Nicotiana benthamiana plants were grown in a growth chamber set at 25 ± 1°C with a 16-h light (70 μmol m^-2 s^-1):8-h dark photoperiod. To generate 35S::sysr1-transgenic plants, the sysr1 coding region was amplified via polymerase chain reactions (PCR) with a forward and reverse primer set (5′-AACACGGGGGACTCTAGAATGATTAACGCCTACGCGGCCC-3′ and 5′-TCGGGGAAATTCGAGCTCTCAATGGCTTAAAACCACACGGT-3′). The amplified fragments were cloned into the XbaI/SacI restriction enzyme sites of pHC30 (Supplementary Figure S2), which was modified from pCAMBIA3300. The resulting pHC30 vector was used to transform N. benthamiana plants, and putative transformants were transferred to soil. DNA isolated from young leaves was used to detect the presence of the transgene via quantitative real-time (qRT)-PCR. Seven independent transgenic lines were established (T₁: Lines 1, 4, and 6–10).

Total RNA was isolated from the collected seedlings using an RNeasy mini kit (Qiagen). Approximately 1 μg DNA-free RNA was used for first-strand cDNA synthesis with M-MuLV reverse transcriptase (Enzynomics). qRT-PCR was conducted using the CFX96 qPCR system (Bio-Rad¹) and SYBR Premix Ex Taq (TaKaRa²), and primers (0.1 μM) were used in a 25-μL final volume. The qRT-PCR protocol was as follows: 95°C for 10 min; 40 cycles of 95°C for 5 s and 60 °C for 20 s. A dissociation curve was subsequently generated. All reactions were completed in triplicate, and details about the qRT-PCR primers are provided in Supplementary Table S1.

Homozygous T₃ sysr1-transgenic N. benthamiana plants (Lines 1, 4, and 7) were analyzed in a salt-stress assay. The seeds of wild-type (WT) and transgenic plants were surface-sterilized and vernalized at 4 °C for 3 days. Samples were then placed in Petri dishes containing Murashige and Skoog (MS) medium (pH 5.7) supplemented with vitamins, 3% sucrose, and 0.4% (w/v) Phytagel. The seeds were incubated at 25 ± 1°C in an illuminated growth chamber. After 2 weeks, the seedlings were transferred to square Petri dishes containing MS agar (0.6% Phyto Agar) medium, which was supplemented with 300 mM NaCl or 400 mM mannitol for salt-stress treatments. After a 4-week incubation, root lengths, numbers of lateral roots, and fresh weights were recorded.

Salt tolerance at the adult stage was evaluated according to the method of Sun et al. (2013). WT and sysr1-overexpressing (sysr1-OX) plants grown on MS agar medium for 4 weeks were transferred to soil, and the samples were then acclimated for 2 weeks. Each plant was then watered with NaCl solution every 3 days. The initial NaCl concentration was 100 mM, and it was then increased in 50-mM increments until a final concentration of 300 mM was reached.

Regarding floating leaf disk assays, 0.8-cm diameter leaf disks (six disks per treatment) were prepared from WT and transgenic leaves at identical developmental stages. The disks were floated on 0 mM (i.e., H₂O) and 300 mM NaCl solutions for 5 days, and they were then treated with 80% aqueous acetone, and the total chlorophyll content was calculated as previously described (Marr et al., 1995). The assays was repeated three times, and mean values were used for analyses.

Z-3-hexenal, E-2-hexenal, and Z-3-hexenol were analyzed using a gas chromatography-mass spectrometry system coupled to a thermal desorption unit (TD-GC-MS). The TD-GC-MS analysis was completed using a GC-MS-QP 2010 Ultra instrument (Shimadzu Corporation, Japan) equipped with an Rtx-5MS column (30 m in length, 0.25 mm internal diameter, and 0.25 μm film thickness; Restek, United States) (Kallenbach et al., 2014). The generated data were processed using GC-MS Solution software (version 4.20, Shimadzu Corporation). E-2-hexenal and Z-3-hexenol were identified based on comparisons with pure standards, while Z-3-hexenal was identified by matching the mass spectrum with data in the NIST14 library and a previously reported retention time (Kallenbach et al., 2014). The peak area of each GLV was normalized based on the peak area at 15.5 min for PDMS tubing pieces, because this peak area was proportional to the PDMS tubing length.

Aliquots of leaf tissue extracts were stored at -80°C until assayed. ADH activity was determined colorimetrically (FLUOstar^® Omega) by quantifying the amount of NADH produced using an Alcohol Dehydrogenase Activity Colorimetric Assay Kit (Biovision).

Two-week-old N. benthamiana plants grown on Murashige and Skoog agar plates (250 cm³) were treated with dichloromethane (DCM; Sigma–Aldrich) or individual GLVs (i.e., E-2-hexenal, Z-3-hexenol, and Z-3-hexenyl acetate; Sigma–Aldrich). Volatiles were diluted with DCM, which does not induce HPL expression. A 2-μL aliquot of 0.1 M volatile solution was applied to 3M^TM Micropore^TM Surgical Tape, which was attached to the inside of the plate cover. The cover was immediately set on the plastic plate, and the plants were incubated for 1 h at 25°C in an illuminated growth chamber (70 μmol m^-2 s^-1). DCM-treated plants were used as controls.

All experiments were repeated three times, and mean values were analyzed with Student’s t-test implemented in the JMPIN program (version: 4.0.4).

To identify components of the salt-tolerance pathway, we compared the gene expression levels of two strains of Synechocystis sp., namely hypersaline lake isolate PCC 6906 (Taxonomy ID 722431) and freshwater isolate PCC 6803 (Taxonomy ID 1148), in response to a high NaCl concentration (data not shown). We isolated salt-responsive gene 1 (designated Synechocystis salt responsive gene 1 (sysr1)) from PCC 6906. A BLAST search against the NCBI non-redundant protein sequence database with the Sysr1 amino acid sequence as the query revealed that Sysr1 is more than 80% identical to AdhA (slr1192 protein). The Synechocystis sp. strain PCC 6803 slr1192 gene encodes a member of the medium-chain ADH family. Additionally, AdhA exhibits NADP-dependent ADH activity, with diverse primary alcohols and aldehydes as substrates (Vidal et al., 2009). An alignment of horse liver ADH (LADH, Equus caballus), AdhA (PCC6803), and Sysr1 (PCC6906) amino acid sequences is presented in Supplementary Figure S1. LADH has previously been used as a standard for comparisons of ADH structures (Eklund et al., 1976). Each ADH sample had the following two major domains: a substrate-binding or catalytic domain, consisting of an N-terminal region with irregular β-coils and a short C-terminal region; and a co-enzyme–binding domain, comprising a duplicated β-sheet known as a Rossmann fold (Rossmann et al., 1974) (Supplementary Figure S1). An analysis of the aligned sequences indicated that the Sysr1 amino acid sequence had characteristics typical of medium-chain ADHs.

The expression of adhA (slr1192) is induced by osmotic (Mikami et al., 2002), salt (Shoumskaya et al., 2005), and heat (Vidal et al., 2009) stresses. To investigate the potential role of Sysr1 in response to salinity stress, transgenic N. benthamiana plants that ectopically express sysr1 were generated (Supplementary Figure S2A), and seven independently transformed tobacco lines were isolated (Supplementary Figure S2B). To assess sysr1 expression levels in transgenic N. benthamiana plants, 3-week-old homozygous T₃ transgenic seedlings were analyzed using qRT-PCR (Figure 1A). We then used an enzyme activity assay to confirm the mRNA data (Figure 1B). Lines 4 and 7 exhibited the highest sysr1 expression levels and ADH activity, so the lines were selected for further analyses (Figures 1A,B). WT and Line 1 plants exhibited similarly low sysr1 expression levels and ADH activity, so Line 1 was used as a control in subsequent experiments (Figures 1A,B). There were no obvious phenotypic differences between the transgenic and WT plants under normal growth conditions (Supplementary Figure S2C).

Plant salt tolerance can be assessed based on the relative plant growth rate after prolonged exposure to a given salt concentration or the plant survival rate after a treatment with a defined salt concentration (Munns, 2002). We analyzed WT and transgenic plants to investigate whether the constitutive expression of sysr1 enhances salt tolerance. Two-week-old seedlings were transferred to plates containing Murashige and Skoog agar medium supplemented with 300 mM NaCl. After 1 month, primary root length and fresh weight data were analyzed. The transgenic seedlings from Lines 4 and 7 grew better than the WT plants (Figures 2A–C). Furthermore, the 300 mM NaCl treatment considerably inhibited the growth of WT and Line 1 control plants, resulting in lower fresh weights than the seedlings of Lines 4 and 7 (Figure 2B). We also assessed plant growth in response to salinity in adult plants grown in soil. Six-week-old WT and sysr1-OX plants grown in the same pot were watered with NaCl solution (100–300 mM) for 1 month. As shown in Figure 3A, WT plants exhibited chlorosis and growth retardation, whereas sysr1-OX tobacco plants grew relatively well, thus demonstrating that ectopic expression of sysr1 significantly enhanced the tolerance of these transgenic plants to salinity. The degree of leaf bleaching provides a visual estimate of the damage caused by salt stress. The effects of salinity stress on chlorophyll content were measured using a floating leaf disk assay. When leaf disks were floated on a 300 mM NaCl solution for 5 days, the disks of WT plants were bleached more intensely than those of sysr1-OX plants (Figures 3B,C). Additionally, decreases in leaf disk chlorophyll levels were greater in WT plants than in sysr1-OX plants (Figures 3B,C). These results indicated that transgenic N. benthamiana plants overexpressing sysr1 were better able to tolerate salinity stress than WT plants. Plant damage caused by high salt concentrations likely varies depending on the age of the plant, and inhibited root growth was clearly observed during the seedling stage. However, in adults, inhibited growth of aerial plant parts and chlorosis of the leaves were more prominent symptoms of salt stress (Figure 3A). These results suggested a positive correlation between ADH activity and salt tolerance in sysr1-OX plants.

To investigate whether this salinity tolerance was attributable to osmotic mechanisms, 2-week-old WT and sysr1-OX seedlings were exposed to mannitol (400 mM). Osmotic stress tolerance was assessed by monitoring primary root elongation 4 weeks later. A significant difference in the root growth rate was observed between transgenic and WT seedlings (Supplementary Figure S3) and these results implied that sysr1 overexpression conferred tolerance to salinity and osmotic stresses.

Previous studies indicated that ADH is responsible for the conversion of C₆ aldehydes to their corresponding alcohols (Bicsak et al., 1982; Longhurst et al., 1990). Therefore, we compared the quality of the emitted GLVs between WT and sysr1-OX plants. The volatiles from transgenic leaves with enhanced ADH levels were analyzed using GC-MS (Kallenbach et al., 2014) to determine the effects of sysr1 overexpression on the relative amount of volatile aldehydes and alcohols in wounded leaves. The results indicated that Sysr1 was involved in the interconversion of aldehydes and alcohols in transgenic N. benthamiana leaves. The WT control plants produced more Z-3-hexenal than Z-3-hexenol within 1 h of wounding (Figure 4). However, leaf Z-3-hexenol levels were approximately twofold higher in sysr1-OX plants than in WT plants within 1 h of wounding. Results of the GC-MS analysis revealed that the conversion of Z-3-hexenal (C₆ aldehyde) to its corresponding alcohol, Z-3-hexenol (C₆ alcohol), was at least partially mediated by Sysr1 in transgenic leaves. These data indicated that Sysr1 functions as an ADH in sysr1-OX plants.

The expression of some stress-related genes was analyzed using qRT-PCR to determine how sysr1 overexpression increased salt tolerance. Genes encoding key signaling factors for abiotic stress response pathways [e.g., dehydration-responsive element binding protein 2a (DREB2A), heat shock protein 17.6 (HSP17.6), responsive to desiccation 29 (RD29B), and HPL] were more highly expressed in transgenic plants than in WT plants under normal growth conditions and after treatment with 300 mM NaCl (Figure 5). We sampled plants 1 h after salt treatment to highlight the phenotypes of sysr1-OX plants. Even in the absence of salt stress, greater accumulation of transcripts associated with stress-related genes was observed in sysr1-transgenic plants compared to WT plants. This phenomenon is similar to the priming effect observed in plants exposed to salt stress in advance. Therefore, these observations implied that the greater salt tolerance of sysr1-OX plants was relevant to the elevated expression levels of stress response genes.

Environmental stresses increase the quantity and quality of volatile organic compounds (VOCs) emitted by plants (Loreto and Schnitzler, 2010). A recent study concluded that salt-responsive A. thaliana VOCs induce salt tolerance in neighboring plants (Lee and Seo, 2014). To determine whether salt stress promotes the emission of VOCs from sysr1-OX plants to enhance salt tolerance in neighboring WT plants, we treated WT and sysr1-OX plants with 300 mM NaCl and investigated whether VOCs released from the sysr1-transgenic plants induced salt tolerance in WT plants. Two-week-old WT and sysr1-OX seedlings were transferred to MS agar medium supplemented with 300 mM NaCl. We used two-compartment plates, which contained WT seedlings in one compartment and sysr1-OX or WT seedlings in the other. The plates, which allowed the exchange of airborne signals between compartments, were sealed and incubated at 25 ± 1°C for 4 weeks. We then examined the growth of WT plants that were grown with WT or sysr1-transgenic plants exposed to 300 mM NaCl. The salt tolerance of WT plants was enhanced in the presence of sysr1-transgenic plants, suggesting that the VOCs emitted from the sysr1-transgenic plants enhanced the salt tolerance of neighboring WT plants (Figures 6A–C). We also assessed the volatile effects of sysr1-OX plants on salt tolerance in adult plants grown in soil. WT and sysr1-OX plants grown in the same pot received water supplemented with NaCl (100–300 mM). Unfortunately, we did not observe acquired salt tolerance in WT plants that were grown together with adult sysr1-OX plants in the same pot (Figure 3A).

Even in the absence of salt stress, HPL transcripts accumulated more in sysr1-transgenic plants than in WT plants (Figure 5). HPL is important for GLV biosynthesis in N. attenuata (Allmann et al., 2010). Additionally, sysr1 overexpression modifies the balance between Z-3-hexenal and Z-3-hexenol in transgenic leaves. Therefore, we speculated that GLVs might be airborne signals. Furthermore, sysr1-OX and WT plants may differ with regard to the quality or quantity of GLVs emitted in response to high-salt conditions. To elucidate the molecular mechanisms underlying the induction of salt tolerance in plants neighboring sysr1-OX seedlings, we compared the effects of GLVs on the expression of defense-related genes. GLVs comprise a family of C₆ compounds, including E-2-hexenal, Z-3-hexenol, and hexenyl derivative Z-3-hexenyl acetate. WT plants were treated with pure vaporized C₆ compounds (10 nmol cm^-3 for 1 h). The seedlings were collected 0.5 and 1 h after initiating treatment, because the focus of this study was early transcriptional changes induced by GLVs. Earlier studies revealed that DREB2A transcript levels were highest 0.5 h after samples were exposed to E-2-hexenal (Yamauchi et al., 2015). We observed a transient increase in DREB2A transcript abundance at 0.5 h, and a subsequent decrease was detected after 1 h in WT plants treated with vaporized E-2-hexanal. In contrast, DREB2A transcription levels in DCM-treated control plants remained low (Figure 7). We also tested the effects of other GLVs on selected stress-related transcript, and vaporized Z-3-hexenol and Z-3-hexenyl acetate induced the expression of DREB2A, RD29B, and HPL (Figure 7). Moreover, Z-3-hexenol and Z-3-hexenyl acetate upregulated the expression of selected stress-related genes more than E-2-hexenal (Figure 7).

We assessed GLV-induced salt tolerance based on root length and the number of lateral roots, because GLV treatments enhanced the expression of HPL, DREB2A, and RD29B (Figure 7), which contribute to abiotic stress tolerance (Sakuma et al., 2006a). After 2 weeks of growth on basal Murashige and Skoog agar plates, WT seedlings were treated with DCM (i.e., vaporized solvent control) and three synthetic GLVs for 1 h, and samples were then transferred to vertical Murashige and Skoog agar plates supplemented with 300 mM NaCl. Regarding primary root length and the number of lateral roots, seedlings pretreated with Z-3-hexenol or Z-3-hexenyl acetate grew better than solvent control-treated seedlings, thus indicating the physiological importance of GLVs in salt-stress responses. However, the E-2-hexenal pretreatment did not enhance salt tolerance (Figure 8), and these results were consistent with the expression patterns of stress-related genes in GLV-treated seedlings (Figure 7). Comparing salt-induced growth inhibition after treatments with three different GLVs indicated that C₆ alcohol and C₆ ester forms of GLVs increased salt tolerance more than the C₆ aldehyde form.

Besides well-established ADH functions that occur during seed development and responses to flooding stress, there is mounting evidence that ADH mediates tolerance to other abiotic stresses. Several previous studies of model plants confirmed that ADH expression is influenced by stress (Matton et al., 1990; Christie et al., 1991; Ingersoll et al., 1994; Bucher et al., 1995), and it is also linked to changes in secondary metabolism (Bicsak et al., 1982; Longhurst et al., 1990; Speirs et al., 1998). Although ADH expression is induced by salt stress (Vidal et al., 2009; Zhang et al., 2016), the role of ADH during salt-tolerance signaling has not been established. In this study, we tested whether transgenic plants carrying Synechocystis sp. ADH exhibit salinity tolerance. We also attempted to characterize the mechanism responsible for the correlation between salinity tolerance and ectopic sysr1 expression (Figures 2, 3). Our findings clearly indicated that sysr1 overexpression increases the salt tolerance of N. benthamiana plants.

The results of our gene expression analyses may help explain increased salt tolerance of sysr1-OX plants. The overexpression of sysr1 in N. benthamiana plants upregulated the expression of the abiotic stress-related genes DREB2A, RD29B, HSP17.6, and HPL (Figure 5). The enhanced expression of DREB2A reportedly increases rice tolerance to dehydration and salt stress conditions (Mallikarjuna et al., 2011), and RD29A and RD29B are specific targets of the DREB2A transcription factor (Sakuma et al., 2006a,b). These homologous genes are highly sensitive to various abiotic stressors. For example, cold, drought, and salt stresses induce RD29A and RD29B expression. However, the RD29A promoter is more responsive to drought and cold stresses, and the RD29B promoter is highly responsive to salt stress (Msanne et al., 2011). HSP17.6A encodes a small heat-shock protein belonging to the A. thaliana cytosolic class II family, and it is expressed during development and stress responses. Furthermore, overproduction of HSP17.6A and NtHSP70-1 can increase salt or drought tolerance in plants (Sun et al., 2001; Cho and Hong, 2006). Because the expression levels of abiotic stress-related genes increased in sysr1-OX plants, we speculated that stress-induced signal transduction occurs faster in transgenic plant cells, resulting in faster and stronger activation of salt-tolerance-related responses. Upon exposure to salt stress, a set of signaling proteins is activated, thus augmenting salt-stress responses.

A distinct characteristic of sysr1-transgenic plants is the exhibition of greater ADH activity than the WT plants. Consequently, Z-3-hexenol was more abundant than Z-3-hexenal (Figure 4). Previous reports indicated that adh1 mutant plants released less hexanol and Z-3-hexenol than WT plants, but more E-2-hexenal was produced (Chang and Meyerowitz, 1986; Strommer, 2011). The overexpression of ADH in tomato plants changes the balance between the C₆ aldehydes and alcohols in ripened fruits (Speirs et al., 1998). We speculated that the increase in Z-3-hexenol content in sysr1-OX N. benthamiana plants may influence the transcription of abiotic stress-related genes. Short-chain leaf volatiles (e.g., E-2-hexenal) can strongly induce the expression of abiotic stress-related transcription factor genes such as DREB2A (Yamauchi et al., 2015). However, we observed that the expression levels of abiotic stress-related genes were more than twofold higher in plants treated with vaporized Z-3-hexenol and Z-3-hexenyl acetate than in plants exposed to E-2-hexenal (Figure 7). Thus, our data indicate that GLVs formed in sysr1-transgenic plants can upregulate gene expression, leading to stronger effects of Z-3-hexenol than E-2-hexenal.

Green leaf volatiles are produced in reactions catalyzed by HPL, which is a component of the lipoxygenase pathway. In the lipoxygenase/HPL pathway, the plant first produces C₆ aldehydes, which are then converted to C₆ alcohols (e.g., Z-3-hexenol) by ADH (Matsui, 2006). In plant communities, GLVs are important infochemicals that mediate plant–plant and plant–insect interactions. In particular, Z-3-hexenol and Z-3-hexenyl acetate are associated with plant–plant communication (Engelberth et al., 2004; Farag et al., 2005; Heil and Silva Bueno, 2007; Frost et al., 2008; Heil et al., 2008). Airborne Z-3-hexenol from wounded plants may trigger pre-defense reactions in neighboring healthy plants, enabling faster and stronger responses during subsequent attacks. This phenomenon is called plant–plant communication or the priming effect of volatiles (Wei and Kang, 2011), but the physiological and molecular mechanisms responsible for GLV-induced priming have not been characterized. Priming often results in the enhanced transcription of defense-related genes (Thulke and Conrath, 1998; van Wees et al., 1999; Zimmerli et al., 2000; Kohler et al., 2002). Thus, transcription factors are important for the regulation of priming effect initiation (Van der Ent et al., 2009).

Unfortunately, we were unable to identify the salt-induced GLVs released in WT and sysr1-transgenic plants, and this likely occurred because GLVs were released at very low levels. However, the results of our experiment using two-compartment plates suggested that airborne signals from salt-stressed sysr1-transgenic plants enhanced the salt tolerance of neighboring plants (Figures 6A–C). The priming effect was observed because plants were located in a small enclosed space, and neighboring plants were exposed to relatively high concentrations of volatile components for an extended period. In contrast, WT and sysr1-OX plants grown in soil (in an open space) did not exhibit a priming effect, because they were only briefly exposed to relatively low concentrations of volatile components. Therefore, exposure to sufficient concentrations of volatile components for an adequate period is required for the induction of a priming effect between neighboring plants. However, we observed that vaporized Z-3-hexenol and Z-3-hexenyl acetate considerably increased salt tolerance in neighboring WT plants (Figure 8). Interestingly, E-2-hexenal had relatively little priming effects. Shiojiri et al. (2012) exposed healthy A. thaliana plants to 140 ppt GLVs from wounded neighboring plants twice per week for 3 weeks, and this concentration triggered a response in the healthy plants. Although we were unable to measure salt-induced GLVs, the aforementioned results suggest that a very low concentration of salt-induced GLVs can trigger salt tolerance in neighboring plants. Therefore, increases in ADH activity may affect the salt tolerance of neighboring plants by changing the balance between emitted Z-3-hexenol and E-2-hexenal.

In plants, ADH enzymes have multiple functions related to anaerobic and aerobic fermentation as well as the production of scents that discourage predation, attract pollinators, and facilitate seed dispersal. In particular, sysr1 overexpression affects the quality of stress-inducible GLVs, resulting in the upregulation of expression of stress-related genes. These changes may be associated with observed enhanced salt tolerance of sysr1-OX plants and neighboring plants. Our results suggested that the increased salt tolerance of sysr1-OX plants may have resulted from increased expression of stress-related genes, which was caused by enhanced Z-3-hexenol production. In this study, we could not explain why the priming effect associated with the induction of salt tolerance in neighboring plants was only observed in seedlings cultivated in an airtight container. Experiments designed to fully characterize the molecular mechanisms associated with the regulation of the salt tolerance priming effect of sysr1-OX plants are currently in progress.