Regulation of ABI5 expression by ABF3 during salt stress responses in Arabidopsis thaliana
- 233 Downloads
Basic region/leucine zippers (bZIPs) are transcription factors (TFs) encoded by a large gene family in plants. ABF3 and ABI5 are Group A bZIP TFs that are known to be important in abscisic acid (ABA) signaling. However, questions of whether ABF3 regulates ABI5 are still present.
In vitro kinase assay results showed that Thr-128, Ser-134, and Thr-451 of ABF3 are calcium-dependent protein kinase phosphorylation sites. Bimolecular fluorescence complementation (BiFC) analysis results showed a physical interaction between ABF3 and 14-3-3ω. A Thr-451 to Ala point mutation abolished the interaction but did not change the subcellular localization. In addition, the Arabidopsis protoplast transactivation assay using a luciferase reporter exhibited ABI5 activation by either ABF3 alone or by co-expression of ABF3 and 14-3-3ω. Moreover, chromatin immunoprecipitation-qPCR results showed that in Arabidopsis, ABI5 ABA-responsive element binding proteins (ABREs) of the promoter region (between − 1376 and − 455) were enriched by ABF3 binding under normal and 150 mM NaCl salt stress conditions.
Taken together, our results demonstrated that ABI5 expression is regulated by ABF3, which could contribute to salt stress tolerance in Arabidopsis thaliana.
KeywordsABF3 Calcium 14-3-3 ABI5 Salt Abscisic acid Phosphorylation
Salt stress has recently become a serious problem that causes decreased crop yields worldwide and is caused by global climate change. The components of salt stress signaling in plants have, therefore, become important topics in recent years. The Salt Overly Sensitive (SOS) pathway was activated to confer salt tolerance in plants under conditions of salt stress (Zhu 2001; Munns and Tester 2008). In addition, many genes have been shown to be transcriptionally up-regulated under salt or osmotic stress conditions (Shinozaki and Yamaguchi-Shinozaki 2007). These signal transduction pathways included the abscisic acid (ABA)-independent (Shinozaki and Yamaguchi-Shinozaki 2007) and ABA-dependent pathways (Shinozaki and Yamaguchi-Shinozaki 2007; Fujita et al. 2011). In the ABA-dependent pathway, the ABA receptor serves the first line of ABA signal perception (Kline et al. 2010). This pathway induces the bZIP transcription factor (TF), which binds to the ABA-responsive element (ABRE) for the up-regulation of downstream genes, such as RD29B (Shinozaki and Yamaguchi-Shinozaki 2007). By contrast, the ABA-independent pathway induces expression of a transcription factor gene, DREB2. DREB2 belongs to the DRE BINDING PROTEIN/C-REPEAT BINDING FACTOR family that binds to the dehydration-responsive element/C-repeat (DRE/CRT) element. RD29A gene activation is both ABA-dependent and ABA-independent. DREB transcription factor binds to the DRE/CRT element of a downstream gene, such as the RD29A gene, which results in up-regulation of this gene under drought stress, which eventually leads to enhanced salt or osmotic stress tolerance in plants.
bZIP transcription factors are found in animals, yeast, and plants. In Arabidopsis, there are 75 bZIP members in the bZIP family (Jakoby et al. 2002). The bZIP family can be divided into ten groups (A, B, C, D, E, F, G, H, I, and S). In maize, 125 bZIP genes encode 170 bZIP proteins, and based on phylogenetic analysis results; these can be divided into 11 groups (Wei et al. 2012). Based on the primary structure, each bZIP TF has a basic region for DNA binding, and a leucine zipper domain (Jakoby et al. 2002). bZIP TFs are G-box binding factors (GBFs), which can bind the G-box motif on DNA with ACGT cis-elements (Foster et al. 1994; Sibéril et al. 2001). The ABRE motif belongs to the G-box family (Fujita et al. 2011). bZIP TFs can dimerize to form homodimers and heterodimers (Deppmann et al. 2004; Vinson et al. 2006), and the homodimers can be visualized using bimolecular fluorescence complementation (BiFC) (Walter et al. 2004). bZIP TFs can also interact with the 14-3-3 scaffold signaling protein to provide signals (Sibéril et al. 2001; Eckardt et al. 2001; Schoonheim et al. 2007; de Boer et al. 2013; Vysotskii et al. 2013).
bZIP genes have been reported to be involved in the abiotic stress response (Uno et al. 2000). bZIP transcription factors can be membrane-bound and released into the cytosol during stress responses (Seo et al. 2008). In Arabidopsis, ABF3 and ABF4 are involved in ABA signaling (Kang et al. 2002) and the salt stress response (Kim et al. 2004). Activated AtbZIP17 was shown to enhance salt tolerance (Liu et al. 2008). In rice (Oryza sativa), constitutive OsbZIP46 activation conferred drought stress tolerance (Tang et al. 2012). Overexpression of the soybean (Glycine max) GmbZIP1 gene improved high salt stress tolerance in transgenic plants (Gao et al. 2011), and a maize (Zea mays) ZmbZIP72 gene conferred salt stress tolerance in Arabidopsis transgenic plants (Ying et al. 2012). Overexpression of the rice OsbZIP23 gene and the tomato (Solanum lycopersicum) SIAREB gene improved drought and high salt stress tolerance in the respective transgenic plants (Xiang et al. 2008; Hsieh et al. 2012). In lotus (Nelumbo nucifera) plants, the LrbZIP gene was shown to be important in the salt resistance of roots (Cheng et al. 2012). However, the molecular mechanisms of the bZIP genes in salt stress responses is still not completely understood.
The abscisic acid responsive element-binding factor 3 (ABF3) is a member of the group A bZIP TFs. ABF3 overexpression in Arabidopsis showed an ABA hypersensitive phenotype (Kang et al. 2002) and increased drought stress tolerance in both rice and alfalfa (Oh et al. 2005; Wang et al. 2016). Ectopic Arabidopsis ABF3 expression conferred drought tolerance in soybeans (Kim et al. 2018) and cotton (Gossypium hirsutum) (Kerr et al. 2018). However, the target genes of ABF3 have not been thoroughly studied. Therefore, in this study, we investigated the transcriptional regulation of ABI5, another group A bZIP member, by the ABF3 and 14-3-3 proteins in Arabidopsis. The bimolecular fluorescence complementation (BiFC) assay was used to confirm the interaction between ABF3 and 14-3-3ω, and the transactivation assay was used to investigate if ABF3 regulates ABI5. Promoter deletion was also tested in the transactivation assay. Moreover, chromatin immunoprecipitation-qPCR was also introduced. Our results showed regulation of ABI5 expression by ABF3 in response to salt stress in Arabidopsis thaliana.
Materials and methods
Plant materials, growth conditions, and salt stress treatment
Arabidopsis thaliana ecotype Columbia was used in the present study. A T-DNA insertion mutant line, abi5, was obtained from Dr. Hsu-Liang Hsieh‘s lab at the National Taiwan University. A T-DNA insertion mutant line, abf3 (SALK_096965) of ABF3 gene (At4g34000), was ordered from the Arabidopsis Biological Resource Center (ABRC) (Additional file 1). Plant transformation was performed in Arabidopsis Col-0 using the floral dip method (Clough and Bent 1998). To perform the chromatin immunoprecipitation (ChIP) analysis, AtABF3 overexpression lines were generated. Full-length AtABF3 CDS driven by the 35S promoter in pEarleyGate103 vector was transformed into the Col-0 WT. AtABF3 overexpression lines were isolated using the plant selection maker, BASTA. Seeds were surface sterilized and stratified at 4 °C for 3 days in the dark, then propagated and grown on 1/2 Murashige–Skoog (MS) agar medium containing 0.8% sucrose (21 °C, 16 h light). Seven-day-old seedlings were treated with 100 mM or 150 mM NaCl by transferring the seedlings to plates containing 1/2 MS medium and NaCl, and the plants were incubated for either 0 h, 0.5 h, 1 h, or 3 h before RNA extraction.
RNA extraction and real-time PCR analysis
Seedlings (10–100 mg) were ground into a powder with liquid nitrogen, and 1 ml RezolTM C&T (Omics Bio, Taipei, Taiwan) with 200 μl chloroform was added. The samples were centrifuged at 12,000×g (Sigma 1–15 K, USA) for 15 min and moved to a new 1.5 ml tube. Five hundred microliters of isopropanol were added, and the samples were centrifuged at 12,000×g (Sigma 1–15 K, USA) for 10 min. The pellets were washed with 75% EtOH and resolved using DEPC-H2O. The contaminating DNA was removed using TURBO DNA-free™ DNase according to the manufacturer’s instructions (Ambion, California, USA). Isolated RNA was used for cDNA synthesis using the iScript™ cDNA Synthesis Kit (BIO-RAD, California, USA). Real-time PCR was performed using CFX and CFX manager software (BIO-RAD, California, USA). SsoFast™ EvaGreen Supermix (BIO-RAD, California, USA) was used for amplifications. Arabidopsis ACTIN2 was used as a quantitative control.
Isolation of Arabidopsis leaf protoplasts
Arabidopsis protoplast extraction was carried out as previously described and modified (Yoo et al. 2007). Arabidopsis plants were grown in the soil in an environmentally-controlled chamber with a relatively short photoperiod (8 h light at 22 °C/16 h dark at 22 °C). Well-expanded leaves were chosen from 3-week-old plants. One mm leaf strips were cut from the middle part of a leaf using a fresh sharp razor blade. Leaf strips were gently transferred into the prepared enzyme solution (1% cellulose R10, 0.25% macerozyme R10, 0.4 M mannitol, 20 mM KCl, 20 mM MES pH 5.7, 10 mM CaCl2, 5 mM β-mercaptoethanol, and 0.1% BSA) by dipping both sides of the strips. The tissues were then placed under a vacuum for 30 min. The digestion reaction continued for at least 3 h at room temperature with shaking. The enzyme solution should turn green after a gentle swirling motion, which indicates the release of protoplasts. A clean filter paper was washed with a W5 solution (154 mM NaCl, 125 mM CaCl2, 5 mM KCl, 2 mM MES pH 5.7, and 5 mM Glucose) before protoplast filtration. The enzyme solution containing protoplasts were filtered after wetting the filter paper. The flow-through was centrifuged at 100g (2420, KUBOTA, Japan) to pellet the protoplasts in a 15 ml round-bottomed tube for 3 min. The supernatants were removed, and protoplast pellets were washed by gentle swirling with the W5 solution. Protoplasts were resuspended in the W5 solution after counting cells under the microscope using a hemocytometer (BX40, OLYMPUS, USA). The protoplasts were kept on ice for 30 min and then resuspended in an MMG solution (4 mM MES pH 5.7, 0.4 M mannitol, 15 mM MgCl2) to a concentration of 2.5 × 105 protoplasts/ml.
Plasmid construction, and the transformation of plasmids for bimolecular fluorescence complementation (BiFC) analysis
BiFC analyses were carried out by a modified method, as previously described (Yoo et al. 2007; Liu et al. 2012). The fluorescence signal of the yellow fluorescence protein (YFP) was measured for protein–protein interactions. The open reading frame of AtABF3 was amplified from cDNA. The amplified open reading frame was inserted into the pSAT4-YN or pSAT5-YC vector (from Dr. Kai-Wun Yeh’s lab), driven by the 35S promoter and fused to the YFP-N or YFP-C in frame. The YN and YC fragments of YFP were fused to the C-terminus of the full-length ABF3 cDNA, and 14-3-3ω, respectively. The paired plasmids were transfected into Arabidopsis protoplasts (ABF3N/14-3-3ωC, ABF3C/14-3-3ωN). ACS7 was used as a positive control for the 14-3-3 interaction (Huang et al. 2013). Moreover, the transformation of EmptyN/14-3-3ωC, ACS7C, Di19-2C, ABF3C, and EmptyC/14-3-3ωN, ACS7 N, Di19-2N, ABF3N were used as negative controls (Additional file 2). Ten microgram plasmids (YFP-N and YFP-C) and 100 μl protoplasts were added into a 15 ml round-bottomed tube and gently mixed. One hundred and ten microliters of a polyethylene glycol solution were added and incubated at room temperature for 10 min. The polyethylene glycol solution containing protoplasts was diluted with 1 ml of the W5 solution and gently mixed. Protoplasts were centrifuged at 100×g (KUBOTA 2420, Japan) for 3 min to pellet the protoplasts. The supernatants were removed, and the protoplasts were washed twice with the W5 solution. The protoplasts were resuspended with 100 μl of the W5 solution in an Eppendorf tube at room temperature. After 12–16 h, YFP fluorescence was detected using a confocal microscope (TCS SP5, Leica).
The transactivation assay using Arabidopsis leaf protoplasts
For the reporter gene construct, the 5× GAL4, 4X GCC, and TATA box in the 5× GAL4-4X GCC-TATA-LUC-Nos M13Fprimer vector was replaced by the ABI5 promoter and fused to the firefly Luc gene. For effector plasmids, the AtERF gene in the pUC vector was replaced by the coding regions of 14-3-3ω, ABF3 wild type, mutated ABF3 (T451A), ABF3 (T128A), ABF3 (S126A), and ABF3 (S134A) that were constructed into the pRTL2 vector using the Gateway LR Clonase™ II Enzyme Mix (Invitrogen, California, USA). The PRL plasmid containing the Renilla Luc gene driven by the CaMV35S promoter was used as an internal control for the transactivation assay. Arabidopsis protoplasts were isolated and transfected by a modified polyethylene glycol method, as previously described (Abel and Theologis 1994). Ten micrograms of a reporter plasmid and 5 μg of an effector plasmid were co-transfected into protoplasts with 10 μg of the internal control plasmid, PRL. The transfected cells were incubated at 22 °C for 20 h under light. Protoplasts were harvested by centrifugation at 500×g for 1 min (Z223 M-2, HERMLE, Germany). Cells were assayed for luciferase activity using the Dual-Glo™ Luciferase Assay System (Promega) following the manufacturer’s instructions.
The in vitro kinase assay
The fusion peptides used in the kinase assay
Chromatin immunoprecipitation (ChIP) assays
Two-week-old seedlings grown on vertically oriented plates with MS medium were collected (~ 2 g) for the ChIP assays (Gendrel et al. 2005). After fixation with formaldehyde, the chromatin was sheared to an average length of 500 bp by sonication and then immunoprecipitated with Protein G Mag Sepharose Xtra magnetic beads (GE) and an anti-YFP antibody (catalog: 66002-1-lg, proteintech). After the cross-linking was reversed, the number of precipitated DNA fragments and amount of input DNA was detected by quantitative real-time PCR using the specific primers. The percentage of input DNA was calculated by determining 2−ΔCt(= 2−[Ct(ChIP)−Ct(Input)]). ACTIN2 and UBQ10 were used as the negative control.
The survival rate of the abf3 mutant is decreased under 150 mM NaCl salt stress condition
Altered ABI5 expression in the abf3 mutants under salt stress condition
ABF3 phosphorylation is catalyzed by CDPK3 and CDPK16 in vitro
Nuclear protein–protein interactions between ABF3 and 14-3-3ω were detected using BiFC analysis
To identify the ABF3 binding site, the mutation of a predicted 14-3-3 binding site to Ala (T451A) in ABF3 was tested to see if there was an interaction with 14-3-3. No interaction between ABF3 (T451A) and 14-3-3 was detected (Fig. 4). The transient expression results confirmed that the binding site was the T451 of ABF3, which is consistent with the results of Sirichandra et al. (2010), who showed that the T451 of ABF3 is the 14-3-3 binding site.
ABI5 activation by either ABF3 alone or ABF3 co-expressed with 14-3-3ω in a transactivation assay
To determine if ABI5 activation was induced by 14-3-3ω, mutated ABF3 (T451A) was used as an effector and luciferase expression was observed. Compared with the wild type ABF3, mutated ABF3 (T451A) combined with 14-3-3ω did not activate ABI5 (Fig. 6). Therefore, T451 phosphorylation might not be required for ABF3-promoted ABI5::LUC activity under normal conditions. T451 phosphorylation of ABF3 by the ABA-activated kinase, OST1, was previously shown to be required for ABF3 stability in Arabidopsis (Sirichandra et al. 2010). When the T451 site on of ABF3 was point-mutated into an Ala, ABF3 stability was affected (Sirichandra et al. 2010). Since our results showed that T451 is not important for ABF3 transcriptional activation, T451 phosphorylation and 14-3-3 binding could exclusively function in the regulation of ABF3 protein stability (Sirichandra et al. 2010). Taken together, these results indicate that ABF3 regulates ABI5 expression under normal conditions.
A promoter deletion assay identified ABRE cis-elements of the ABI5 promoter in a transactivation assay
ABF3 binds to the ABI5 promoter in vivo under salt stress condition
The 14-3-3 binding site, T451, is phosphorylated by CDPKs in vitro
ABFs have been reported to be phosphorylated by many kinases. In rice, TRAB1 (a rice ABF) has been shown to be phosphorylated in an ABA-dependent manner (Kobayashi et al. 2005). In potatoes (Solanum tuberosum), StABF1 is phosphorylated in response to ABA and salt stress. Specifically, a potato CDPK isoform (StCDPK2) phosphorylated StABF1 in vitro (Muniz Garcia et al. 2012). In Arabidopsis, AtCPK32 phosphorylated ABF4 in vitro. Serine-110 of ABF4 was phosphorylated by AtCPK32 (Choi et al. 2005). Moreover, a previous study showed that ABF3 could be phosphorylated by SnRK2E/SnRK2.6 in Arabidopsis (Sirichandra et al. 2010). In Fig. 3, GST-ABF3 was found to be phosphorylated by recombinant CDPK3 and CDPK16 in vitro. In addition, the results from our kinase assay showed that Thr-128, Ser-134, and T451 of AtABF3 are CDPK phosphorylation sites. Our results are consistent with these reports with one exception being that Thr-128 is a previously uncharacterized site. According to the conserved kinase domain sequence, CDPK belongs to the CDPK-SnRK superfamily. The CDPK-SnRK superfamily consists of seven serine-threonine protein kinase types, namely calcium-dependent protein kinase (CDPKs), CDPK-related kinases (CRKs), phosphoenolpyruvate carboxylase kinases (PPCKs), PEP carboxylase kinase-related kinases (PEPRKs), calmodulin-dependent protein kinases (CaMKs), calcium and calmodulin-dependent protein kinases (CcaMKs), and SNF-related serine/threonine-protein kinases (SnRKs) (Hrabak et al. 2003). To date, three consensus CDPK phosphorylation motifs have been identified (Jaspert et al. 2011). The phosphorylation site of the fusion peptide, P4, is consistent with the CDPK phosphorylation (φ-5-X-4-Basic-3-X-2-X-1-S,) and SnRK2 phosphorylation motifs (LXRXX(S/T) (Sirichandra et al. 2010). However, whether AtABF3 is a CDPK substrate, and Thr-128, Ser-134, and Thr-451 are the in vivo phosphorylation sites requires further studies.
The ABI5 promoter is regulated by ABF3 binding
ABA signal transduction, perceived from environmental cues to physiologic responses, involves many components, including ABA receptors, protein kinases, phosphatases, transcription factors, and ABA-induced genes containing conserved G–box like cis-acting elements (ABREs) in their promoter regions (Hernandez Sebastia et al. 2004). Most ABA-regulated genes contain conserved ABA-responsive elements as the determinant cis-elements in their promoters. In general, a single ABRE copy is not able to induce ABA-mediated transcription. Successful ABA-induced gene expression requires either additional copies of ABREs or coupling elements (Shen 1996). Recently, however, a frequency distribution approach has shown that ABRE–ABRE pairs are major cis-elements in Arabidopsis and rice (Gomez-Porras et al. 2007). Multiple ABREs or a combination of an ABRE with a so-called coupling element (CE) can establish a minimal ABA-responsive complex (ABRC), and thereby, confer ABA responsiveness to a minimal promoter (Gomez-Porras et al. 2007).
In the ChIP assay (Fig. 8), ABF3 bound to the promoter of BOX 2 + BOX 3 and BOX 4, indicating that ABI5 is a direct target gene of ABF3. A previous study indicated that ABI5 and ABF3 in some seedlings have redundant ABA and stress responses, but the relative importance of these genes varies among responses (Finkelstein et al. 2005). For example, ABI5 is a much more critical determinant of germination sensitivity to ABA or other stresses, consistent with its much stronger expression in mature seeds. Alternatively, ABF3 is more important for the ABA sensitivity of seedling root growth (Finkelstein et al. 2005). By contrast, our data showed that ABF3 could directly regulate ABI5 expression, which has not been previously reported. In addition, ABF3 could bind to the ABI5 promoter under normal and 150 mM NaCl salt stress conditions (Fig. 8). Under salt stress conditions, the fold enrichment is higher than normal conditions. Our results supported that ABRE–ABRE pairs could be involved in the regulation of ABI5 gene expression by ABF3 in response to salt stress.
Based on our results, it is possible that AtCDPK3 and AtCDPK16 are activated under salt stress condition. The activated CDPKs may phosphorylate AtABF3 at T451 followed by 14-3-3 binding. The 14-3-3 binding stabilizes ABF3 and ABF3 binds to the promoter region of AtABI5, which in turn activates the ABI5 gene in response to salt stress. In summary, our results showed that ABF3 is an in vitro CDPK substrate. In addition, ABF3 could bind to ABI5 ABRE elements to activate ABI5 gene expression in Arabidopsis in response to salt stress.
We thank John Cushman and Jeffrey Harper, University of Nevada, Reno for providing the CDPK3 plasmid. We appreciated funding support (MOST # 105-2313-B-002-006, 104-2311-B-002-005, 103-2311-B-002-006, 107-2313-B-002-002 and 108-2311-B-002-002) from the Ministry of Science and Technology, Taiwan (107L893106, 108L893106). We also thank technical support from TechComm, National Taiwan University. The manuscript was English-edited by BioMed Proofreading, LLC.
HCC conducted gene expression analyses, transactivation assay, CDPK kinase assay. MCT conducted survival rate test and BiFC analyses. SSW contributed to revisions of phenotyping and ChIP assay in Figs. 1 and 8. IFC led research direction, organized the data, and finalized the manuscript. All authors read and approved the final manuscript.
Ethics approval and consent to participate
Consent for publication
The authors declare that they have no competing interests.
- Choi HI, Park HJ, Park JH, Kim S, Im MY, Seo HH, Kim YW, Hwang I, Kim SY (2005) Arabidopsis calcium-dependent protein kinase AtCPK32 interacts with ABF4, a transcriptional regulator of abscisic acid-responsive gene expression, and modulates its activity. Plant Physiol 139:1750–1761CrossRefGoogle Scholar
- Kobayashi Y, Murata M, Minami H, Yamamoto S, Kagaya Y, Hobo T, Yamamoto A, Hattori T (2005) Abscisic acid-activated SNRK2 protein kinases function in the gene-regulation pathway of ABA signal transduction by phosphorylating ABA response element-binding factors. Plant J 44:939–949CrossRefGoogle Scholar
- Uno Y, Furihata T, Abe H, Yoshida R, Shinozaki K, Yamaguchi-Shinozaki K (2000) Arabidopsis basic leucine zipper transcription factors involved in an abscisic acid-dependent signal transduction pathway under drought and high-salinity conditions. Proc Natl Acad Sci USA 97:11632–11637CrossRefGoogle Scholar
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.