1. Introduction

Soil salinity, which is a major agricultural problem, limits crop productivity and threatens ecological security worldwide (Zhou et al., 2023). Salinity stress is a primary abiotic factor that leads to osmotic stress, oxidative damage, and ionic imbalance, adversely affecting plant growth and development (Zhao et al., 2020). To cope with salinity stress, plants have evolved intricate mechanisms involving altered physiological processes and growth adjustment (Munns and Tester, 2008; Zhao et al., 2020; Zhou et al., 2023). Phytohormones play crucial roles in plant responses to salinity stress (Yu et al., 2020). For example, the ethylene signaling pathway positively regulates Arabidopsis thaliana salt tolerance by activating ETHYLENE INSENSITIVE3 (EIN3) to prevent the excessive accumulation of reactive oxygen species (ROS) (Peng et al., 2014). The abscisic acid (ABA) pathway positively regulates salt tolerance by activating SNF1-RELATED PROTEIN KINASE2.6 (SnRK2.6), which modulates stomatal aperture by regulating ion transport (Osakabe et al., 2013). Other phytohormones, including salicylic acid and jasmonic acid, also contribute to plant salinity stress responses (Toda et al., 2013; Jayakannan et al., 2015).

Jasmonic acid and its derivatives, collectively called jasmonates (JAs), are widely distributed in the plant kingdom (Staswick and Tiryaki, 2004; Wasternack and Song, 2017). Previous studies showed that JA regulates various biological processes, including seed germination, primary root elongation, root hair development, anthocyanin accumulation, and stamen development (Shan et al., 2009; Chen et al., 2011; Song et al., 2011; Han et al., 2020, 2023a; Pan et al., 2022). Additionally, JA is a crucial defense signal protecting plants from biotic and abiotic stresses (Lorenzo et al., 2004; Schweizer et al., 2013; Hu et al., 2017; Mao et al., 2017; Howe et al., 2018; Xiang et al., 2022). Environmental stresses trigger the synthesis of jasmonic acid-isoleucine (JA-Ile), which is the most bioactive endogenous JA; JA-Ile is subsequently recognized by the F-box protein CORONATINE INSENSITIVE1 (COI1), which is a component of the SCF^COI1 protein complex (Xie et al., 1998; Xu et al., 2002; Yan et al., 2009; Dar et al., 2015; Howe et al., 2018; Hu et al., 2023). Upon perception of JA-Ile, the SCF^COI1 complex targets JASMONATE ZIM-DOMAIN (JAZ) proteins, which are repressors of JA signaling, for degradation via the 26S proteasome pathway (Chini et al., 2007; Thines et al., 2007; Yan et al., 2009). This degradation leads to the release of downstream transcription factors, enabling the activation of JA-responsive genes (Fernandez-Calvo et al., 2011; Kazan and Manners, 2013; Wasternack and Song, 2017).

JAZ proteins negatively regulate JA signaling by inhibiting the functions of multiple transcription factors through physical interactions (Fernandez-Calvo et al., 2011; Qi et al., 2011; Zhu et al., 2011; Kazan and Manners, 2013; Jiang et al., 2014; Han et al., 2023b). For example, JAZ proteins target MYC2, a basic helix-loop-helix (bHLH) transcription factor, to modulate primary root elongation (Chini et al., 2007; Chen et al., 2011), whereas they target the R2R3-MYB transcription factors MYB21 and MYB24 to regulate stamen development (Song et al., 2011). JAZ proteins also inhibit the activity of the transcription factor ROOT HAIR DEFECTIVE6 (RHD6) to negatively regulate root hair development (Han et al., 2020). In addition to their roles in developmental processes, JAZ proteins are extensively involved in stress responses. They negatively regulate MYC3 and MYC4, which are phylogenetically close to MYC2 and act additively with MYC2 to protect against bacterial pathogens and insect herbivory (Fernandez-Calvo et al., 2011). The interactions between the bHLH transcription factor INDUCER OF CBF EXPRESSION-C-REPEAT BINDING FACTOR/DRE BINDING FACTOR1 (ICE1) and JAZ proteins influence the regulation of freezing tolerance (Hu et al., 2013). Furthermore, in rice, RICE SALT SENSITIVE3 (RSS3) combines with JAZ proteins and class C bHLH transcription factors to form a ternary complex that regulates root cell elongation during adaptation to salinity (Toda et al., 2013).

Previous studies revealed a correlation between salinity stress and JA signaling (Zhao et al., 2014; Chen et al., 2016, 2017; Valenzuela et al., 2016). Specifically, the expression of JA-responsive genes is induced by NaCl treatments (Chen et al., 2016, 2017; Valenzuela et al., 2016). Moreover, mutations to JA biosynthesis and signaling genes (AOS, COI1, and MYC2/3/4) result in mutants with a longer root elongation zone than wild-type (WT) plants under saline conditions (Valenzuela et al., 2016), implying salt-induced root growth inhibition depends on the JA pathway. Despite these earlier findings, the molecular mechanism underlying the enhancement of JA signaling in response to salinity stress in Arabidopsis remains poorly understood. Hence, this study was conducted to elucidate this mechanism. We confirmed that salinity stress enhances JA-mediated root growth inhibition, anthocyanin accumulation, and the upregulation of JA-responsive genes. Moreover, we determined the importance of the COI1-mediated JA signaling for salinity stress-enhanced JA responses. Additionally, phenotypic analyses showed the negative regulatory effects of JAZ proteins on salinity stress-enhanced JA signaling. Further analyses detected physical interactions between several JAZ proteins and members of the basic leucine zipper (bZIP) domain family, including ABSCISIC ACID-RESPONSIVE ELEMENT BINDING FACTOR1 (ABF1), AREB1/ABF2, ABF3, and AREB2/ABF4. These transcription factors respond to salinity stress and positively regulate salt tolerance in plants (Uno et al., 2000; Yoshida et al., 2015; Du et al., 2023). The overexpression of ABF3 increased the sensitivity to salinity stress-enhanced JA responses, while loss-of-function mutations to ABFs attenuated JA-related phenotypes under saline conditions. Genetic evidence indicated that ABF3-promoted JA signaling following an exposure to salinity stress is compromised by JAZ1. Moreover, ABF3 was observed to indirectly activate ALLENE OXIDE SYNTHASE (AOS) transcription, but this activation was suppressed by JAZ1. In addition, ABF3 competes with MYC2 for the binding to JAZ1. Collectively, our findings suggest that ABF3 positively regulates JA signaling under salinity stress. Thus, this study provides mechanistic insights into the enhancement of JA signaling in response to salinity stress.

2. Materials and methods 2.1. Materials and growth environment

Phanta Max Super-Fidelity DNA polymerase was purchased from Vazyme Biotech, restriction enzymes were obtained from Thermo Fisher Scientific, methyl jasmonate (MeJA) was acquired from Sigma–Aldrich and common chemicals were form Shanghai Sangon Biotechnology. The wild-type (WT) and mutant Arabidopsis thaliana in current study were in the Columbia (Col-0) background. The mutants or transgenic plants coi1-1 (Xie et al., 1998), coi1-2 (Xu et al., 2002), coi1-16 (Ellis and Turner, 2002), jazQ (jaz1 jaz3 jaz4 jaz9 jaz10) (Campos et al., 2016), JAZ1-Δ3A (Thines et al., 2007), areb2 abf3 abf1-2, areb1 areb2 abf3 and areb1 areb2 abf3 abf1-2 (Yoshida et al., 2010, 2015) were described previously. Two transgenic lines overexpressing ABF3 from Du et al. (2023), ABF3-OE-L5 and ABF3-OE-L6, were selected for phenotypic analyses. ABF3-OE-L5 JAZ1-Δ3A plants were generated via standard genetic hybridizations.

For phenotypic analyses, Arabidopsis seeds were surface-sterilized for 8 min in 20% (v/v) sodium hypochlorite, sown on half-strength Murashige and Skoog (MS) medium supplemented with 0.8% (w/v) agar and 1% (w/v) sucrose, stratified at 4 ℃ for 1 day before germination. All Arabidopsis seedlings used for phenotypic analyses and propagation were grown in a growth chamber at 22 ℃ under long-day conditions, with a 16-h light (100 μE m⁻² s⁻¹, white fluorescent bulbs, full light spectrum): 8-h dark cycle. Arabidopsis plants for dual-luciferase (Dual-LUC) assays and Nicotiana benthamiana plants for bimolecular fluorescence complementation (BiFC) assays were cultivated at 22 ℃ under short-day conditions (8-h light: 16-h dark).

2.2. MeJA and NaCl treatments

Although MeJA lacks biological activity, it can be cleaved by esterase and converted to JA-Ile by JAR1 in vivo (Staswick and Tiryaki, 2004; Stitz et al., 2011). For phenotypic analyses, the MeJA stock solution was diluted (5 or 10 μM final concentration) in half-strength MS medium with or without 125 mM NaCl. Seeds were germinated on half-strength MS medium containing MeJA, NaCl, or both. As the control, seeds were germinated on half-strength MS medium without MeJA and NaCl. The seedlings used for a reverse transcription quantitative PCR (RT-qPCR) analysis were immersed in the double-distilled water, solution with 100 μM MeJA, 250 mM NaCl, or solution containing both 100 μM MeJA and 250 mM NaCl.

2.3. Root length and anthocyanin content measurements

To measure the primary root length of seedlings, seeds were sown on half-strength MS medium with/without NaCl (NaCl+/NaCl−) or MeJA (MeJA+/MeJA−) in plates, stratified at 4 ℃ for 1 day, and germinated. The plates were maintained in a vertical position under long-day conditions for 7 days. At least three biological replicates with similar results were used. Specifically, the primary root length was measured for 20 seedlings per biological replicate, after which the mean length was calculated for the analysis of variance (ANOVA).

To measure the anthocyanin content of seedlings, seeds were sown on half-strength MS medium with/without NaCl (NaCl+/NaCl−) or MeJA (MeJA+/MeJA−) in plates, stratified at 4 ℃ for 1 day, and germinated. The plates were maintained in a horizontal position under long-day conditions for 5 days. The anthocyanin content was expressed as (A₅₃₀ − A₆₀₀) per gram fresh weight. This analysis was repeated three times, with similar results.

2.4. RNA extraction and RT-qPCR

Seven-day-old seedlings were immersed for 4 h in double-distilled water, 100 μM MeJA solution, 250 mM NaCl solution, or 100 μM MeJA and 250 mM NaCl solution. The treated seedlings were immediately frozen in liquid nitrogen and stored at −80 ℃. Total RNA was extracted from the frozen samples using TRIzol reagent (Invitrogen) as described previously (Pan et al., 2020). Purified RNA was reverse transcribed to cDNA using the PrimeScript™ RT Reagent Kit (Takara). To conduct an RT-qPCR analysis, 1 μl cDNA was used along with SYBR qPCR SuperMix (Novoprotein) and a LightCycler 480 real-time PCR instrument. ACTIN2 (AT3G18780) was used as the internal reference gene. The gene-specific RT-qPCR primers in this study are listed in Supplemental Data Set1.

2.5. Yeast two-hybrid (Y2H) assays

To construct ABF bait vectors (BD-ABF1, BD-AREB1, BD-ABF3, and BD-AREB2) for Y2H assays, each full-length coding sequence (CDS) was inserted into a separate pGBKT7 vector. To verify ABF-JAZ interactions, each of 12 JAZs (JAZ1, JAZ2, JAZ3, JAZ4, JAZ5, JAZ6, JAZ7, JAZ8, JAZ9, JAZ10, JAZ11, JAZ12) was introduced into separate prey vectors (pGADT7) as described previously (Pan et al., 2020). To investigate the critical region for protein interaction, multiple truncated ABF3 and JAZ1 sequences were cloned into pGBKT7 and pGADT7, respectively. Yeast strain AH109 cells were co-transformed with bait and prey vectors and then cultured at 28 ℃ on a dropout medium lacking tryptophan (Trp) and leucine (Leu) or medium lacking Trp, Leu, histidine (His), and adenine (Ade). This experiment was performed at least three times. The primers for vector construction are listed in Supplemental Data Set1.

2.6. Bimolecular fluorescence complementation (BiFC) assays

The full-length ABF3 CDS and sequences encoding N-terminal amino acids 1–114 and 115–201 were cloned into pFGC-cYFP to obtain ABF3-cYFP, ABF3¹⁻¹¹⁴-cYFP, and ABF3^115−201-cYFP, respectively. In addition, the full-length JAZ1 CDS and sequences encoding amino acids 122–154 and 202–253 were fused to the sequence encoding the YFP N-terminus. As described by Li et al. (2023), Agrobacterium tumefaciens strain EHA105 cells were transformed with the recombinant plasmids for the subsequent transient infiltration of Nicotiana benthamiana leaves. The infected tobacco plants were incubated under dark and humid conditions for 48 h. The primers for vector construction are listed in Supplemental Data Set1.

2.7. Dual-Luciferase (Dual-LUC) assays

The putative AOS promoter sequence was amplified by PCR and inserted into the pGreenII 0800-LUC vector to generate the ProAOS: LUC reporter construct. The full-length JAZ1, ABF3, and GFP CDSs were amplified by PCR and incorporated into separate pGreenII 62-SK vectors (i.e., effectors). Different recombinant plasmid combinations were inserted into WT and ABF3-OE-L6 leaf mesophyll protoplasts as described previously (Sheen, 2001). Transfected protoplasts were cultured under low-light conditions for 16 h before analyzing the relative LUC activity using a Dual-LUC Reporter Assay system, which measures the firefly LUC and Renilla reniformis (internal control) LUC (REN) activities. The primers for vector construction are listed in Supplemental Data Set1.

2.8. Protein extraction, immunoblotting and co-immunoprecipitation (Co-IP) assay

Proteins were extracted from protoplasts using an extraction buffer containing 50 mM Tris-HCl (pH 7.4), 1 mM EDTA, 150 mM NaCl, 10% (v/v) glycerol, 0.1% (v/v) Triton X-100, 5 mM dithiothreitol, 1 mM PMSF, and 1 × complete protease inhibitor cocktail (Roche). Immunoprecipitation experiments were conducted using Protein A/G Plus-Agarose beads (Santa Cruz Biotechnology). Briefly, cell lysates were precleared using Protein A/G Plus-Agarose beads, which were then incubated with the anti-MYC antibody in the extraction buffer at 4 ℃ overnight. The beads were washed three times with extraction buffer, after which the co-immunoprecipitated proteins were detected by immunoblotting using an anti-MYC or anti-HA antibody.

For immunoblotting, 1 × Laemmli SDS-PAGE protein loading buffer was added to the extracts, which were then heated at 95 ℃ for 10 min. Samples were centrifuged (12, 000 g for 5 min) and then a 10 μl aliquot of each supernatant was analyzed by 10% SDS-PAGE gel electrophoresis. The separated proteins were transferred to a PVDF membrane (Immobilon), which was blocked for 1 h in 5% (w/v) reconstituted nonfat skim milk powder in TTBS buffer at room temperature and then incubated with an anti-MYC antibody (1:250; catalog no. A7470, Sigma–Aldrich) or an anti-HA antibody (1:10, 000; catalog no. F7425, Sigma–Aldrich) at 4 ℃ overnight. Next, the membrane was incubated with an HRP-conjugated goat anti-mouse secondary antibody (1:10, 000; D110087, Sangon Biotech, Shanghai, China). Luminescence was detected using Chemiluminescent HRP substrate (Millipore).

For the Co-IP assay, full-length JAZ1 and ABF3 CDSs fused with sequences encoding MYC-tag and HA-tag, respectively, were inserted into separate pGreenII 62-SK vectors. Arabidopsis leaf mesophyll protoplasts were transformed with different recombinant plasmid combinations as described previously (Sheen, 2001). The primers for vector construction are listed in Supplemental Data Set1.

2.9. Yeast one-hybrid (Y1H) assays

The Matchmaker Yeast One-Hybrid System Kit (Clontech) was used to conduct Y1H assays. The full-length ABF3 CDS was cloned into pGADT7 to generate the AD-ABF3 construct. Putative AOS promoter fragments were cloned into the pAbAi vector to generate pAbAi-pAOS.1, pAbAi-pAOS.2, and pAbAi-pAOS.3, which were linearized by BstBⅠ prior to the transformation of Y1HGold yeast cells. The transformed cells were grown on a dropout medium lacking uracil (Ura) for 3 d. Next, the AD-ABF3 construct was inserted into cells containing linearized pAbAi-pAOS. The transformed cells were selected on dropout medium lacking Leu. Co-transformed cells were cultured for 3 d on a dropout medium lacking Leu, but supplemented with 200 μg/L aureobasidin A (AbA). Positive clones were obtained for several yeast cell concentrations from 10⁰ (OD₆₀₀ = 0.8) to 10⁻³. The primers for vector construction are listed in Supplemental Data Set1.

2.10. RNA-sequencing (RNA-seq)

The samples used for RNA-seq analyses were prepared as described in section 2.4. The RNA extraction and sequencing were completed by OE Biotech, Inc., Shanghai, China. The raw data generated by the Illumina NovaSeq 6000 platform were filtered and processed using the fastp software to obtain high-quality clean data. The HISAT2 software was used for calculating fragments per kilobase million (FPKM) values (Supplemental Data Set2). The HTSeq-count software was used to determine the number of reads (counts) for each gene. Differential expression was analyzed using the DESeq2 software. Differentially expressed genes (DEGs) are provided in Supplemental Data Set2. A Gene Ontology (GO) analysis was performed using OECloud tools (https://cloud.oebiotech.com/task/). The results of the functional categorization of salt stress-responsive genes are shown in Supplemental Data Set2. A gene expression heatmap was generated using R (https://www.r-project.org/) of the OECloud platform.

2.11. Data analysis

Data were analyzed using Student's t-test or one-way or two-way ANOVA (SPSS Statistics 26); P < 0.05 was set as the threshold for significance.

3. Results 3.1. Salinity stress enhances jasmonate-induced root growth inhibition and anthocyanin accumulation

Previous research showed that salinity stress activates the expression of JA-responsive genes (Valenzuela et al., 2016; Chen et al., 2017). To further investigate the potential enhancement of plant responses to exogenous JA under saline conditions, we examined the phenotype of wide type (WT) seedlings grown on half-strength MS medium containing MeJA, NaCl, or both MeJA and NaCl, with seedlings grown on normal half-strength MS medium serving as the control. Inhibition of root elongation is a typical response to MeJA in plants (Chen et al., 2011; Han et al., 2023a). We observed that seedlings treated with NaCl exhibited shorter primary roots when exposed to MeJA, compared to the mock-treated seedlings (Fig. 1A and B). Moreover, MeJA-induced root growth inhibition was further enhanced under saline conditions (Fig. 1D). In addition to root growth, we measured the content of anthocyanin, another characteristic phenotype of jasmonate, in seedlings subjected to MeJA and NaCl treatment. However, 7-d-old WT seedlings growth on half-strength MS media containing 5 or 10 μM MeJA displayed no significant difference in anthocyanin content compared to the control (Fig. S1). In contrast, 5-d-old seedlings exposed to 5 or 10 μM MeJA exhibited a significant difference (Fig. S1). Consequently, we adjusted the treatment to 10 μM MeJA for 5 d for a pronounced effect on anthocyanin accumulation. As shown in Fig. 1C, the NaCl-treated seedlings accumulated more anthocyanin in presence of MeJA compared to the mock-treated seedlings. Furthermore, MeJA-induced anthocyanin accumulation was further enhanced when applied NaCl (Fig. 1E). These findings indicated that salinity stress enhances JA response.

3.2. Salinity stress activates jasmonate signaling pathway

Genes responsive to salinity and MeJA were analyzed on the basis of an RNA-seq analysis of WT seedlings treated with 100 μM MeJA, 250 mM NaCl, or both for 4 h. A total of 2447 DEGs were responsive to exogenous MeJA, 8873 DEGs were responsive to NaCl, and 9534 DEGs were responsive to the combined MeJA and NaCl treatment (Fig. 2A). Additionally, 1570 DEGs were responsive to both MeJA and NaCl, suggestive of a crosstalk between JA and salinity stress responses. We performed a GO analysis of the DEGs upregulated by salinity stress (Fig. 2B), which indicated some pathways, including those associated with responses to ABA, wounding, salt stress, osmotic stress, and jasmonic acid, were significantly upregulated by salinity stress. A gene expression heatmap of the DEGs regulated by JA or salinity stress (Fig. 2C) showed that the NaCl treatment upregulated the expression of certain JA-inducible genes involved in JA synthesis (e.g., LOX3 and OPR1) and JA signaling (e.g., JAZ5 and MYC2).

To verify these findings, we conducted an RT-qPCR analysis of several JA-responsive genes, including ALLENE OXIDE SYNTHASE (AOS), ALLENE OXIDE CYCLASE2 (AOC2), LIPOXYGENASE2 (LOX2), LIPOXYGENASE3 (LOX3), JAZ1, JAZ5, JAZ8, JAZ10, and VEGETATIVE STORAGE PROTEIN2 (VSP2). The results showed that the induction of JA-responsive genes by exogenous MeJA was further enhanced by salinity stress (Fig. S2). Collectively, these results provide the evidence that salinity stress enhances JA signaling in plants.

3.3. Salinity stress enhances COI1-mediated jasmonate signaling

To clarify the molecular mechanism underlying salinity stress-enhanced JA signaling, we investigated whether the necessary components of the JA signaling pathway are involved. The F-box protein COI1 is a JA receptor that positively regulates JA signaling (Yan et al., 2009, 2013). Previous studies showed that coi1 mutants are insensitive to exogenous MeJA (Shan et al., 2009; Song et al., 2011; Huang et al., 2014; He et al., 2023). Consistent with these earlier findings, the coi1-1, coi1-2, and coi1-16 mutants had longer roots and accumulated less anthocyanin than the WT control when treated with exogenous MeJA (Fig. 3A–C). The changes in root length and anthocyanin accumulation under saline conditions were smaller for coi1-1, coi1-2, and coi1-16 than for the WT control (Fig. 3D and E), reflecting the importance of the COI1-mediated JA signaling for the activation of JA responses following an exposure to salinity stress. Moreover, mutations to COI1 eliminated the induction of JA-responsive genes (JAZ5, JAZ8, and VSP2) in samples treated with NaCl alone (Fig. 3F). Considered together, these results imply that salinity stress enhances the COI1-mediated JA signaling.

3.4. JAZ proteins negatively regulate jasmonate signaling under saline conditions

JAZ proteins function as negative regulators of JA signaling (Howe et al., 2018). This led us to question whether JAZ proteins play a role in regulating the salinity-induced enhancement of JA signaling. We performed phenotypic analyses of jazQ quintuple mutants (jaz1 jaz3 jaz4 jaz9 jaz10) and JAZ1-Δ3A transgenic plants (heterozygous) overexpressing JAZ1 lacking Jas domain, which encodes amino acids 202–228. In accordance with previous studies (Thines et al., 2007; Campos et al., 2016; He et al., 2023), the jazQ quintuple mutants exhibited increased sensitivity to MeJA, whereas JAZ1-Δ3A plants were relatively insensitive to MeJA (Fig. 4A–C), similar to the coi1 mutants. To explore the regulatory effects of JAZ proteins on the enhancement of JA signaling due to salinity stress, we examined the JA-related phenotypes of jazQ and JAZ1-Δ3A plants treated with both MeJA and NaCl. As anticipated, compared with the WT control, the jazQ and JAZ1-Δ3A plants were more and less sensitive to MeJA, respectively, when NaCl was included in the treatment (Fig. 4A–C). In addition, MeJA-induced primary root growth inhibition and anthocyanin accumulation increased in jazQ mutant plants treated with NaCl (Fig. 4D and E). However, the change in JAZ1-Δ3A seedling root length was unaffected by NaCl (Fig. 4D). There was a slight change in anthocyanin accumulation in JAZ1-Δ3A seedlings treated with NaCl (Fig. 4E). These findings provide evidence that JAZ proteins negatively regulate the enhancement of JA signaling under saline conditions. To further confirm our findings, we conducted an RT-qPCR analysis of JAZ5, JAZ8, and VSP2 expression in jazQ and JAZ1-Δ3A plants; the results were consistent with the observed phenotypes (Fig. 4F). Overall, these findings demonstrate the negative regulatory effects of JAZ proteins on the enhancement of JA signaling under saline conditions.

3.5. JAZ proteins physically interact with ABF transcription factors

Having determined the regulatory effect of JAZ proteins on the enhancement of JA signaling induced by salinity stress, we investigated the underlying molecular mechanism. JAZ proteins interact with multiple factors affecting diverse processes, serving as nodes for signaling crosstalk (Howe et al., 2018). On the basis of the Y2H assay results, we detected physical interactions between JAZ proteins and ABFs (Fig. S3). ABFs are crucial transcription factors responsive to abiotic stress and exogenous ABA (Choi et al., 2000; Uno et al., 2000; Furihata et al., 2006; Yoshida et al., 2015). We subsequently determined the specific region of JAZ1 required for the interaction with ABF3. Multiple sequences encoding truncated JAZ1 fragments were inserted into pGADT7. Y2H analysis indicated that the ZIM domain of JAZ1 was essential for the interaction with full-length ABF3 (Fig. 5A). We also determined that the N terminal-containing C1 domain of ABF3 is required for the interaction with full-length JAZ1 (Fig. S4). To further validate the interaction between JAZ1 and ABF3 in plant cells, we conducted BiFC assays in N. benthamiana. Consistent with the Y2H assay results, JAZ1 interacted with full-length ABF3. In plant cells, the ZIM domain of JAZ1 interacted with ABF3, with the Jas domain of JAZ1 serving as the negative control (Fig. 5B). Moreover, sequences encoding different ABF3 segments were fused to the sequence encoding the C-terminal fragment of YFP. The results of the co-transformation with JAZ1 were in accordance with the Y2H assay results (Fig. 5B). Additionally, the interaction between JAZ1 and ABF3 was further verified by a Co-IP assay, in which ABF3 was immunoprecipitated by anti-MYC agarose beads in Arabidopsis protoplasts co-expressing JAZ1-MYC and ABF3-HA (Fig. 5C). Hence, JAZ1 can physically interact with ABF3 in vivo.

3.6. ABF3 positively regulates the jasmonate response under salinity stress

To further clarify the role of ABF3 in JA-induced root growth inhibition and anthocyanin accumulation under saline conditions, we used two transgenic lines overexpressing ABF3 from Du et al. (2023). We measured the root length and anthocyanin content of ABF3-OE-L5 and ABF3-OE-L6 seedlings grown on half-strength MS medium containing MeJA, NaCl, or both. The ABF3-OE-L5 and ABF3-OE-L6 seedlings were sensitive to NaCl, with shorter primary roots than the WT plants (Fig. 6A and B), which is in accordance with the findings of an earlier study (Kang et al., 2002). Unexpectedly, compared with the WT plants, the ABF3-OE-L5 and ABF3-OE-L6 seedlings had longer primary roots, but accumulated more anthocyanin, following the MeJA treatment (Fig. 6B and C). However, when NaCl was included in the treatment, the ABF3-OE-L5 and ABF3-OE-L6 seedlings were more sensitive to MeJA than the WT plants in terms of the root length and anthocyanin content (Fig. 6D and E). According to the RT-qPCR analyses, ABF3 overexpression upregulated the expression of the JA-inducible genes JAZ5, JAZ8, and VSP2 (Fig. 6F). Accordingly, ABF3 appears to positively regulate JA signaling when exposed to salinity stress.

To further assess the regulatory effects of ABF transcription factors on JA signaling under saline conditions, we examined the phenotypes of loss-of-function ABF mutants. Because ABF1, ABF3, AREB1, and AREB2 have overlapping functions during the response to salinity stress (Yoshida et al., 2010, 2015), we analyzed the following triple and quadruple mutants: areb2 abf3 abf1-2, areb1 areb2 abf3, and areb1 areb2 abf3 abf1-2. In terms of the root length, the mutants were unaffected by the treatment with MeJA or NaCl alone, but these mutants had longer root than WT plants when treated with both MeJA and NaCl (Fig. 7A and B). According to Fig. 7C, areb1 areb2 abf3 and areb1 areb2 abf3 abf1-2 had a lower anthocyanin content than WT plants. Additionally, compared with the WT plants, these mutants had less extensive changes in root length and anthocyanin accumulation under saline conditions (Fig. 7D and E). Further RT-qPCR analyses demonstrated that the mutations decreased the expression of JA-responsive genes induced by salinity stress (Fig. 7F). However, the expression levels of JAZ5, JAZ8, and VSP2 still increased in these mutants treated with MeJA and NaCl, indicating unknown regulatory proteins are involved in this process. In summary, we conclude that ABF transcription factors positively regulate the JA pathway during the response to salinity stress.

3.7. ABF3-enhanced jasmonate signaling under salinity stress is compromised by JAZ1

To elucidate the relationship between JAZ1 and ABF3, we analyzed whether ABF3 overexpression can rescue the JA-related phenotypes of the JAZ1-Δ3A seedlings. We generated transgenic plants overexpressing JAZ1 and ABF3 via genetic crossing and examined the expression of JAZ1 and ABF3 by RT-qPCR (Fig. S5). Phenotypic examinations revealed that ABF3-OE-L5 JAZ1-Δ3A seedlings and JAZ1-Δ3A seedlings treated with MeJA alone had similar root lengths (Fig. 8A and B). However, the anthocyanin content of ABF3-OE-L5 JAZ1-Δ3A seedlings was in between that of ABF3-OE-L5 and JAZ1-Δ3A plants (Fig. 8C). Moreover, ABF3 overexpression partially rescued the insensitivity of JAZ1-Δ3A plants to MeJA under saline conditions (Fig. 8D and E). This finding was further supported by the RT-qPCR data (Fig. S6). On the basis of these observations, we conclude that JAZ1 suppresses the effects of ABF3 on the enhancement of JA signaling.

Given that JAZ1 interacts with ABF3, we explored the functional implications of this interaction by investigating the regulatory effects of JAZ1 on the ABF3 function. To examine the potential regulatory role of ABF3, we initially investigated the promoters of JA-responsive genes, which revealed the presence of G-box (CACGTG). Earlier research showed this sequence is a potential binding site for ABFs (Choi et al., 2000). The Y1H assays performed to explore the potential binding of ABF3 to the AOS promoter in yeast cells indicated that ABF3 does not bind directly to promoter fragments containing G-boxes (Fig. S7).

Considering the increased expression of JA-responsive genes in ABF3-OE plants, we speculated that ABF3 can induce AOS expression. We conducted Dual-LUC assays using effector constructs containing ABF3, JAZ1, or GFP under the control of the 35S promoter and a reporter construct comprising the AOS promoter fused to the LUC gene (Fig. S8A). LUC expression driven by the AOS promoter was higher when ABF3 was co-expressed with GFP than when GFP was expressed alone, indicating that ABF3 is involved in the activation of JA-responsive gene expression (Fig. 8F). Interestingly, the presence of JAZ1 attenuated the regulatory effect of ABF3 on the AOS promoter (Fig. 8F). To exclude the possibility JAZ1 modulates the ABF3 protein level and vice versa, we determined the JAZ1 and ABF3 contents in protoplasts. Neither protein affected the abundance of the other protein (Fig. S8B). To further validate these findings, we conducted experiments involving ABF3-OE-L6 plants to analyze whether ABF3 induces AOS expression. As expected, the AOS promoter-driven LUC expression level was higher in ABF3-OE-L6 plants than in WT plants (Fig. 8G). However, the presence of JAZ1 suppressed this enhancement, further supporting the modulatory effect of JAZ1 on ABF3-activated AOS transcription (Fig. 8G). Thus, ABF3 positively regulates JA signaling in a process that is inhibited by JAZ1.

Because of the reported interaction between JAZ1 and MYC2, we explored the regulatory relationships among JAZ1, ABF3, and MYC2. BiFC assays were performed to assess the interaction between JAZ1 and MYC2 with or without ABF3. When ABF3 was co-expressed with JAZ1-cYFP and MYC2-nYFP, YFP fluorescence decreased in Nicotiana benthamiana leaf cells, with β-glucuronidase (GUS) co-expressed with JAZ1-cYFP and MYC2-nYFP serving as the negative control (Fig. 8H). These results suggest that ABF3 competes with MYC2 for the binding to JAZ1. Previous studies showed that the expression of genes encoding JAZ-interacting transcription factors may be induced by MeJA (Qi et al., 2011; He et al., 2023), so we investigated the potential regulatory effects of JA on ABF3. Specifically, we analyzed the ABF3 transcript level in WT seedlings. Surprisingly, ABF3 transcription was not significantly affected by the MeJA treatment (Fig. S9). Overall, these results indicate that ABF3-promoted JA signaling under saline conditions is disrupted by JAZ1.

4. Discussion

Earlier research showed that salinity stress activates JA biosynthesis and signaling in plants (Toda et al., 2013; Zhao et al., 2014; Chen et al., 2016; Valenzuela et al., 2016), suggesting a correlation between salinity stress and JA signaling. In this study, RNA-seq analyses detected the overlap between the salinity stress-regulated transcriptome and the JA-regulated transcriptome (Fig. 2A). Li et al. (2024) recently conducted GO analyses and reported that salt stress activates JA biosynthesis and the ABA-activated signaling pathway. In the current study, we analyzed the JA-responsive pathway, while also focusing on some abiotic stress-induced biological processes and the expression of related genes (Fig. 2B and C). Our gene expression heatmap revealed the positive relationship between the salt stress response and the JA response. More specifically, the expression of some JA-regulated genes (e.g., OPR1, JAR1, and MYC2) was further induced by the NaCl treatment, while the expression of some salt stress-responsive genes (e.g., MKK9 and RD22) was activated by the exogenous MeJA treatment (Fig. 2C).

Salinity stress limits crop growth and decreases agricultural production. In wheat, the overexpression of the JA biosynthesis gene TaOPR1 or the application of exogenous JA promotes salinity stress tolerance (Dong et al., 2013). Similarly, the expression of a wheat AOC1-encoding gene was reported to increases salinity tolerance. The TaAOC1 gene contributes to α-linolenic acid metabolism, which is involved in JA synthesis (Zhao et al., 2014). The constitutive expression of TaAOC1 in Arabidopsis leads to increased salinity tolerance, but this tolerance is inhibited by a non-functional AtMYC2, implying JA signaling is essential for regulating salinity tolerance (Zhao et al., 2014). Additionally, exogenous MeJA treatments of soybean can alleviate salinity-induced growth inhibition (Yoon et al., 2009). A recent study showed that a loss-of-function mutation to JAZ8 increases the survival rate of salt-stressed plants (Li et al., 2024). All of these earlier studies reflect the positive regulatory effects of JA on salinity tolerance. In the present study, we observed that salt stress modulates root growth and anthocyanin accumulation through the COI1–JAZ-mediated JA signaling pathway. Anthocyanin is an antioxidant that alleviates oxidative stress caused by salinity in plants (Kim et al., 2017; Dabravolski and Isayenkov, 2023). Notably, JA can also promote root hair development (Han et al., 2020), thereby improving nutrient acquisition. On the basis of these findings, we propose that in plants, the activation of JA signaling alters root growth to enhance survival and promotes the accumulation of anthocyanin for scavenging ROS generated by salinity stress.

ABF transcription factors are also known as ABA response element (ABRE)-binding factors, which are responsive to salt, drought, and ABA treatments, and positively regulate plant abiotic stress resistance (Kim et al., 2004; Kim, 2005; Hossain et al., 2010; Yoshida et al., 2010; Liu et al., 2022; Song et al., 2022; Du et al., 2023). In Arabidopsis, ABFs interact with the INDETERMINATE DOMAIN (IDD) transcription factor IDD14 and cooperatively regulate drought tolerance (Liu et al., 2022). A recent study demonstrated the antagonistic effects of CONSTANS (CO) on ABFs in plants exposed to salinity stress under long-day conditions (Du et al., 2023). Considering the interaction between JAZ proteins and ABFs, we assumed that ABFs are involved in JA signaling. Our analyses of ABF3 overexpressing transgenic plants and multiple ABF mutants indicate that ABF3 positively regulates JA signaling under saline conditions (Fig. 6, Fig. 7). However, when exposed to exogenous MeJA alone, ABF3 overexpression lines exhibited longer roots (Fig. 6A and B), possibly due to the distinct regulatory mechanisms of ABF3 on root length under exogenous MeJA compared to salinity stress. According to the recent study of Du et al. (2023), the 20-d-old soil-grown areb1 areb2 abf3 abf1-2 mutants exhibited decreased tolerance to salinity stress, indicating the positive role of ABFs in plant salinity tolerance. However, our phenotypic analyses demonstrated no significant difference in root length of ABF mutants compared to the WT plants under saline conditions (Fig. 7A and B). Interestingly, another bZIP transcription factor ABI5, highly homogenous with ABFs, acts redundantly with ABF3 in regulating root growth when exposed to salt stress (Finkelstein et al., 2005). Given the overlapping function between ABFs and ABI5, and considering that ABI5 primarily expresses in seeds and during post-germination growth (Collin et al., 2021), we speculate that the root growth of abi5 areb1 areb2 abf3 abf1-2 quintuple mutant seedlings may be insensitive to salinity. However, this hypothesis will require validation in future studies. Moreover, members from all of the major transcription factor families (such as bZIP, MYB, ERF/AP2, NAC, and WRKY) have been found to be involved in the salt stress response, but master transcriptional regulators for salt stress have not been identified (Zhao et al., 2020). Hence, unknown mechanisms and factors may contribute to the salt stress-regulated root growth. During the response to osmotic stress, ABFs, which are also ABA-dependent transcription factors, are activated by the phosphorylation mediated by SnRK2.2/3/6 (Collin et al., 2021; Yoshida et al., 2015). Thus, whether SnRK2-activated ABFs in the ABA pathway are required for the enhancement of JA under salinity stress should be investigated. Interestingly, ABA also inhibits primary root elongation (Yoshida et al., 2015). Future studies will need to determine whether ABFs integrate ABA signaling with JA pathways under saline conditions.

MYC transcription factors have been extensively characterized regarding their activation of JA-responsive genes (Chen et al., 2011; Kazan and Manners, 2013; Wasternack and Song, 2017). Thus, we speculated whether ABF3 functions similarly. Our results demonstrated that ABF3 activates AOS transcription (Fig. 8F and G), but it cannot bind directly to the AOS promoter fragments containing G-boxes in yeast cells (Fig. S7). The possibility that ABF3 can bind to the AOS promoter fragments containing other motifs remains investigated in future study. Interestingly, the observed interaction between ABF3 and JAZ1 (Fig. 8H) may decrease the inhibitory effect of JAZ1 on MYC2, resulting in the activation of JA-responsive genes regulated by MYC2. Therefore, we speculate that ABF3 can directly or indirectly activate JA-responsive genes. In a recent study, the transcription factors PHR1 and MYC2 synergistically modulated JA-responsive genes (He et al., 2023). However, the potential association between ABF and MYC transcription factors during this process remains to be assessed. Previous research showed that JAZ proteins interact with downstream transcription factors and suppress their functions (Wasternack and Song, 2017; Han et al., 2020; He et al., 2023), suggesting JAZ1 may have repressive effects on ABF3. Our biochemical analysis indicated that the presence of JAZ1 prevented ABF3 from activating AOS expression (Fig. 8G and H). Moreover, ABF3 overexpression partially restored the insensitivity of JAZ1-Δ3A plants to exogenous MeJA under saline conditions (Fig. 8D and E), indicating that JAZ1 negatively regulates salinity-enhanced root growth inhibition and anthocyanin accumulation partially through the modulation of ABF3.

The JA signaling pathway is regulated by both endogenous and exogenous signals, and its molecular mechanisms vary depending on environmental conditions. Under normal growth conditions, JAZ proteins interact with and repress downstream transcription factors, inhibiting the JA response (Wasternack and Song, 2017; Howe et al., 2018). However, when plants perceive endogenous developmental or environmental signals, JA biosynthesis increases substantially; the binding of JA to the receptor COI1 promotes the ubiquitination and degradation of JAZ proteins (Kramell et al., 1995; Pedranzani et al., 2003; Huang et al., 2022). Specific environmental conditions, such as inorganic phosphate (Pi) deficiency, can also activate JA signaling. A recent study showed that a key Pi signaling-related transcription factor (PHR1) interacts with MYC2 and then they synergistically activate JA-responsive genes, but this activation is inhibited by JAZ1 (He et al., 2023), thereby promoting root system remodeling and anthocyanin accumulation in response to Pi deficiency. When plants are under shade conditions (low red: far-red light ratio), MYC2 is destabilized by the RING-finger E3 ubiquitin ligase COP1, whereas most JAZ proteins are stabilized, thereby inhibiting JA-mediated defense responses and promoting the reallocation of resources from defense to growth processes (Chico et al., 2014). Under salinity stress, JAZ1 is destabilized in a proteasome-dependent manner (Valenzuela et al., 2016). Furthermore, a recent study indicated that salt stress promotes the ubiquitination and degradation of JAZ8, resulting in the release of the NF-YA1-YB2-YC9 transcription factor complex that is inhibited by JAZ8, ultimately leading to increased salinity tolerance (Li et al., 2024). In the current study, we determined JAZ1 can interact with ABF3. Moreover, JAZ1 has an inhibitory effect on ABF3. However, the biological significance of this interaction is unclear. We speculate that under saline conditions, JAZ1 is degraded, which releases ABF3, enabling it to promote the JA response. Conversely, under normal conditions, JAZ1 interacts with ABF3 and inhibits its ability to activate transcription.

According to earlier studies, the transcription of JAZ-interacting transcription factor genes is induced by MeJA (Qi et al., 2011; He et al., 2023). Therefore, we investigated whether exogenous MeJA can upregulate ABF3 expression. Our results showed that MeJA failed to induce the transcription of ABF3 regardless of the presence of NaCl (Fig. S9). The expression levels of other ABF-encoding genes were also unaffected by MeJA according to our RNA-seq analyses (Fig. 2C). After elucidating the mechanism promoting JA activation in plants exposed to salinity stress, we wondered how ABF3 is activated during the response to excessive salinity. We detected the upregulated expression of ABF3 under saline conditions (Fig. S9), but the transcription factors responsible for this upregulation will need to be identified. Previous research showed that ABA upregulates ABF expression, with ABF transcription factors contributing to this upregulation (Wang et al., 2018). Therefore, we propose that in response to salinity stress, ABFs are phosphorylated by SnRK2s, after which the activated ABFs induce the expression of ABF genes. However, this hypothesis will need to be experimentally verified. Collectively, our findings have clarified the modulatory effects of ABF3 on JA, while providing mechanistic insights into salinity stress-enhanced JA signaling. This study also elucidated the biological consequences of the activation of JA signaling under saline conditions.

Acknowledgements

We thank Drs. Daoxin Xie (Tsinghua University), Gregg A. Howe (Michigan State University), Fumiyuki Soma (University of Tokyo), Kazuko Yamaguchi-Shinozaki (University of Tokyo), Jigang Li (China Agricultural University), for sharing research materials. We thank Liwen Bianji (Edanz) (www.liwenbianji.cn) for editing the English text of a draft of this manuscript. We also thank the Central Laboratory of Xishuangbanna Tropical Botanical Garden, Chinese Academy of Sciences for technical supports. This work was supported by the Natural Science Foundation of China (32270613, 31922009, and 31870259), the Yunnan Fundamental Research Projects (202201AS070051, 202001AV070009, 2019FI006, 202001AT070118, and 202101AW070005, 202401AT070220), the CAS "Light of West China" Program (to X.H.), the Youth Innovation Promotion Association of the of Chinese Academy of Sciences (Y201973 and 2022399).

Data availability

Data will be made available on request.

CRediT authorship contribution statement

Qi Zhang: Writing – review & editing, Writing – original draft, Visualization, Investigation, Data curation. Jiancan Du: Writing – review & editing, Methodology, Conceptualization. Xiao Han: Writing – review & editing. Yanru Hu: Writing – review & editing, Methodology, Conceptualization.

Appendix A. Supplementary data

Supplementary data to this article can be found online at https://doi.org/10.1016/j.pld.2024.05.003.