Regulation of drought tolerance by gene manipulation of 9-cis-epoxycarotenoid dioxygenase, a key enzyme in abscisic acid biosynthesis in Arabidopsis


For correspondence (fax +81 298 36 9060; e-mail


Abscisic acid (ABA), a plant hormone, is involved in responses to environmental stresses such as drought and high salinity, and is required for stress tolerance. ABA is synthesized de novo in response to dehydration. 9-cis-epoxycarotenoid dioxygenase (NCED) is thought to be a key enzyme in ABA biosynthesis. Here we demonstrate that the expression of an NCED gene of Arabidopsis, AtNCED3, is induced by drought stress and controls the level of endogenous ABA under drought-stressed conditions. Overexpression of AtNCED3 in transgenic Arabidopsis caused an increase in endogenous ABA level, and promoted transcription of drought- and ABA-inducible genes. Plants overexpressing AtNCED3 showed a reduction in transpiration rate from leaves and an improvement in drought tolerance. By contrast, antisense suppression and disruption of AtNCED3 gave a drought-sensitive phenotype. These results indicate that the expression of AtNCED3 plays a key role in ABA biosynthesis under drought-stressed conditions in Arabidopsis. We improved drought tolerance by gene manipulation of AtNCED3 causing the accumulation of endogenous ABA.


Drought is one of the major environmental stresses that limit the growth of plants and the production of crops. Plants respond to dehydration at the cellular and molecular levels (Bray, 1997; Ingram and Bartels, 1996; Shinozaki and Yamaguchi-Shinozaki, 1997; Shinozaki and Yamaguchi-Shinozaki, 1999). Drought stress induces the expression of various genes that are involved in stress tolerance and response. Experiments using abscisic acid (ABA) mutants in Arabidopsis show that the drought signal is mediated through both ABA-dependent and ABA-independent pathways to regulate the expression of various drought-inducible genes (Shinozaki and Yamaguchi-Shinozaki, 2000). The transcription factors DREB1/CBF and DREB2 specifically interact with a cis-acting element, DRE/CRT (dehydration-responsive element/C-repeat), and induce the expression of stress tolerance genes (Liu et al., 1998; Stockinger et al., 1997). Overexpression of DREB1A cDNA activated the expression of stress tolerance genes involved in an ABA-independent pathway, and improved tolerance to drought, salt and freezing stresses (Jaglo-Ottosen et al., 1998; Kasuga et al., 1999; Liu et al., 1998). However, it was not reported whether activation of the expression of stress tolerance genes involved in ABA-dependent pathways confers tolerance to drought stress.

ABA is a plant hormone that is involved in stress responses, and is quickly accumulated by many plant species when exposed to drought stress. Application of ABA promotes stomatal closure (Leung and Giraudat, 1998) and induces the expression of stress-related genes such as rab18, kin1 and rd29B (Kurkela and Franck, 1990; Lang and Palva, 1992; Yamaguchi-Shinozaki and Shinozaki, 1993).

It has been proposed that ABA is synthesized from carotenoids (C40) in plants (Zeevaart and Creelman, 1988). Three genes that participate in ABA biosynthesis have been isolated. They encode zeaXanthin epoxidase (ZEP; Marin et al., 1996); 9-cis-epoxycarotenoid dioxygenase (NCED; Schwartz et al., 1997); and abscisic aldehyde oxidase (AAO; Seo et al., 2000). ZEP catalyses the epoxidation of zeaXanthin to produce epoxycarotenoid; NCED catalyses the cleavage reaction of epoxycarotenoids to produce xanthoxin (the first C15 intermediate); and AAO catalyses the final step of ABA biosynthesis, which converts ABA aldehyde to ABA. We previously reported that the cowpea NCED gene VuNCED1 is strongly induced by drought stress in leaves, but the cowpea ZEP gene VuABA1 is not (Iuchi et al., 2000). The induction of VuNCED1 occurs slightly earlier than that of ABA accumulation by drought stress (Iuchi et al., 2000). The expression of several NCED genes is also induced by drought stress in maize, bean and tomato (Burbidge et al., 1997; Qin and Zeevaart, 1999; Schwartz et al., 1997). Recently, the transcript of AAO was shown to accumulate in response to drought stress, but the product of AAO was not (Seo et al., 2000). Biochemical studies indicate that a key step in ABA biosynthesis is the cleavage of 9-cis-epoxycarotenoid (Kende and Zeevaart, 1997; Sindhu and Walton, 1988). Moreover, ectopic expression of a tomato NCED gene causes overproduction of ABA in tomato and tobacco (Thompson et al., 2000), which also suggests a key regulatory role of NCED in ABA biosynthesis.

At least seven genes are homologous to NCED in the Arabidopsis genome, but little is known about their roles in the accumulation of ABA in various processes during plant growth and stress responses. Moreover, the drought tolerance of transgenic plants that overexpress an NCED gene has not been evaluated. We report here that AtNCED3, an Arabidopsis NCED gene, is induced by drought stress and is responsible for the production of ABA under drought stress. Overexpression of the AtNCED3 cDNA in Arabidopsis enhanced the expression of drought-inducible genes and decreased leaf transpiration through the accumulation of ABA. These AtNCED-overexpressing transgenic plants showed improved drought tolerance. This study provides a novel method for the protection of crops from drought stress.


Isolation and characterization of NCED genes from Arabidopsis

To isolate NCED genes from Arabidopsis, we screened an Arabidopsis cDNA library using cowpea VuNCED1 cDNA as a probe and isolated an Arabidopsis cDNA for NCED, named AtNCED3 (Figure 1a). We found seven independent sequences that are highly homologous to that of NCED in the public DNA database (Figure 1b). AtNCED3 is identical to one of these sequences (EMBL accession number AB028617). To compare the expression of the seven genes by drought stress, we amplified the six remaining genes from Arabidopsis genomic DNA or from an Arabidopsis cDNA library by PCR, cloned them, and designated them AtNCED1, 2, 4, 5, 6 and 9.

Figure 1.

Sequence analysis of AtNCED genes.

(a) Alignment of deduced amino acid sequences for AtNCED3 and VuNCED1. Positions with a consensus residue are boxed.

(b) Phylogenetic tree of AtNCED genes and VuNCED1. Analysis performed using genetyx-mac.

Northern blot analysis revealed that AtNCED3 was induced by drought stress, but AtNCED1, 2, 4, 5, 6 and 9 were not (Figure 2). This result suggests that AtNCED3 is responsible for ABA biosynthesis under drought stress. To examine the enzyme activity, the AtNCED genes (except AtNCED7) were expressed in Escherichia coli to obtain glutathione S-transferase (GST)-fused recombinant proteins, and the recombinant proteins were incubated with various carotenoids. Analysis of the reaction mixture by HPLC indicated that AtNCED3 protein catalyses the cleavage reaction of two 9-cis-epoxycarotenoids, 9-cis-violaXanthin and 9′-cis-neoXanthin (Figure 3a). Recom binant proteins of AtNCED2, 6 and 9 also catalysed the same reaction (data not shown).

Figure 2.

Northern blot analysis of the induction of AtNCED genes by drought stress treatment (dry) and water as a control (water).

Wild-type (Columbia) Arabidopsis plants were grown on agar plates for 3 weeks, then dehydrated on filter paper. Total RNA was isolated from the plants at the indicated hours after treatment. Each lane was loaded with 10 µg total RNA. The RNA was fractionated on a 1% agarose gel, blotted onto a nylon membrane, and probed with [32P]-labelled full-length cDNA of AtNCED.

Figure 3.

Enzyme activity of AtNCED3 recombinant protein.

(a) HPLC profiles of the reaction products from 9-cis-neoxanthin incubated with GST or GST-AtNCED3 fusion protein. Putative C25 product was eluted at the retention time of 6.7 min. Incubation of 9-cis-violaxanthin with GST-AtNCED3 also gave similar chromatogram.

(b) Full-scan mass spectrum of predicted cis-xanthoxin produced from 9-cis-neoxanthin. A similar spectrum was obtained in the GC–MS analysis of predicted cis-xanthoxin produced from 9-cis-violaxanthin.

The enzyme activity of the GST-AtNCED3 fusion protein was further confirmed by the identification of cis-xanthoxin by GC–MS as a product from 9-cis-epoxycarotenoids (Figure 3b). These results confirm that AtNCED3 encodes an active NCED enzyme.

Creation of transgenic lines that express sense or antisense AtNCED3

To further analyse the role of AtNCED3 in ABA accumulation under drought stress, we generated transgenic Arabidopsis plants that overexpressed sense (S) or antisense (AS) AtNCED3 transcripts. We selected two lines each from sense (S1, S2) and antisense (AS1, AS2) transgenic plants for further analyses. Strong expression of AtNCED3 was detected in sense transgenic lines under both stressed and non-stressed conditions, whereas expression of AtNCED3 in antisense plants was not detected, even under drought-stressed conditions (Figure 4).

Figure 4.

Characterization of transgenic plants and T-DNA insertional mutant.

(a) Expression of AtNCED3, rab18, kin1 and rd29B in sense (S) and antisense (AS) transgenic plants and wild-type (WT) plants. Each lane was loaded with 10 µg total RNA prepared from 3-week-old transgenic plants that had been dehydrated for 0 or 10 h. The membranes were probed with [32P]-labelled AtNCED3 antisense RNA fragments of the 3′ region of AtNCED3 cDNA. DNA fragments of the full-length kin1 and rab18 and the 3′-terminal-specific DNA fragment of rd29B were used as probes. EtBr, ethidium bromide-stained gel profiles.

(b) Expression of AtNCED3 in the plants of wild-type, transgenic and T-DNA insertional mutant (T-5004). Plants were grown on soil for 3 weeks, then 14-day drought treatment was applied. Each lane was loaded with 10 µg total RNA prepared from the plants, and the membrane was probed with a DIG-labelled RNA probe of the 3′ region of AtNCED3.

ABA level and drought tolerance of AtNCED3 transgenic plants and the T-DNA-tagged mutant of AtNCED3

We measured ABA levels of transgenic plants with sense or antisense AtNCED3 cDNA under normal growth conditions, and found an increase in ABA level in the sense transgenic plants (Figure 5a). We isolated a T-DNA insertional mutant line for AtNCED3 (named as T5004) by PCR screening. Molecular characterization of this mutant line revealed that the T-DNA was inserted between the ATG codon and putative TATA box of AtNCED3 (−78 bp), and the expression level of AtNCED3 was much lower than the wild type, even under water-stressed conditions (Figure 4b). The ABA levels of antisense plants and T-5004 were lower than those of wild-type plants under normal growth conditions (Figure 5a). These ABA levels were consistent with the level of expression of AtNCED3 mRNA.

Figure 5.

ABA levels (a) and transpiration rates (b) of 3-week-old transgenic and wild-type plants under normal growth conditions. Error bars represent standard errors.

To examine whether altered expression of AtNCED3 affected drought tolerance in transgenics, the AtNCED3 transgenic plants were grown for 3 weeks under normal conditions, then exposed to drought stress by stopping the water supply. After 14 days following the termination of water supply, the AtNCED3 antisense plants (AS1, AS2) and the T-DNA insertion mutant (T5004) appeared to be more sensitive to drought stress than wild-type plants (Figure 6a). When plants were rehydrated after 14 days’ drought treatment, all sense and wild-type plants recovered, but none of the antisense plants or T-DNA-tagged mutants did so (data not shown). After 18 days’ drought treatment, the AtNCED3 sense plants (S1, S2) clearly showed stronger tolerance to drought stress than wild-type plants (Figure 6b). Thus a correlation occurred between drought tolerance and the levels of expression of AtNCED3 and endogenous ABA.

Figure 6.

Drought tolerance of transgenic and wild-type plants.

Separate samples of the 3-week-old plants grown on soil were exposed to drought stress. Drought stress was imposed by withholding water for 14 (a) or 18 (b) days.

Transpiration rate of AtNCED3 transgenic plants

ABA induces stomatal closure to reduce water loss by transpiration. To elucidate the effect of altered expression of AtNCED3 on the stomatal aperture, we measured transpiration from leaves in AtNCED3 transgenic plants, T-DNA-tagged mutants, and wild-type plants that had been grown under normal conditions (Figure 5b). The transpiration rates of the antisense transgenics and the T-DNA-tagged mutants were significantly higher than for wild-type plants. By contrast, the transpiration rate of the sense plants was lower than for wild-type plants.

Expression of drought-inducible genes in AtNCED3 transgenic plants

To analyse whether altered expression of AtNCED3 causes altered expression of drought-inducible genes, we examined the expression of three drought-inducible genes, rab18, kin1 and rd29B, in AtNCED3 transgenic plants (Figure 4a). kin1 and rab18 contain both DRE- and ABA-responsive elements (ABRE) as cis-acting elements in their promoter sequences, and their expression is controlled by both ABA-dependent and ABA-independent pathways. In contrast, rd29B has only ABRE in the promoter and is controlled by ABA under drought-stressed conditions (Yamaguchi-Shinozaki and Shinozaki, 1994). The expression of rab18 and kin1 was enhanced in the unstressed sense transgenic plants. The induction of rab18, kin1 and rd29B by drought stress in the sense plants was more enhanced than in wild-type plants (Figure 4a). By contrast, in the antisense plants the expression of rd29B was completely repressed under drought-stressed conditions, whereas the expression of rab18 and kin1 was not.


The 9-cis-epoxycarotenoid cleavage reaction is thought to be a rate-limiting step in ABA biosynthesis. This reaction is catalysed by NCED. To examine the role of NCED in ABA biosynthesis under drought-stressed condition, we isolated seven NCED genes (AtNCED) from the Arabidopsis genome and identified a drought-inducible NCED gene, AtNCED3. The expression of AtNCED3 was strongly induced by drought stress (Figure 2). The expression of AtNCED9 was weakly induced by drought stress, but the other AtNCED genes were not. Recombinant AtNCED3 protein catalysed the conversion of 9-cis-epoxycarotenoids to xanthoxin. Thus AtNCED3 should be the major NCED gene in Arabidopsis involved in the regulation of ABA levels under drought stress. It has been reported that the expression of AtNCED1 was induced by drought stress in Arabidopsis thaliana ecotype Landsberg (Neill et al., 1998). However, in this study AtNCED1 was not significantly induced in A. thaliana ecotype Columbia. This discrepancy may be due to the differences in ecotypes and stress treatments.

Drought-inducible NCED genes have been also reported in maize (Schwartz et al., 1997), tomato (Burbidge et al., 1997), bean (Qin and Zeevaart, 1999), cowpea (Iuchi et al., 2000) and avocado (Chernys and Zeevaart, 2000). As NCED is thought to function in the rate-limiting step of ABA biosynthesis, drought-inducible homologues for NCED may be responsible for the accumulation of ABA under drought-stressed conditions. It is noteworthy that overexpression of LeNCED1, a tomato NCED, elevates endogenous ABA in transgenic tobacco leaves (Thompson et al., 2000). In the present study we show that altered expression of AtNCED3 in transgenic Arabidopsis plants affects levels of endogenous ABA. Overexpression of AtNCED3 cDNA upregulated the endogenous ABA level in transgenics (Figure 5a). By contrast, repression of AtNCED3 downregulated the endogenous ABA level in antisense transgenics and T-DNA-tagged AtNCED3 knockout mutants (Figure 5a). Our results indicate that endogenous ABA levels can be manipulated by controlling the expression levels of NCED genes in transgenic plants.

Next we examined whether altered expression of AtNCED3 affected tolerance to drought stress. AtNCED3 sense transgenic plants were more resistant to drought stress than wild-type plants (Figure 6). By contrast, AtNCED3 antisense transgenic plants and T-DNA-tagged mutant (T5004) plants were more sensitive to drought stress than wild-type plants (Figure 6). The leaf transpiration rate was lower in the sense plants, but higher in the antisense and T-DNA insertion mutants, than in the wild-type plants (Figure 5b). ABA promotes stomatal closure and prevents water loss under drought-stressed conditions. ABA-deficient mutants of Arabidopsis and tobacco and ABA-insensitive mutants of Arabidopsis have reduced stomatal closure in response to water deficit, and show a wilty phenotype (Burbidge et al., 1999; Giraudat et al., 1994; Grill and Himmelbach, 1998; Leon-Kloosterziel et al., 1996; Marin et al., 1996; Sagi et al., 1999). Con versely, ABA-hypersensitive mutants of Arabidopsis show a reduced transpiration rate and enhanced drought tolerance (Pei et al., 1998). The enhanced drought tolerance of the AtNCED3 sense plants may be due to a reduction in the stomatal aperture caused by the elevated ABA level. By contrast, the decreased drought tolerance of the AtNCED3 antisense transgenic plants and T-DNA insertion mutants may be due to stomatal disclosure caused by reduced ABA.

Overexpression of AtNCED3 cDNA led to overexpression of the drought-inducible genes rab18, kin1 and rd29B (Figure 4). The expression of rd29B is induced by drought stress and the application of ABA, but not by cold stress (Yamaguchi-Shinozaki and Shinozaki, 1994). Promoter analysis of rd29B showed that two ABREs function in the dehydration-responsive transcription of rd29B (Uno et al., 2000). Complete repression of rd29B in the AtNCED3 antisense transgenic plants also supports the ABA-dependent expression of rd29B under drought-stressed conditions. kin1 and rab18 have both ABRE and DRE in their promoters, and are induced by both drought and cold (Kurkela and Franck, 1990; Lang and Palva, 1992). These results suggest that stress-inducible genes containing ABRE in their promoters are overexpressed in AtNCED3 sense transgenic plants, causing drought tolerance in the transgenics.

In conclusion, endogenous ABA could be controlled by gene engineering of AtNCED3 in transgenic Arabidopsis plants. Accumulation of ABA in the transgenic plants activated ABA signal transduction pathways and stomatal closure, resulting in enhanced drought tolerance. These results point the way to molecular breeding of drought-tolerant crops using a key gene involved in ABA biosynthesis.

Experimental procedures

Plant materials and growth conditions

Arabidopsis thaliana (L.) Heynh. ecotype Columbia was used in this study. Seeds were sown on GM agar plates or soil, and plants were grown at 22°C. Whole plants were used for molecular and biochemical analyses.

Dehydration treatment

Plants harvested from agar plates were dehydrated on Whatman 3MM filter paper at room temperature and approximately 60% humidity under dim light. Control plants were collected from the agar plates and grown in deionized water. Plants grown on soil were subjected to drought stress by withholding water.

Isolation of NCED genes from Arabidopsis

NCED cDNA of A. thaliana was isolated from a cDNA library prepared from dehydrated plants by using VuNCED1 fragments as a probe, and named AtNCED3 (Iuchi et al., 2000). Genomic DNA was isolated from A. thaliana plants. Synthetic oligonucleotides for PCR were designed based on the sequences of putative AtNCED genes in the DNA database. The sequences of synthetic oligonucleotides were as follows: 5′-AAGAATTCATGGCGGAGA AACTCAGTGATGGCAGC-3′ and 5′-AAAAGAATTCGGCTTATATA AGAGTTTGTTCCTGG-3′ (set A, accession numbers AL163818 and AJ005813; Neill et al., 1998); 5′-CCCGGGATCCCTCAAGCCTCT CTACTATACCG-3′ and 5′-CCCGGGATCCTTTATACGGATTCTGA GGGAG-3′ (set B, accession number AL021710); 5′-ATTGAA TTCATGGACTCTGTTTCTTCTTCTTCC-3′ and 5′-ATTGAATTCTTA AAGCTTATTAAGGTCACTTTCC-3′ (set C, accession numbers AL021687 and AL161550); 5′-ATGACGATATAATAACCATTA TTTCTGGTATG-3′ and 5′-CTAACACAAAGCTTGCTTCGATAAAT CTTCC-3′ (set D, accession number AC013430); 5′-CGGGAT CCATGCAACACTCTCTTCGTTCTGATCTTCTTC-3′ and 5′-CGG GATCCTCAGAAAACTTGTTCCTTCAACTGATTCTCGC-3′ (set E, accession number AB028621); 5′-ATGGCTTGTTCTTACATA TTAACACCAAACCC-3′ and 5′-TTAAGCCTGGTTTAACATATCCGC CGAATTCACG-3′ (set F, accession number AC074176). DNA fragments of putative NCED genes from A. thaliana, named AtNCED 2, 4, 5, 6 and 9, were amplified from total cDNAs of stressed and unstressed plants by PCR using the synthetic oligonucleotide sets B, C, D, E and F, respectively. The DNA fragments were cloned into the pBluescript II SK+ cloning vector (Stratagene, La Jolla, CA, USA) and sequenced using an ABI3700 sequencer to confirm that no sequence error was introduced by the PCR. The AtNCED1 cDNA was obtained by screening the Arabidopsis cDNA library using the genomic DNA fragment of AtNCED1 (amplified by PCR with the oligonucleotide set A) as a probe.

Northern analysis

Total RNA was isolated according to the method described by Nagy et al. (1988), fractionated in a 1% agarose gel containing formaldehyde, and blotted onto a nylon filter (Sambrook et al., 1989). The filter was hybridized with [32P]-labelled cDNA or RNA fragments at 42°C. The filter was washed twice with 0.1 × SSC, 0.1% SDS, at 65°C for 15 min, and autoradiographed. A DIG-labelled RNA probe prepared from the 3′ region of AtNCED3 was prepared and used for the Northern analysis shown in Figure 4(b), according to the manufacturer's instructions (Roche Diagnostics, Mannheim, Germany).

Expression of a glutathione S-transferase (GST)-AtNCED fusion protein

Full-length DNA of AtNCED14, 6 and 9 was amplified by PCR. The PCR products were cloned into the EcoRV site of pBluescript II SK+ cloning vector, amplified in Escherichia coli, and sequenced. Full-length DNAs thus prepared were then digested from the plasmids by EcoRI and inserted into the EcoRI site of pGEX4T (Amersham Pharmacia Biotech, Uppsala, Sweden) to yield pGST-AtNCED14, 6 and 9. Cells of E. coli strain JM109 were transformed with pGST-AtNCED14, 6 and 9 and grown in LB Luria-Bertani medium (L broth) at 37°C. When the OD600 reached 0.5, isopropyl β-d-thiogalactopyranoside was added to the culture, and the cells were further incubated at 17°C for 12 h. Then the cells were harvested, washed and suspended in the extraction buffer (10 mm Tris–HCl pH 8.0, 5 mm MgCl2, 5% (v/v) glycerol, 0.1 mm phenylmethylsulfonyl fluoride, 0.1 mm dithiothreitol). Fusion proteins were purified from the crude extracts according to the manufacturer's instructions (Amersham Pharmacia Biotech).

Assay of NCED activity

The procedure used for the assay of NCED enzyme activity was as described previously (Iuchi et al., 2000). Each recombinant protein was incubated with 9-cis- or all trans-epoxycarotenoids (neoxanthin and violaxanthin) under 100 mm bis-Tris {2-[bis (hydroxyethyl)amino]-2-(hydroxymethyl)-1-propane-1,3-diol; pH 6.7}, 0.05% (v/v) Triton X-100, 10 mm ascorbate and 5 mm FeSO4 at room temperature for 1 h (total volume 100 µl). After the addition of 1 ml water, the reaction mixture was extracted twice with 1 ml ethyl acetate. The ethyl acetate fractions were combined, concentrated and analysed by HPLC on a column of Senshu Pak ODS H 3151 (150 mm length, 8 mm inner diameter; Senshu Scientific, Tokyo, Japan). The column was eluted with a linear gradient between solvent A (85 : 15 v/v, methanol : water) and solvent B (1 : 1 v/v, chloroform : methanol) at a flow rate of 1.5 ml min−1. The concentration of solvent B was increased from 10 to 50% in 25 min, and kept at 50% for 5 min. The elution from the column was monitored with a UV/visible detector at 440 nm wavelength in order to detect substrate carotenoid and C25 products.

To identify cis-xanthoxin, the ethyl acetate extract of the reaction mixture was submitted to HPLC purification using the column described above. The column was eluted with 50% (v/v) aqueous methanol at a flow rate of 1.5 ml min−1, and the elution from the column was monitored with a UV detector at 260 nm wavelength. The predicted cis-xanthoxin fraction was collected and submitted to GC–MS analysis.

GC–MS analysis

An AUTOMASS mass spectrometer (JEOL Ltd, Akishima, Japan) equipped with a 5890 gas chromatograph (Hewlett-Packard, Palo Alto, CA, USA) was used. Analytical conditions were as follows: ionization, EI 70 eV; column, DB-5 (15 m length, 0.25 mm inner diameter, 0.25 µm film thickness; J&W Scientific, Folsom, CA, USA); carrier gas, He (1 ml min−1); injection temperature, 250°C; initial oven temperature, 80°C. Starting 1 min after injection, the oven temperature was increased to 170°C at a rate of 30°C min−1, followed by further increment to 280°C at a rate of 10°C min−1. cis-Xanthoxin was eluted from the column at the retention time of 7.9 min.

Construction of transgenic plants

To generate transgenic plants, we constructed a chimeric gene in which the coding sequence of the AtNCED3 cDNA was fused in a sense or antisense orientation between the 35S promoter from the cauliflower mosaic virus and the nos terminator sequences of the expression vector pBE2113Not (Liu et al., 1998; Mitsuhara et al., 1996), to produce plasmids 35S-AtNCED3 (sense) or 35S-anti-AtNCED3 (antisense). We introduced these plasmids into wild-type Arabidopsis seedlings. Among the T2 plants, lines homozygous with respect to the transgene were selected by examining their kanamycin resistance (in T3 seeds) after self-pollination. T3 seeds were used for subsequent experiments.

Isolation and characterization of an insertional mutant in the AtNCED3 gene

A T-DNA-tagged insertional mutant in AtNCED3 was isolated by screening a collection of 12 000 independently T-DNA-transformed lines by PCR, using a pair of synthetic oligonucleotides, 5′-CTTCTTCATCACGTATATGTATTTTGTGCG-3′ and 5′-ATAACG CTGCGGACATCTAC-3′.

A positive line, T5004, was used for further characterization. Because T5004 had multiple T-DNA insertion sites in its genome, we could not find a line with a single T-DNA insertion site in its progeny. However, the phenotype (low tolerance to water deficit) was linked with disruption of AtNCED3. The expression level of AtNCED3 was examined by Northern hybridization to confirm that expression was almost absent in this mutant.

Measurement of ABA level

Endogenous ABA was measured using exactly the same procedure as previously described (Iuchi et al., 2000).

Measurement of leaf transpiration rate

The transpiration rate was measured in fully expanded leaves with a portable photosynthesis system (model LI-6400, Li-Cor, Lincoln, NE, USA) under the following conditions: 100 µmol m−2 sec−1, 350 p.p.m. CO2, 22°C, 70% relative humidity.


We thank Ms Keiko Maeda and Ms Setsuko Kawamura for their excellent technical assistance. This work was supported in part by the Special Coordination Fund of the Science and Technology Agency of the Japanese Government; by a Grant-in-Aid from the Ministry of Education, Science and Culture of Japan to K.S.; and by the Program for Promotion of Basic Research Activities for Innovative Biosciences. This work was also supported in part by the Special Postdoctoral Researchers’ Program and the President's Special Research Grant from RIKEN to S.I.