Plant Molecular Biology

, Volume 75, Issue 1, pp 107–127

Functional characterization of four APETALA2-family genes (RAP2.6, RAP2.6L, DREB19 and DREB26) in Arabidopsis

Authors

  • Sowmya Krishnaswamy
    • Department of Agricultural, Food and Nutritional ScienceUniversity of Alberta
  • Shiv Verma
    • Department of Agricultural, Food and Nutritional ScienceUniversity of Alberta
  • Muhammad H. Rahman
    • Department of Agricultural, Food and Nutritional ScienceUniversity of Alberta
    • Department of Agricultural, Food and Nutritional ScienceUniversity of Alberta
Article

DOI: 10.1007/s11103-010-9711-7

Cite this article as:
Krishnaswamy, S., Verma, S., Rahman, M.H. et al. Plant Mol Biol (2011) 75: 107. doi:10.1007/s11103-010-9711-7

Abstract

APETALA2 (AP2) transcription factors (TFs) play very important roles in plant growth and development and in defense response. Here, we report functional characterization of four AP2 TF family genes [(RAP2.6 (At1g43160), RAP2.6L (At5g13330), DREB 26 (At1g21910) and DREB19 (At2g38340)] that were identified among NaCl inducible transcripts in abscisic acid responsive 17 (ABR17) transgenic Arabidopsis in our previous microarray analyses. DREB19 and DREB26 function as transactivators and localize in the nucleus. All four genes were abundant in early vegetative and flowering stages, although the magnitude of the expression varied. We observed tissue specific expression patterns for RAP2.6, RAP2.6L, DREB19 and DREB26 in flowers and other organs. RAP2.6 and RAP2.6L were responsive to stress hormones like jasmonic acid, salicylic acid, abscisic acid and ethylene in addition to salt and drought. DREB19 and DREB26 were less responsive to stress hormones, but the former was highly responsive to salt, heat and drought. Overexpression of RAP2.6 in Arabidopsis resulted in a dwarf phenotype with extensive secondary branching and small siliques, and DREB26 overexpression resulted in deformed plants. However, overexpression of RAP2.6L and DREB19 enhanced performance under salt and drought stresses without affecting phenotype. In summary, we have demonstrated that RAP2.6, RAP2.6L, DREB26 and DREB19 are transactivators, they exhibit tissue specific expression, and they participate in plant developmental processes as well as biotic and/or abiotic stress signaling. It is possible that the results from this study on these transcription factors, in particular RAP2.6L and DREB19, can be utilized in developing salt and drought tolerant plants in the future.

Keywords

ArabidopsisAP2Transcription factorsStress tolerance

Introduction

Drought and salinity are major abiotic stress factors that affect plant productivity and can reduce average yield for most crops by 50% or more (Bray et al. 2000; Bartels and Sunkar 2005). Areas under drought and salinity are increasing worldwide (Burke et al. 2006) and, therefore, it is important to develop crops that can perform better when subjected to such environmental stresses. To date, many genes have been evaluated for stress tolerance and it has been shown that transcription factors (TFs) are highly effective in engineering stress tolerant plants (Sakuma et al. 2006a, b; Bhatnagar-Mathur et al. 2007; Khong et al. 2008). TFs are DNA-binding proteins and more than 1500 TF genes are present in Arabidopsis thaliana which constitute over 5% of its genome (Riechmann et al. 2000). TFs regulate expression of many genes and, therefore, manipulation of the expression of even a few of these regulatory genes can lead to remarkable changes in plant traits (Martin 1996; Liu et al. 1999; Udvardi et al. 2007).

The APETALA2 (AP2) gene family is one of the largest TF gene families of Arabidopsis containing 145 loci (Sakuma et al. 2002). These DNA binding proteins have characteristic AP2 domain, which contains 68 amino acids and is also referred to as AP2/ethylene responsive element binding factor domain (AP2/ERF) (Hao et al. 1998; Riechmann and Meyerowitz 1998). AP2/ERF genes can be grouped into two classes based on the number of AP2-DNA binding domains. The first class is AP2-like TFs, which encode proteins with two AP2 domains (Riechmann and Meyerowitz 1998). Examples of proteins belonging to this class are AP2, AINTEGUMENTA (ANT), baby boom (BBM) and Glossy15 (GL15). The second class is ERF-like TFs which encode proteins with only one AP2 domain and includes C-repeat/dehydration responsive element binding factors (CBFs/DREBs), LePtis, ERFs, TINY, abscisic acid insensitive (ABI4), and RAV (related to ABI3/VP1) proteins (Riechmann et al. 2000; Sakuma et al. 2002). Based on their DNA-binding regions, AP2/ERF genes have been classified into five subfamilies: AP2, RAV, DREB, ERF and others (Sakuma et al. 2002). The ERF and DREB subfamily proteins regulate many stress responsive genes by binding to defined cis-regulatory sequence (Guo et al. 2005). The ERF subfamily proteins bind to ethylene response elements (ERE) or GCC box found in the promoters of ethylene inducible pathogenesis related genes (Ohme-Takagi and Shinshi 1995), while the DREB subfamily proteins bind to C- repeat or dehydration response element (DRE) in the promoters of low temperature and/or water deficit responsive genes (Stockinger et al. 1997; Gilmour et al. 1998).

The AP2/ERF family proteins have been implicated in various growth events like plant growth, flower development, meristem determinancy and organ identity, as well as abiotic/biotic stress tolerance (Saleh and Pagés 2003). For instance, AP2, AINTEGUMENTA, TINY, DRN, BD1 genes are involved in floral morphogenesis, organ identity and growth regulation (Kunst et al. 1989; Klucher et al. 1996; Wilson et al. 1996; Chuck et al. 1998; Kirch et al. 2003). The genes DREB1A, DREB2A, WXP1, CaPF1, Pti, CaERFLP1 and NtERF5 have been reported to be involved in biotic and abiotic stress tolerance (Liu et al. 1998; Gu et al. 2002; Yi et al. 2004; Lee et al. 2005; Fischer and Dröge-Laser 2004; Zhang et al. 2007). Some of the AP2 TFs like ABI4, AtERF4, ABR1 and DDF1 are also involved in abscisic acid (ABA), ethylene (ET), gibberellic acid (GA) and brassinosteroid response signaling (Finkelstein et al. 1998; Hu et al. 2004; Magome et al. 2004; Yang et al. 2005; Pandey et al. 2005). Despite the important roles played by AP2 TFs in many physiological aspects of plant, the precise functions of many members of this family are still unknown (Nakano et al. 2006). Nevertheless, there are reports of improving plant’s response to stress through modifying the expression of AP2 TFs (Nakano et al. 2006; Sakuma et al. 2006a, b).

The pea (Pisumsativum) abscisic acid-responsive protein ABR17 is a member of pathogenesis related protein 10 (PR10) families and is also referred as PR10.4 (Iturriaga et al. 1994; Srivastava et al. 2006). ABR17 is significantly homologous to intracellular pathogenesis related (IPR) proteins and has been demonstrated to possess ribonuclease activity (Iturriaga et al. 1994; Srivastava et al. 2007). In addition to ribonuclease activity, members of this family have demonstrated binding properties with phytohormones like cytokinins and brassinosteroids, and therefore have been implicated in hormone signaling and suggested to have function as general hormone carriers (Carpin et al. 1998; Mogensen et al. 2002; Markovic-Housley et al. 2003; Pasternak et al. 2006). Abundance of ABR17 protein has been observed in salt treated pea plants (Kav et al. 2004) and constitutive overexpression of pea ABR17 in Arabidopsis and Brassica has resulted in phenotype with early flowering, increased number of lateral branches and siliques and had elevated levels of CKs compared to WT (Srivastava et al. 2006, 2007; Dunfield et al. 2007). Furthermore, plants overexpressing ABR17 have exhibited enhanced seed germination and seedling vigor under multiple abiotic stresses including salinity stress (Srivastava et al. 2006). In our microarray analyses of ABR17-mediated modulation of gene expression (Krishnaswamy et al. 2008), we observed that transcript abundance of four putative AP2 TF genes, RAP 2.6, RAP 2.6L, DREB19 and DREB26 were up-regulated significantly in salt treated ABR17-transgenic plants compared to unstressed transgenic plants, while only RAP2.6 and RAP2.6L transcripts were observed to increase in abundance significantly in salt treated wild type (WT) plants compared to unstressed WT plants (Krishnaswamy et al. 2008). In addition, transcript abundance of RAP2.6 was significantly higher in salt treated ABR17-transgenic plants compared to salt treated WT plants (Krishnaswamy et al. 2008). However, there were no significant differences in expression of these AP2 genes between WT and ABR17-transgenic plants under normal/unstressed conditions (Krishnaswamy et al. 2008). It was speculated that the observed enhanced stress tolerant phenotype of ABR17-transgenic Arabidopsis compared to WT (Srivastava et al. 2006) could be, at least in part, due to increased expression of AP2 TF genes together with the expression modulation of other important genes (Krishnaswamy et al. 2008). AP2 family genes are known to play important roles in abiotic stress response and based on transcript abundance of RAP2.6, RAP2.6L, DREB19 and DREB26 in salt treated WT and/or ABR17-trasngenic Arabidopsis plants, we hypothesize that these AP2 genes might participate in plant defense response against salt stress and therefore overexpression of these genes in Arabidopsis might enhance tolerance to salt and related stress conditions like drought.

RAP2.6 and RAP2.6L belong to ERF subfamily, while DREB19 and DREB26 belong to DREB subfamily (Guo et al. 2005), and all four of them code for proteins with single AP2 domain (Fig. 1). RAP2.6 is activated by the CBF (C repeat binding factor) expression (Fowler and Thomashow 2002) is involved in plant stress and has been shown to code for protein that possess transcription activator function (He et al. 2004; Zhu et al. 2010). RAP2.6L has been demonstrated to be involved in gene regulation during shoot regeneration from root explants (Che et al. 2006) as well as in disease resistance (Sun et al. 2010). However, there is no information available on DREB19 and DREB26 genes. Here we report and discuss the results from functional characterization of RAP2.6, RAP 2.6L, DREB19 and DREB26 with respect to overexpression, localization/transactivation, spatial/temporal expression and stress/hormonal response experiments. In addition, overexpressed transgenic lines are evaluated under salt and drought stress conditions and the utility of these AP2 genes in engineering plants for abiotic stress tolerance is discussed.
https://static-content.springer.com/image/art%3A10.1007%2Fs11103-010-9711-7/MediaObjects/11103_2010_9711_Fig1_HTML.gif
Fig. 1

Sequence alignment of AP2 domains from AP2 TF proteins (RAP2.6-At1g43160; RAP2.6L-At5g13330; DREB19-At2g38340; DREB26-At1g21910) showing YRG and RAYD elements. Residues in black box represent conserved amino acid residues between the AP2 TF proteins. Sequence alignment was done using MegAlign (DNASTAR Lasergene8) software

Materials and methods

Subcellular localization

RNA was extracted from A. thaliana (ecotype WS) using the RNeasy® Plant Mini Kit (Qiagen Sciences), reverse transcribed (iScript® cDNA synthesis kit, Bio-Rad laboratories) and the cDNA was used as a template to amplify AP2 TF genes (DREB19 and DREB26) using polymerase chain reaction (PCR; High Fidelity PCR system, Roche Diagnostics Corp.). Gene specific primers were as follows: DREB19 (forward: 5′-CATGCCATGGAAAAGGAAGATAACGGATCGAAACAGAGCTCC-3′, reverse: 5′-CATGCCATGGCAGCTCCACCTCCACCTCCGAATCTGAAATACTCAAAATATGAATAGAATC-3′) and DREB26 (forward: 5′-CATGCCATGGTGAAACAAGAACGCAAGATCCAAACCAGC-3′, reverse: 5′-ATGCCATGGCAGCTCCACCTCCACCTCCATTGAAACTCCAAAGCGGAATG-3′). The AP2 genes were amplified using the following thermocycling parameters: DREB19 (94°C for 2 min; 35 cycles at 94°C for 1 min, 62°C for 1 min, 72°C for 2 min; and a final extension of 72°C for 7 min) and DREB26 (94°C for 2 min; 10 cycles at 94°C for 30 s, 40°C for 30 s, 72°C for 45 s; 30 cycles at 94°C for 30 s, 48°C for 1 min, 72°C for 50 s; and a final extension of 72°C for 5 min). Amplified PCR products were gel purified (QIAquick® gel extraction kit, Qiagen Sciences), restriction digested using NcoI or BspHI (New England Biolabs) and cloned into pCsGFPBT (GenBank: DQ370426). A Gly-Ala-rich peptide linker was used between coding sequence and synthetic green fluorescent protein (sGFP) while generating the fusion protein (Jiang and Deyholos 2009). Sequences of the constructs were verified by DNA sequence analysis and, along with empty vector controls (VC), transformed into Agrobacterium tumefaciens GV3101 using the freeze–thaw method (Weigel and Glazebrook 2002). Agrobacterium strains carrying recombinant pCsGFPBT (with DREB19 and DREB26) and VC were transformed into A. thaliana (ecotype WS) using the floral dip method (Clough and Bent 1998). T0 seeds were screened on half-strength Murashige and Skoog (MS) medium (Murashige and Skoog 1962) containing 50 mg/L hygromycin B (Sigma–Aldrich). Roots from seven-day-old T1 plants transformed with genes of interest along with VC seedlings were stained with DAPI (4′, 6-diamidino-2′-phenylindole, dihydrochloride; 0.5 μg/ml) for ten minutes, washed twice with distilled water and mounted on slides. The slides were visualized under florescence microscope (Zeiss fluorescence microscope) or confocal microscope (Leica DM IRBE, Leica Microsystems Inc.) for the sGFP and DAPI signals. At least five independent T1 plants from each construct were used in these studies.

Trans-activation assay

The coding sequences of AP2 TF genes (RAP2.6, RAP2.6L, DREB19 and DREB26) were PCR amplified (High Fidelity PCR system, Roche Diagnostics Corp.) using cDNA of A. thaliana (ecotype WS) as template. Gene specific primers used for the amplification were as follows: RAP2.6 (forward: 5′-GCGGCCGGAATTCATGGTGTCTATGCTGACTAATGTTGT-3′, reverse: 5′-GCGGTCGGTCGACTTAACCAAAAGAGGAGTAATTGTAT-3′), RAP2.6L (forward: 5′-GATCTCGGAATTCATGGTCTCCGCTCTCAGCCGTGTCAT-3′, reverse 5′-GCGGCCGCTGCAGTTATTCTCTTGGGTAGTTATAATAA-3′), DREB19 (forward: 5′-GCGGCCGGAATTCATGGAAAAGGAAGATAACGGATCG-3′, reverse: 5′-GCGGCCGGTCGACCTAGAATCTGAAATACTCAAAATATG-3′) and DREB26 (forward: 5′-GCGGCCGGAATTCATGGTGAAACAAGAACGCAAGATCC-3′, reverse: 5′-GCGGTCGGTCGACTTAATTGAAACTCCAAAGCGGAAT-3′). PCR thermocycling parameters for DREB19 and DREB26 were as described previously. RAP2.6 and RAP2.6L genes were amplified using the following thermocycling parameters: RAP2.6 (94°C for 2 min; 35 cycles at 94°C for 1 min, 62°C for 1 min, 72°C for 2 min; and a final extension of 72°C for 7 min) and RAP2.6L (94°C for 2 min; 35 cycles at 94°C for 30 s, 63°C for 1 min, 72°C for 50 s; and a final extension of 72°C for 7 min). The amplified fragments were gel purified (QIAquick® gel extraction kit, Qiagen Sciences), and double digested using restriction enzymes EcoRI-SalI (for RAP2.6, RAP2.6L and DREB19) and EcoRI -PstI (for DREB26) (New England Biolabs). The digested fragments were cloned into pBD-GAL4 Cam vector (Stratagene) and their sequences were confirmed by DNA sequence analysis. The sequenced recombinant plasmids (carrying RAP2.6, RAP2.6L, DREB19 or DREB26 gene) and empty vector controls (VC) were transformed into yeast strain YRG-2 (Stratagene), according to the manufacturer’s instructions. Positive yeast colonies were selected on synthetic drop out (SD) medium for tryptophan (Sigma–Aldrich®). The YRG-2 strain has the auxotrophic marker hitidine (his3) as a reporter for detection of trans-activation activity. The positive yeast colonies, confirmed by PCR, were streaked on synthetic drop-out medium for histidine (Sigma–Aldrich®) for determining the trans-activation activity, along with the controls (yeast without vector and with empty pBD-GAL4 Cam plasmid).

Overexpression constructs

The coding sequences of AP2 TF genes (RAP2.6, RAP2.6L,DREB19 and DREB26) were amplified as described previously using cDNA of A. thaliana (ecotype WS) as template. Gene specific primers used in the experiment were as follows: RAP2.6 (forward: 5′-GAGGCGCTCGAGATGGTGTCTATGCTGACTAATGTTGTCTC-3′, reverse : 5′-GCCGGCGTCTAGATTAACCAAAAGAGGAGTAATTGTATTGATCATATTC-3′), RAP2.6L (forward: 5′-TAATTAGAAGCTTATGGTCTCCGCTCTCAGCCGTGTCATAG-3′, reverse: 5′-GGCCGCGTCTAGATTATTCTCTTGGGTAGTTATAATAATTGTAAC-3′), DREB19 (forward: 5′-GCGGCGTCTAGACTAGAATCTGAAATACTCAAAATATGAATCGAATC-3′, reverse: 5′-GTGTCGAAGCTTATGGGACGATCACCGTGTTGTGAGAAGAAG-3′) and DREB26 (forward: 5′-GCGCCGAAGCTTATGGTGAAACAAGAACGCAAGATCC-3′, reverse: 5′-GCGCGCGTCTAGATTAATTGAAACTCCAAAGCGGAATGTC-3′). The PCR conditions used to amplify AP2 genes were as described previously. The amplified products were gel purified (QIAquick® gel extraction kit, Qiagen Sciences) and inserted between cauliflower mosaic virus 35S (CaMV35S) promoter and rbcS32 terminator in the binary vector pKYLX-71 (Schardl et al. 1987), using restriction enzymes, XhoI-XbaI (for RAP2.6) and HindIII-XbaI (for RAP2.6L, DREB19 and DREB26) (New England Biolabs). The sequenced recombinant plasmids and empty vectors were transformed into A. tumefaciens GV3101 using freeze–thaw method (Weigel and Glazebrook 2002) and subsequently transformed into A. thaliana (WS) as described previously. T0 seeds were screened for transformants on half strength MS medium (Murashige and Skoog 1962) containing kanamycin (50 mg/L), 1.5% sucrose and 0.7% agar. T1 seeds were screened for 3:1 ratio and bulked homozygous T2 seeds were used for further studies. To confirm the presence of transgenes, the homozgygous T2 plants were grown for a month and the leaf tissue was used to extract RNA (RNeasy® Plant Mini Kit, Qiagen Sciences). cDNA was synthesized (iScript® cDNA synthesis kit, Bio-Rad laboratories) and used as template for RT-PCRs. The PCRs were carried out using gene specific forward primer and vector specific reverse primer. Plant Actin (forward: 5′-TGTTGCCATTCAGGCCGTTCTTTC-3′ and reverse: 5′-TGGAACCACCACTGAGAACGATGT-3′) or 18srRNA (forward: 5′-CCAGGTCCAGACATAGTAAG-3′ and reverse: 5′-GTACAAAGGGCAGGGACGTA-3′) primers were used as internal controls. The genes 18srRNA and Actin were amplified using the following thermal conditions: 94°C for 2 min; 15 cycles at 94°C for 1 min, 55°C for 1 min, 72°C for 1 min; and a final extension of 72°C for 10 min.

Promoter activity by GUS fusion

Sequences upstream of the ATG codon from RAP2.6 (930 bp), RAP2.6L (999 bp), DREB19 (781 bp) and DREB26 (992 bp) and were amplified by PCR (High fidelity Pfu polymerase kit, Fermentas Life Sciences), using genomic DNA of wild type (WS) WT A. thaliana (ecotype WS) as template. Primers used to amplify upstream sequences of the AP2 genes were as follows: RAP2.6 (forward: 5′-GCGGCCGAAGCTTGTTGTTGTCTTTTCTTCCAAGGAAG-3′, reverse: 5′-GCGGTCGTCTAGAGTTTGAAATTGCGGTGGTAGACAAG-3′), RAP2.6L (forward: 5′-GTGGTCGATCGATGCAGTTTAGTACCTGACTAATCTTGCAGCTTTTA-3′, reverse: 5′-ATATCAGGGATCCGGCGGTGACATCAGTCTCGTTCCAAGACGAATT-3′), DREB19 (forward: 5′-GCGGCCGAAGCTTAGTAAATTACAAAAAAGTACAAAGTC-3′, reverse: 5′-GCGGCCGGGATTCTGGAAAAACACAACACGTACAAACTGTAG-3′) and DREB26 (forward: 5′-GCGGCCGAAGCTTAAGAAAATTGATATCTCACAACC-3′, reverse: 5′-GTGGTCGGGATCCGGTAATGTTGTTGTGTACGTACAGGCT-3′). The promoter sequences were amplified using the following thermocycling parameters: RAP2.6 (94°C for 2 min; 10 cycles at 94°C for 30 s, 45°C for 30 s, 72°C for 45 s; 30 cycles at 94°C for 30 s, 55°C for 1 min, 72°C for 50 s; and a final extension of 72°C for 5 min), DREB26 (94°C for 2 min; 10 cycles at 94°C for 30 s, 40°C for 30 s, 72°C for 45 s; 30 cycles at 94°C for 30 s, 60°C for 1 min, 72°C for 50 s; and a final extension of 72°C for 5 min), RAP2.6L and DREB19 (94°C for 2 min; 10 cycles at 94°C for 30 s, 40°C for 30 s, 72°C for 45 s; 30 cycles at 94°C for 30 s, 55°C for 1 min, 72°C for 50 s; and a final extension of 72°C for 5 min). The amplified fragments were gel purified (QIAquick® gel extraction kit, Qiagen Sciences), and double digested using restriction enzymes HindIII-XbaI (for RAP2.6), ClaI-BamHI (for RAP2.6L) and HindIII-BamHI (for DREB19 and DREB26) (New England Biolabs). The CaMV 35S promoter of the binary vector pBI121 (GenBank: AF48578) was replaced with AP2 TF gene promoter to express β-glucuronidase (GUS) gene and the sequence of the recombinant plasmids verified. The sequenced recombinant or empty plasmids (pBI121) were transformed into Arabidopsis as described previously. The T0 transgenic plants were selected on half strength MS medium containing kanamycin (50 µg/ml), 1.5% sucrose and 0.7% agar. The presence of transformed promoter was confirmed in T1 plants by PCR using genomic DNA as template. Forward promoter specific and reverse GUS gene specific primers were used in PCRs.

T1 plants were used for analyzing promoter activity, and at least 5 independent transgenic lines in each promoter construct were used in the study. Promoter activity was considered in terms of GUS activity that leads to blue color formation by reacting with the substrate X-Gluc (5-bromo-4-chloro-3-indolyl β-d-lucuronide; Sigma–Aldrich ®) (Jefferson et al. 1987). For GUS activity assays, the following samples were used: germinated seeds, 7-day-old seedlings, 14-day-old seedlings, rosette leaves, inflorescence, immature and mature pods. The samples were permeabilised in cold 90% acetone for 1 h at −20°C and washed twice for 5 min with 100 mM phosphate buffer (pH 7.6). The samples were incubated overnight at 37°C in GUS staining buffer (2 mM X-Gluc, 2 mM K4 [Fe (CN) 6].3H20 and 2 mM K3 [Fe (CN) 6]). The samples were washed with 70% ethanol and scored for dark blue staining. Small samples like seed, flower and immature pods were photographed using dissecting microscope (Wild model M8, Wild Leitz Canada Ltd.) equipped with a digital camera (Nikon DXM1200).

Plant growth conditions

For studying spatial and temporal expression patterns of AP2 TF in A.thaliana, WT plants were grown in 6″ pots containing Metro Mix 290 (Grace Horticultural products) in the green house (22°C day, 18°C night, 16 h photoperiod). The plants were fertilized once every 2 week (Peters 20-20-20, Plant Products) containing micronutrients. Tissue was collected at different growth stages of Arabidopsis (according to Boyes et al. 2001) from: seedling above ground (growth stage 1.1, 10 rosette leaves >1 mm in length), rosette leaves and stem (growth stage 3.7, rosette is 70% of final size), early floral buds (growth stage 5.1 when plants start to bolt), inflorescence (growth stage 6.1, 10% of flowers to be produced have opened) and mature siliques (growth stage 7, filled siliques). Tissues were flash frozen in liquid nitrogen and stored at −80°C, which was later used for quantitative real time-PCR (qRT–PCR) for studying the expression profile of AP2 TF genes.

For observing the differences in the phenotype among AP2 TF overexpression lines and controls (WT and VC), the lines were grown in 6″ pots as mentioned earlier. The flowering time was recorded and the plants were photographed at different stages of growth. The experiment was repeated three times, in each biological replication with 10 plants per line. For recording the time of flowering, at least 30 plants/line/biological replication was used. The data was analyzed using statistical analysis software (SAS) version 9.1 (SAS Institute Inc.).

Hormone treatments of WT Arabidopsis

WT Arabidopsis plants were grown in plastic trays, in the greenhouse for 3 weeks as described earlier. Jasmonic acid (JA; 50 μM), salicylic acid (SA; 1 mM) and abscisic acid (ABA; 50 μM) (Sigma–Aldrich®) were made in 0.1% (v/v) ethanol and applied on Arabidopsis plants with a hand-held spray bottle. The ethylene (ET) treatment was performed in an air-tight acrylic chamber (1.5 m × 0.6 m × 0.6 m) placed in the greenhouse, into which 100 ppm ethylene gas in air (Praxair) was passed at the rate of 2 L/min. The control treatment was performed on plants in another chamber into which air (Praxair) was passed at the same rate. Leaves and shoots were collected and pooled after 6 and 24 h post-treatment and flash frozen in liquid nitrogen, and stored at −80°C. qRT–PCR was performed with these samples to study the response of AP2 TF genes to different stress related hormones. The entire experiment was repeated three times and there were at least 25 plants per treatment in every biological replication.

Imposition of stresses on WT Arabidopsis

The response of AP2 TF genes to different stresses was studied by imparting stress to WT Arabidopsis plants. For drought stress, the plants were grown for 2 weeks in the greenhouse as mentioned earlier. After 2 weeks, watering was withheld and plants were allowed to wilt (which took another 9–10 days). Control plants were well-watered till the tissues were collected. Leaf samples from wilted and well watered-plants were flash frozen in liquid nitrogen and stored at −80°C. The experiment was repeated three times and there were 30 plants per treatment in each biological replication. For heat and freezing stresses, WT Arabidopsis seeds were seeded on petri dish containing half strength MS (Murashige and Skoog 1962) medium (1.5% sucrose and 0.7% agar) and grown for 2 weeks (at RT and light intensity 40 μEM2s−1). Heat stress was imposed by placing the petri dishes containing seedlings at 48°C for 2 h. The plates were subsequently incubated at 22 ± 1°C for another 6 h. For inducing freezing stress, the plates were placed at −5°C for 4 h and returned to 22 ± 1°C for 6 h. For salt stress, WT Arabidopsis seeds were seeded in petri dish containing MS medium (1.5% sucrose, 0.7% agar and 100 mM NaCl) and grown for 2 weeks (RT and light intensity 40 μEM2s−1). The seedlings from salt stress, heat stress (6 h post treatment), freezing stress (6 h post treatment) and control treatments were flash frozen in liquid nitrogen and stored at −80°C. qRT-PCR was performed to study the response of AP2 TF genes to different stresses.

Evaluation of transgenic Arabidopsis lines against salt and drought stresses

To evaluate the performance of AP2 transgenic Arabidopsis under different stresses, the following lines were used: wild type (WT), vector control (VC), RAP2.6 (lines A2, A6 and A39), RAP2.6L (lines C23, C28 and C31) and DREB19 (lines D1, D5 and D12). For salt stress, seeds from different lines were seeded on half strength MS (Murashige and Skoog 1962) medium supplemented with 1.5% sucrose, 0.7% agar and NaCl (0 mM, 125 mM or 150 mM) and incubated for 3 weeks (at 22 ± 1°C, light intensity of 18 μEM2s−1 and 12 h photoperiod). Germination counts were recorded every week and the plates were photographed after 3 weeks. The experiment was repeated three times and there were 6 plates (14 seeds per plate) per line in each biological replicate.

For inducing salt stress in the green house, after 8 DAS (days after sowing) control and transgenic seedlings were watered with salt water (200 mM NaCl) on alternate days till 25 DAS. Data on number of plants flowered and number of plants with pods were recorded. For inducing drought stress, on 8 DAS, the trays were watered to saturation and excess water was allowed to drain. After this, the plants were not watered till they wilted (which took another 15–17 days) and, subsequently, re-watered. Data on number of plants wilted, number of plants recovered 1 day after re-watering, number of plants with flowers and number of plants with pods, was recorded. Salt and drought stressed plants were photographed along with the unstressed plants. These experiments were repeated three times, and there were 16 plants/line/treatment in each biological replicate. The data was statistically analyzed using SAS—version 9.1 (SAS Institute Inc.). During data analysis, the percentages in each observation class (e.g. percent germination) were calculated based on the number of seedlings at the start of the experiment.

qRT-PCR

qRT-PCR was performed to study the response of AP2 TF genes to different stresses and hormones, and to determine their expression at different stages of plant growth, and also to measure their expression levels in overexpressed transgenic plants. RNA was extracted from the pooled tissue as described earlier and was treated with RNase-free-DNase (Qiagen Sciences). RNA was quantified using NanoDrop ND-1000 (NanoDrop Technologies, Inc.), and was electrophoresed on 1.2% agarose gel in order to test the integrity and reverse transcribed to synthesize cDNA, which was used as template in qRT-PCR (iScript® cDNA synthesis kit, Bio-Rad laboratories). Primers for qRT-PCR were designed using PrimerExpress3.0 (Applied Biosystems) targeting an amplicon size of 80–150 bp. Primer specificity was tested by performing BLAST analysis (http://www.ncbi.nlm.nih.gov/). Primers used in the qRT-PCR analysis were as follows: RAP2.6 (forward: 5′- GAGAGGCCAAAAAAATATAGAGGAGTAA-3′, reverse:5′- GCCTTGTGTGGGTCTCGAA-3′), RAP2.6L (forward: 5′- CAAGGCCCTACTACCACCACAA-3′, reverse: 5′- GGTCGAGGAGGAGGTGAGTTC-3′), DREB19 (forward: 5′- GCTTGGCACGTTTGCTACTG-3′, reverse: 5′- TGGCATAGGGTCCGTACATGA-3′), DREB26 (forward: 5′- GGGCACCAAATCAAAAGACAA-3′, reverse: 5′- GTGCAACATCGTAAGCTCTAGCA-3′), actin (forward: 5′-CCACCATGTTCCCAGGAATT-3′, reverse: 5′-TTTCTCTCTGGCGGTGCAA-3′), Glyceraldehyde-3-phosphate dehydrogenase (GAPDH; forward: 5′-TTGGTGACAACAGGTCAAGCA-3′, reverse: 5′-AAACTTGTCGCTCAATGCAATC-3′) and ubiquitin-conjugating enzyme 21 (UBC21; forward: 5′-CTGCGACTCAGGGAATCTTCTAA 3′; reverse: 5′-TTGTGCCATTGAATTGAACCC- 3′). qRT-PCR analysis was performed using SYBR green system (Yang et al. 2007) on ABI StepOne thermocycler (Applied Biosystems Inc.). The delta-delta method (Livak and Schmittgen 2001) was used to calculate the relative gene expression using either actin, GAPDH or UBC21 as an endogenous control. Reactions were performed in triplicate using samples from each biological replicate.

Results

Gene isolation

We isolated RAP2.6 (At1g43160), RAP2.6L (At5g13330), DREB19 (At2g38340) and DREB26 (At1g21910) genes from wild type A. thaliana (ecotype WS). Sequence analysis showed differences in coding sequence of RAP2.6 and DREB26 compared to available sequence from Columbia genotype (Accession numbers AY062847 and BT024616, respectively). The coding sequence of RAP2.6 had three substitutions (at positions 61, 405 and 420), however, when translated changed only tryptophan 20 to arginine 20 (W20R). The coding sequence of DREB26 had three extra bases at nucleotide position 114 and the resulting translated product had one extra amino acid (serine) at the 38th position.

Subcellular localization

Based on consensus sequence analysis, RAP2.6, RAP2.6L, DREB19 and DREB26 proteins were deduced to contain a single DNA binding AP2 domain (Fig. 1) and were therefore expected to act as TFs and therefore should localize in nucleus. Nuclear localization of RAP2.6 and RAP2.6L has been reported (Che et al. 2006; Zhu et al. 2010). In order to confirm the presumed nuclear localization of DREB19 and DREB26, the coding regions were translationally fused to the N-terminus of synthetic green fluorescent protein (sGFP) under the control of cauliflower mosaic virus (CaMV) 35S promoter, and expressed in Arabidopsis. Localization was determined by visualizing root samples employing fluorescence confocal microscopy. As shown in Fig. 2, sGFP was uniformly distributed throughout the cell in control, whereas AP2: sGFP fusion proteins (sGFP: DREB19 and sGFP: DREB26) were detected exclusively in the nucleus, suggesting that these proteins are indeed constitutively nuclear localized. In addition, DAPI and GFP were co-localized in sGFP: DREB19 and sGFP: DREB26 Arabidopsis roots (Fig. S1) confirming their nuclear localization.
https://static-content.springer.com/image/art%3A10.1007%2Fs11103-010-9711-7/MediaObjects/11103_2010_9711_Fig2_HTML.jpg
Fig. 2

Roots from 1 week old (T2) transgenic Arabidopsis plants showing nuclear localization of a DREB19, b DREB26 and c control pCsGFPBT under confocal microscope. Left panel is bright field, the middle panel is GFP florescence, and the right one is overlay of the two images

Transactivation assay

AP2 TF proteins can function either as transcriptional activators or as repressors based on the presence of a conserved EAR (ethylene-responsive element-binding factors-associated amphillic repression) motif (Stockinger et al. 1997; Fujimoto et al. 2000; Ohta et al. 2001). The AP2 genes RAP2.6, RAP2.6L, DREB19 and DREB26 lack an EAR motif and therefore were expected to act as transcriptional activators. To verify this, a transactivation assay was performed using yeast one hybrid system. The full-length coding region of AP2 genes (RAP2.6, RAP2.6L, DREB19 and DREB26) were fused with GAL4 binding domain using pBD-Gal4 Cam plasmid and tested for the expression of HIS3 reporter gene in yeast (Fig. 3). Yeast cells carrying pBD-Gal4 Cam -AP2 TF genes activated the expression of downstream HIS3 reporter gene, enabling them to grow on synthetic drop-out/-histidine medium (Fig. 3a). Yeast cells with or without empty pBD-Gal4 Cam plasmid did not grow on synthetic drop-out/-histidine medium (Fig. 3a). These results suggest that RAP2.6, RAP2.6L, DREB19 and DREB26 genes indeed code for transcriptional activators.
https://static-content.springer.com/image/art%3A10.1007%2Fs11103-010-9711-7/MediaObjects/11103_2010_9711_Fig3_HTML.gif
Fig. 3

a Transactivation assay of AP2 TF genes, and b a schematic representation of yeast-one- hybrid system with HIS3 reporter. Controls: YRG-2 and YRG-2 with pBD-GAL4-Cam. Growth of the transformants on SD/-his medium indicates that the corresponding gene encodes protein with transactivation activity. The genes RAP2.6, RAP2.6L, DREB19 and DREB26 code for transactivators

Spatial and temporal expression pattern

AP2/ERF family proteins have been reported/shown to play a key role in plant growth and development (Saleh and Pagés 2003). In order to explore the possibility of the involvement of RAP2.6, RAP2.6L, DREB19 and DREB26 in growth processes, their expression patterns were investigated employing qRT-PCR and GUS reporter fusion system. For qRT-PCR, tissue was collected from the following growth stages (Boyes et al. 2001) of Arabidopsis: seedling above ground (stage 1.1, 10 rosette leaves >1 mm in length), rosette leaves and stem (stage 3.7, rosette is 70% of final size), early floral buds (stage 5.1 when plants start to bolt), inflorescence (stage 6.1, 10% of flowers to be produced have opened) and mature siliques (stage 7, filled siliques). qRT-PCR was used to examine the transcript abundance of AP2 genes in different tissues compared to their levels in rosette leaves and stems (stage 3.7). RAP2.6 mRNA was more abundant in seedling as compared to rosette leaves and stem, early floral buds and inflorescence (Fig. 4a), while transcripts of RAP2.6L were more abundant in inflorescence and early floral buds as compared to rosette leaves and seedlings (Fig. 4b). DREB19 mRNA was more abundant in seedlings and inflorescence (Fig. 4c), and DREB26 mRNA was more abundant in inflorescence as compared to other tissues (Fig. 4d). Transcript abundance of all the studied AP2 genes decreased from seedling stage to rosette leaf stage, but subsequently increased during flowering, except for RAP2.6, whose transcripts were more abundant in seedlings than in any other tissues sampled (Fig. 4). None of the examined AP2 transcripts were detected in mature siliques (data not shown).
https://static-content.springer.com/image/art%3A10.1007%2Fs11103-010-9711-7/MediaObjects/11103_2010_9711_Fig4_HTML.gif
Fig. 4

Spatial and temporal expression pattern of AP2 TF genes, aRAP2.6, bRAP2.6L, cDREB19, and dDREB26 at different stages of growth: RL & S (rosette leaves and stem from growth stage 3.7), SL (seedling, growth stage 1.1), EFB (early floral buds) and I (inflorescence). qRT-PCR analysis was performed to compare the transcript abundance of AP2 TF genes in different tissues relative to RL & S

In order to investigate the tissue specific expression pattern of AP2 genes, germinated seeds, 7-day-old seedlings, 14 days old seedlings, flowers and silliques of T1Arabidopsis containing AP2 TF promoters and β-glucuronidase (GUS) reporter gene fusion constructs (pRAP2.6-GUS, pRAP2.6L-GUS, pDREB19-GUS, pDREB26-GUS and control pCaMV35S-GUS) were tested for GUS activity. As shown in Fig. 5, strong levels of GUS expression was detected in germinated seeds, seedlings, flowers and siliques of control Arabidopsis plants (pCaMV35S-GUS). In plants bearing pRAP2.6-GUS fusions, the GUS expression was detected in roots of 7-day-old seedlings, in petals and carpels and in the valves of immature silique (Fig. 5). The GUS gene expression was detected in anthers, specifically in pollens of plants with pRAP2.6L-GUS fusion construct (Fig. 5). GUS expression was detected in tip of the cotyledonary leaves in germinated seeds and in a region where laves emerge from shoots in 7-day-old seedlings and 14 days old seedlings of plants carrying pDREB19-GUS reporter gene fusion (Fig. 5). In addition, the GUS expression was detected in xylem tissues and also in stigma, anther and in the region where sepals and petals attach the peduncle in pDREB19-GUS Arabidopsis plants (Fig. 5). A strong expression of GUS was detected in cotyledonary leaves of 7 days old seedlings, ovules and seeds in immature siliques (Fig. 5) of plants carrying pDREB26-GUS. In addition, a weak level of GUS expression was also detected in 14 days old seedlings containing pDREB26-GUS (Fig. 5). All the studied AP2 genes (RAP2.6, RAP2.6L, DREB19 and DREB26) were found to be expressed in Arabidopsis flowers, with very specific expression patterns as detected by promoter-GUS fusions (Fig. 5).
https://static-content.springer.com/image/art%3A10.1007%2Fs11103-010-9711-7/MediaObjects/11103_2010_9711_Fig5_HTML.jpg
Fig. 5

Promoter activity of AP2 TF genes as determined by GUS reporter expression. ae represents a germinated seeds, b 7 days old seedlings, c 14 days old seedling, d flower, and e immature silique of plants containing GUS transgene with different AP2 promoters (CaMV-35S, RAP2.6, RAP2.6L, DREB19 and DREB26)

Response to different stresses and stress hormones

In addition to their involvement in plant growth and development, AP2 TF genes have been implicated in biotic and abiotic stress response (Saleh and Pagés 2003; Nakano et al. 2006). We used qRT-PCR to investigate the responses of AP2 genes (RAP2.6, RAP2.6L, DREB19 and DREB26) in Arabidopsis seedlings to different abiotic stresses like salt (100 mM NaCl), heat (48°C), freezing (−5°C) and drought (Fig. 6, left panel). Transcript abundance of AP2 genes in stressed seedlings was compared to their controls. We observed that the tested AP2 transcripts exhibited expression modulation following exposure to NaCl, heat, freezing and drought, although the level of the response differed between stresses (Fig. 6). RAP2.6 transcripts significantly increased following exposure to NaCl, heat, drought but was unaffected by freezing (Fig. 6a). Transcript abundance of RAP2.6L was unaffected by heat stress but significantly increased during NaCl and drought stress, and significantly decreased following freezing stress (Fig. 6b). DREB19 transcripts were significantly increased in abundance on exposure to NaCl, heat and drought, but did not change in abundance after freezing stress (Fig. 6c). The transcript abundance of DREB26 did not alter as a result of exposure to NaCl or drought, although their abundance was significantly decreased following exposure to heat and freezing stress (Fig. 6d). All the tested AP2 genes except DREB26 exhibited an increase in transcript abundance on exposure to NaCl and drought (Fig. 6). Our results suggest an important role for these TFs in mediating plant responses to abiotic stresses.
https://static-content.springer.com/image/art%3A10.1007%2Fs11103-010-9711-7/MediaObjects/11103_2010_9711_Fig6_HTML.gif
Fig. 6

qRT-PCR analysis of aRAP2.6, bRAP2.6L, cDREB19, and dDREB26 transcript abundance in 14 day-old Arabidopsis seedlings exposed to various (left panel) stresses and (right panel) stress hormones showing transcript abundance in treated plants relative to control. The gene expression levels in control samples have been normalized to 1. Asterisks indicate a statistically significant difference (P < 0.05) in transcript abundance compared to control. Mean values are calculated from three biological replicates, error bars represent the standard error of mean (SEM)

In addition to different stresses, we also investigated the response of AP2 genes (RAP2.6, RAP2.6L, DREB19 and DREB26) in Arabidopsis upon exposure to stress hormones SA, JA, ABA and ET (Fig. 6, right panel). qRT-PCR was used to compare the transcript abundance of AP2 genes in plants exposed to a variety of hormones to that of mock treated control plants at 6 and 24 h after exposure. RAP2.6 transcripts were significantly increased in abundance at 6 h after exposure to both JA and SA, and although decreased at 24 h of exposure, they were still significantly high in JA treated tissue (10 times higher than control) (Fig. 6a). Transcript abundance of RAP2.6 was not altered at 6 h after exposure to ABA, however significantly increased at 24 h after exposure to ABA (Fig. 6a). RAP2.6 did exhibit alterations in transcript abundance in response to ET (Fig. 6a). The transcript abundance of RAP2.6 L was significantly high at 6 h after exposure to SA, JA, ABA and ET but decreased to normal level by 24 h post-exposure in SA, ABA, but in case of JA treatment, the transcript levels, although decreased but was still 10 times higher than control (Fig. 6b). In case of ET treated tissue, the transcript abundance of RAP2.6L increased from 6 to 24 h after exposure (Fig. 6b). DREB19 did not exhibit statistically significant (P < 0.05) alteration of transcript abundance in response to any of the tested hormones (Fig. 6c), while DREB26 showed a moderate increase in transcript abundance at 24 h after exposure to JA, and 6 h after exposure to SA (Fig. 7d). Thus RAP2.6 and RAP2.6L were most responsive to different stress hormones compared to DREB19 and DREB26 (Fig. 6).
https://static-content.springer.com/image/art%3A10.1007%2Fs11103-010-9711-7/MediaObjects/11103_2010_9711_Fig7_HTML.jpg
Fig. 7

Phenotype of representative of a adult plants, and b silliques of control (WT and VC) and AP2 TFs overexpressed genotypes

Overexpression of RAP2.6, RAP2.6L, DREB19 and DREB26

The AP2 TF genes (RAP2.6, RAP2.6L, DREB19 and DREB26) were characterized by separately overexpressing each gene under the control of CaMV35S promoter in A. thaliana (WS). In case of DREB26, we obtained 12 independent T0 transgenic lines, of which only 4 lines set seeds (T1) and very few at that. The presence of transgene was confirmed using RT-PCR in all transgenic lines (Fig. S2), and the DREB26 expression levels were measured in eight T0DREB26 overexpressed Arabidopsis plants using qRT-PCR. The transcript abundance of DREB26 was significantly higher (ranged from 20 to 120 folds higher) in transgenic plants compared to WT and VC (data not shown). DREB26 transgenic plants (T0) exhibited abnormal morphology with tiny leaves, few or no secondary branches and deformed flowers (Fig. 7a). In addition, the T1DREB26 plants died early in vegetative stage, and therefore, we were unable to characterize them any further.

In case of RAP2.6, RAP2.6L and DREB19, we selected three independent transgenic lines in each gene based on initial screening for phenotype and confirmed the presence of transgene using RT-PCR in 2 weeks old homozygous T2 plants (Fig. S2). In addition, the expression levels of RAP2.6, RAP2.6L and DREB19 genes were quantified using qRT-PCR in transgenic RAP2.6, RAP2.6L and DREB19Arabidopsis lines, respectively (Table 1). The expression of RAP2.6 was higher (ranged from 31,019.10 ± 14,694.86 to 78,003.17 ± 26,651.89 in three lines) in RAP2.6 overedxpressed lines (A2, A6 and A39) compared to controls (Table 1). Similarly, the transcript abundance of RAP2.6L was higher (ranged from 8,907.49 ± 512.38 to 16,631.41 ± 896.90 in three lines) in RAP2.6L overexpressed lines (C23, C28 and C31) compared to controls (Table 1). The expression levels of DREB19 were higher (ranged from 134.83 ± 24.28 to 967.47 ± 235.16 in three lines) in DREB19 overexpressed Arabidopsis lines (D1, D5 and D12) compared to controls (Table 1). These results (Table 1) demonstrate that AP2 genes are indeed getting overexpressed in respective transgenic plants and their transcript abundance are significantly higher compared to WT and VC.
Table 1

Transcript abundance of AP2 genes in AP2 TFs overexpressed, wild type (WT), vector control (VC) genotypes as detected by qRT-PCR

Gene

Genotypes

Line

Fold changea

Standard error

RAP2.6

RAP2.6-OX

A2

78,003.17

26,651.89

RAP2.6-OX

A6

31,019.10

14,694.86

RAP2.6-OX

A39

44,975.16

18,392.29

VC

1.14

0.30

WT

 

1

RAP2.6L

RAP2.6L-OX

C23

8,907.49

512.38

RAP2.6L-OX

C28

16,631.41

896.90

RAP2.6L-OX

C31

15,977.00

1,512.19

VC

1.55

0.79

WT

1

DREB19

DREB19-OX

D1

967.47

235.16

DREB19-OX

D5

383.61

94.71

DREB19-OX

D12

134.83

24.28

VC

1.19

0.57

WT

1

aExpression levels of AP2 genes in different genotypes were calculated relative to WT, and expression levels in WT was normalized to 1, and mean values are from three biological replicates

The differences in phenotype between WT and homozygous T2 transgenic (RAP2.6, RAP2.6L and DREB19) Arabidopsis plants was studied in greenhouse. Representative pictures of adult plants and siliques of control and RAP2.6, RAP2.6L and DREB19 transgenic Arabidopsis plants are shown in Fig. 7. RAP2.6L and DREB19 did not show any phenotypic differences compared to controls except for flowering time, while RAP2.6 showed observable phenotypic difference compared to controls (WT and VC) (Fig. 7; Table 2). RAP2.6 transgenic plants were dwarf with many secondary branches and shorter siliques, compared to their controls (Fig. 7). However, no differences were found between RAP2.6 transgenic plants and the controls in terms of germination, growth and morphology up to bolting stage (data not shown). Apart from this, significant differences were observed in flowering time between RAP2.6 transgenic lines and controls (Table 2). RAP2.6 transgenic lines (A2, A6 and A39) flowered 2–3 days earlier than the controls (Table 2). A significant difference in flowering time was also observed in RAP2.6L and DREB19 transgenic Arabidopsis lines compared to controls (WT and VC) (Table 2). RAP2.6L transgenic lines (C23, C28 and C31) flowered 3–4 days earlier and DREB19 transgenic lines (D1, D5 and D12) flowered nearly 3 days earlier than controls (Table 2). In summary, overexpression of RAP2.6, RAP2.6L, DREB19 and DREB26 altered the phenotypes in terms of growth and appearance and/or flowering time in Arabidopsis.
Table 2

Days required for floral initiation in wild type (WT), vectors control (VC) and AP2 TFs overexpressed Arabidopsis genotypes under normal conditions

Genotypes

Line

Number of observations

Average number of days required for floral initiationa

SEMb

RAP2.6-OX

A2

65

23.74*

0.32

A6

65

22.52*

0.23

A39

63

22.25*

0.17

RAP2.6L-OX

C23

63

22.57*

0.29

C28

65

21.94*

0.21

C31

65

22.69*

0.31

DREB19-OX

D1

65

23.38*

0.25

D5

65

23.12*

0.21

D12

65

22.92*

0.27

WT

85

26.00

0.20

VC

78

26.41

0.35

* Indicates a significant difference (P < 0.05) when compared to WT

aMean values from three biological replicates

bSEM standard error of mean

Evaluation of RAP2.6, RAP2.6L and DREB19 transgenic plants for abiotic stress tolerance

Transgenic RAP2.6 lines (A2, A6 and A39), RAP2.6L lines (C23, C28 and C31) and DREB19 lines (D1, D5 and D12) were evaluated for abiotic stress tolerance (Figs. 8, 9, 10). Salt tolerance in early vegetative stage was studied by plating transgenic and control (WT and VC) seeds on MS medium containing 0 mM NaCl, 125 mM or 150 mM NaCl (Fig. 8). Without stress, 3 week old RAP2.6, RAP2.6L and DREB19 Arabidopsis plants appeared developmentally advanced in terms of growth and floral bud initiation compared to WT and VC seedlings (Fig. 8a). All RAP2.6, RAP2.6L and DREB19 transgenic lines appeared developmentally more advanced than control (WT and VC) seedlings even at 125 mM NaCl stress (Fig. 8a). The AP2 transgenic plants had more shoot and root mass and also the true leaves appeared earlier in transgenic plants compared to control (Fig. 8a). In addition, at 150 mM NaCl stress, RAP2.6L transgenic lines were developmentally advanced compared to WT (Fig. 8a).
https://static-content.springer.com/image/art%3A10.1007%2Fs11103-010-9711-7/MediaObjects/11103_2010_9711_Fig8_HTML.gif
Fig. 8

a Photographic representation of representative AP2 TF overexpressing Arabidopsis plants along with wild type (WT) and vector control (VC) on MS medium containing NaCl (0 mM, 125 mM or 150 mM), and percent germination of AP2 TF overexpressing Arabidopsis lines and controls, on MS medium with b 125 mM NaCl, and c 150 mM NaCl. Asterisks indicate a significant difference (P < 0.05) when compared to WT. Mean values from three biological replicates are shown. Error bars are the standard error of mean (SEM) and N = 252

https://static-content.springer.com/image/art%3A10.1007%2Fs11103-010-9711-7/MediaObjects/11103_2010_9711_Fig9_HTML.gif
Fig. 9

a Performance of wild type (WT), vector control (VC) and AP2 TF transgenic Arabidopsis plants at 200 mM NaCl stress in greenhouse showing percent of plants that not survived, flowered and set pods. All the percentages were calculated based on the total number of initial plants. None of the dead plants had flowers or pods as they died before the onset of flowering. All the survived plants did not flower and some of the flowered plants did not set pods, b performance of WT, VC and AP2 TF transgenic plants under drought stress in terms of wilting and recovery, and c performance of WT, VC and AP2 TF transgenic plants under drought stress in terms of flowering and pod set. Asterisks indicate a significant difference (P < 0.05) when compared to WT. Mean values from three biological replicates are shown. Error bars are the standard error of mean (SEM) and N = 48

https://static-content.springer.com/image/art%3A10.1007%2Fs11103-010-9711-7/MediaObjects/11103_2010_9711_Fig10_HTML.jpg
Fig. 10

Photographic representation of WT, VC and AP2 TF overexpressing Arabidopsis plants a without stress, b at 200 mM NaCl stress, c drought stress, and d recovery 1 day after re-watering

Although there were no differences in germination rate between any of the transgenic lines and controls without stress (data not shown), significant differences were observed in percent germination of RAP2.6L (lines C23, C28 and C31) and RAP2.6 (lines A6 and A39) transgenic plants compared to WT, after 1 week on 125 mM NaCl medium (Fig. 8b). Percent germination was higher in RAP2.6L transgenic lines (ranged from 80 to 84% in three lines) and lower in RAP2.6 transgenic lines (ranged from 40 to 50% in three lines) compared to WT (65%) (Fig. 8b). The germination percent was significantly high in all three RAP2.6L transgenic lines (ranged from 88 to 95% in three lines) compared to WT (70%) even after 3 weeks on medium containing 125 mM NaCl (Fig. 8b). Percent germination was significantly low (44%) in one of the RAP2.6 transgenic lines (A6) compared to WT after 3 weeks on 125 mM NaCl medium (Fig. 8b). Other than this, no significant differences were observed between WT and VC or other tested AP2 transgenic lines for percent germination in 125 mM NaCl stress (Fig. 8b). Although, the percent germination was less in RAP2.6 transgenic lines in at least 2 lines (A6 and A39) at 125 mM NaCl stress, the seedlings appeared to be developmentally advanced in terms of shoot and root growth compared to controls (Fig. 8a). The percent germination was very less after 1 week of seeding on medium containing 150 mM NaCl in all the tested Arabidopsis lines (Fig. 8c). However, nearly 5% germination was observed in at least two RAP2.6L transgenic lines (C23 and C28) and the difference was significant when compared with WT (Fig. 8c). At 150 mM NaCl stress, 3 weeks after plating (Fig. 8c), RAP2.6L transgenic lines had significantly higher percent germination (ranged from 30 to 40% in three lines) compared to WT (15%). There were no significant differences in percent germination or differences in appearance between WT and VC or other transgenic lines (RAP2.6 and DREB19 lines) at 150 mM NaCl stress.

The AP2 transgenic lines (RAP2.6, RAP2.6L and DREB19) were also evaluated for salt stress tolerance in the greenhouse, by watering on alternate days with salt water (200 mM NaCl) from 8 days after seeding (DAS) to 25 DAS. Many seedlings died and growth was retarded in NaCl treated plants (Fig. 9a). However, there were no significant differences between any of the transgenic lines and WT in terms of percent death (Fig. 9a). Nevertheless, significantly higher percent flowering and percent pod set was observed in AP2 transgenic lines compared to WT under salt stress (Fig. 9a). RAP2.6L transgenic lines had the highest percent flowering (ranged from 80 to 90% in three lines) followed by DREB19 lines (ranged from 40 to 60% in three lines) and RAP2.6 lines (ranged from 30 to 50% in three lines). RAP2.6L transgenic lines had the highest percent pod set (ranged from 50 to 80% in three lines) followed by DREB19 transgenic lines (ranged from 25 to 37% in three lines) and RAP2.6 transgenic lines (ranged from 20 to 27% in three lines). In contrast, the control genotypes (WT and VC) set very few or no pods (Fig. 9a).

The transgenic plants (RAP2.6, RAP2.6L and DREB19) were also evaluated for drought tolerance by imposing drought stress in greenhouse. As shown in Fig. 9b, more than 85% of the plants wilted in WT, VC, RAP2.6 (lines D1, D5 and D12) and RAP2.6L (lines C23, C28 and C31) transgenic lines while only 60–65% wilted in the DREB19 transgenic lines (D1, D5 and D12). However, all the tested genotypes recovered (>85%) within a day when re-watered except for two of RAP2.6 transgenic lines (A2 and A39) which had significantly less percent recovery (Fig. 9b). There were significant differences in terms of percent flowering and percent pod set between AP2 transgenic lines and WT (Fig. 9c). More than 95% of the plants flowered and 70–90% of the plants set pods in AP2 transgenic lines (RAP2.6, RAP2.6L and DREB19 transgenic lines), whereas only 75–80% of the plants flowered and only 8% of the plants set pods in controls (WT and VC) following exposure to drought stress (Fig. 9c).

In absence of any stress, transgenic plants flowered earlier as previously observed (Table 2) and had a higher number of secondary branches compared to control plants (Fig. 10a). At 200 mM NaCl stress condition, RAP2.6L and DREB19 transgenic lines were taller and produced more flowers and pods compared to WT, VC and RAP2.6 plants (Fig. 10b). Similarly, RAP2.6L and DREB19 plants looked stronger and had higher number of flowers and pods than control plants (WT and VC) and RAP2.6 transgenic lines under drought stress (Fig. 10c–d). In summary, green house stress studies demonstrated the enhanced performance of RAP2.6L and DREB19 transgenic plants under salt and drought conditions compared to WT plants.

Discussion

Salinity and drought are the two major environmental constraints in crop production and more than 10% of the World’s arable land is affected by salinity and drought (Bray et al. 2000; Jenks et al. 2007). Since the completion of the Arabidopsis genome project and subsequent ongoing efforts in genomic research, many genes have been functionally characterized for stress tolerance. TFs represent most important molecular targets in genetic engineering of crop plants for stress tolerance (Nakashima and Yamaguchi-Shinozaki 2006; Khong et al. 2008). This is due to the fact that a single TF can regulate the expression of numerous genes including its own gene and activates the adaptation process of organism to changed environment (Khong et al. 2008). Some examples of application of TF in stress tolerance include, AtMYB44 and GhDREB which conferred enhanced abiotic stress tolerance in Arabidopsis and wheat when overexpressed (Jung et al. 2008; Gao et al. 2009). Similarly, a stress responsive TF gene SNAC1, when overexpressed, enhanced drought tolerance in rice (Hu et al. 2006). Furthermore, overexpression of AP2 TF genes OsDREB1F and HARDY enhanced multiple abiotic stress tolerance in both Arabidopsis and Rice (Karaba et al. 2007; Wang et al. 2008). These examples also illustrate that knowledge obtained from research on model plant Arabidopsis can be applied in improving crop plants. Nevertheless, the biological role of many Arabidopsis TF genes is yet to be explored and many of them may possibly be very useful in engineering crop plants for stress tolerance. In this study, we have made an attempt to investigate the biological role of two ERF (RAP2.6 and RAP2.6L) and two DREB (DREB19 and DREB26) subfamily AP2 TF genes. We chose to study these genes because of their increased transcript abundance in ABR17-overexpressed Arabidopsis compared to WT, under NaCl stress (Krishnaswamy et al. 2008). ABR17-transgenic Arabidopsis plants have demonstrated enhanced tolerance to salt and other abiotic stresses (Srivastava et al. 2006). It was speculated that the higher expression of RAP2.6, RAP2.6L, DREB19 and DREB26 genes in ABR17-transgenic plants under NaCl stress could be partially responsible for the observed salt tolerant phenotype (Srivastava et al. 2006; Krishnaswamy et al. 2008). In this study we have tested if higher expression of these AP2 genes (RAP2.6, RAP2.6L, DREB19 and DREB26) enhances salt and drought tolerance, by overexpressing them in Arabidopsis. Results from functional assay, expression analysis as well as overexpression studies of these AP2 genes are discussed below.

RAP2.6, RAP2.6L, DREB19 and DREB26 are transcription factors

The genes RAP2.6, RAP2.6L, DREB19 and DREB26 code for proteins with one AP2 DNA binding domain (Fig. 1) and sequence analysis suggests that the proteins do not contain EAR motif seen in AP2 transcriptional repressors (Stockinger et al. 1997; Fujimoto et al. 2000; Ohta et al. 2001; Dong and Liu 2010). They are therefore expected to localize in nucleus and act as transcriptional activators. Recently, nuclear localization and transcriptional activity has been demonstrated for RAP2.6 and RAP2.6L (Che et al. 2006; Sun et al. 2010; Zhu et al. 2010). However, the subcellular localization and function of DREB19 and DREB26 proteins are not known. In order to study the subcellular localization of DREB19 and DREB26, they were expressed as GFP fusion proteins in Arabidopsis (Fig. 2 and Fig. S1) and our results demonstrate that DREB19 and DREB26 proteins localize in nucleus (Fig. 2 and Fig. S1) and therefore these proteins might act as transcription factors. However, all nuclear localizing proteins are not transcription factors, therefore we carried out transactivation assay using yeast one hybrid assay with HIS3 reporter gene (Fig. 3) to investigate the role of DREB19 and DREB26 in transcriptional regulation. Our results indicate that DREB19 and DREB26 are indeed transactivators (Fig. 3). We also verified transcriptional activation of RAP2.6 and RAP2.6L using HIS3 reporter gene and the results were consistent with the recent reports (Sun et al. 2010; Zhu et al. 2010). Our study (Figs. 2, 3) and previous studies indicate that the putative AP2 like proteins RAP2.6, RAP2.6L, DREB19 and DREB26 act as TFs.

RAP2.6, RAP2.6L, DREB19 and DREB26 might be important in early vegetative as well as reproductive stages of plants growth

Studying spatial/temporal as well as tissue specific expression pattern of any gene would give information on importance of that gene in different growth phases, growth transitions as well as tissue/organ development. We carried out spatial/temporal expression studies of RAP2.6, RAP2.6L, DREB19 and DREB26 genes by quantifying their transcript abundance using qRT-PCR in WT Arabidopsis plants (Fig. 4). We observed both similar and divergent expression pattern among four genes in different stages of plant growth (Fig. 4). For instance, transcripts of all four genes were enriched in seedlings compared to the rosette leaves stage (Fig. 4). RAP2.6 was most abundant in seedlings compared to any other tissue assayed (Fig. 4). Consistent with this, RAP2.6 expression has been reported to be high in stem compared to flowers (Zhu et al. 2010). Unlike RAP2.6, transcript abundance of RAP2.6L, DREB19 and DREB26 transcripts were most abundant in inflorescence compared to any other tissue assayed (Fig. 4). The transcript abundance of all the four genes increased from floral bud initiation stage to the inflorescence stage indicating their importance in flower development (Fig. 5) and again, there was no expression of these four AP2 genes in fully matured silliques (data not shown). These results indicate that all the four genes, especially RAP2.6, might be very important in early vegetative stage. In addition, all of them might be important in transition from vegetative stage to reproductive stage as well as in flower development than in silique maturation. In fact, RAP2.6L has been implicated in shoot regeneration since RAP2.6L knockdown mutants reduced the efficiency of shoot formation in tissue culture of roots (Che et al. 2006).

Tissue specific expression pattern of RAP2.6, RAP2.6L, DREB19 and DREB26 was studied indirectly by detecting GUS gene expression in Arabidopsis plants containing promoter-GUS fusion system for these genes (Fig. 5). In case of germinated seeds, GUS expression was observed only for DREB19 in cotyledonary leaves. Consistent with qRT-PCR expression analysis, GUS expression was also observed in 7-day-old seedlings with RAP2.6, DREB19 and DREB26 promoters and 14 days old seedlings with DREB19 and DREB26 promoters once again supporting the importance of these genes in early vegetative stage (Fig. 5). DREB19 expression was confined to only region where leaves emerge from the stem and also in xylem tissues in roots, while DREB26 expression was detected in cotyledonary leaves and true leaves, and RAP2.6 expression was seen only in roots (Fig. 5). DREB19 might be involved in leaf emergence as well as in regulation of genes involved in nutrient/water uptake by xylem tissue and DREB26 might be involved in leaf and plant development. Although, qRT-PCR showed enriched expression of all the four genes in seedlings, GUS expression was not detected in 7 and 14 days old plants with RAP2.6L promoter as well as in 14 days old plants with RAP2.6 promoter (Fig. 5). This could be because the elements needed for the expression of gene in such stages might be absent within the cloned promoter region. However, abundant GUS expression has been observed in seedlings when more than 1 kb RAP2.6L promoter was cloned (Che et al. 2006). Promoter-GUS fusion studies together with qRT-PCR studies suggest the importance of these four AP2 genes in early vegetative stages. Furthermore, consistent with our qRT-PCR results, GUS expression was observed in flowers with promoters of all the four genes although each of the AP2 genes tested demonstrated unique expression pattern within flower (Fig. 5). For instance, RAP2.6 was detected in petals and carpels, while RAP2.6L was detected in pollen grains, whereas DREB26 was detected in ovules, and DREB19 was detected in stigmatic surface (Fig. 5). Although, there was no expression of these genes in mature silliques as detected by qRT-PCR, GUS expression was detected in developing young slilliques with promoters of RAP2.6 and DREB26 (Fig. 5). GUS expression was detected in valves of the siliques with RAP2.6 promoter while it was detected in early seeds with DREB26 promoter. These results suggest that they may have very specific roles in flower and silique development. RAP2.6 might be important in sepal, carpel and overall silique development, while RAP2.6L might be important for pollen grain development and function. Similarly, DREB26 might have essential role during seed development. In fact, genes from AP2 TF family are known for their key role in floral morphogenesis and seed development (Kunst et al. 1989; Jofuku et al. 1994; Klucher et al. 1996).

RAP2.6, RAP2.6L, DREB19 and DREB26 are involved in plant defense response

Expression of RAP2.6, RAP2.6L, DREB19 and DREB26 in response to different abiotic stresses and stress hormones were measured in order to find the involvement of these genes in plant stress signaling. Results indicate that RAP2.6 and RAP2.6L are responsive to both abiotic stresses and hormones JA, SA, ABA and ET (Fig. 6). Phytohormone ABA is involved in abiotic stress signaling whereas hormones JA, SA and ET are part of biotic stress response (Fujita et al. 2006). This suggests the participation of ERF subfamily genes, RAP2.6 and RAP2.6L in both biotic and abiotic stress signaling. Indeed, RAP2.6 has been associated with signal transduction during infection of Arabidopsis with Pseudomonassyringae (He et al. 2004) and role of RAP2.6L in bacterial resistance has been demonstrated by mutating RAP2.6L in Arabidopsis (Sun et al. 2010). It has been shown that among different TF families, ERF family is most responsive to JA and Alternaria brassicola (McGrath et al. 2005). The gene ERF1, a member from the ERF subfamily has been suggested to integrate JA and ET signaling pathways in Arabidopsis and has also been demonstrated to confer resistance to fungal pathogen when overexpressed (Berrocal-Lobo et al. 2002; Lorenzo et al. 2003). In case of DREB genes, DREB19 was not responsive to stress hormones while it was found to be most responsive to salt, heat and drought (Fig. 6). Salt and drought responsive genes DREB2A and DREB2B are members of group A-2 of DREB subfamily to which the gene DREB19 belongs (Sakuma et al. 2006b). DREB2A and DREB2B are also reported to be highly responsive to salt, heat and drought, and less responsive to phytohormones like ABA, JA and SA (Liu et al. 1998; Sakuma et al. 2006b). Transcripts abundance of DREB26 moderately changed on exposure to JA and SA and did not alter in response to abotic stresses (Fig. 6). It appears that DREB19 is more involved in abiotic stress compared to DREB26. The different responses of AP2 genes to different stress and stress hormones suggest that they have very specific physiological roles.

Ovexpression of RAP2.6, RAP2.6L, DREB19 and DREB26 alters phenotype in terms of plant development and/or flowering time

RAP2.6, RAP2.6L, DREB19 and DREB26 were overexpressed in Arabidopsis using CaMV35S promoter in order to investigate their roles in plant growth and development as well as in abiotic stress tolerance. Our overexpression studies indicate that all the four genes tested might have very essential roles in plant growth and development as overexpression lead to the altered phenotype with respect to growth or/time of flowering (Fig. 7 and Table 2). Transgenic T0DREB26 lines had altered/deformed phenotype in Arabidopsis; over expressed lines being abnormal, dwarf with thin stem, very few leaves and less/no secondary branches (Fig. 7). Only few lines set seeds, and the seedlings from those seeds died during germination. Our expression studies (Figs. 4, 5) have demonstrated that DREB26 is expressed in cotyledonary leaves, true leaves in seedling stage as well as in flowers and developing seeds (Figs. 4, 5). Although, the expression of DREB26 appears to be important in these stages, a balanced expression might be very essential for the appropriate plant development as overexpression leads to deformed plants with no/leaves, deformed flowers and poor pod set. In addition, DREB26 was less responsive to stress and stress hormones (Fig. 6) which suggests that DREB26 might have major role in growth and development, rather than in defense response. Indeed, our qRT-PCR expression studies also indicated less/no response to abiotic stresses and stress related hormones (Fig. 6). Dwarf phenotype has also been previously reported in AP2 TF overexpressing transgenic Arabidopsis plants (Magome et al. 2004; Tong et al. 2009). For instance, molecular analyses of gibberellin deficient mutant dwarf and delayed flowering 1 (ddf1) revealed increased expression of putative AP2 TF (Magome et al. 2004). Furthermore, overexpression of AP2 TF gene DDF2 that is closely related to DDF1 resulted in dwarf phenotype in Arabidopsis (Magome et al. 2004). Overexpression of chrysanthemum DREB1B in Arabidopsis resulted in expression of a GA deactivation enzyme GA2ox7 and dwarfism (Tong et al. 2009). Whether dwarfism and poor growth of DREB26 transgenic plants is because of altered GA biosynthesis is not clear since overexpressed DREB26 plants did not survive and thus we were unable to perform further studies. Expression of DREB26 under the control of stress inducible promoter and also loss-of-function analysis might shed more light on the importance of DREB26 in plant development and stress response.

In case of other AP2 TF (RAP2.6, RAP2.6L and DREB19) overexpressed plants, we characterized three independent transgenic lines in each of RAP2.6, RAP2.6L and DREB19 overexpressed plants. We observed comparatively altered phenotype in RAP2.6 transgenic lines, being dwarf and having many secondary branches compared to controls with small siliques (Fig. 7). In addition, pod size was comparatively smaller than WT (Fig. 7). There were no obvious differences between WT and RAP2.6 transgenic Arabidopsis plants till the secondary branches started to emerge. Once the secondary branching starts, RAP2.6 transgenic plants developed more secondary branches and became dwarf (Figs. 7, 10a). Our expression analysis (Figs. 4, 5) studies with WT Arabidopsis had suggested the importance of RAP2.6 in vegetative stage and silique development. However, higher expression of RAP2.6 appears to inhibit apical dominance and promote lateral branching and inhibit silique development. Our qRT-PCR expression analysis showed more than 1,000 fold higher expression of RAP2.6 in RAP2.6 transgenic lines (A2, A6 and A39) compared to WT (Table 1). In addition to the altered morphology, RAP2.6 overexpressing lines flowered earlier than WT (Table 2). Indeed, early flowering phenotype was also observed in RAP2.6L and DREB26 transgenic Arabidopsis lines (Table 2). Although few members of AP2 family genes have been reported to be involved in regulation of flowering time, they are known to regulate negatively. For instance, it has been demonstrated that AP2 genes are targets of miR172, and overexpression of miR172 down regulates AP2 genes (AP2, TOE1 and TOE2) and promotes early flowering (Aukerman and Sakai 2003). In addition, delayed flowering has been observed in overexpressed TOE1Arabidopsis and early flowering has been observed in ap2 mutants suggesting the function of TOE1 and AP2 as floral repressors (Aukerman and Sakai 2003; Ohto et al. 2005). Furthermore, overexpression of another AP2 family gene CBF2 results in delayed bolting and flowering in Arabidopsis (Schwager et al. 2010). Enriched transcript abundance of RAP2.6, RAP2.6L and DREB19 in early floral buds and inflorescence as well as GUS expression pattern driven by their promoters (Figs. 4, 5) suggest a role for these AP2 genes in flower development. However, the early flowering phenotype of RAP2.6, RAP2.6L and DREB19 overexpressed plants could be an indirect effect of upregulation of stress related genes which are likely to promote flowering similar to the one observed in stressed plants. The AP2 TF genes RAP2.6, RAP2.6L and DREB19 belong to ERF and DREB subfamilies whose members are known to bind defined cis-elements present in the promoters of pathogenesis related proteins, low temperature and water deficit responsive genes to regulate their expression (Stockinger et al. 1997; Gilmour et al. 1998; Guo et al. 2005). Furthermore, early flowering phenotype has been observed in plants overexpressing stress related genes. For example, overexpression of stress related gene phosphatidylinositol-phospholipase C2 in canola promotes early flowering (Fawzy et al. 2009). Indeed, early flowering phenotype has been observed in both ABR17- transgenic Arabidopsis and Brassica compared to WT under normal conditions in addition to enhanced stress tolerance (Srivastava et al. 2006; Dunfield et al. 2007). However, studying RNAi or T-DNA insertion lines would confirm the role of RAP2.6, RAP2.6L and DREB19 in flowering time.

Overexpression of RAP2.6L and DREB19 enhance salt and drought tolerance

RAP2.6, RAP2.6L and DREB19 transgenic Arabidopsis plants were evaluated under abiotic stresses to investigate the importance and utility of these genes in abiotic stress tolerance. The RAP2.6L transgenic lines performed better than any transgenic and WT genotypes under NaCl stress in Petri plate experiments. They germinated earlier and had high seedling vigor with enhanced rooting compared to WT (Fig. 8). In addition, the RAP2.6L transgenic lines performed better than WT by exhibiting increased percent flowering and percent pod set under NaCl stress in green house conditions (Figs. 9, 10). Although there were no differences with respect to wilting and recovery, RAP2.6L transgenic lines had higher percent flowering and pod set compared to WT, even under drought stress (Figs. 9, 10). Therefore, these results suggest that RAP2.6L might have a major role in salt tolerance although it appears to participate also in drought tolerance. In addition, significantly higher expression of RAP2.6L in response to salt and drought stress (Fig. 6) in the present study as well as its upregulation in our previous salt microarray studies (Krishnaswamy et al. 2008) further supports the role of RAP2.6L in salt and drought stress.

Similar to RAP2.6L transgenic plant, DREB19 overexpressing lines exhibited high seedling vigor compared to WT at 125 mM NaCl stress although there was no difference in germination rate (Fig. 8), and also had high percent flowering and pod set compared to WT under NaCl stress in greenhouse conditions (Fig. 9). Furthermore, under drought stress, DREB19 transgenic lines performed better than WT or any other transgenic lines tested with less percent wilting in addition to high percent flowering and pod set (Figs. 9, 10). These results suggest that although DREB19 is involved in salt tolerance, it appears to be more important in drought tolerance. Detection of DREB19 promoter driven GUS expression in xylem tissues of roots (Fig. 5) and also the significantly increased expression of DREB19 in response to drought and salt stress (Fig. 6) suggests a role for it in drought and salt stress (Fig. 6). In fact, among four AP2 studied, DREB19 was the most responsive to drought stress (Fig. 6). Furthermore, other genes (DREB2A and DREB2B) from the same A-2 group of DREB subfamily have been demonstrated to impart drought and salt tolerance (Sakuma et al. 2006a, b) suggesting that DREB19 might be one of the important genes involved in drought signaling.

Unlike RAP2.6L and DREB19 transgenic plants, the performance of RAP2.6 transgenic plants was comparable to that of WT under salt stress, although RAP2.6 and RAP2.6L belong to the same group B-4 of ERF subfamily. In fact, two of the RAP2.6 lines (A6 and A39) exhibited reduced germination compared to WT when grown on medium containing 125 mM NaCl, although the seedling vigor was higher than WT (Fig. 8). However, they had significantly higher percent flowering and pod set compared to WT under salt stress in green house conditions, where stress was induced after germination (Figs. 9, 10). Our observations suggest that overexpression of RAP2.6 affect germination under salt stress but not seedling growth once germinated. RAP2.6 transgenic plants did not perform better either under drought stress as at least two of the RAP2.6 transgenic lines (A2 and A39) had less recovery than WT following drought stress although they had higher percent flowering and pod set (Figs. 9, 10). The differences in performance between three independent transgenic RAP2.6 lines under salt and drought stress might be due to the position effect as it could not be correlated with the expression levels of RAP2.6 (Table 1). Although, expression analysis of RAP2.6 in response to stress and stress hormones (Fig. 6) as well as previous studies (Fowler and Thomashow 2002; Krishnaswamy et al. 2008) suggest a role of RAP2.6 in plant stress signaling, overexpression of RAP2.6 was not helpful in getting stress tolerant phenotype. However, very high expression levels of RAP2.6 were observed in CaMV35S-RAP2.6 overexpressed lines (Table 1) which may not be ideal for the plants as they also show negative effect on plant growth under normal conditions (Fig. 7). A combination of stress inducible promoter and RAP2.6 would give better stress tolerance with no/less negative effect on phenotype as previously been reported in another AP2 family gene DREB1A (Kasuga et al. 1999, 2004). Expression of DREB1A with sress inducible promoter rd29A gave rise to greater tolerance to stress conditions with minimal effect on plant growth than with CaMV35S promoter (Kasuga et al. 1999, 2004).

Zhu et al. (2010) have reported that RAP2.6 overexpressed lines are hypersensitive to NaCl and ABA compared to WT. However, we did not observe sensitivity of RAP2.6 in any of the three independent RAP2.6 transgenic lines to NaCl although we did see low germination percentage in two lines (A6 and A39) at 125 mM NaCl (Fig. 8). Furthermore, we also did not observe hypersensitivity of RAP2.6 overexpressing lines to ABA (Fig. S3). In addition, Zhu et al. (2010) have not reported any phenotype differences between overexpressed RAP2.6 lines compared to WT under normal conditions, which were very much evident in our study (Fig. 7). These differences could be due to positional effect or due to differences in ecotype and sequence. In the present study, three independent RAP2.6 transgenic lines have been used compared to use of only one or two independent transgenic lines in the aforementioned study (Zhu et al. 2010). We have isolated RAP2.6 from ecotype WS which has one amino acid difference (W20R) from the reported sequence of ecotype Columbia (Zhu et al. 2010), and we have overexpressed in WS background unlike the Columbia ecotype used in their study (Zhu et al. 2010). These differences could also be due to posttranscriptional modifications that convert inactive form to active form like it was observed in another DREB family gene DREB2A (Sakuma et al. 2006a).

In summary, the results from our study suggest that: (1) DREB19 and DREB26 localize in nucleus and act as transcription activators similar to RAP2.6 and RAP2.6L, (2) these AP2 genes have divergent physiological roles as they have different expression pattern and invoke varied responses when subjected to abiotic stresses and stress hormones, (3) they play very important role both in plant development and stress responses since overexpression leads to altered phenotypes and altered responses to abiotic stresses, and (4) increased transcript abundance of RAP2.6L and DREB19 enhance abiotic stress tolerance as speculated based on our previous salt microarray study of ABR17-transgenic plants (Srivastava et al. 2006; Krishnaswamy et al. 2008). Early germination, high seedling vigour, early flowering and maturity traits observed in RAP2.6L and DREB19 transgenic plants under salt and/or drought stresses are the characters of stress tolerant plants as they contribute to escape or avoidance of stress conditions (Munns et al. 2000; Price et al. 2002). Similarly, better root growth observed in these transgenic plants may help in sequestration of toxic ions and enhance tolerance to salt. For example, salt tolerance in barley has been linked with early flowering, fast development and better root growth (Munns et al. 2000). In addition, direct positive yield component parameters like higher germination rate, flowering and pod set and better development that were observed in stressed RAP2.6L and DREB19 overexpressed plants (Figs. 8, 9, 10) are the traits that are considered while engineering seed plants for abiotic stress tolerance (Basra et al. 2003; Munns et al. 2006; Zadeh and Naeini 2007; Blum 2009). Therefore, future studies on overexpressing RAP2.6 and DREB19 in crop plants for developing salt and drought tolerant plants could be a worthwhile endeavor.

Acknowledgments

Financial support from NSERC (Natural Sciences and Engineering Research Council) of Canada and the University of Alberta is gratefully acknowledged. We thank Shiv Ganesh for help in statistical analyses and Dr. Elizabeth-France Marillia, Lipid Biotechnology, PBI/NRC, Saskatoon, Canada, for providing vector pBI 121.

Supplementary material

11103_2010_9711_MOESM1_ESM.pdf (159 kb)
Supplementary material 1 (PDF 158 kb)

Copyright information

© Springer Science+Business Media B.V. 2010