Tumor necrosis factor receptor-associated factor (TRAF) proteins play crucial roles in plant development and response to abiotic stress. Here, we present genetic evidence that SEVEN IN ABSENTIA 2 (SINA2), a TRAF-like family protein, is involved in abscisic acid (ABA)-related drought stress signaling in Arabidopsis.
Gene expression, protein subcellular localization, protein–protein interaction, and a transient transcription dual-luciferase assay were performed. The drought tolerance of SINA2 loss-of-function mutants and SINA2-overexpressing plants was investigated.
In Arabidopsis, SINA2 was significantly induced by ABA and drought treatment. The SINA2-YFP fusion protein was predominately localized in the nuclei and cytoplasm. Loss of function of SINA2 (sina2) reduced drought tolerance, whereas overexpression of SINA2 increased stomatal closure, decreased water loss, and therefore improved drought resistance in transgenic plants. Upon ABA treatment, expression of some key ABA- and stress-responsive genes decreased in the sina2 mutant, but increased in SINA2-overexpressing plants. Furthermore, SINA2 was induced in the ABA-deficient mutant by ABA, but not by drought stress. Thus, the drought response of SINA2 was ABA-dependent. ProSINA2::LUC expression in Arabidopsis protoplasts further revealed that ABA-responsive element (ABRE) binding (AREB) protein 1 (AREB1) AREB2 and ABRE-binding factor 3 (ABF3) might regulate SINA2 expression at the transcriptional level.
Our results indicate that SINA2 functions as a positive molecular link between drought tolerance and ABA signaling in Arabidopsis.
Drought is one of the fundamental abiotic stresses that limit crop production world-wide (Luo, 2010). Physiological and biochemical responses to drought in plants are controlled by a serious of stress-dependent signal transduction pathways (Xiong et al., 2002; Yang et al., 2008). As a response to stress, plant cells trigger a network of signaling events at both transcriptional and post-transcriptional levels, including gene expression in the nucleus and metabolic dynamics in the cytosol, that enable environmental adaptation (Leung & Giraudat, 1998; Finkelstein et al., 2002; Xiong et al., 2002; Zhu, 2002; Guo et al., 2011).
The plant hormone abscisic acid (ABA) plays important roles in plant development and environmental adaptation (Knight & Knight, 2001). Many key components involved in ABA signal transduction have been identified through the characterization of a series of ABA-deficient and ABA-insensitive mutants in Arabidopsis. Whereas the phosphatase 2C (PP2C) family proteins ABA-INSENSITIVE1 (ABI1) and ABI2 are negative regulators in ABA signaling (Merlot et al., 2001; Yu & Setter, 2003), ABI transcriptional factors ABI3 (B3 type), ABI4 (AP2 type) and ABI5 (basic domain/Leu zipper type) are positive regulators that share redundant functions during seed germination and early seedling development (Giraudat et al., 1992; Finkelstein et al., 1998; Finkelstein & Lynch, 2000; Nambara et al., 2010; Li et al., 2011). Recently, Pyrabactin Resistance (PYR)/PYR1-like (PYL), also called Regulatory Components of ABA Receptor (RCAR), was identified as an ABA receptor that interacts with these PP2C proteins (Li et al., 2011). The ABA-dependent interaction between these two families of proteins down-regulates the phosphatase activity of ABI1/ABI2 and then relieves their inhibition of the downstream target protein kinases. This leads to the accumulation of phosphorylated SNF1-related protein kinases (SnRK2s) and subsequent phosphorylation of a variety of ABA-responsive factors, including ion channels and transcriptional factors such as ABA-responsive element (ABRE) binding (AREB) protein 1 (AREB1), AREB2 and ABRE-binding factor 3 (ABF3) (Fujii et al., 2009; Ma et al., 2009; Melcher et al., 2009; Miyazono et al., 2009; Park et al., 2009; Santiago et al., 2009; Cutler et al., 2010; Yoshida et al., 2010; Chai et al., 2011).
Tumor necrosis factor receptor-associated factor (TRAF)-like family proteins have been characterized by their conserved C-terminal coiled-coil domain (the TRAF domain), which is required for homo- or heterodimerization and interaction with their cognate receptors or cytoplasmic signaling proteins (Rothe et al., 1994; Pullen et al., 1998; Leo et al., 1999; Han et al., 2003). The SEVEN IN ABSENTIA (SINA) family proteins are a clade of TRAF-like family proteins. The SINA protein, which was first isolated in Drosophila (Carthew & Rubin, 1990) and subsequently found in the plant kingdom, functions in proteasome-mediated regulation in various developmental processes (Wang et al., 2008). For example, SINA of Arabidopsis thaliana (SINAT5) participates in ubiquitin-mediated degradation of NAC1 ((NO APICAL MERISTEM (NAM), the arabidopsis transcription factor 1 (ATAF1) and CUP-SHAPED COTYLEDON2 (CUC2)), and regulates auxin-associated lateral root development (Xie et al., 2002; He et al., 2005). SINAT2 acts as an RELATED TO APETALA2 (AP2) 2 (RAP2.2) interaction partner (Welsch et al., 2007). Recently, another SINA family member, Oryza sativa drought-induced SINA protein 1 (OsDIS1), was identified as a negative regulator in the drought tolerance of rice (Oryza sativa) (Ning et al., 2011).
To gain more insight into the molecular mechanisms of the functions of SINA proteins, we performed phylogenetic and computational analyses to identify a TRAF-like family gene, SINA2. SINA2 shares high homology with the Arabidopsis SINATs except that it lacks the Really Interesting New Gene (RING) finger domain that is required for the degradation of its target protein. We found that SINA2 knock-out mutants were less drought tolerant and, conversely, SINA2-overexpressing plants were more drought tolerant than wild-type plants. To investigate the possible mechanisms involved in the SINA2-mediated drought stress pathway, the expression levels of SINA2 in both ABA-deficient and ABA-signaling mutants were examined. Our results demonstrated that SINA2 functions as a positive component in drought stress through ABA signaling transduction in Arabidopsis.
Materials and Methods
Plant growth conditions and sina2 mutant isolation
Except for abi2-1, which is in the Landsberg erecta (Ler) ecotype background, all the accessions of Arabidopsis thaliana (L.) Heynh. used in this study are the Columbia-0 (Col-0) ecotype. The T-DNA insertion mutants were obtained from the Arabidopsis Biological Resource Center and the Nottingham Arabidopsis Stock Centre. The abi1-1C and abi5-7 mutants were provided by Dr Kazuo Shinozaki and Dr Eiji Nambara from the RIKEN Plant Science Center. abi3-8 was provide by Prof. Chuanyou Li (Institute of Genetics and Developmental Biology, Chinese Academy of Sciences, Beijing, China), and ABA DEFICIENT 1 (aba1) and abi2-1 were kindly provided by Dr Jiankang Zhu (Department of Horticulture and Landscape Architecture, Purdue University). Five-day-old seedlings grown on Murashige & Skoog (MS) medium (Sigma-Aldrich) supplemented with 2% (w/v) sucrose and 0.8% (w/v) agar were transplanted into soil and then grown under 12 h : 12 h, light : dark cycles in the glasshouse at 22°C (light) or 19°C (dark). To identify the T-DNA insertion in sina2-1 (SALK_129594) and sina2-2 (SALK_077325), the T-DNA borders of sina2-1 and sina2-2 alleles were confirmed using the T-DNA left-border primer LBa1 and two SINA2 gene-specific primers, SINA2F and SINA2R (Table S1). The T-DNA insertion in sina2 was investigated by genomic PCR to confirm the disruption of the endogenous gene. Homozygous sina2 mutants were identified by RT-PCR to confirm that gene expression was disrupted. ACTIN2 was employed as an internal control.
Drought treatment and water loss assays
To assay plant drought tolerance, 7-d-old seedlings germinated on MS medium were transferred to soil and grown for 3 wk in the glasshouse. They were then subjected to progressive water stress by withholding water for 3 wk. Their survival rates were determined 2 d after re-watering. Plants with a fresh green meristem (shoot tip) were scored as survivors. The same number of seedlings from different genotype plants were grown in the same tray to minimize experimental variations.
To assay leaf water loss, 10 leaves of the same size and developmental stage per individual wild-type, sina2 mutant and SINA2-overexpressing plant were excised from 4-wk-old plants. The fresh weights were determined at designated time intervals. The percentage of initial fresh weight at each time-point was used to determine the amount of water loss. These measurements were taken three times.
Measurements of stomatal aperture
For stomatal aperture measurements, epidermal strips were peeled from the rosette leaves of 4-wk-old wild-type, sina2-1 and SINA2-overexpressing plants incubated in a solution containing 10 mM KCl, 10 mM MES-Tris and 50 μM CaCl2 (pH 6.15) and exposed to light (100 μmol m−2 s−1) for 3 h. Subsequently, 0 or 50 μM ABA was added to the solution. After 2 h, samples were examined under 40 × magnification using a Nikon microscope. Widths and lengths of stomatal apertures were measured. Mean ratios of width to length ± SE of three independent experiments (n =30–50) were calculated. The significance of differences was determined using Student's t-test (*, 0.01 < P <0.05; **, P <0.01).
Leaves of 4-wk-old wild-type, sina2-1 and SINA2-overexpressing plants exposed to drought conditions for 3 wk were used for proline quantification. Proline concentration was ascertained using methods described previously (Bates, 1973).
Plant treatments, RT-PCR and quantitative real-time PCR analyses
To analyze levels of gene expression, quantitative real-time PCR (qRT-PCR) was performed with RNA samples isolated from 2-wk-old seedlings grown on MS medium harvested at the indicated times after exposure to drought conditions, 100 μM ABA, 200 mM NaCl or 300 mM mannitol. Total RNA was extracted with Trizol reagent (Takara, Kyoto, Japan) and treated with RNase-free DNase (Promega, Shanghai, China). The first-strand cDNA was synthesized with ReverTra Ace (Toyobo, Osaka, Japan). PCR was performed on 96-well optical reaction plates by iQ5 (Bio-Rad, Richmond, CA, USA) after pre-incubation for 5 min at 95°C, followed by 40 cycles of denaturation at 95°C for 15 s, annealing at 55°C for 20 s, and extension at 72°C for 30 s. Amplification of the detected genes was monitored every cycle using SYBR green fluorescence. All the experiments were carried out using the iQ5 real-time PCR detection system and SYBR Green Realtime PCR Master Mix (Toyobo). Results were normalized to the reference gene ACTIN2 using the ΔΔCt method, and each experiment was repeated three times. The primers used in this study are listed in Supporting Information Table S1.
Generation of SINA2 promoter-β-galactosidase (GUS) construct and histochemical analyses
To generate the SINA2 promoter–GUS construct, the 629-bp 5′-flanking DNA of the SINA2 coding region was amplified with the SINA2 promoter-specific primers PrSINA2F and PrSINA2R (Table S1). To confirm the sequence, the amplified 629-bp PCR fragment was cloned into the pBlueScript SK- vector. The BamHI-SmaI fragment from the pBlueScript SK- subclone was inserted into the same sites in the pCAMBIA1381Z vector (pCAMBIA) to obtain the ProSINA2::GUS vector. The construct was transformed into wild-type Arabidopsis (Col-0 ecotype) plants, as described previously (Clough & Bent, 1998), and hygromycin-resistant transformants were selected.
For histochemical analysis, different tissues from seedlings at different developmental stages were collected and stained with 5-bromo-4-chloro-3- indolyl-d-glucuronide for 24 h. They were then incubated in 75% ethanol to remove chlorophyll using methods described previously (Jefferson et al., 1987).
Confocal microscopy analyses and 4′,6-diamidino-2-phenylindole (DAPI) staining
To determine the subcellular localization of SINA2, the encoding region that did not contain a stop codon was cloned into the BamHI and SpeI sites of the pA7-YFP vector to yield the final plasmid pA7-SINA2-YFP (Liu et al., 2010). Transiently transformed Arabidopsis protoplasts (Yoo et al., 2007) were analyzed for SINA2-YFP expression using a confocal microscope at 514 nm wavelength (LSM 510 META; Zeiss). pA7-YFP was used as a positive control (Zhu et al., 2009).
For DAPI staining, 5-d-old seedlings of transgenic 35S::SINA2-YFP were soaked in 1 μg ml−1 DAPI solution. A confocal laser scanning microscope (FluoView1000 confocal microscope; Olympus, Center Valley, PA, USA) was used to observe nuclear localization.
Transgenic vector construction and plant transformation
To construct the vector for sina2 mutant complementation, the 1.7-kb genomic DNA sequence including the 629-bp DNA fragment upstream of the predicted ATG start codon of SINA2, the full-length genomic sequence of SINA2, and the 247-bp 3′ UTR was cloned into the pCAMBIA1300 binary vector between the EcoRI and SalI sites. To construct the SINA2-overexpressing vector, the open reading frame of SINA2 from the pBlueScript SK- subclone was inserted into a modified pCAMBIA1301 vector via the BamHI and SalI restriction sites driven by two copies of the cauliflower mosaic virus (CaMV) 35S promoter. Full-length cDNA of SINA2 without the stop codon was fused upstream of YFP under the control of the CaMV 35S promoter in pCAMBIA1300 to obtain the 35::SINA2-YFP vector. For the construction of the ProSINA2::SINA2-YFP vector, the 35S promoter was replaced with the SINA2 promoter in the previously generated 35::SINA2-YFP vector via the EcoRI and SalI restriction sites. The resultant constructs were introduced into the Agrobacterium tumefaciens GV3101 strain. Wild-type or mutant Arabidopsis plants were transformed as described previously (Clough & Bent, 1998). Transgenic plants were screened on MS medium supplemented with 50 μg ml−1 hygromycin. Homozygous complementation and overexpression plants from T3 progeny were used for all the experiments.
Transient transcription dual-luciferase assays
To generate the PrSINA2::LUC reporter construct for the dual-luciferase assays, the promoter region of SINA2 was digested from the pBlueScript SK- subclone and inserted into the BamHI and SalI sites of pGreenII0800-LUC. To generate the CaMV 35S promoter-driven transcriptional factor effector constructs, AREB1, AREB2, ABF3 and ABI5 were digested from the pBlueScript SK- subclones and inserted into the BamHI and SalI sites of pGreenII62-SK, respectively.
Transient dual-luciferase assays in Arabidopsis protoplasts were performed as described previously (Hellens et al., 2005) and checked using dual-luciferase assay reagents (Promega), with some modifications. Briefly, after the isolation of Arabidopsis protoplasts, cells were set at 107 per milliliter, and 100 μl of protoplasts was used for each transformation as described previously (Yoo et al., 2007). Then the protoplasts were kept in the dark for 12 h and homogenized in 100 μl of passive lysis buffer. The crude extract (20 μl) was mixed with 100 μl of luciferase assay buffer and the firefly luciferase (LUC) activity was measured using a GLOMAX 20/20 luminometer (Promega). Stop and Glow Buffer (100 μl) was then added to the reaction and the renillia luciferase (REN) activity was measured (Li et al., 2010). Each sample was measured three times.
Yeast two-hybrid, bimolecular fluorescence complementation (BiFC) and co-immunoprecipitation (Co-IP) assays
For the yeast two-hybrid assays, the coding sequence of SINA2 was introduced into the pGBKT7 bait plasmid (Clontech, Shanghai, China) to produce a fusion protein with the galactose pathway gene GAL4 DNA-binding domain. SINA2 and five SINAT cDNAs (SINAT1, SINAT2, SINAT3, SINAT4 and SINAT5) were inserted into the pGADT7 prey plasmid which contained the GAL4 activation domain. Plasmids of different combinations from pGADT7 and pGBKT7 were introduced into the yeast strain AH109 by the lithium chloride–polyethyleneglycol method according to the manufacturer's instructions (Clontech). Transformants were selected on SD/–Leu/–Trp plates at 30°C. The interactions were tested on SD/–Leu/–Trp/–His/–Ade plates (Yu et al., 2012).
BiFC assays were performed on Nicotiana benthamiana leaves as described previously (Gou et al., 2011). The firefly LUC enzyme was divided into an N-terminal part (LUCn) and a C-terminal part (LUCc), and SINA2 was fused with LUCn and LUCc, respectively. The resultant constructs LUCn-SINA2 and SINA2-LUCc and the empty vector control were introduced into Agrobacterium tumefaciens strain GV3101. After the infiltration, plants were placed at 22°C for 3 d and infiltrated with 1 mM luciferin before being observed. An exposure time of 5 min with 3 × 3 binning was used for all images taken.
For Co-IP assays, the coding region of SINA2 was fused upstream of Flag and downstream of Myc in the pCAMBIA1300 vector driven by the CaMV 35S promoter. The resultant constructs were transformed into Agrobacterium strain GV3101 and then co-infiltrated into the leaves of 4-wk-old N. benthamiana plants. A 50-μl volume of anti-Flag agarose (Sigma-Aldrich) was used to capture the protein complex from the total proteins prepared from the infiltrated leaves. The prepared samples were washed with 1 ml of TBS buffer five times and then the immunoprecipitation products were detected by immunoblot analyses. Anti-Myc (Sigma-Aldrich) or anti-Flag (Sigma-Aldrich) antibodies were applied, and the chemiluminescence signal was detected by autoradiography.
Identification and characterization of SINA2
Ubiquitination-mediated protein degradation in plant development and the abiotic stress response has been widely studied (Stone & Callis, 2007; Lyzenga & Stone, 2012). Previously, the TRAF-like SINA family protein SINAT5 was identified as a RING-type E3 ubiquitin ligase involved in lateral root development through degradation of its target NAC1 (Xie et al., 2002). However, abiotic stress responses mediated by SINAT family proteins are largely unclear. To explore the possible roles of SINAT family proteins in plant responses to abiotic stresses, we searched the TAIR10 database (www.arabidopsis.org) using SINAT5. In addition to the five SINATs, a TRAF-like family protein, SINA2, was identified (Figs 1a, S1).
As a first step to understanding the possible biological functions of SINA2, we examined its expression pattern in wild-type Arabidopsis by qRT-PCR. SINA2 was ubiquitously expressed in seedlings and various tissues, including stems, flowers, cauline leaves, rosette leaves and roots, although the expression of SINA2 in seedlings was much lower than that in the tissues of adult plants (Fig. 2a). The most abundant SINA2 mRNA was present in roots and senescent leaves of adult plants (Fig. 2a,b). We also generated the SINA2 promoter (629 bp)-β-glucuronidase (GUS) reporter vector and introduced it into wild-type Arabidopsis plants. Consistent with the qRT-PCR results, GUS was expressed in seedlings and various tissues (Fig. 2c–l), with higher expression in the root, mature rosette leaf and guard cell, and senescent leaves (Fig. 2g,i,j,m). High expression of GUS in guard cells may suggest the potential function of SINA2 in stomata-mediated water loss control in Arabidopsis.
Published microarray data have shown that transcription of SINA2 can be induced by ABA and drought conditions. Therefore, we performed qRT-PCR to detect the transcription level of SINA2 after different stress treatments (Fig. 1b). qRT-PCR indicated that expression of SINA2 was strongly induced by ABA and drought, but not by NaCl. This observation was also confirmed by GUS staining experiments (Fig. 1c).
SINA2 is mainly localized in the nuclei and cytoplasm
Cellular and subcellular localization of a protein may indicate how it functions and/or where it might function. Therefore, we fused the coding region of SINA2 in-frame with YFP at the C-terminal as pA7-SINA2-YFP. Transient expression in Arabidopsis protoplasts demonstrated that the SINA2-YFP fusion protein was mainly localized in the nuclei and cytoplasm (Fig. 3a). Transgenic Arabidopsis plants expressing SINA2-YFP under the CaMV 35S or SINA2 promoter were also examined under confocal microscopy. SINA2-YFP fluorescence was observed in the elongation zone cells of roots (Fig. 3b) and in different tissue cells (Fig. S2). We also examined the localization of SINA2-YFP in 45-d-old senescent rosette leaves and ABA-treated seedlings and no significant localization difference was observed when compared with the observations shown in Fig. S2 (data not shown).
Interaction of SINA2 with itself and SINAT family proteins
Similar to the previously reported SINA and some other TRAF-like family proteins, SINA2 contains a conserved C-terminal TRAF-like domain which is required for homo- or heterodimerization (Rothe et al., 1994; Pullen et al., 1998; Leo et al., 1999). Therefore, we performed yeast two-hybridization experiments. As expected, only yeast cells co-transformed with the combination of BD-SINA2 (BD, pGBDT7, Clontech) and AD-SINA2, or BD-SINA2 and different AD-SINATs (AD, pGADT7, Clontech) could grow on SD/–Leu/–Trp/–His/–Ade media (Fig. 4a). To examine whether the SINA2 protein interacts with itself in vivo, we further carried out a BiFC assay in N. benthamiana (Chen et al., 2008; Yu et al., 2012). Luminescence was detected in the leaves infiltrated with LUCn-SINA2 and SINA2-LUCc, but not in those infiltrated with LUCn and SINA2-LUCc, LUCn-SINA2 and LUCc or LUCn and LUCc (Fig. 4b). We also verified the specific in vivo self-interaction of SINA2 by an immunoprecipitation assay in N. benthamiana through the co-expression of Myc-SINA2 and SINA2-Flag, or Myc and SINA2-Flag. Antibodies to Myc detected the proteins immunoprecipitated by Flag antibodies for the co-expression of Myc-SINA2 and SINA2-Flag, but not Myc and SINA2-Flag (Fig. 4c). These in vivo results demonstrate that SINA2 is a putative TRAF-like family protein.
SINA2 positively regulates drought response in Arabidopsis
To further determine the biological function of SINA2, we isolated two mutant lines (sina2-1 and sina2-2) that harbored a T-DNA insertion in the third exon of SINA2 (Fig. 5a). This insertion abolished the expression of SINA2 in both sina2-1 and sina2-2, as confirmed by RT-PCR analyses using SINA2-specific primers (Fig. 5b; Table S1). When grown under well-watered conditions, both sina2-1 and sina2-2 plants did not show any visible phenotypic change or developmental abnormality (Figs 5c, 6a). However, sina2 mutants, especially sina2-1, were hypersensitive to drought stress compared with wild-type plants (Figs 5c, 6a, S3a). The enhanced drought sensitivity of the sina2-1 mutant may result from the increased water loss by transpiration. Therefore, we examined the fresh weight (FW) of detached leaves over different time intervals as an indicator of transpiration water loss (Ko et al., 2006). The most rapid water loss happened within the first hour after detachment (Fig. 5d). Leaves from sina2-1 plants lost c. 38% of their fresh weight in 1 h, whereas leaves from wild-type and complemented gSINA2/sina2-1 plants had only 22% water loss during the same time period (Fig. 5d).
To further understand the function of SINA2 in drought tolerance, we produced transgenic Arabidopsis plants that overexpressed SINA2. Expression of SINA2 in transgenic plants was driven by the CaMV 35S promoter (Yang et al., 2008). At least 20 independent transgenic lines (T1 generation) were produced; ten lines were grown to produce seeds. Four transgenic lines homozygous for the 35S::SINA2 transgene (4, 6, 7 and 8) were chosen for further experiments. RT-PCR analyses showed that SINA2 was overexpressed in all transgenic lines tested (Fig. S3b). Water was withheld from 4-wk-old wild-type, sina2-1, sina2-2 and 35S::SINA2 transgenic plants (lines 4, 6 and 7) for 3 wk. At the end of treatment, 35S::SINA2 plants appeared less dehydrated, while sina2-1 and sina2-2 appeared more dehydrated than wild type (Fig. 6a,b). Upon re-watering, the survival rates of 35S::SINA2 transgenic plants were significantly higher (Fig. 6b). Detached leaves from 35S::SINA2 transgenic plants also lost water more slowly than those from wild-type and sina2-1 plants (Fig. 6c).
SINA2 regulates ABA-mediated stomatal closure
The observation that SINA2 is prominently expressed in stomata (Fig. 2j), and 35S::SINA2 plants showed enhanced drought tolerance and decreased water loss (Fig. 6a–c), impelled us to investigate whether SINA2 would regulate ABA-mediated stomatal closure, one of the processes that determine the transpiration rate (Li et al., 2011). Even under normal conditions, the stomatal aperture of 35S::SINA2-4 was smaller than that of wild type. Upon treatment with ABA, the stomatal apertures of wild type, sina2-1 and 35S::SINA2-4 all decreased, with a less significant decrease in sina2-1 and a more significant decrease in 35S::SINA2-4 compared with the wild-type plants (Fig. 6d,e). These data suggest that overexpression of SINA2 could boost ABA-mediated stomatal closure in transgenic plants.
Altered expression of stress-responsive genes in sina2-1 and 35S::SINA2 transgenic plants
To identify the possible genes involved in the SINA2-mediated drought-responsive pathway, we performed qRT-PCR to examine the expression levels of a number of ABA- and stress-responsive genes, such as ABI1, ABI2, PP2C, response to dehydration 29B (RD29B), response to ABA 18 (RAB18) and lipid transfer protein 3 (LTP3), in sina2-1 and 35S::SINA2 transgenic plants. qRT-PCR analyses demonstrated that all these genes were significantly induced by ABA in wild-type, sina2-1 and 35S::SINA2 transgenic plants. Consistent with the drought-tolerant phenotype, ABA-induced expression of ABI1, ABI2, PP2C, RD29B and LTP3 was induced to a higher level in 35S::SINA2 transgenic plants. In sina2-1, ABI2, RD29B, RAB18 and LTP3 showed weaker induction in response to the same ABA treatment, suggesting that SINA2 acts as a positive factor in regulating the expression of these stress-responsive genes (Fig. 7). The different induction patterns of these genes also indicate that SINA2 positively regulates the ABA response by modulating the expression of diverse stress-related genes in Arabidopsis.
Drought-induced SINA2 expression is ABA dependent
We also investigated the expression of SINA2 in both ABA biosynthesis- and ABA signaling-deficient mutants after treatment with ABA or drought stress. As shown in Fig. 8, unlike the wild-type plants, drought-induced expression of SINA2 was blocked in all tested mutants (aba1, aba2-1, abi1-1C and abi2-1), but ABA-induced expression of SINA2 was only blocked in abi1-1C and abi2-1 (Fig. 8). This implies that ABI1- and ABI2-mediated ABA signaling transduction is required for the drought- and ABA-induced expression of SINA2.
We further checked ABA-induced SINA2 expression in other abi mutants. The transcriptional induction of SINA2 by ABA was also dramatically reduced in abi4-1 (Finkelstein et al., 1998) and especially in abi5-7 (Nambara et al., 2002; Tamura et al., 2006), but not in abi3-8 (Nambara et al., 2002; Tamura et al., 2006), suggesting that ABI4 and ABI5 also positively regulate SINA2 expression (Fig. 9a).
Previously, it was reported that AREB1, AREB2 and ABF3 are the master transcriptional factors that cooperatively control the ABA-responsive element (ABRE)-dependent ABA signaling involved in the drought stress tolerance in Arabidopsis (Yoshida et al., 2010). We found that ABA-induced SINA2 expression was significantly impaired in the areb1 areb2 abf3 triple mutant (Fig. 9b), and the promoter region of SINA2 contains one ABRE motif. To understand whether SINA2 functions as a target of these transcriptional factors, we performed transient transcriptional activity assays in Arabidopsis protoplasts using ProSINA2::LUC (firefly luciferase-coding gene driven by the SINA2 promoter) as a reporter, and AREB1, AREB2, ABF3 or ABI5 driven by the CaMV 35S promoter as an effector (Fig. 9c). We found that AREB2, ABF3, and especially AREB1 induced the expression of LUC, but ABI5 did not (Fig. 9d). These data suggest that AREB1, AREB2 and ABF3 could induce the expression of SINA2.
The biological functions of SEVEN IN ABSENTIA (SINA) of Arabidopsis thaliana 2 (SINAT2), SINAT5 and OsDIS1 have been studied in Arabidopsis and rice (Xie et al., 2002; Welsch et al., 2007; Ning et al., 2011). We demonstrate here that, unlike SNAT2, SINAT5 and OsDIS1, SINA2 is similar to animal TRAF1 proteins which lack a conserved N-terminal RING domain (Fig. S4; Leo et al., 2001). Therefore, SINA2 does not have E3 ligase activity which is required for the degradation of its target protein. However, like other TRAF-like family proteins (Fig. 1a), it has the conserved C-terminal TRAF-like domain which may allow it to function through homo- or heterodimerization (Rothe et al., 1994; Pullen et al., 1998; Leo et al., 1999). This may be supported by the observation that SINA2 interacts with itself and all the five SINATs from Arabidopsis in yeast (Figs 4a–c, S5).
To date, numerous drought stress-inducible genes have been identified, and many of them are activated by ABA (Yamaguchi-Shinozaki & Shinozaki, 2006; Yoshida et al., 2010). When senescence occurs, ABA synthesis- and ABA signal-associated genes are up-regulated, and the endogenous ABA concentration increases (Gepstein & Thimann, 1980; Weaver et al., 1998; He et al., 2001; Van der Graaff et al., 2006; Zhang et al., 2012). ABA concentrations also increase in response to environmental stresses such as drought, salt, and high temperature which, in turn, induce leaf senescence (Guo & Gan, 2005). The observation that SINA2 is significantly induced by drought and ABA (Fig. 1b,c) and is highly expressed in senescent leaves (Fig. 2m) suggests that SINA2 may work as a functional component of ABA signaling involved in the drought stress response in adult plants only. Indeed, when we compared the growth phenotypes of 5-d-old seedlings of wild type (Col-0), sina2-1 and 35S::SINA2-4 on MS media supplemented with different concentrations of ABA, no significant difference was observed in seed germination, early seedling development, and root growth (Fig. S6). Some possible explanations are that SINA2 functions by interacting with or is regulated by another partner protein(s) which works in adult plants, or SINA2 is degraded or modified in the early growth period of plants. As in mammals, TRAF1 interacts with other partners to transduce downstream signals (Leo et al., 2001). The interaction of SINA2 with itself and the five SINATs (Fig. 4a–c) coincides with observations that the TRAF-like family proteins function through homo- or heterodimerization (Rothe et al., 1994; Pullen et al., 1998; Leo et al., 1999). The interaction of SINA2 with SINAT1, SINAT2, SINAT3, SINAT4 and SINAT5 also implies that SINA2 may perform its function by cooperating with some other stress response genes in drought stress, as most SINATs, especially SINAT3, are stress responsive (Fig. S7). We performed yeast two-hybrid screening to search for the interactive partner(s) of SINA2; unfortunately, no abiotic stress-associated genes of interest were identified.
Stomata could sense water stress and reduce water evaporation by closing their guard cells. GUS staining revealed that SINA2 is preferentially expressed in leaf guard cells (Fig. 2j). Therefore, SINA2 may function in drought tolerance by regulating stomatal closure. Indeed, both sina2-1 and sina2-2 are more sensitive, whereas transgenic plants constitutively expressing SINA2 are more resistant to drought stress compared with the wild-type plants (Fig. 6a–c). Consistent with this, ABA-mediated stomatal closure was greater in 35S::SINA2-4 than in sina2-1 and wild type (Figs 6d,e, S8), which is in accordance with the previous observation that detached leaves from 35S::SINA2-4 lost water more slowly than those from sina2-1 and wild type (Fig. 6c). Therefore, more rapid wilting after water withholding in sina2-1 and wild type could be a result of their inability to efficiently close the stomata and reduce transpiration. Actually, the stomatal aperture of 35S::SINA2-4 was smaller than that of sina2-1 and wild type even without ABA treatment (Fig. 6d,e). As ABA is a key regulator of stomatal aperture and transpiration, SINA2 may function by modulating stomatal movement through the ABA-mediated drought stress response.
As the core component of the ABA signal pathway, ABI1 regulates many drought- and ABA-responsive genes (Yamaguchi-Shinozaki & Shinozaki, 2006). abi1-1C has been identified as a dominant mutation which could block ABA signal transduction (Nishimura et al., 2007). In many cases, exogenous ABA-induced expression of stress-responsive genes is impaired in ABA-deficient mutants (Strizhov et al., 1997; Chak et al., 2000). Using the dominant mutants abi1-1C and abi2-1, as well as the ABA-deficient mutants aba1 and aba2-1, we found that expression of SINA2 in aba1 and aba2-1 was induced by ABA, but not by drought (Fig. 8). Therefore, the ABA signaling pathway appears to be required for the drought-induced expression of SINA2.
In the promoter regions of many drought stress-induced and ABA-regulated genes, conserved cis-elements, designated ABREs, control gene expression via basic leucine zipper (bZIP)-type AREB/ABF transcription factors (Yoshida et al., 2010). We found that the SINA2 promoter also contains one ABRE motif, and ABA-induced SINA2 expression was blocked significantly in areb1 areb2 abf3 (Fig. 9b). Our study with the ProSINA2::LUC system showed that AREB1, AREB2 and ABF3 (but not ABI5, which also functions as a transcription factor) all induced the expression of LUC (Fig. 9d), suggesting that AREB1, AREB2 and ABF3 are both necessary and sufficient, whereas ABI5 is necessary but not sufficient, for ABA-induced SINA2 expression. One of the possible reasons for this is that ABI5 may need to undergo a post-transcriptional modification to trigger the ABA-induced expression of SINA2. Indeed, the ABI5 protein can be phosphorylated, dephosphorylated and ubiquitinated in plants (Stone et al., 2006; Miura et al., 2009; Nakashima et al., 2009; Lee et al., 2010; Dai et al., 2013).
As proline is an important dehydrin produced in response to abiotic stresses, its concentration is one of the indicators of drought tolerance. Proline accumulations in the leaves of wild type, sina2-1, and 35S::SINA2-4 were also examined before and after the treatment with drought stress. No significant difference was detected (Fig. S3c). These data indicate that SINA2 may function through regulating the closure of stomata and therefore decreasing water loss by transpiration, not by the accumulation of dehydrins such as proline.
The induction of ABA- and stress-related genes has been taken to be a hallmark of stress adaptation in plants (Thomashow, 1999; Zhu et al., 2010). We found that overexpression of SINA2 triggered the expression of various ABA- and stress-responsive genes in transgenic lines (Fig. 7). Our observations suggest that SINA2 could function in one or several signal transduction pathways by affecting the activity of stress-related genes. Although the exact mode of action of SINA2 in plant responses to ABA and drought stress is still intangible, the results of our study provide direct evidence that altered expression of SINA2 can significantly modify resistance to drought in transgenic plants. Further study of the possible target(s) that interacts with SINA2 will help to elucidate its molecular functions.
We thank the Arabidopsis Biological Resource Center and the Nottingham Arabidopsis Stock Centre for providing us with the T-DNA insertion mutants; Dr Kazuo Shinozaki and Dr Eiji Nambara from the RIKEN Plant Science Center for providing the abi1-1C and abi5-7 mutants; Dr Chuanyou Li (Institute of Genetics and Developmental Biology, Chinese Academy of Sciences, Beijing, China) for providing the abi3-8 mutant. aba1 and abi2-1 were kindly provided by Dr Jiankang Zhu (Department of Horticulture and Landscape Architecture, Purdue University, USA). The BiFC assay vectors were kindly provided by Dr Jiawei Wang (I Shanghai Institute of Plant Physiology and Ecology, Chinese Academy of Sciences, Shanghai, China). We are also grateful to Prof. Hai Huang (Shanghai Institute of Plant Physiology and Ecology, Chinese Academy of Sciences, Shanghai, China) and Prof. Andrea Gershon (University of Toronto, Toronto, Canada) for critical comments on the manuscript. This work was supported by the following grants: National Natural Science Foundation of China 31000120, 31000288, 31171169, 31100212, 31270314, 31371228, 31370670, 30872044, and 30800880; the National Basic Research Program of China, 2010CB126600; the National Mega Project of GMO Crops, 2013ZX08001003-007, 2013ZX08004002-006, and 2014ZX0800942B; and the Strategic Priority Research Program of the Chinese Academy of Sciences, XDA08030108.