a Tyr residue include the location of the Tyr in a loop, the existence of a basic amino acid flanking the Tyr in the primary sequence, the spatial proximity of a negatively charged amino acid in the 3D structure, and the conservation between homologs (25). On the basis of these parameters in PYR1, we predicted that Tyr58 and Tyr120 were the likeliest potential targets of nitration (table S1). We performed in vitro nitration of purified PYR1 followed by liquid chromatography–tandem mass spectrometry (LC-MS/MS) to identify nitrated peptides. We detected peptides corresponding to nitration of Tyr23, Tyr58, and Tyr120 residues (Table 1 and table S2).

Table 1 LC-MS/MS–based identification of nitrated peptides in recombinant PYR1 purified proteins exposed to SIN-1.

Three independent samples were searched for carbamidomethyl (C), nitro (Y), and oxidation (M) as variable modifications with three maximum missed cleavages and monoisotopic mass values with fragment mass tolerance of 0.5 dalton for run of the first sample (1st run), 20 milli-mass unit (mmu) for the run of the second sample (2nd run), and 10 mmu for run of the third sample (3rd run). See table S2 for a complete data set of peptides identified in the LC-MS/MS analysis.

View this table:

We used site-directed mutagenesis to define the contribution of the different Tyr residues of PYR1 to nitration of the receptor and also the importance of these residues in mediating ABA-induced phosphatase inhibition. First, we confirmed that purified PYR1 with single Tyr-to-Ala mutations in any of the four Tyr residues was nitrated in vitro by exposure to SIN-1 (Fig. 2A). Moreover, combinations of two Tyr-to-Ala mutations, such as PYR1Y23,58A and PYR1Y58,120A, also yielded nitrated proteins upon exposure to SIN-1 (Fig. 2A), thus suggesting that at least three of four Y residues were susceptible to nitration in vitro. PYR1Y23,58,120,143A contains no Tyr residues, and we detected no signal in the Western blot with the antibody recognizing nitrated Tyr (Fig. 2A). We found that as the number of Y-to-A mutations in the PYR1 molecule increased, we obtained less protein, suggesting that these Tyr residues may be essential in maintaining the proper conformation of PYR1, which may also have a potential effect on the activity of the mutated receptors in the in vitro assay independent from the effect nitration.

Fig. 2 Effect of Tyr nitration on the ABA receptor function of different PYR1 mutants.

(A) Tyr nitration of the PYR1 Tyr-to-Ala mutants. The indicated PYR1 mutants were either treated (+) or not (−) with 0.5 mM SIN-1 in 0.1xTBS or with 0.1xTBS alone. Data are representative of three experiments. (B) ABA-induced activity of the PYR1 mutants and the effect of Tyr nitration in ABA-induced phosphatase inhibition assay. Data are presented as in Fig. 1. n = 3 independent experiments. The 100% HAB1 activity was 45.86 ± 1.80 nmol min−1 mg−1 protein. (C) Tyr nitration and HAB1 phosphatase activity–coupled assay for PYR1Y120F and PYR1Y143F. Western blot data represent three experiments. HAB1 inhibition data are presented as in Fig. 1. n = 3 independent experiments. The 100% HAB1 phosphatase activity was 45.86 ± 1.80 nmol min−1 mg−1 protein.

To assess the potential effects of the Tyr-to-Ala mutations on basal receptor function, as well as in Tyr nitration–mediated inactivation, we analyzed ABA-induced receptor function in the phosphatase-coupled assay with either untreated or nitrated receptors (Fig. 2B). Whereas the activity profile of unmodified and nitrated PYR1Y23A was the same as that of wild-type PYR1, PYR1Y120A or any combination including this mutation failed to inhibit HAB1 activity whether exposed to SIN-1 or not (Fig. 2B), which is in agreement with previously reported data showing that this residue is essential for ABA binding (26). Even in the absence of SIN-1, PYR1Y143A also exhibited less capacity to inhibit HAB1 activity than unmodified wild-type PYR1 (Fig. 2B). However, PYR1Y58A exhibited nitration-regulated ABA-induced phosphatase inhibition, although its basal, unmodified activity was lower than that of unmodified wild-type PYR1 (Fig. 2B). These data indicated that Tyr120 and Tyr143 are necessary to maintain the phosphatase-inhibiting activity of PYR1 in response to ABA. Combined with the structural data, these residues are likely required for ABA binding. Consequently, those mutations cannot provide insight into the effect of Tyr nitration–mediated inactivation of the ABA receptor.

To overcome this functional inactivation of the Tyr120- and Tyr143-to-Ala mutants, we generated PYR1Y120F and PYR1Y143F versions of the receptor with these tyrosines mutated to the more structurally related Phe. In contrast to PYR1Y120A and PYR1Y143A, which were inactive in ABA perception (Fig. 2A), PYR1Y120F and PYR1Y143F mediated ABA-dependent phosphatase inhibition and were susceptible to nitration-mediated inactivation (Fig. 2C). Therefore, we concluded that Tyr23 and Tyr58 were dispensable for the Tyr nitration–mediated PYR1 inactivation and also that this mode of inactivation likely required the nitration of more than one Tyr residue, with at least one of them being Tyr120 or Tyr143.

Nitration of PYR/PYL/RCAR proteins occurs in plants

To determine whether nitration of PYR1 occurred in planta, we generated transgenic Arabidopsis plants expressing a hemagglutinin (HA)–tagged version of PYR1 (35S::3xHA-PYR1). We infiltrated the leaves of 35S::3xHA-PYR1 plants with SIN-1, collected samples at different times after SIN-1 addition, isolated Tyr-nitrated proteins by immunopurification with the 3NY antibody recognizing nitrated Tyr residues, and detected HA-tagged PYR1 by Western blotting. We could detect nitrated PYR1 even at zero time point after SIN-1, and then we observed a reduction in the amount of nitrated PYR1, which was undetectable by 3 hours after treatment (Fig. 3A). To assess whether the accumulation of nitrated PYR1 was subjected to proteasomal degradation, we pretreated plants for 12 hours with the proteasome inhibitor MG-132 before the addition of SIN-1. Nitrated PYR1 isolated from leaves pretreated with MG-132 was stable (Fig. 3A). Proteasomal-mediated degradation of PYR1 only affected the nitrated fraction of the protein, because the amounts of total HA-tagged PYR1 protein were not altered throughout the experiment in either MG-132–treated or untreated plants (Fig. 3A). This result also suggested that only a small fraction of the total HA-PYR1 was nitrated.

Fig. 3 In planta detection of Tyr nitration of PYR1 and effect of the proteasome inhibitor on the amount of nitrated protein.

(A) Nitrated HA-PYR1 isolated from plants exposed to SIN-1. 35S::3xHA-PYR1 plants were either pretreated for 12 hours with 0.1 mM proteasome inhibitor (+MG-132) or not (−MG-132) and was then treated with 2 mM SIN-1, and total protein extracts were obtained at the indicated times. The Tyr-nitrated fraction, isolated by immunopurification (IP) with α-3NY–coated magnetic beads (top panel), was detected by Western blot (WB) with the antibody recognizing HA (α-HA). The bottom panel shows the amount of HA-tagged proteins in 20 μg of lysates used for immunopurification with the 3NY antibody. The left panels are representative of three independent experiments. Quantification of the relative amounts of proteins from the three replicates is shown in the right panels. (B) Nitrated HA-PYR1 from plants exposed to ABA. Tyr-nitrated HA-PYR1 was immunopurified from plants after treatment of 35S::3xHA-PYR1 plants with 0.1 mM ABA for the indicated times in the absence or presence of MG-132 pretreatment. The left panels are representative of three independent experiments. Quantification of the relative amounts of proteins from the three replicates is shown in the right panels. (C) Polyubiquitylation of Tyr-nitrated HA-PYR1. 35S::3xHA-PYR1 plants were pretreated with MG-132 for 24 hours before treatment with 0.1 mM ABA, and then Tyr-nitrated proteins were immunopurified from samples that were collected at the indicated times after ABA treatment. HA-PYR1 was detected with the HA antibody, and ubiquitin was detected with the antibody recognizing ubiquitin (α-Ubq). Molecular mass in kD is indicated.

We showed that ABA also induced Tyr nitration and the subsequent degradation of PYR1 in vivo. We exposed 35S::3xHA-PYR1 plants that were or were not pretreated with the proteasome inhibitor to ABA, isolated nitrated HA-tagged PYR1 with the antibody recognizing Tyr-nitrated proteins, and blotted for PYR1 with the HA antibody. We found that the amount of nitrated HA-tagged PYR1 protein was high both at the beginning of the experiment and 30 min after ABA addition and then decreased, becoming undetectable at 3 hours in the samples from plants that had not been pretreated with MG-132 (Fig. 3B). For ABA to trigger nitration and thereby trigger degradation of nitrated PYR1, ABA must induce the production of peroxynitrite. We detected both NO and superoxide anion in roots by 30 min after ABA treatment, thus ensuring the formation of peroxynitrite (fig. S5).

Because proteasomal degradation of proteins is preceded by polyubiquitylation (27), we tested whether polyubiquitylated forms of PYR1 were detected in the immunopurified nitrated fraction from plants pretreated with MG-132 and then treated with ABA. Three hours after ABA treatment, the monomeric unmodified form of nitrated PYR1 was accompanied by a ladder of higher–molecular mass proteins cross-reacting with the HA antibody, and the antibody against ubiquitin revealed multiple cross-reacting bands indicative of polyubiquitylated versions of PYR1 (Fig. 3C).

In vivo identification of NO-related PTMs of PYR/PYL/RCAR proteins allows mapping functional target residues

We used LC-MS/MS to identify posttranslationally modified sites in ABA receptors isolated from plants expressing HA-tagged versions of PYR1, PYL4 (28), or PYL8 (29) under the control of the 35S promoter. We analyzed the samples for the presence of nitrated Tyr residues, S-nitrosylated Cys residues, and polyubiquitylated Lys residues. The comparison of the amino acid sequence of PYR/PYL/RCAR proteins indicates the conservation of three Tyr residues, two Cys residues, and several Lys residues that flank conserved Tyr or Cys residues (fig. S3A). After a pretreatment of 24 hours with MG-132, plants expressing HA-tagged versions of PYR1, PYL4, or PYL8 were treated with 0.1 mM ABA, and samples were collected at 0, 0.5, and 3 hours after ABA treatment (fig. S6). Protein extracts from 20 plants for each genotype and condition were pooled and immunopurified with HA antibody–coated magnetic beads under nonreducing conditions and analyzed by LC-MS/MS. We detected each of these PTMs (nitrated Tyr residues, S-nitrosylated Cys residues, and polyubiquitylated Lys residues) in peptides from the PYR/PYL/RCAR proteins analyzed (Fig. 4 and table S3). We mapped the Tyr nitration, S-nitrosylation, and Lys ubiquitylation sites along the amino acid sequences of PYR1, PYL4, and PYL8 by scoring as modified only those residues that were present in modified peptides in the three LC-MS/MS proteomic analyses that we performed with independent biological replicates (Fig. 4). PYR1 displayed nitration only in Tyr58 and Tyr120. The PYL4 Tyr80 and PYL8 Tyr60 residues, which are the equivalent of PYR1 Tyr58, were also nitrated in vivo. We also detected nitration of PYL8 at Tyr158 (Fig. 4). We detected peptides with S-nitrosylated Cys corresponding to Cys67 and Cys157 in PYL8, but only peptides for a single S-nitrosylated Cys residue in PYR1 (Cys65) and in PYL4 (Cys185). Nitrosylated peptides of PYL8 were also identified with the greatest number for Cys67, which is conserved among all members of the PYR/PYL/RCAR family (fig. S3A).

Fig. 4 A schematic representation of the nitration, S-nitrosylation, and ubiquitination sites identified in planta in PYR1, PYL4, and PYL8.

Empty and filled boxes located along the lines representing the PYR/PYL/RCAR indicate nonmodified and posttranslationally modified residues, respectively. Presence of a modification is indicated by a filled symbol if it was detected in three independent proteomic analyses. Orange, blue, and purple colors correspond to Tyr nitration, S-nitrosylation of Cys, and Lys ubiquitylation, respectively. See table S3 for LC-MS/MS data.

We also searched for peptides with ubiquitylated Lys residues, which would be detected as peptides with the K-GG tag resulting after trypsin digestion of ubiquitylated proteins. Peptides indicative of Lys ubiquitylation were detected for 5 of 8 Lys residues for PYR1, 5 of 8 for PYL4, and 7 of 10 for PYL8 (Fig. 4). In PYL8, Lys61, Lys70, and Lys84 flank Tyr60 and Cys67, for which we detected peptides with these residues nitrated or nitrosylated, respectively. We observed a similar pattern of enrichment in ubiquitination sites flanking nitrated Tyr or nitrosylated Cys residues for PYL4 and to a lesser extent for PYR1. The MS/MS spectra of every identified peptide with a nitrated Tyr residue, S-nitrosylated Cys, or ubiquitylated Lys can be visualized in the PRoteomics IDEntification (PRIDE) database, where the complete data set corresponding to our proteomic experiments have been deposited with the identifiers PXD002396 and 10.6019/PXD002396.

Accessibility of the residues may affect their ability to be posttranslationally modified. An in silico analysis of PYR1 based on the 3D structure of the dimeric form of this receptor [Protein Data Bank (PDB) ID: 3K3K] suggested that amino acid residues along the molecule had a different solvent accessibility (fig. S7). Differences in solvent accessibility were likely relevant for the posttranslationally modified Lys residues that we identified (fig. S7). However, Tyr residues were similarly accessible to solvent (fig. S7), thus suggesting that Tyr nitration is controlled by other factors.

Nitrated PYL8 accumulates in ABA-treated plants

To determine whether the proteasome-dependent loss of detectable Tyr-nitrated PYR1 was a common feature of other Tyr-nitrated PYR/PYL/RCAR family members, we analyzed nitration of PYL8 in 35S::3xHA-PYL8 plants. The amount of HA-PYL8 detected by Western blot was similar in samples prepared from plants exposed to MG-132 (Fig. 5). However, contrary to that observed for PYR1 (Fig. 3B), the nitrated forms of PYL8 were undetectable in the absence of proteasomal inhibition (Fig. 5). Compared with PYR1, in samples from plants pretreated with MG-132, nitrated PYL8 required longer to accumulate, becoming detectable 1 hour after exogenous ABA treatment (Fig. 5). These data suggested that kinetics of ubiquitylation and proteasomal degradation triggered by ABA and the effect of nitration may differ between different receptors of the family. Alternatively, ABA might prevent proteasomal degradation of nitrated PYL8 by yet unknown mechanisms (30, 31).

Fig. 5 In planta detection of Tyr nitration of PYL8 and effect of proteasome inhibitor on the amounts of nitrated protein.

35S::3xHA-PYL8 plants were either pretreated for 12 hours with 0.1 mM proteasome inhibitor (+MG-132) or not (−MG-132) and were then treated with 0.1 mM ABA, and total protein extracts were obtained at the indicated times. The Tyr-nitrated proteins were isolated by immunopurification with α-3NY–coated magnetic beads, and PYL8 was detected by Western blotting with the antibody recognizing HA. The panels below show the total amount of PYL8 by Western blotting with HA antibody in 20 μg of lysates used for immunopurification in the data shown in the top panels. Western blot data are representative of three experiments, and quantification of the relative amount of proteins in all three experiments is shown below.


Although essential for activating abiotic stress–triggered plant responses, ABA function has to be finely tuned to avoid the detrimental effects of a constitutive or sustained response. Plants modulate ABA function at two levels, either by controlling its biosynthesis and metabolism or by modulating its perception and downstream signaling (32). Data presented here suggested that inactivation of ABA receptors by NO-mediated PTM might be a rapid way to attenuate ABA signaling. The PTM of PYR/PYL/RCAR proteins by nitration of Tyr residues rendered the proteins inactive in an in vitro assay. Tyr nitration requires peroxynitrite as the nitrating agent. Abiotic stress conditions that increase the endogenous ABA content also cause enhanced production of NO and ROS (33), thus representing a scenario for the synthesis of peroxynitrite. By the LC-MS/MS of the recombinant protein nitrated in vitro, as well as of immunopurified tagged proteins from plant extracts, we found that several ABA receptors were Tyr-nitrated. Because NO and derivatives, such as S-nitrosoglutathione (GSNO), also accumulate in the microenvironment where peroxynitrite is produced, S-nitrosylation of target proteins may also occur in plants. We found that PYR/PYL/RCAR proteins were S-nitrosylated on Cys residues. Contrary to that observed for Tyr nitration, S-nitrosylation of ABA receptors did not alter their function in vitro. NO also inhibits ABA signaling by S-nitrosylation of the kinase SnRK2.6 in guard cells (34), thus suggesting that different NO-related PTMs may act on diverse targets throughout the ABA signaling cascade to attenuate ABA-triggered responses. The inactivation of the ABA receptors specifically by Tyr nitration suggested that antagonism exerted by NO on ABA signaling at the level of hormone perception might be restricted to conditions and/or microenvironments where both oxidative and nitrosative stress coexist. The rapid kinetics of oxidative and nitrosative burst in many processes and plant responses controlled by ABA, including seed germination or stomata closure, are consistent with the operation of a fast mechanism based on NO-related PTM of the ABA receptors.

An analysis of the amino acid sequence flanking Tyr and Cys residues of the PYR/PYL receptors that were nitrated or nitrosylated, respectively, showed no consensus sequence, thus suggesting that both modifications are guided by 3D structural determinants rather than by primary amino acid sequence. We analyzed the in vitro and in planta nitration of PYR1 by proteomic analyses. Peptides containing nitrated Tyr residues corresponding to two of four Tyr residues of PYR1 were found, with Tyr58 the most frequently detected nitrated residue. However, we detected only one peptide containing nitrated Tyr120 and not a single peptide containing nitrated Tyr143 either in recombinant PYR1 nitrated in vitro or in HA-tagged protein isolated from plants. Structural modeling indicated that Tyr120 and Tyr143 were as exposed to solvent as other Tyr residues of PYR1 (fig. S7). We speculate that the Tyr143 residue is not a target for nitration and Tyr120 is susceptible but not a preferred nitration target in PYR1 or that PYR1 containing those nitrated Tyr residues become rapidly degraded. The analysis of Tyr mutants of PYR1 in the in vitro phosphatase assay suggested that nitration-mediated inactivation of PYR1 involved the simultaneous nitration in more than one site.

Analysis of the Lys ubiquitylation pattern in the receptors indicated preferential ubiquitylation of Lys residues flanking either nitrated Tyr or nitrosylated Cys residues. In PYL8 and PYL4, polyubiquitylated Lys residues flanked the Tyr and Cys residues in the peptides that were detected by LC-MS/MS and contained the NO-modified forms of Tyr and Cys residues, suggesting that NO modification and ubiquitylation may be interconnected PTMs. However, the pattern of ubiquitylation and NO modification in PYR1 differed from that of PYL4 and PYL8. In PYR1, only one of the NO-modified residues, Tyr58, was flanked by ubiquitylated Lys residues. Further work will be necessary to assess whether nitration of Tyr residues or S-nitrosylation of Cys residues is a primary event guiding PYR/PYL/RCAR to polyubiquitylation and proteasomal degradation, or whether ubiquitylation marks sites for NO-related PTM.

PYL8 undergoes proteasomal degradation mediated by the CRL4 complex and the substrate adaptor DDA1 (30). Whereas our work suggested that exogenous ABA promoted PYR1 polyubiquitylation and proteasomal degradation (Fig. 3B), ABA limits PYL8 polyubiquitylation, thereby reducing DDA1-mediated destabilization of PYL8 (30). However, our results suggested that Tyr nitration of PYL8 may render the protein very unstable, because we could not detect Tyr-nitrated PYL8 in the absence of a proteasome inhibitor. Thus, it seems unlikely that Tyr nitration of PYL8 is involved in the ABA-mediated stabilization of PYL8 through inhibition of DDA1. The differences that we observed for Tyr-nitrated PYR1 and PYL8 may result from different ubiquitin ligases acting on different members of the PYR/PYL/RCAR family, and ABA may have different effects on these processes, creating additional potential complexity. Further work is necessary to understand the relationship between different PTMs and PYR/PYL/RCAR function.

Despite the apparent differences exerted by NO-related PTMs in plants, the effect of Tyr nitration on receptor activity in vitro was consistently negative for all tested ABA receptors. Because we have used transgenic plants overexpressing tagged versions of the ABA receptors to analyze the potential in vivo effect of nitration on receptor function, Tyr nitration might be a homeostatic or compensatory mechanism resulting from receptor overexpression. Although in another study 9 hours of treatment with 50 μM ABA increased the amount of HA-tagged PYL8 threefold over the amount in untreated plants (29), we found no changes in total PYL8 protein up to 3 hours after ABA treatment (Fig. 5). Additionally, we found that ABA did not alter the frequency of detecting nitrated peptides by LC-MS/MS in 35S::3xHA-PYL8 plants. Thus, our results indicated that an increase in the PYR/PYL/RCAR due to overexpression would not be sufficient to induce nitration-mediated inactivation reported in this work.

Our findings suggested that NO-related PTMs of PYR/PYL/RCAR represent regulatory mechanisms controlling the hormone perception, downstream signaling, and stability of ABA receptors. Our in vitro data suggested that Tyr nitration and not S-nitrosylation was specifically involved in controlling ligand perception or phosphatase coupling, such that Tyr-nitrated receptors were inactive. We propose that Tyr nitration–mediated inactivation of ABA receptors represents a rapid mechanism to locally limit the effects of ABA. This mechanism would be complementary to the well-established role of NO together with hydrogen peroxide in promoting ABA catabolism (35, 36), which involves the enhanced expression of genes coding for catabolic enzymes of ABA, protein synthesis, and metabolism of the endogenous ABA pool. The rapid regulatory mechanism mediated by Tyr nitration would require integrating the action of NO and the reactive superoxide anion, both of which are produced in response to ABA and form the reactive molecule peroxynitrite, which would nitrate Tyr residues of ABA receptors and decrease sensitivity to ABA (Fig. 6). After the rapid reduction in ABA responsiveness, the cell would eliminate the Tyr-nitrated receptors by polyubiquitylation and proteasomal degradation. Further work will reveal whether this rapid Tyr nitration–dependent regulatory mechanism operates in different plant processes and also will elucidate the biological role of S-nitrosylation of ABA receptors.

Fig. 6 The proposed model of NO-responsive switch in ABA sensitivity.

Fast mechanism of regulation of ABA signaling based on peroxynitrite (ONOO)–mediated inactivation through Tyr nitration of PYR/PYL/RCAR versus slow NO-induced metabolic-based alteration of ABA content. Dashed lines are used to represent the regulatory mechanism proposed in this work; solid lines represent the previously established slow NO-induced regulatory model of ABA signaling. Nitrated and polyubiquitylated residues are represented by Y-NO2 and K-(Ubq)n, respectively. TF, transcription factors.


Plant material and growth conditions

A. thaliana wild-type ecotype Col-0 was used in this work. Seeds were grown in soil mixture or MS (Murashige and Skoog) medium as previously described (37). Full-length PYR1 cloned into the ALLIGATOR2 destination vector (38) was used to transform Agrobacterium tumefaciens C58C1 (pGV2260), and then the transformed Agrobacterium was used to transform wild-type plants by floral dip transformation (39) to generate 35S::3HA-PYR1 transgenic plants used for in planta analyses of PTMs. Transgenic plants expressing HA-tagged PYL4 and PYL8 were previously described (28, 29).

Recombinant protein production and purification

Tyr-to-Ala mutated versions of PYR1 were generated by polymerase chain reaction–based site-directed mutagenesis (40), cloned into pCR8-GW-TOPO, and recombined by LR reaction to the indicated destination vectors. pETM11 or pET28a vectors were used to generate N-terminal 6xHis-tagged recombinant proteins. Cloning of 6xHis-tagged constructs for ΔNHAB1 (lacking residues 1 to 178), PYR1, PYL1, PYL4, PYL5, PYL6, and PYL8 was previously described (28, 29, 41, 42). The different PYR1-mutated versions were cloned into pETM11 vector (PYR1Y58A, PYR1Y120A, PYR1Y143A, and PYR1Y58,120A) or pDEST17 vector (Invitrogen) (PYR1Y23A, PYR1Y23,58A, and PYR1Y23,58,120,143A). BL21(DE3) cells transformed with the corresponding constructs were grown at 30°C in Luria-Bertani medium to an OD600 (optical density at 600 nm) of 0.4 to 0.6, then 1 mM isopropyl-β-D-thiogalactopyranoside (IPTG) was added, and the cells were harvested after overnight incubation at 16°C. Bacterial pellets were resuspended in 50 mM tris-HCl (pH 7.6), 250 mM KCl, 10% glycerol, and 0.1% Tween 20, and the cells were disrupted by a Branson Sonifier. The clear lysate obtained after centrifugation was purified by the Ni-NTA purification system (Qiagen). A washing step was performed with 50 mM tris, 250 mM KCl, 20% glycerol, 30 mM imidazole, and 10 mM β-mercaptoethanol washing buffer, and the protein was eluted using 50 mM tris, 250 mM KCl, 20% glycerol, and 250 mM imidazole elution buffer without β-mercaptoethanol, except for ΔNHAB1 that required β-mercaptoethanol during elution to keep it active (26, 43).

In vitro nitration or S-nitrosylation of proteins

About 30 μg of purified recombinant proteins was nitrated or S-nitrosylated in darkness by incubation with 0.5 mM 3-morpholinosyndnonimine (SIN-1; Invitrogen) or 0.25 mM GSNO (Merck) for 3 hours at 37°C and 1 hour at room temperature, respectively. When indicated, other concentrations of nitrating or nitrosylating agents were used. Reaction mixtures were further cleaned and concentrated by successive centrifugation through Amicon Ultra 10K filters (Millipore). Selective scavenging of peroxynitrite with epicatechin was used to assess the specificity of protein nitration (44).

Type 2C phosphatase–coupled assay for ABA receptor function

PYR/PYL/RCAR activity was assessed by a colorimetric assay coupled to PP2C activity in a 100-μl reaction volume containing 25 mM tris-HCl (pH 7.5), 2 mM MnCl2, and 10 mM p-nitrophenyl phosphate (Sigma-Aldrich). A mixture of 0.5 μM PP2C phosphatase ΔHAB1 and PYR/PYL recombinant receptors was incubated for 10 min in the absence or presence of 10 μM(+)-ABA. Assays contained between 4.2 and 5.1 mg of the different His6-tagged PYR/PYL proteins to obtain a ratio of 1:4 phosphatase/receptor. Measurements of absorbance at 405 nm were performed every 30 s over 30 min with a Multiskan GO Microplate Spectrophotometer (Thermo Scientific). Values represent the ratio of the activity in the presence of ABA versus that obtained in the absence of ABA for every sample. The mean ± SE of at least three independent experiments is shown.

Immunopurification and Western blot detection

Total protein extracts were obtained from whole 14-day plants grown in MS medium supplemented with 1% sucrose under long-day photoperiodic conditions. The HA-tagged versions of PYR/PYL/RCAR proteins were immunopurified by incubation with μMACS magnetic beads coated with HA antibodies (Miltenyi Biotec) or with nitrotyrosine antibody magnetic bead conjugate (Millipore) with continuous shaking overnight at 4°C. Immunopurified proteins were eluted from magnetic beads under nondenaturing, nonreducing conditions using 0.1 mM triethylamine (pH 11.8) (Sigma), and proteins were collected in 1 M MES buffer (pH 3.0) for neutralization of the eluted immunopurified proteins. Proteins were separated by SDS–polyacrylamide gel electrophoresis, blotted onto nitrocellulose membrane, stained with Ponceau S, and probed with primary antibodies at the following dilutions: monoclonal anti–3-nitroY (1:1000; Cayman Chemical), anti-His6 (1:500; Roche), and polyclonal anti-HA (1:5000; Abcam). Secondary antibody was anti-mouse or anti-rabbit for monoclonal or polyclonal primary antibody, respectively, coupled to horseradish peroxidase (GE Healthcare) at 1:5000 and 1:10,000 dilutions, respectively. The ECL kit (Pierce) was used for visualization of proteins. S-nitrosylated proteins were detected by a switch assay using a nonbiological iodoTandem Mass Tag (iodoTMT), and further detection was carried out with an antibody recognizing TMT (1:1000) as indicated by the manufacturer with slight modifications (Pierce S-Nitrosylation Western Blot Kit) (45).

LC-MS/MS–based proteomic analyses

Protein samples immunopurified under nonreducing conditions were precipitated with six volumes of cold acetone and digested overnight, under nonreducing conditions, with trypsin (mass spectrometry grade, Promega Corp.; 1:10 enzyme/substrate ratio) at 37°C in 6 M urea, 200 mM ABC buffer. The resulting peptides, desalted with UltraMicroSpin columns, were fragmented by collision-induced dissociation in a Thermo LTQ Orbitrap Velos Pro mass spectrometer. Fragmented peptides were separated in a packed nanocapillary column (NTCC-360/75-1.9-25L Nikkyo Technos). Raw data were processed and analyzed by using the Mascot Server v2.4 (Matrix Science) database [false discovery rate (FDR) < 5%, A. thaliana UniProt database including 13,140 proteins, with a mass tolerance of 7 ppm for the precursors and 0.5 dalton for fragments]. A customized database with the 864 proteins identified in the original search was then reanalyzed with SEQUEST searching for the N-terminal modification by acetylation (+42.011 daltons) and the dynamic modifications of S-nitrosylation of Cys (+28.990 daltons), nitration of Tyr (+44.985 daltons), amination of Tyr (+15.011 daltons), oxidation of Met (+15.995 daltons), and GG of Lys (+114.043 daltons) as a mark of ubiquitination after trypsin digestion. Identified peptides were filtered by XCorr (z = 2 XCorr > 0.9, z = 3 XCorr > 1.2, z > 4 XCorr > 1.5) and deltaCN > 0.15. The maximum number of missed cleavages was 3, and the MS and MS2 tolerances were 0.5 and 0.7, respectively. To rule out the chance of false identification of nitration or S-nitrosylation modification during the MS/MS analysis, we ran control protein samples immunopurified and digested with trypsin under reducing conditions. No nitrated Tyr– or nitrosylated Cys–containing peptides were identified from these samples from plants expressing tagged PYR1 (table S4). We performed MS/MS analyses, one for each of the three genotypes exposed to each of the three ABA conditions (table S3). Each sample was prepared by pooling about 20 14-day-old plants, which were extracted and immunopurified, and the resulting purified samples were analyzed by LC-MS/MS.

In silico analyses of primary amino acid sequences and 3D structure models

The 3D protein models were generated by homology modeling at the SWISS MODEL workspace (46) using the coordinates of the PYR1 receptor in complex with ABA [PDB code 3K90; (47)] and the crystal structure of dimeric PYR1 with ABA-bound closed lid and ABA-free open lid subunits (PDB code 3K3K) (48). The 3D models were visualized and manipulated with Yasara ( or PyMol ( Prediction of the solvent exposure based on the 3D structure of PYR1 (3K3K.pdb) was performed with of the ProtSA web application [; (49)]. Data of Solvent Accessible Surface Area (SASA) values for the different amino acid residues in the unfolded and folded conformations were calculated.


Fig. S1. Dose-dependent nitration of PYR/PYL/RCAR proteins by the peroxynitrite donor SIN-1.

Fig. S2. Effect of increasing concentrations of the nitrosylating reagent GSNO on PYR/PYL/RCAR function.

Fig. S3. Amino acid sequence alignment of PYR/PYL/RCAR from A. thaliana and other plants.

Fig. S4. Models of 3D structures of PYR1.

Fig. S5. Simultaneous production of NO and superoxide anion in roots treated with ABA.

Fig. S6. Schematic representation of the experimental system designed for in planta proteomic analyses.

Fig. S7. In silico analysis of solvent accessibility in PYR1.

Table S1. Spatial parameters of PYR1 Tyr residues.

Table S2. Proteomic analyses of in vitro nitrated PYR1.

Table S3. Peptides and PTMs identified in the in vivo proteomic analyses.

Table S4. Peptides and PTMs identified from HA-tagged PYR1 proteins processed under reducing conditions.


Acknowledgments: We thank M. Holdsworth (University of Nottingham, UK) for the critical reading of the manuscript. Thanks also to the Proteomics Unit of the Center of Genomic Regulation (CRG)–Universitat Pompeu Fabra (UPF) for their technical support in the proteomic analyses. Funding: This work was supported by the Ministry of Science and Innovation (MICINN) (Spain) grants BIO2008-00839 and BIO2011-27526 (to J.L.) and BIO2011-23446 (to P.L.R.). Author contributions: J.L. and J.L.-J. designed the research; M.-C.C., J.L.-J., M.G.-G., L.R., and J.L. performed the research; M.-C.C., J.L.-J., M.G.-G., L.R., P.-L.R., and J.L. analyzed the data; and J.L. wrote and edited the article. Competing interests: The authors declare that they have no competing interests. Data and materials availability: The mass spectrometry proteomics data have been deposited to the ProteomeXchange Consortium (50) through the PRIDE partner repository with the data set identifiers PXD002396 and 10.6019/PXD002396. Materials used in this study are available upon request.
View Abstract

Stay Connected to Science Signaling

Navigate This Article