Nitro-Fatty Acids in Plant Signaling: Nitro-Linolenic Acid Induces the Molecular Chaperone Network in Arabidopsis

Endogenous nitro-fatty acids support signaling in defense against abiotic-stress.

Abstract

Nitro-fatty acids (NO₂-FAs) are the product of the reaction between reactive nitrogen species derived of nitric oxide (NO) and unsaturated fatty acids. In animal systems, NO₂-FAs are considered novel signaling mediators of cell function based on a proven antiinflammatory response. Nevertheless, the interaction of NO with fatty acids in plant systems has scarcely been studied. Here, we examine the endogenous occurrence of nitro-linolenic acid (NO₂-Ln) in Arabidopsis and the modulation of NO₂-Ln levels throughout this plant’s development by mass spectrometry. The observed levels of this NO₂-FA at picomolar concentrations suggested its role as a signaling effector of cell function. In fact, a transcriptomic analysis by RNA-seq technology established a clear signaling role for this molecule, demonstrating that NO₂-Ln was involved in plant defense response against different abiotic-stress conditions, mainly by inducing heat shock proteins and supporting a conserved mechanism of action in both animal and plant defense processes. Bioinformatics analysis revealed that NO₂-Ln was also involved in the response to oxidative stress conditions, mainly depicted by H₂O₂, reactive oxygen species, and oxygen-containing compound responses, with a high induction of ascorbate peroxidase expression. Closely related to these results, NO₂-Ln levels significantly rose under several abiotic-stress conditions such as wounding or exposure to salinity, cadmium, and low temperature, thus validating the outcomes found by RNA-seq technology. Jointly, to our knowledge, these are the first results showing the endogenous presence of NO₂-Ln in Arabidopsis (Arabidopsis thaliana) and supporting the strong signaling role of these molecules in the defense mechanism against different abiotic-stress situations.

Over the last few years, significant progress has been made in understanding the function of nitric oxide (NO) in higher plants. In this regard, NO is the major reactive nitrogen species, which is involved in different physiological and stress conditions in plants, such as seed germination (Libourel et al., 2006), development, senescence (Leshem et al., 1998; Corpas et al., 2004), and the response to biotic (Delledonne et al., 1998; Feechan et al., 2005) or abiotic (Barroso et al., 2006; Chaki et al., 2011a, 2011b; Begara-Morales et al., 2014a, 2014b) stress.

Because plants are sessile organisms, they are exposed to several biotic and abiotic stressful situations. In response to these adverse conditions, plants have evolved various mechanisms, both constitutive and inducible, designed to defend themselves. Among these mechanisms, fatty acids constitute an important source of reserve energy and are essential components of membrane lipids. A number of studies have revealed the role of lipids and lipid metabolites during plant-pathogen interactions throughout the lipoxygenase pathway with the production of jasmonic acid (JA), which is an important signaling molecule in defense regulation (Wasternack and Hause, 2013). It has also been recently shown that linolenic acid (Ln), the precursor of JA, has an important role in the response against abiotic and oxidative stress conditions through the induction of several antioxidant systems such as galactinol synthase or Met sulfoxide reductase enzymes (Mata-Pérez et al., 2015). Fatty acids are also involved in the response to biotic and abiotic stresses by remodeling of membrane lipid composition (Upchurch, 2008) and through the very long chain fatty acid pathway. Very long chain fatty acids are fatty acids containing 20 to 36 carbons, and which are required for the biosynthesis of plant cuticle and the generation of sphingolipids that can be bioactive molecules on their own (Raffaele et al., 2009). Moreover, both 16- and 18-carbon fatty acids participate in defense to modulate basal, effector-triggered, and systemic immunity in plants (Kachroo and Kachroo, 2009), confirming the key role of fatty acids in plant physiology. Regarding abiotic stress situations, plants have developed other molecular defense mechanisms besides the regulation of membrane fluidity. It consists of the induction of heat shock genes, encoding for heat shock proteins (HSPs; Richter et al., 2010). These HSPs can play a crucial role in protecting plants against stress by re-establishing normal protein conformation and thus cellular homeostasis (Wang et al., 2004). They are known to be expressed in plants not only when they experience high temperature stress but also in response to a wide range of other environmental insults, such as water stress, salinity, cold, and oxidative stress (Vierling, 1991; Scarpeci et al., 2008).

More recently, plant research has begun to focus on posttranslational modifications mediated by NO such as protein nitration or S-nitrosylation (Lindermayr et al., 2006; Begara-Morales et al., 2013; Begara-Morales et al., 2014a, 2014b). Although protein nitration is one of the most studied processes, the interaction of NO with other important macromolecules such as lipids has been scarcely studied in plant systems (Sánchez-Calvo et al., 2013; Fazzari et al., 2014). In this respect, nitro-fatty acids (NO₂-FAs) stem from the reactions of NO, NO-derived oxides of nitrogen [e.g. nitrogen dioxide (NO₂) and peroxynitrite (ONOO−)], and oxygen-derived inflammatory mediators [e.g. superoxide (O₂^.−), hydrogen peroxide (H₂O₂), and lipid peroxyl radicals (LOO·)] in animal systems (Baker et al., 2004; Schopfer et al., 2005a, 2005b; Freeman et al., 2008; Jain et al., 2008). However, the mechanism of fatty acid nitration in vivo remains unknown (Rubbo, 2013). The potential relevance of NO₂-FAs as powerful signaling molecules in higher plants is based on a proven signaling capacity mediating antiinflammatory functions and cardiovascular benefits in animal systems (Cui et al., 2006; Kansanen et al., 2011). In this sense, NO₂-FAs act as signaling effectors because small amounts function as effective mediators of potent signal transduction cascades, being in concordance with the low amounts of NO₂-FAs detected in animal systems (nanomolar or picomolar concentrations; Tsikas et al., 2009; Salvatore et al., 2013). Research in this field has demonstrated that NO₂-FAs could occur in vivo as free fatty acids, esterified in complex lipids in hydrophobic compartments, and adducted with proteins—the latter two being the most abundant reservoirs (Schopfer et al., 2005a, 2005b; Trostchansky and Rubbo, 2008; Rudolph et al., 2009). Among NO₂-FAs biological properties, the ability to release NO into aqueous environments has been considered a signaling mechanism of NO₂-FAs as potent NO donors (Schopfer et al., 2005; Gorczynski et al., 2007). Furthermore, NO₂-FAs predominantly act via posttranslational modification (Schopfer et al., 2009; Geisler and Rudolph, 2012) in which the high electronegativity of NO₂ substituents, when bound to an alkenyl carbon of fatty acids, confers an electrophilic nature upon the adjacent carbon and enables a reversible Michael addition reaction with nucleophiles such as protein His and Cys residues by a process termed “nitroalkylation” (Baker et al., 2007). In this respect, the study of the potential role of NO₂-FAs in animal systems has been widely analyzed, having shown that electrophilic fatty acid derivatives such as nitro-oleic (NO₂-OA) or nitro-linoleic (NO₂-LA) acids can mediate antiinflammatory and prosurvival signaling reactions (Cui et al., 2006; Schopfer et al., 2011). In this sense, the only transcriptomic analysis to explore signaling pathways and genes modulated by NO₂-FAs has been performed by microarray technology with NO₂-OA in cultures of human endothelial cells (Kansanen et al., 2009). This study identified heat shock response (HSR) as the main pathway activated by this nitro-fatty acid in an Nrf2-independent way.

Although very recently the presence of endogenous nitro-conjugated linoleic acid (NO₂-cLA) has been reported in extra virgin olive oil and NO₂-OA-Cys adducts in fresh olives, and supports their potential antiinflammatory role (Fazzari et al., 2014), the presence and physiological function of NO₂-FAs can be considered an unexplored area in higher plant research (Sánchez-Calvo et al., 2013). Here, we report the endogenous occurrence of nitro-Ln (NO₂-Ln) in Arabidopsis (Arabidopsis thaliana) and the modulation of NO₂-Ln content throughout the development of this plant. Moreover, transcriptomic analysis by RNA-seq technology demonstrated that NO₂-Ln was involved in a plant-defense response against different abiotic-stress conditions mainly through the induction of HSPs. Finally, mass spectrometry analysis revealed that NO₂-Ln levels were significantly induced under several abiotic-stress conditions analyzed, thus validating the results found by RNA-seq technology. Taken together, to our knowledge, these are the first results showing the endogenous presence of NO₂-Ln in Arabidopsis and supporting the important signaling role of these molecules in the defense mechanism against different abiotic-stress situations.

RESULTS

Lipid Composition of Arabidopsis

Fatty acid composition of Arabidopsis lipid extracts from seeds, whole 14-d-old seedlings, leaves from 30- and 45-d-old plants, and Arabidopsis cell suspension cultures (ACSCs) is shown in Supplemental Table S1. Linolenic acid (18:3) was the most abundant fatty acid in 14-d-old seedlings, leaves from 30- and 45-d-old plants, and ACSCs (31–49%), followed by palmitic acid (16:0) and linoleic acid (18:2), whereas the most abundant fatty acid in seeds was linoleic acid (37%) followed by linolenic acid (23%). Moreover, in some tissues such as leaves from 30- and 45-d-old plants and 14-d-old seedlings, the presence of hexadecatrienoic acid (16:3) was also observed (12–30%). However, the occurrence of this fatty acid in seeds and ACSCs was not detected. These findings are consistent with previous data reported for Arabidopsis (Miquel et al., 1998; Bonaventure et al., 2003).

Synthesis and Characterization of NO₂-Ln Standard

NO₂-Ln has not previously been described and is not commercially available, so it was synthesized and characterized to obtain a standard for its detection as an endogenous compound in Arabidopsis. Thus, a procedure similar to that reported for linoleic acid nitration was followed (Alexander et al., 2006; see “Materials and Methods”), and the NO₂-Ln structure determined was then analyzed by NMR and liquid chromatography-electrospray ionization tandem mass spectrometry (LC-MS/MS; see Supplemental Figs. S1–S3 and Supplemental Information S1).

Detection of Endogenous NO₂-Ln through Arabidopsis Development

Arabidopsis samples obtained from lipids extracts, as indicated in “Materials and Methods,” were analyzed by LC-MS/MS, showing the presence of one peak with multiple-reaction monitoring transitions m/z 322/275 characteristic of NO₂-Ln in whole Arabidopsis 14-d-old seedlings and in ACSCs (Fig. 1C). However, it was not possible to detect the presence of other NO₂-FAs (Fig. 1, A and B). The occurrence of NO₂-Ln in Arabidopsis was also confirmed by incubating the lipid extracts from ACSCs with an excess of β-mercaptoethanol (β-ME) and by seeking the covalent nitroalkylated adduct formed by the reaction of NO₂-Ln and the thiol of β-ME, as previously described by Bonacci et al. (2012) for nitro-linoleic acid (Supplemental Fig. S4A). Moreover, to assure whether those peaks detected in several Arabidopsis samples with multiple-reaction monitoring transition 324/277 corresponded to NO₂-LA, nitroalkylation analysis was carried out (Supplemental Fig. S4B). Incubation with β-ME of lipid extracts did not produce the disappearance of these peaks, assuring they did not correspond to electrophilic NO₂-LA.

Figure 1.

Detection of endogenous NO₂-Ln in different Arabidopsis samples. Lipid extracts from whole 14-d-old seedlings and cell-suspension cultures (ACSCs) of Arabidopsis were obtained, as indicated in “Materials and Methods,” and analyzed by LC-MS/MS. The following MRM transitions were analyzed: m/z 326/279 for nitro-oleic acid (NO₂-OA), m/z 324/277 for nitro-linoleic acid (NO₂-LA), and 322/275 for NO₂-Ln. A, Peak corresponding to NO₂-OA standard and the absence of this nitro-fatty acid in both samples analyzed. B, Peak for NO₂-LA not displaying any peak in the retention time of this nitro-fatty acid. C, NO₂-Ln standard and the presence in both plant samples of this nitro-fatty acid with the same retention time. Peaks refer to a total ion intensity of 1.74 e4, 6.00 e4, and 3.00 e3 for NO₂-OA, NO₂-LA, and NO₂-Ln, respectively. Vertical dashed lines, Peaks with the same retention time.

NO₂-Ln content through Arabidopsis development was analyzed. Quantification of NO₂-FAs in several samples analyzed was carried out using ¹³C-labeled nitro-oleic acid (¹³C₁₈-NO₂OA) as an internal control. For this study, seeds, 14-d-old seedlings, and 30- and 45-d-old leaves (senescent plants) from this species were used. The results showed a fall in the NO₂-Ln levels over the development, with the highest levels appearing in seeds (11.18 ± 1.68 pmol/g FW), followed by 14-d-old seedlings (3.84 ± 0.44 pmol/g FW) and 30- and 45-d-old Arabidopsis plants with the lowest NO₂-Ln levels (0.36 ± 0.04 and 0.54 ± 0.06 pmol/g FW, respectively; Fig. 2).

Figure 2.

Detection of NO₂-Ln throughout Arabidopsis developmental process. Lipid extracts from seeds, 14-d-old seedlings, and 30-d-old and 45-d-old leaves from Arabidopsis were obtained, as indicated in “Materials and Methods,” and analyzed by LC-MS/MS. NO₂-Ln values are expressed as the mean ± se from at least 10 independent experiments. A, NO₂-Ln levels detected in 14-d-old seedlings, 30-d-old and 45-d-old (senescent) leaves from Arabidopsis, were significant (P < 0.05) compared to those found in seeds. B, NO₂-Ln levels detected in 30-d-old and 45-d-old leaves from Arabidopsis were significant (P < 0.05) compared to those obtained in 14-d-old seedlings.

Transcriptomic Analysis of NO₂-Ln-Responsive Genes in ACSCs

The endogenous levels of NO₂-Ln detected in ACSC were 0.28 pmol/g FW. In this regard, ACSCs were treated with 10 and 100 μm NO₂-Ln, Ln, methanol (vehicle), and distilled water (control) under nonoxidative stress conditions (data not shown). In this regard, in the only transcriptomic analysis made to date by microarray technology with NO₂-FAs, specifically NO₂-OA, 3 μm of this nitro-fatty acid was used in cultures of human endothelial cells (Kansanen et al., 2009). The detected levels of NO₂-OA in plasma of healthy individuals are about 1 nm (Tsikas et al., 2009), thus being the concentration used in that transcriptomic analysis about 3,000-fold higher than the physiological one. In this sense, 10 and 100 μm treatments are equivalent to 10 and 100 nmol/g FW, being those treatments are about 1,000 and 10,000-fold higher than the physiological concentrations detected in several Arabidopsis samples. Therefore, treatments carried out in this study allowed us to observe a clear gene expression response due to NO₂-Ln treatment and were in concordance with previous studies carried out in the field of nitro-fatty acids. Different ACSCs exposed to diverse treatments were harvested and used for total RNA isolation. Then, paired-end libraries were assembled and sequenced as described in “Materials and Methods.” Firstly, the gene-expression profile of control and vehicle ACSC groups was compared. This analysis was used to eliminate genes responding to methanol, filtering them by a fold change (FC) of 1.5 up and down; this is henceforth referred to as a control. Then, a new comparison between ACSCs incubated with 100 μm Ln and 100 μm NO₂-Ln was carried out, considering an expression level of these genes greater than 0. At this point, it bears remarking that the gene-expression level for ACSCs treated with 10 μm of Ln and NO₂-Ln was analyzed, showing the modulation of the same subset of genes observed in 100 μm treatments, although with a lower FC in those 10 μm treatments (see Supplemental Fig. S5). In this regard, A and B of this figure display the line plot and heat map for up-regulated genes while C and D show the line plot and heat map for down-regulated genes by NO₂-Ln treatment, evidencing a dose-dependence response of this nitro-fatty acid. Due to these results, the transcriptomic analysis was focused on the gene response in the 100 μm treatments. The analysis indicated that NO₂-Ln induced significant changes (95% of matches, P < 0.05) in the gene-expression level of 1,308 ACSC genes, 437 being up-regulated and 871 down-regulated. From this set of genes, the two FCs up and down NO₂-Ln-responsive genes with a gene-expression level greater than 0 were selected, showing the modulation of 316 genes, from which 129 were up- and 187 were down-regulated. At this point, unknown proteins were discarded for the rest of the study, resulting in the up-regulation of 103 genes and the down-regulation of 156 genes. In this sense, 1 to 3 from Figure 3 display the scatter plot, line plot, and heat map, respectively, for up-regulated genes, and 4 to 6 show a scatter plot, line plot, and heat map, respectively, for down-regulated genes whose expression changes significantly in response to NO₂-Ln, evidencing this behavior. To further confirm this dose-dependence response at low concentrations of NO₂-Ln, the gene expression profile from several up- and down-regulated genes by NO₂-Ln was analyzed in ACSCs treated with 1, 10, and 100 μm of NO₂-Ln by quantitative real-time reverse transcription-PCR (qRT-PCR). In this sense, gene expression analysis demonstrated a dose-dependence response in both up- and down-regulated genes (Supplemental Fig. S6) at 1, 10, and 100 μm of NO₂-Ln. Dose dependence was previously observed in cultures of human endothelial cells treated with 0.5, 1, and 5 μm of NO₂-OA (Kansanen et al., 2009), thus confirming the dose-dependence response of nitro-fatty acids in both animal and plant systems. Regarding the results, the FC of up-regulated genes was much higher than the FC of the down-regulated genes, reflecting that the FC of the most-induced gene by NO₂-Ln was close to 4,000 (AT4G27670) and the FC for the rest of the up-regulated genes was approximately 165. On the other hand, the FC of the most repressed gene by NO₂-Ln was close to 326 (AT2G16580) and the FC for the rest of the down-regulated genes was close to 8. Furthermore, these results from RNA-seq analysis were tested by the random assignment of several NO₂-Ln-responsive genes to conduct the expression analysis by qRT-PCR. Figure 3B shows the comparison between the qRT-PCR and RNA-seq analysis, showing that all the NO₂-Ln-responsive genes tested and previously identified by RNA-seq were confirmed by qRT-PCR. The results showed a positive correlation between the two approaches (with a correlation coefficient of 0.91), indicating that the RNA-seq expression analysis performed is highly reliable.

Figure 3.

Scatter plot, line plot, and heatmap for NO2-Ln 100 μm-responsive genes with 2FCs using DNAStar software and qRT-PCR validation of RNA-seq results. A, DNAStar analysis. (1) Scatter plot, (2) line plot, and (3) heatmap for overexpressed genes. (4) Scatter plot, (5) line plot, and (6) heatmap for repressed genes. Distilled water, methanol, and Ln-responsive genes were used to filter the results of NO₂-Ln treatment. All graphs show two fold-change genes with 95% significant differential expression (in log2 scale) obtained by Student’s t test from whole NO₂-Ln-responsive genes. B, qRT-PCR validation of RNA-seq results. Twenty genes identified previously as NO₂-Ln-responsive genes by RNA-seq (white bar) in ACSCs were randomly selected to analyze, by qRT-PCR, the differential expression changes (red bars). Comparison of FC of RNA-seq and qRT-PCR shows a correlation coefficient of 0.91, indicating that RNA-seq results are reliable. Results are average data of two independent experiments in triplicate with sd. BT1 (AT1G04570), Arabidopsis folate-biopterin transporter; LCR29 (AT2G10535), Arabidopsis low-molecular, secreted, Cys-rich protein; HSFA2 (AT2G2615), Arabidopsis heat-stress transcription factor A-2; APX2 (AT3G09640), Arabidopsis l-ascorbate peroxidase 2; HSP40 (AT3G14200), Arabidopsis chaperone DnaJ-domain superfamily protein; DUF1645 (AT3G27880), Arabidopsis unknown-function protein; DUF 604 (AT4G11350), Arabidopsis unknown-function protein; PMP22 (AT4G33905), Arabidopsis peroxisomal membrane protein 22; HSP20 (AT5G37670), Arabidopsis HSP20-like chaperones superfamily protein; TPM1 (AT3G61920), Arabidopsis unknown protein involved in N-terminal protein myristoylation; CAF1A (AT3G44260), Arabidopsis putative CCR4-associated factor 1; TPS8 (AT1G70290), Arabidopsis α-trehalose-phosphate synthase (UDP-forming) 8; BBD1 (AT1G75380), Arabidopsis bifunctional nuclease 1; BBX14 (AT1G68520), Arabidopsis B-box type zinc finger containing protein 14; NUDT21 (AT1G73540), Arabidopsis nudixhydrolase 21; RLK1 (AT2G37710), Arabidopsis receptor lectin kinase 1; CBSX5 (AT4G27460), Arabidopsis cystathionine β-synthase (CBS) domain containing protein 5; BBX30 (AT4G15248), Arabidopsis B-box-type zinc-finger-containing protein 30; HSP21 (AT4G27670), Arabidopsis chloroplast located sHSP 21; SAUR8 (AT2G16580), Arabidopsis small auxin up-regulated RNA8 protein.

Otherwise and to confirm the results observed with NO₂-Ln in ACSCs, the gene expression profile in whole young (14-d-old seedlings) and senescent (45-d-old) Arabidopsis plants treated with 100 μm NO₂-Ln, depicting different stages of Arabidopsis development, was also analyzed by qRT-PCR. Results exhibited a similar behavior in gene expression profile in ACSCs,14- and 45-d-old Arabidopsis plants both in up- and down-regulated genes (Supplemental Fig. S7) and thus confirmed that NO₂-Ln launches the same set of genes in all stages of Arabidopsis development.

Gene Ontology Term Enrichment Analysis

On the basis of these results, more specific analysis of biological processes involved in NO₂-Ln treatments was made using the Blast2GO suite. The test was conducted using a filter cutoff value of a false discovery rate (FDR) < 0.001 for up-regulated genes and an FDR < 1e⁻⁶ for down-regulated genes. The results are shown in Figure 4 as the percentage of sequences annotated for each biological process gene ontology (GO) term for both control ACSCs and 100 μm NO₂-Ln-responsive genes. Bars are labeled with their corresponding P values in Fisher’s exact test. The ACSC treatment with 100 μm of NO₂-Ln induced a more significant response in GO terms in up-regulated genes (Fig. 4A) than in down-regulated genes (Fig. 4B). The GO terms of these up-regulated genes were closely related to abiotic stimulus and stress response. In fact, high-light intensity, temperature, and oxidative-stress processes were highly represented in genes induced by the NO₂-Ln treatment. In this regard, the oxidative-stress response was depicted by GO term overrepresentation in hydrogen peroxide (H₂O₂), reactive oxygen species (ROS), and oxygen-containing compound responses. Furthermore, the GO-enrichment analysis revealed the overrepresentation of processes related to inorganic substances, radiation, chemical stimulus, and endoplasmic reticulum stress responses (Fig. 4A). With regard to GO terms of down-regulated genes (Fig. 4B), it was found that the level of overrepresentation of biological processes was lower than in up-regulated genes. These were related to biosynthetic and metabolic processes, involved mainly in nitrogen compounds, nucleic acids, and macromolecule metabolic processes. Moreover, the responses to different compounds such as hormones, lipids, and organic and chemical substances were observed and, finally, a repression in transcription-related processes was detected. These results suggest a close relationship between NO₂-Ln and the involvement in different abiotic-stress conditions.

Figure 4.

GO-term-enriched graph of biological processes of NO₂-Ln-responsive genes. A, Up-regulated genes. The node filter was set at FDR < 0.001. B, Down-regulated genes. The node filter was set at FDR < 1e-6. Bars for up- and down-regulated genes were labeled with their corresponding P values in Fisher’s exact test against expressed control genes. Scale of positive y axis shows the percentage of sequences that annotated for each biological-process GO term for both control and NO₂-Ln 100 μm-responsive genes.

Overexpressed Genes in Response to NO₂-Ln Treatment: Involvement of NO₂-Ln in Abiotic Stress Processes

Due to the dearth of information concerning the role of nitro-fatty acids in plant systems, a network analysis was conducted on the main genes regulated by NO₂-Ln. For this reason, with the use of the GeneMANIA resource (genemania.org), the 10-most-induced genes by NO₂-Ln (representing 10% of total up-regulated genes) were introduced for network graph analysis. In this sense, the transcriptomic study revealed a remarkable induction of HSPs with a representation of about 25% of the total up-regulated genes (Fig. 5A). In this regard, one of the most overrepresented biological processes by NO₂-Ln treatment was the response to heat (Fig. 5B) with 21.25% of total abiotic-stress conditions and closely related to the response to heat acclimation (6.88%) and high-light intensity (19.38%). In this respect, the most up-regulated gene was HSP21 (AT4G27670) with a FC of 3,995.73 (see Supplemental Table S2). A general behavior of the high number of induced HSPs was that most of them were small HSPs (sHSPs) such as HSP17.6II (AT5G12020) or HSP17.6A (AT5G12030), although other members of HSP40, HSP60, HSP70, and HSP90 families were also induced. Another relevant aspect of up-regulated genes by NO₂-Ln treatment was the induction of HSFA2 (AT2G26150) and HSFA7B (AT3G63350), which belong to the heat shock transcription factor (HSF) family. Moreover, among other up-regulated genes, was MBF1C (AT3G24500), which has been proposed as a key transcriptional regulator in plant thermo-tolerance (Suzuki et al., 2008). On the other hand, protein folding and response to endoplasmic reticulum stress were also induced processes, with 21.25 and 11.88%, respectively. HSPs were again involved in these processes, but other important genes such as HOP3 (AT4G12400), TPR10 (AT3G04710), TPR2 (AT1G04130), ROF1 (AT3G25230), and BAG6 (AT2G46240) were induced. Finally, other abiotic processes such as heavy metals, mechanical stimulus, salt stress, and low temperature were increased by the NO₂-Ln treatment.

Figure 5.

NO₂-Ln-induced genes. A, Network graph of principal up-regulated genes by NO₂-Ln treatment. B, Distribution of up-regulated genes for abiotic processes. Inside each box the type of stress is depicted together with a representative gene in that category with its FC. C, Representative immunoblots (top) showing the increase in the expression of HSP17.6, HSP70 family protein, HSP21, and APX after NO₂-Ln treatment. Dilution of 1:6,000 was employed for the HSP17.6, HSP21, and HSP70 family protein and 1:2,000 for APX. A 5 μg aliquot of protein was used per line. (Bottom) Ponceau-stained loading control. ACSCs were treated with 100 μm Ln and NO₂-Ln for 1 h as described in “Materials and Methods.”

As mentioned above, a general phenomenon underlying abiotic-stress situations is oxidative stress (Apel and Hirt, 2004; Miller and Mittler, 2006). The above-described GO-term analysis (Fig. 4A) showed a well-defined response against oxidative-stress conditions, highlighting H₂O₂ (19.38% of total abiotic-stress processes), ROS, and oxygen-containing compound responses (Fig. 5B). An in-depth analysis of genes involved in these situations led to the identification of the up-regulation of APX2 (AT3G09640) and other genes involved in oxidation-reduction processes, such as AT1G60750, CYP81D1 (AT5G36220), CYP707A4 (AT3G19270), AT5G48320, AT4G15070, and AT4G33040.

Because of the involvement of NO₂-Ln in the induction of the molecular chaperone network and for validation of the transcriptional data obtained by RNA-seq analysis, several HSPs previously identified as induced by this nitro-fatty acid were analyzed by immunoblot. NO₂-Ln treatment provoked the higher increase in the expression of HSP17.6 as compared to Ln treatment with 49.6% (Fig. 5C) and followed by the induction of 42.3 and 31.7% from a member of the HSP70 family protein and HSP21, respectively (Fig. 5C). Moreover, because of the relationship between NO₂-Ln and the response to oxidative stress-related processes, protein expression of APX was also evaluated. Results showed a slight induction on the expression of this protein by NO₂-Ln treatment of about 22.4% as compared to nonnitrated Ln (Fig. 5C), thus confirming the transcriptomic outcomes obtained by RNA-seq technology. Bottom of Figure 5C shows ponceau-stained loading control.

After characterizing the behavior of NO₂-Ln in ACSCs, to evaluate whether it is extensible to other NO₂-FAs, the gene expression profile with other major NO₂-FAs such as NO₂-OA and NO₂-LA was conducted by qRT-PCR. Results displayed an increase in gene expression with NO₂-OA and NO₂-LA of the same subset of genes previously identified to be induced by NO₂-Ln (Supplemental Fig. S8A).

Due to NO₂-FAs having the ability to release NO in aqueous medium (Schopfer et al., 2005a, 2005b; Gorczynski et al., 2007), gene expression analyses with several known NO donors were performed. In this regard, S-nitrosoglutathione (GSNO) analysis was included because of this molecule being considered a biological NO donor and a mobile reservoir of NO (Liu et al., 2001; Sakamoto et al., 2002; Leitner et al., 2009; Begara-Morales et al., 2014a, 2014b). S-nitroso-N-acetyl-penicillamine was also included because of its use as a pharmacological NO donor in numerous studies in plant systems (Beligni and Lamattina, 2000; Pagnussat et al., 2002). Results derived from qRT-PCR analyses showed a clear opposite response between the gene expression profile of NO₂-Ln-induced genes and the different NO donors studied (Supplemental Fig. S9A).

Down-Regulated Genes by NO₂-Ln: Decrease in Cell Metabolic Activity

A remarkable result of NO₂-Ln treatment was the potent induction of genes, whereas this did not occur for down-regulated genes (see Supplemental Table S3). Among NO₂-Ln-repressed processes, about 42% were related to metabolism (Fig. 6). In this regard, a decrease in the biosynthesis of sugars, in cell wall components, and in chlorophyll was detected together with a decrease in expression of genes associated with photosynthesis, the electron-transport chain, and metabolite transport. Moreover, transcription- and translation-related processes were also down-regulated in response to NO₂-Ln treatment, and some biological processes related to hormone metabolism and stress processes were down-regulated.

Figure 6.

Genes down-regulated by NO₂-Ln. Distribution of down-regulated genes for biological processes. Inside each box, the type of stress is depicted together with a representative gene in that category with its FC.

As it has previously mentioned with up-regulated genes by NO₂-Ln, the incubation of ACSCs with 100 μm of NO₂-OA and NO₂-LA produced the same behavior in the gene expression profile for down-regulated genes (Supplemental Fig. S8B).

On the other hand, treatment of ACSCs with 100 μm GSNO and 100 μm S-nitroso-N-acetyl-penicillamine generated again an opposite response between NO₂-Ln-repressed genes and genes modulated by these NO donors (Supplemental Fig. S9B).

Increase of NO₂-Ln Levels in Abiotic-Stress Situations: Validation of RNA-Seq Results

Because part of the results found by this RNA-seq analysis evidenced that NO₂-Ln was implicated in the defense response to abiotic stress situations, the levels of this nitro-fatty acid were analyzed in some of these conditions. NO₂-Ln levels in ACSCs subjected to salt stress (5 and 30 min after treatment) and 14-d-old Arabidopsis seedlings exposed to mechanical wounding, heavy metal, and low temperature stresses were studied. The results showed that salt stress (100 mm NaCl) caused a significant increase in NO₂-Ln levels at 5 min after treatment (0.96 ± 0.12 pmol/g FW), tripling the levels of this nitro-fatty acid with respect to control ACSCs (0.28 ± 0.04 pmol/g FW; Fig. 7A). By contrast, a decrease in NO₂-Ln levels after 30 min of salt treatment (0.52 ± 0.06 pmol/g FW) was observed. Mechanical wounding, salt, and low-temperature stresses provoked a noteworthy rise in NO₂-Ln levels with regard to control seedlings. Mechanical wounding was the stress situation with higher levels in this nitro-fatty acid (7.46 ± 1.20 pmol/g FW), followed by cadmium (6.62 ± 0.98 pmol/g FW) and low-temperature stresses (5.75 ± 0.79 pmol/g FW; Fig. 7B) compared to nonstressed seedlings (3.84 ± 0.44 pmol/g FW).

Figure 7.

NO₂-Ln modulation in different abiotic stress conditions. A, NO₂-Ln detection in 9-d-old ACSCs subjected to salt stress and B, NO₂-Ln detection in different abiotic stress conditions such as wounding, cadmium exposure, and low temperature in 14-d-old Arabidopsis seedlings. NO₂-Ln values are expressed as the mean ± se from at least three independent experiments. Asterisks (*) represent that NO₂-Ln-detected levels are significant (P < 0.05) compared to control.

DISCUSSION

NO₂-Ln Is an Endogenous Molecule Whose Levels Are Modulated throughout Arabidopsis Development

Linolenic acid was the major fatty acid in most of Arabidopsis-studied samples, in agreement with previous data on the Arabidopsis lipid profile (Bonaventure et al., 2003). The synthesis and characterization of its nitrated form was carried out, for the first time to our knowledge, by the nitroselenation method, as was previously reported for other nitro-fatty acids (Alexander et al., 2006), thereby obtaining a NO₂-Ln standard. On the basis of these results, the endogenous presence of NO₂-Ln in ACSCs and 14-d-old Arabidopsis seedlings was studied by mass spectrometry analyses. In all cases was possible to detect a single peak having the same chromatographic properties as the NO₂-Ln standard, but it was not possible to detect the presence of nitro-derivatives about major fatty acids such as NO₂-OA and NO₂-LA. Furthermore, NO₂-Ln was also characterized by the formation of the covalent nitroalkylated adduct resulting from the reaction between the electrophilic NO₂-Ln and nucleophilic β-ME. This behavior has been previously described for NO₂-OA and NO₂-LA in animal systems (Schopfer et al., 2009), and it is considered as a posttranslational modification sensitive to fatty acid cell redox state that could act as an important signaling mechanism (Baker et al., 2007; Geisler and Rudolph, 2012).

Besides analyzing the endogenous presence of NO₂-Ln in ACSCs and in 14-d-old Arabidopsis seedlings, the study was extended to whole development of this model plant showing a decrease in NO₂-Ln content throughout this physiological process from seeds to senescent plants. In this regard, it has been established that NO₂-FAs can release NO both in vitro, as has been reported for NO₂-OA and NO₂-LA (Lima et al., 2005; Schopfer et al., 2005a, 2005b) and in vivo for NO₂-Ln (Sánchez-Calvo et al., 2013). In this respect, the higher content of NO₂-Ln in seeds could contribute to the availability of NO, favoring germination and the onset of vegetative development (Beligni and Lamattina, 2000; Bethke et al., 2006; Libourel et al., 2006). Moreover, the decline in NO₂-Ln levels in senescence may be due to the low NO availability observed in senescent and mature organs (Beligni and Lamattina, 2001). These low levels found in Arabidopsis plants support the possibility of a role for this nitro-fatty acid as a signaling molecule, as previously demonstrated for other NO₂-FAs in animal systems (Tsikas et al., 2009; Salvatore et al., 2013).

HSR Is the Principal Pathway Activated by NO₂-Ln

To establish the potential signaling mechanisms in which NO₂-Ln is involved, a transcriptomic analysis by RNA-seq technology was performed by analyzing 10 μm and 100 μm concentrations of this nitro-fatty acid. Bioinformatics results showed a dose-dependence response, as has been previously demonstrated for other NO₂-FAs such as NO₂-cLA in animal systems (Bonacci et al., 2012). Moreover, transcriptomic analyses by qRT-PCR with low concentrations of NO₂-Ln (1 μm represents about 100-fold higher the physiological concentration of NO₂-Ln in Arabidopsis) also produced a dose-dependence response in gene expression profile as has been previously described for NO₂-OA (Kansanen et al., 2009). NO₂-Ln treatments in ACSCs did not produce any oxidative damage, and it allowed us to evaluate a clear gene response due to the treatment and was in concordance with prior studies carried out in the field of nitro-fatty acids (Tsikas et al., 2009; Salvatore et al., 2013). In this sense, the only transcriptomic analysis made to date with NO₂-FAs in animal systems is a microarray study with NO₂-OA in cultures of human endothelial cells (Kansanen et al., 2009). This study showed that HSP-pathway activation by NO₂-OA contributes to cell protection through signaling actions mainly involved in antiinflammatory processes. Additionally, it was identified the HSR as the major activated pathway by NO₂-OA in an Nrf-2 independent manner. This nitro-fatty acid induced the expression of a great number of HSPs that are related to high temperature, chemical exposure, heavy metals, and oxidative stress in animal systems (Westerheide and Morimoto, 2005). Moreover, the HSR is mediated by HSFs, with HSF1 being the most important mediator in cell-stress responses (Anckar and Sistonen, 2011). In this work, a similar behavior was found in different stages of Arabidopsis development including ACSCs, 14-d-old seedlings, and 45-d-old plants treated with NO₂-Ln, evidencing a conserved mechanism of action of these signaling molecules both in animal and plant systems. In this regard, it is remarkable to note that ACSC incubation with other NO₂-FAs such as NO₂-OA and NO₂-LA modulates the same set of genes as NO₂-Ln, confirming the universality of the signaling mechanism that NO₂-FAs are able to start up. In fact, this transcriptomic analysis about NO₂-Ln-responsive genes in ACSCs showed a remarkable induction of HSPs. In this sense, HSPs in plants are expressed in a wide spectrum of abiotic-stress situations that also course with oxidative stress processes such as high light intensity, osmotic stress, low temperature, and salt stress (Wang et al., 2004). Among the induced-HSPs in this survey, a large set corresponded to sHSPs, which have been postulated to act as a primary line of defense against protein aggregation, thus maintaining protein homeostasis in several abiotic-stress situations (Haslbeck and Vierling, 2015). Recently, the role of HSPs acquired a great relevance based on their participation in a wide range of stress situations (Xu et al., 2012), mainly abiotic stress conditions (Wang et al., 2004). Most of these situations are accompanied by protein dysfunction, therefore maintaining the proteins in their native conformations and preventing them from the aggregation are key processes in cell survival (Wang et al., 2004). These results support the idea that NO₂-Ln sets up a key defense mechanism against these adverse conditions. Moreover, a notable induction was observed in several HSFs, proposing again a conserved action mechanism between animal and plant systems. Furthermore, the proteomic induction of several HSPs by NO₂-Ln reinforces the key role of this nitro-fatty acid in setting up the chaperone network and the antioxidant defense system in Arabidopsis. On the other hand, it has been shown by a recent in silico analysis that the up-regulation of HOP3, TPR10, TPR2, and ROF1 is related to a high potential for interacting with HSP90/HSP70 as cochaperones (Prasad et al., 2010), suggesting anew a direct connection between NO₂-Ln and HSR. Finally, a remarkable point obtained from the studies with different well-characterized NO donors is that NO₂-Ln induces the expression of several HSPs by mechanisms that are independent of NO release, and which are mainly based on the electrophilic ability these molecules possess.

NO₂-Ln Could Set Up a Defense Mechanism against Abiotic and Oxidative Stress

Based on the results obtained by GO-enrichment and the GeneMANIA resource, it has been observed that NO₂-Ln is involved in the response to several abiotic stress conditions. As it has been previously mentioned, the response to heat was the principal process that this nitro-fatty acid was able to trigger, although NO₂-Ln was also involved in the response to heavy metals, mechanical stimulus, salt, and low temperature stresses. This behavior reveals a tight relationship between this nitro-fatty acid and the response to different abiotic stress conditions. The implication of NO₂-Ln, in the response to oxidative stress that underlies most stress states, has already been described (Apel and Hirt, 2004; Miller et al., 2008). GO-enrichment bioinformatics analysis by Blast2GO revealed that a large number of NO₂-Ln-induced genes were related to oxidative stress response, mainly depicted by H₂O₂, ROS, and oxygen-containing compound responses. Among the main genes related to this, APX2 was up-regulated by NO₂-Ln treatment. This enzyme is involved in H₂O₂-detoxification processes using ascorbate as an electron donor, and is a key component of plant-defense response. In addition, HSFA2 has been described as being induced in response to high-light intensity and H₂O₂ in Arabidopsis (Nishizawa et al., 2006), whose production is enhanced under many environmental stress situations (Levine et al., 1994; Dat et al., 1998). Additionally, it has been revealed (Nishizawa et al., 2006) that HSFA2 directly controls APX2 levels at high temperatures and under light-intensity stress in Arabidopsis. It has been postulated that a certain accumulation of H₂O₂ during high temperature stress is required for an effective expression of HSR genes (Volkov et al., 2006). Regarding the action mechanism by which NO₂-Ln could modulate APX activity, it is important to bear in mind that APX2 contains three Cys and nine His residues, and hence a nitroalkylation mechanism with NO₂-Ln is feasible, thereby enhancing its activity as a regulator of cellular redox homeostasis. The observed induction of NO₂-Ln levels in different abiotic stress situations was analyzed, and shown to be similar to the levels found in the transcriptomic study for mechanical wounding, cadmium, salt, and low temperature stresses—evidence that this nitro-fatty acid could act as a signaling mediator in the plant-defense mechanism in abiotic-stress situations, setting up a defense response against cell damage arising as a result of stress and mediated mainly by HSP induction. Therefore, these results suggest a close relationship with high temperature, abiotic stress in general, and oxidative stress—with both stress processes tightly regulated by NO₂-Ln.

On the other hand, down-regulated genes were mainly associated to biosynthetic- and photosynthetic-related processes. In this sense and because NO₂-Ln appears to be involved in the defense response against abiotic stress, the observed decrease in metabolic processes leads to a metabolic reconfiguration. This phenomenon promotes a down-regulation in the biosynthesis of new molecules and compounds, counteracting energy costs resulting from the induced defense mechanism, mainly performed through HSP up-regulation. Such findings have previously been described in several stress situations, indicating that this process is required to maintain a balance between the continuation of cell function and survival (Fahnenstich et al., 2008; Mahajan et al., 2014).

CONCLUSION

In summary, this study provides, to our knowledge for the first time, new insights into the nitro-fatty acids plant signaling. The modulation of NO₂-Ln levels throughout development and in response to different abiotic stress conditions highlights the signaling role of this nitro-fatty acid in cell activity functioning and in the induction of antioxidant response against several abiotic stress situations. Moreover, functional transcriptomic studies by RNA-seq technology reveal that NO₂-Ln is involved in oxidative and abiotic stress processes through the modulation of transcript levels of HSPs, showing a conserved mechanism of action in both animal and vegetal systems.

MATERIALS AND METHODS

Plant Material and Growth Conditions

Several plant materials were used: 9-d-old ACSCs, ecotype Columbia seeds, 14-d-old seedlings (young stage), and 30- and 45-d-old (senescence stage) Arabidopsis (Arabidopsis thaliana) plants. Nine-day-old ACSCs were grown as previously described by Mata-Pérez et al. (2015). Arabidopsis ecotype Columbia seeds were surface-sterilized for 5 min in 70% (v/v) ethanol containing 0.1% (w/v) SDS, placed for 20 min in sterile water containing 20% (v/v) bleach and 0.1% (w/v) SDS, and washed four times in sterile water. Arabidopsis 14-d-old seedlings were grown in petri plates according to the method outlined elsewhere (Leterrier et al., 2012). For 30- and 45-d-old Arabidopsis plants, seeds were sown in tubes with 1% phytoagar and were then grown in a culture chamber for 7 d under anaerobic conditions. Afterward, the seeds were transferred to hydroponic cultures under aeration with a specific growth medium (Cellier et al., 2004) for 30 and 45 d. The plants were grown with 16 h light, 22°C/8 h dark, and 18°C under a light intensity of 100 to 120 μE m⁻² s⁻¹. For the analysis of the involvement of NO₂-Ln in the mechanism of gene-expression regulation, 9-d-old ACSCs were incubated with 10 and 100 μm NO₂-Ln (which is equivalent to 0.1 μm mol and 1 μm mol NO₂-Ln/g FW), 10 μm and 100 μm Ln, methanol (fatty acid vehicle), and distilled water, with the last three being used as controls. Samples were designated as control (C), vehicle (MeOH), Ln, and NO₂-Ln ACSCs. The treatments were applied under nonstress conditions. Due to ACSC growth in liquid medium, the first step was the vacuum extraction of liquid and cell pellets, which were then harvested and used for RNA isolation.

Treatments with Different NO₂-FAs and NO Donors

For qRT-PCR approaches, 14-d-old Arabidopsis plants were carefully extracted from phytoagar, placed on a dish plate, and then incubated for 1 h with NO₂-Ln and Ln, MeOH, and distilled water as controls. In all cases, only roots from these seedlings were in touch with treatments. For RNA isolation, whole 14-d-old plants were analyzed. On the other hand, in 45-d-old Arabidopsis plants, the nutrient solution was removed; root systems were washed with distilled water (Begara-Morales et al., 2014a, 2014b) and then incubated for 3 h with NO₂-Ln and Ln, MeOH and distilled water as controls. Thus, the treatments were performed under nonstress conditions. Then, whole 45-d-old plants from each treatment were harvested and used for RNA isolation. ACSCs were treated with 100 μm NO₂-LA and NO₂-OA for 1 h (using the corresponding nonnitrated fatty acids, vehicle, and distilled water as controls) in the same conditions of ACSC growth (see previous section).

ACSCs were otherwise treated with 100 μm GSNO and S-nitroso-N-acetyl-penicillamine for 1 h in the ACSC growth conditions. GSNO treatment was performed in darkness and using dark-treated distilled water as control. On the other hand, DMSO was used as control for S-nitroso-N-acetyl-penicillamine treatment.

Stress Conditions

Healthy and vigorous 14-d-old Arabidopsis seedlings were selected and exposed to different adverse conditions. Briefly, for mechanical wounding and low temperature, seedlings were manipulated as described previously by Chaki et al. (2011a, 2011b). For the experiments with cadmium stress, seeds were grown in the presence of 150 μm CdCl₂ as carried out elsewhere (Corpas and Barroso, 2013). Finally, ACSCs were treated with 150 mm NaCl as described previously by Fares et al. (2011). Specifically, after 5 and 30 min of salt treatment, liquid medium from ACSCs was extracted by vacuum, washed with PBS solution to further eliminate salt excess, and immediately used for subsequent biochemical approaches.

Fatty Acid Analysis of Arabidopsis

The Arabidopsis lipid fraction of seeds, whole organs of 14-d-old seedlings, leaves from 30- and 45-d-old (senescent) plants, and cell-suspension cultures were analyzed as previously described by Mata-Pérez et al. (2015).

Synthesis and Characterization of NO₂-Ln Standard by NMR Spectroscopy and LC-MS/MS

Because NO₂-Ln is not commercially available, it was prepared following a nitroselenation-oxidation-hydroselenoxide elimination sequence similar to that previously described for nitro-oleic (Baker et al., 2005) and nitro-linoleic (Alexander et al., 2006) acid preparations. Thus, solid mercury chloride (0.380g, 1.4 mmol), phenylselenyl bromide (0.260g, 1.1 mmol), and sodium nitrite, (0.075g, 1.1 mmol) were added to commercial linolenic acid (0.300g, 1.1 mmol) in a mixture of tetrahydrofuran-acetonitrile (1:1, v/v, 7.0 mL) in an Ar atmosphere, and the mixture was stirred at room temperature for 4 h. The solids in suspension were removed by filtration and the solvent was eliminated under reduced pressure. The residue was dissolved in tetrahydrofuran (7.0 mL) and cooled at 0°C in a water-ice bath. A 30% hydrogen peroxide solution (1.2 mL, 11.0 mmol) was then added dropwise and the mixture was maintained and stirred in the cooling bath for a further 20 min. The cooling bath was removed and the reaction was allowed to reach room temperature. The reaction crude was extracted with hexane (2 × 20 mL) and the hexane fraction was washed with saturated aqueous sodium chloride, dried over anhydrous magnesium sulfate, filtered, and evaporated to dryness under reduced pressure. The residue was taken up in a hexane/ether/acetic acid mixture (5 mL, 80:20/1, v/v/v) and purified by flash column chromatography (silica gel 60, 230–400 mesh, Fluka, Buches, Switzerland) with a mixture of hexane/ether/acetic acid (80:20/1, v/v/v) and ensuring the purification of mononitrated linolenic acid. The fractions were analyzed by TLC on silica gel 60 plates (25-μm particle size, 0.2 mm thickness, Fluka Alu foils), eluted with a mixture of hexane/ether/acetic acid (70:30:1, v/v/v) and visualized with iodine vapors. Appropriate fractions were pooled to reach 100 mg (33% yield) of chromatographically pure NO₂-Ln, whose structure was further analyzed by NMR and LC-MS/MS. The NMR data were taken using a Bruker Avance 400 spectrometer (Billerica, MA) operating at 400.13 MHz for ¹H and 100.61 MHz for ¹³C; spectra processing and calculation were performed with ACD/Labs software version 12.01, 2009 (Advanced Chemistry Development, Toronto, Ontario, Canada).

NO₂-LA was synthesized as previously described by Alexander et al. (2006) and NO₂-OA was commercially acquired from Cayman Chemical (Ann Arbor, MI). Synthesis of the standard of deuterated nitro-linolenic acid (D-NO₂-Ln) was carried out by a nitroselenation process as previously described for NO₂-Ln.

Preparation of Lipid Extracts from Arabidopsis

Lipid extracts from Arabidopsis (seeds, whole organs of 14-d-old seedlings, leaves of 30- and 45-d-old plants, and cell-suspension cultures) were prepared using the Bligh and Dyer method (Bligh and Dyer, 1959). Then, the lipid extracts were vacuum-dried (Concentrator Plus, Eppendorf, Hamburg, Germany) and 10 nm of ¹³C₁₈-NO₂-OA was added and used as an internal standard. The lipid fraction was dissolved in 50 μL of methanol and 50 mm phosphate buffer, pH 7.4, treated with 1,000 U/mL pancreatic lipase, 80 U/mL phospholipase A2 from porcine pancreas (Sigma-Aldrich, St. Louis, MO), and 80 U/mL phospholipase A1 from Thermomyces lanuginosus (Sigma-Aldrich). The reaction mixture was incubated at 37°C for 2.5 h with stirring and then samples were displayed at 30% MeOH for solid-phase extraction (C18-SPE). The columns were activated with 6 mL of methanol and preconditioned with 6 mL of 30% methanol. The sample was added to the column, washed with 30% methanol, vacuum-dried for 10 min, and the lipid fraction was eluted with 2 mL of methanol. The resulting extract was evaporated as above and the lipids were dissolved in methanol for LC-MS/MS analysis.

To discard artifactual fatty acid nitration, deuterated-linolenic acid (D-Ln) added to plant material at the beginning of lipid extraction was used as control (Supplemental Fig. S10). This analysis did not show the formation of nitrated D-Ln derivatives (D-NO₂-Ln).

Detection, Identification, and Characterization of Endogenous NO₂-Ln in Arabidopsis

The identification of NO₂-Ln by LC-MS/MS was carried out using a triple quadrupole mass spectrometer (Water Xevo TQS, Manchester UPLC Acquity H-Class) in negative-ion mode. Lipid extracts were separated using an Acquity UPLC BEH C18 column (2.1 mm × 50 mm, 1.7 μm). The nitrated products were eluted from the column at a flow rate of 0.4 mL/min using a solvent system consisting of A (H₂O/0.1% formic acid) and B (acetonitrile/0.1% formic acid) with the following solvent gradient: 10 to 95% B (0–10 min); 95% B (10–11 min); 95 to 10% B (11–13 min). MS analysis was conducted using collision energy ranging from 10 eV to 15 eV and desolvation temperature was set at 400°C. NO₂-Ln was detected using the multiple-reaction monitoring (MRM) scan mode with specific MRM transitions corresponding to nitrated lipids derived from Ln (Baker et al., 2005; Bonacci et al., 2012). In all cases, the data were collected, analyzed, and processed using MassLynx Mass Spectrometry Software (Waters Corporation, Pleasanton, CA).

Nitroalkylation of NO₂-Ln by β-ME in Arabidopsis

Electrophilic activation of NO₂-Ln in Arabidopsis was confirmed by determining the formation of a covalent adduct between NO₂-Ln and β-ME. The resultant lipid extracts were incubated at 37°C for 2 h with 500 mm β-ME and 50 mm phosphate buffer, pH 7.4 (1:1, v/v). Samples were directly analyzed by LC-MS/MS, as described previously, by monitoring the loss of the β-ME transition m/z 78 (Schopfer et al., 2009; Bonacci et al., 2012). β-ME-adducted NO₂-Ln was analyzed in the MRM scan mode, with the β-ME adduct being detected by monitoring for molecules that undergo a M−/[M−β-ME]− transition of m/z 400/322, which corresponds to β-ME-adducted NO₂-Ln.

RNA Sample Preparation and High-Throughput Sequencing

Total RNA from each ACSC experimental condition was pooled and obtained using Trizol Reagent (Gibco-BRL, Life Technologies, Grand Island, NY), as described in the manufacturer’s manual. Two biological replicates of each ACSC condition were sequenced on different lanes in the flow cell. RNA was then purified using a Spectrum Plant Total RNA kit (Sigma-Aldrich), following the manufacturer’s instructions. Any DNA contamination was removed by DNase I treatment on column (Roche, Basel, Switzerland). The quality and the quantity of the total RNA were determined using a Bioanalyzer 2100 and Qubit 2.0. Poly(A)+ mRNA fraction was isolated from total RNA and cDNA libraries were assembled following Illumina’s recommendations. Briefly, poly(A)+ RNA was isolated on poly-T oligo-attached magnetic beads and chemically fragmented prior to reverse transcription and cDNA generation. The cDNA fragments then went through an end-repair process, the addition of a single A-base to the 3′ end and then ligation of the adapters. Finally, the products were purified and enriched with PCR to create the indexed final double-stranded cDNA library. The quality of the libraries was analyzed in a Bioanalyzer 2100, High Sensitivity assay; the quantity of the libraries was determined by real-time PCR in a LightCycler 480 (Roche). Prior to cluster generation in cbot (Illumina, San Diego, CA), an equimolar pooling of the libraries was performed. The pool of the cDNA libraries was sequenced by paired-end sequencing (100 × 2) in an Illumina HiSEquation 2000 sequencer by GeneSystems (Valencia, Spain).

Bioinformatic Analysis

Quality control of sequencing was carried out using FastQC software (V0.11.1, Babraham Bioinformatics, Cambridge, UK). Gene expression was studied using DNAStar (ArrayStar 9) Qseq software for RNA-seq analysis (www.dnastar.com) with the TAIR10 database as a template. For mapping purposes, we used the k-mer = 63 and 95% of matches parameters and the reads-per-kilobase per million-mapped-reads default normalization method. The GO terms were loaded in the Blast2GO suite V.3.0 (Conesa et al., 2005; Conesa and Götz, 2008) to statistically analyze GO-term enrichment. Blast2GO integrated the Gossip package for the statistical assessment of differences in GO-term abundance between two sets of sequences (Blüthgen et al., 2004). This package uses Fisher’s exact test and corrects for multiple testing. A one-tailed Fisher’s exact test was carried out using a FDR with a filter value of <0.1. Blast2GO returns GO terms overrepresented at a specified significance value (Conesa and Götz, 2008). The results were saved in a Microsoft Excel datasheet (Redmond, WA), and charts were generated. Furthermore, the GeneMANIA resource (www.genemania.org) was used to generate network graphs (Warde-Farley et al., 2010).

Crude Extracts of ACSCs

All operations were performed at 0 to 4°C. ACSCs were ground to a powder in a mortar with liquid nitrogen and were suspended in extraction buffer composed by 100 mm Tris-HCl buffer, pH 7.5, containing 0.1 mm EDTA, 7% (w/v) PVPP, 5% Suc, 0.0005% Triton X-100, 1 mm PMSF, 15 mm DTT, and a commercial cocktail of protease inhibitors (AEBSF, 1,10-phenantroline, pepstatin A, leupeptine, bestatine, and E-64 from Sigma-Aldrich; 1/2, FW/v). Homogenates were filtered through one layer of Miracloth (EMD Millipore; Calbiochem, San Diego, CA) and centrifuged at 3,000 × g for 10 min. Total protein content was analyzed by Bradford assay.

Electrophoretic Methods and Immunoblot Analyses

Polypeptides were separated by 12% and 15% SDS-PAGE, and proteins were transferred to PVDF membranes (Immobilon P, Millipore, Bedford, MA) using a Semi-Dry Transfer System (Bio-Rad, Hercules, CA), as described by Corpas et al. (1998). For immunodetection of HSP21 and HSP17.6, commercial rabbit polyclonal antibodies against HSP21 and HSP17.6 obtained from Agrisera (Vännäs, Sweden) and diluted 1:6,000 were used. For APX immunodetection, membrane was incubated with a rabbit polyclonal antibody against cucumber APX (Corpas and Trelease, 1998) diluted 1:2,000. Finally, an antibody against cucumber PMP72 (diluted 1:6,000) that recognizes the HSP70-family protein was used (Corpas and Trelease, 1997; Chaki et al., 2011a, 2011b). Immunoreactive bands were detected using a photographic film (Hyperfilm, Amersham Pharmacia Biotech, GE Healthcare Bio-Sciences, Pittsburgh, PA) with an enhanced chemiluminescence kit (ECL-PLUS, Amersham Pharmacia Biotech).

qRT-PCR

Total RNA from each experimental condition was pooled and isolated as described above and first-strand cDNA was synthesized using the First Strand cDNA Synthesis kit (Roche) at a final volume of 20 μL following the manufacturer’s instructions. Real-time PCR was performed in a CFX384 real-time PCR Detection System (Bio-Rad). Amplifications were carried out in 5 μL of total volume containing 5 ng of cDNA, 2 μm of specific primers (see Supplemental Table S4), and SsoFast EvaGreen Supermix (Bio-Rad). PCR conditions used consisted of an initial denaturation at 98°C for 30 s, followed by 39 cycles at 98°C, 5 s and 60°C, 30 s. After cycling, melting curves of the reaction were run from 72°C to 82°C. Results were normalized using Actin12 (AT3G46520), 18S rRNA (AT2G01010), and L2 (AT2G44065) as internal controls.

Data Availability

The Illumina-sequenced read data reported in this article have been deposited in the National Center for Biotechnology Information Sequence Read Archive and are available under the Accession Numbers Bioproject ID: PRJNA278518 and SRP Study Accession: SRP056239.

Supplemental Data

The following supplemental materials are available.

• Supplemental Information S1. Characterization of NO₂-Ln standard by NMR spectroscopy and LC-MS/MS spectrometry and nitroalkylated adduct formation.

• Supplemental Figure S1. ¹H-NMR of NO₂-Ln.

• Supplemental Figure S2. ¹³C-NMR of NO₂-Ln.

• Supplemental Figure S3. Mass spectrometry analysis of NO₂-Ln.

• Supplemental Figure S4. Nitroalkylation by β-ME of nitro fatty acids in Arabidopsis cell-suspension cultures.

• Supplemental Figure S5. Dose-dependence gene response of NO₂-Ln treatment in ACSCs.

• Supplemental Figure S6. Dose-dependence gene response of NO₂-Ln treatment in ACSCs by qRT-PCR.

• Supplemental Figure S7. Transcriptome profiling of NO₂-Ln-responsive genes at different stages of Arabidopsis development.

• Supplemental Figure S8. Gene expression analysis with other nitro-fatty acids by qRT-PCR.

• Supplemental Figure S9. Gene expression analysis with NO donors by qRT-PCR.

• Supplemental Figure S10. Artifactual nitration analysis during sample preparation.

• Supplemental Table S1. Fatty acid composition of Arabidopsis samples.

• Supplemental Table S2. Up-regulated NO₂-Ln-responsive genes.

• Supplemental Table S3. Down-regulated NO₂-Ln-responsive genes.

• Supplemental Table S4. Oligonucleotides used for qRT-PCR.

Glossary

β-ME

β-mercaptoethanol

ACSC

Arabidopsis cell suspension culture

FC

fold change

FDR

false-discovery rate

GO

gene ontology

GSNO

S-nitrosoglutathione

HSF

heat shock transcription factor

HSP

heat shock protein

HSR

heat shock response

JA

jasmonic acid

LC-MS/MS

liquid chromatography-electrospray ionization tandem mass spectrometry

Ln

linolenic acid

MRM

multiple-reaction monitoring

qRT-PCR

quantitative real-time reverse transcription PCR

RPKM

reads per kilobase per million mapped reads

ROS

reactive oxygen species