ABA - Wiley Online Library

J. M. Barrero et al .

Plant, Cell and Environment (2006) 29, 2000–2008

doi: 10.1111/j.1365-3040.2006.01576.x

Both abscisic acid (ABA)-dependent and ABA-independent pathways govern the induction of NCED3, AAO3 and ABA1 in response to salt stress JOSÉ MARÍA BARRERO1, PEDRO L. RODRÍGUEZ2, VÍCTOR QUESADA1, PEDRO PIQUERAS1, MARÍA ROSA PONCE1 & JOSÉ LUIS MICOL1 1 División de Genética and Instituto de Bioingeniería, Universidad Miguel Hernández, Campus de Elche, 03202 Elche, Alicante, Spain, and 2Instituto de Biología Molecular y Celular de Plantas, Universidad Politécnica de Valencia-Consejo Superior de Investigaciones Científicas, Camino de Vera, 46022 Valencia, Spain

ABSTRACT

INTRODUCTION

The responsiveness of plants to osmotic stress is critically mediated by the increase in abscisic acid (ABA) levels. Osmotic stress induces the biosynthesis of ABA, whose increased levels subsequently exert a positive feedback on its own biosynthetic pathway. As only qualitative or semiquantitative analyses were performed to test the inducibility of ABA biosynthetic genes in Arabidopsis thaliana, we used quantitative reverse transcriptase-polymerase chain reaction to re-examine the induction of the ABA1, ABA2, ABA3, NCED3 and AAO3 genes by NaCl and ABA. Quantitative gene expression data obtained from wild-type plants and severely ABA-deficient mutants support the prevailing notion that the 9-cis-epoxycarotenoid cleavage reaction is the key regulatory step in NaCl-induced ABA biosynthesis. Interestingly, strong induction by NaCl of NCED3 was still observed in severe ABA-deficient mutants, pointing to an ABA-independent induction pathway for NCED3 that is NaCl-dependent. Therefore, in the absence of the ABA-mediated positive feedback on ABA biosynthesis, the ABA-independent pathway makes a major contribution to the induction of key ABA biosynthetic genes, such as NCED3, AAO3 and ABA1. In addition, and in contrast to some previous reports, our data do not support the limited ability of ABA to induce NCED3 expression. Under our experimental conditions, the induction of NCED3 by ABA, either in wild-type plants or ABAdeficient mutants, was predominant over that of other ABA biosynthetic genes. Natural variability was found in the induction by NaCl and ABA of NCED3 and ABA1 expression in different Arabidopsis accessions, although NCED3 expression was clearly predominant.

The plant hormone abscisic acid (ABA; Ohkuma et al. 1963; Cornforth et al. 1965; Addicott et al. 1968) has long been known to be involved in the responsiveness of plants to various environmental stresses, particularly drought and salinity (reviewed in Zhu 2002). ABA biosynthetic genes and enzymes have been identified in different plant species (reviewed in Taylor, Burbidge & Thompson 2000; Milborrow 2001; Seo & Koshiba 2002; Nambara & Marion-Poll 2005). One of these enzymes is zeaxanthin epoxidase (ZEP; Marin et al. 1996; Audran et al. 1998, 2001; Xiong et al. 2002), which is encoded by the ABA1 gene of Arabidopsis thaliana and catalyses the epoxidation of zeaxanthin, the first oxygenated precursor of the ABA biosynthetic pathway, to antheraxanthin and all-trans-violaxanthin (Fig. 1). The latter is then converted into the cis-isomers of neoxanthin and violaxanthin, which are cleaved by a 9-cis-epoxycarotenoid dioxygenase (NCED) to form xanthoxin (Schwartz et al. 1997; Qin & Zeevaart 1999; Iuchi et al. 2000). The enzymes catalysing the conversion of all-trans-violaxanthin to 9′-cis-violaxanthin or 9-cisneoxanthin have not been identified. A short-chain dehydrogenase/reductase (SDR) encoded by the ABA2 gene in Arabidopsis catalyses the conversion of xanthoxin to abscisic aldehyde (Cheng et al. 2002; González-Guzmán et al. 2002), which is oxidized into ABA by the Arabidopsis aldehyde oxidase 3 (AAO3; Seo et al. 2000a,b). The activity of the AAO3 enzyme requires a sulfurated molybdenum cofactor (MoCo), which is converted from the desulfo- to the sulfo-form by the MoCo sulfurase ABA3 (Xiong et al. 2001; Bittner, Oreb & Mendel 2001). In vegetative tissues, the concentration of ABA increases up to 40-fold under drought and salt stress (Zeevaart & Creelman 1988). Mutants with impaired ABA biosynthesis or perception are more sensitive to environmental changes, while transgenic plants producing high levels of this hormone display more tolerance to those forms of abiotic stress than the wild type (Iuchi et al. 2001; Qin & Zeevaart 2002). Several analyses of the temporal and spatial expression profiles of different ABA biosynthetic genes have been

Key-words: Arabidopsis thaliana; ABA biosynthesis; accessions; QRT-PCR; salt tolerance.

Correspondence: J. L. Micol. Fax: 34 96 665 85 11; e-mail: [email protected]

2000

© 2006 The Authors Journal compilation © 2006 Blackwell Publishing Ltd

QRT-PCR analysis of ABA biosynthesis genes 2001

H3C

CH 3

H3C

CH 3

CH 3

CH 3

OH

O

H3C

CH 3

CH 3 HO

CH 3

Zeaxanthin ABA1

H3C

H3C

CH 3

CH 3 HO

OH

H3C

CH 3

CH 3

CH3

CH 3

CH 3

Antheraxanthin ABA1 H3C H3C

CH 3

CH 3

CH 3 O

CH 3

CH 3 HO

OH

O

H3C

CH3

CH 3

All-trans-violaxanthin CH3

H3C

CH3

H 3C H 3C

CH 3

O O

HO

H3C

CH 3

OH

CH3 H 3C H 3C CH 3

H3C

9-Cis-neoxanthin

9’-Cis-violaxanthin

CH 3 H 3C

CH 3

O

H3C

CH3

H3C

HO OH H 3C

OH

NCED3 CH3

CH 3

H3C

O CHO HO

CH 3

Xanthoxin ABA2 H3C

CH 3

CH 3

OH CHO

O

AAO3, ABA3 CH 3

CH 3

OH COOH O

MATERIALS AND METHODS

CH 3

Abscisic aldehyde

H3 C

ABA3 and AAO3 are key enzymes regulating the ABA biosynthesis pathway in Arabidopsis and other plant species, given that expression of the corresponding genes is induced either by drought or exogenous ABA (Qin & Zeevaart 1999; Seo et al. 2000a; Iuchi et al. 2001; Xiong et al. 2001, 2002; Tan et al. 2003; Xiong & Zhu 2003). On the contrary, the ABA2 gene is not significantly induced by either osmotic stress or exogenous ABA (Cheng et al. 2002; González-Guzmán et al. 2002). Because previous studies of genes encoding ABA biosynthetic enzymes were mostly qualitative or semi-quantitative, we used quantitative reverse transcriptase polymerase chain reaction (QRT-PCR) to examine the expression of ABA biosynthetic genes. NCED3 induction represents an early limiting step in controlling osmotic stress-induced ABA biosynthesis, although the currently accepted model for stress induction of ABA biosynthesis states that ABA has a limited ability to induce the expression of NCED3 (Xiong et al. 2002). Other genes, such as AAO3, ABA1 and ABA3, would be more sensitive to the ABA-mediated positive feedback on ABA biosynthesis (Xiong et al. 2002). In contrast to this notion, we report here that under our experimental conditions, the induction of NCED3 by ABA either in the wild-type or ABA-deficient mutants was predominant over other ABA biosynthetic genes. Finally, data obtained in this work for the NCED3, AAO3 and ABA1 genes in both wild-type and severe ABAdeficient mutants reveal a major contribution of an ABA-independent pathway on the induction of ABA biosynthetic genes.

CH 3

ABA Figure 1. Biosynthesis of abscisic acid (ABA) from carotenoid precursors in higher plants. Adapted from Cutler & Krochko (1999).

published (reviewed in Xiong & Zhu 2003), some of them producing contrasting results. For instance, in the study of Audran et al. (2001) in Arabidopsis, ABA1 transcript levels were increased by drought stress only in root tissues, and the drought-induced upregulation of ABA1 was unaffected both in several ABA-deficient mutants and the ABAinsensitive abi1-1 mutant. In contrast, the study of Xiong et al. (2002) revealed osmotic induction of ABA1 both in wild-type roots and shoots as well as impaired induction of ABA1 in ABA-deficient mutants and the ABAinsensitive abi1-1 mutant. The analysis of the regulation of ABA biosynthesis is crucial for understanding ABA actions in plant physiology. Previous studies indicated that ZEP (ABA1), NCED3,

Plant material, growth conditions and germination assays Wild-type (Col-0) and mutant (all of them in a Col-0 genetic background) plants of A. thaliana L. Heynh. were grown in sterile conditions (in 150 mm Petri dishes containing 100 mL of agar medium), at 20 ± 1 °C, 60–70% relative humidity and continuous illumination of 7000 lx, as described in Ponce, Quesada & Micol (1998). For ABA treatment, 15-day-old plants were sprayed directly with 100 µM ABA (Sigma-Aldrich A1049; Sigma-Aldrich, St. Louis, MO, USA), or with water in the case of the controls (Xiong et al. 2002). For the NaCl treatment, 15-day-old plants were transferred into Petri dishes containing filter paper saturated with 300 mM NaCl, or with water for the controls (Xiong et al. 2002). Germination rates were determined in triplicate by plating seeds on a water suspension, using a Pasteur pipette, at a density of 100 regularly spaced seeds per plate, in 150 mm Petri dishes filled with 100 mL of agar medium supplemented with 0–450 mM NaCl, as described in Quesada et al. (2002). Each germination assay was made in triplicate by using seeds that had been simultaneously collected from several plants and then pooled. Germination percentages were scored 18 d after sowing. We considered as

© 2006 The Authors Journal compilation © 2006 Blackwell Publishing Ltd, Plant, Cell and Environment, 29, 2000–2008

2002 J. M. Barrero et al. germinated those seeds whose radicle had emerged through the seed coat, irrespective of subsequent survival. In fact, development of the germinated seeds was arrested in the most saline conditions.

RNA isolation and quantification For both the ABA and NaCl treatments, plants were collected after 5 h of incubation in a growth chamber, frozen in liquid N2 and ground in RNase-free conditions. RNA was extracted from either whole plants or from separately collected roots and aerial tissues. RNA was extracted by using a Qiagen RNeasy Plant Mini Kit, treated with DNase I and finally resuspended in 40 µL of RNase-free water. Three to five micrograms of the RNA solution obtained was reversetranscribed using random primers and the SuperScript II Reverse Transcriptase (Gibco BRL, Paisley, UK) following the protocol provided by the manufacturer, to finally obtain a 40 µL cDNA solution. QRT-PCR amplifications and measurements were performed mainly as described in Cnops et al. (2004) and PérezPérez, Ponce & Micol (2004), using an ABI PRISM 7000 sequence detection system. For each of the genes under study, a primer pair was designed to obtain a PCR amplification product of approximately 100 bp. The 5′ and 3′ halves of one of the oligonucleotides of each primer pair corresponded to the sequences of two exons flanking an intron, so that genomic DNA could not be amplified. The sequences of the primers used for QRT-PCR amplifications of ABA1 (At5g67030), ABA2 (At1g52340), ABA3 (At1g16540), NCED3 (At3g14440), AAO3 (At2g27150) and OTC (At1g75330) transcripts were the following: NCED3-F, 5′-CGGTGGTTTACGACAAGAACAA-3′; NCED3-R, 5′-CAGAAGCAATCTGGAGCATCAA-3′; AAO3-F, 5′GGAGTCAGCGAGGTGGAAGT-3′; AAO3-R, 5′-TG CTCCTTCGGTCTGTCCTAA-3′; ABA1-F, 5′-GGCA TTTGGTC TAAGGTGAGAA-3′; ABA1-R, 5′-CAGACT CGATATCCGCTGGTA-3′; ABA2-F, 5′-TTCTCTTCCTA GTCAAAGGCTTT-3′; ABA2-R, 5′-GCAGACTTTGGC ACCGTGCT-3′; ABA3-F, 5′-CAAAAGGAAGAGTCAA GAGGAAA-3′; ABA3-R, 5′-TTTCTTTCATCAA CTT CACCAGAT-3′; OTC-F, 5′-TGAAGGGACAAAGGTTGTGTATGTT-3′; OTC-R, 5′-CGCAGACAAAGTGGA ATGGA-3′. QRT-PCR amplifications were performed in 25 µL reaction mixes by adding 12.5 µL of the SYBR-Green PCR Master Kit (Applied Biosystems, Foster City, CA, USA), 10 pmol of each primer and 1 µL of the 40 µL cDNA solution obtained as described earlier. Relative quantification of gene expression data was carried out using the 2-∆∆CT or comparative CT method (Livak & Schmittgen 2001). All reactions were made in triplicate by using three aliquots from a single RNA sample and expression levels were normalized using the CT values obtained for the housekeeping ornithine transcarbamilase (OTC) gene (Quesada, Ponce & Micol 1999), which was used as an internal reference gene (Pérez-Pérez et al. 2004; Cnops et al. 2004). All results were referred to those obtained for the Col-0 wild type under

non-supplemented conditions, to which a value of 1 was given. Finally, we designed our oligonucleotides to avoid simultaneous amplification of more than one gene family member and the presence of a single QRT-PCR product was further verified by dissociation analysis in all amplifications.

RESULTS Mutants used in this work Extremely hypomorphic or null alleles of ABA biosynthetic genes were used in this study to provide severe ABA-deficient genetic backgrounds. One of them, the aba1-101 mutant, initially named sre3 (González-Guzmán et al. 2002), contains a deletion of a T in position 396 of the ABA1 gene, generating a stop codon after amino acid 131. Both the phenotypic and molecular characterizations of aba1-101 suggest that it represents a null alelle of the ABA1 gene (Barrero et al. 2005). Additionally, the aba1101 mutation is likely to affect stability of the ABA1 transcript, whose level is severely diminished in the aba1-101 mutant (see Fig. 2a). The aba2-14 and aba3-101 mutations (initially named sañ3-2 and sañ4-1, respectively; Quesada, Ponce & Micol (2000) were induced by fast neutron bombardment and suffered severe deletions in their transcription units, abolishing expression of the corresponding genes (Quesada et al. 2000; González-Guzmán et al. 2002; see also Fig. 2b,c). Finally, the aao3-2 allele carries a T-DNA insertion that abolishes the expression of the gene, as determined by northern blot analysis, and leads to impaired ABA biosynthesis (González-Guzmán et al. 2004; see also Fig. 2d,i). We found the ABA content in these mutants strongly reduced both in the presence and absence of NaCl. Thus, ABA concentration in the absence of NaCl was 30 and less than 20% of the wild-type value in aao3-2 and aba1101, respectively, as well as 15 and 2% after exposure to NaCl (González-Guzmán et al. 2004; Barrero et al. 2005; González-Guzman and Rodríguez, unpublished results). Although we did not quantify ABA in the aba2-14 mutant, another null allele of ABA2, aba2-11, displayed 20% of the wild-type level under unstressed conditions, and less than 5% when exposed to salt stress (González-Guzmán et al. 2002). No ABA content measurements were made in aba3-101.

QRT-PCR analyses of ABA biosynthetic genes after ABA and NaCl treatments in whole plants Figure 2 shows the results of QRT-PCR analyses of the expression of the ABA1, NCED3, ABA2, AAO3 and ABA3 genes in 15-day-old whole plants of the Col-0 wild-type and aba1-101, aba2-14, aba3-101 and aao3-2 homozygous mutants, either under control conditions or exposed to ABA or NaCl. The analyses of the expression of the ABA1, ABA2, ABA3 and AAO3 genes in the aba1-101, aba2-14, aba3-101 and aao3-2 genetic backgrounds, respectively,



1.90

1.86

1.06

1.94

1.03

1.03

0.06

0.20

0.92

0.78 1.07

1.15

0.00

0.00

1.00

0.57

1.00

1.00

0.77

0.87

0.51

0.78

0.00 1.32

0.00

0.98

1.00

0.93

0.80

0.02

0.02

0.86

(h)

1.19

1.68

1.14

2.21

1.91

0.60

1.00

(c)

0.14

0.41

0.01

0.77

0.70

0.03

1

0.34

1.00

0.1

1.92

ABA2 expression

(g)

(b)

1

0 10

ABA3 expression

1.00 1.51

0.1

0.01 10

0.1

0.01

3.89

2.79

1.03

1.14

2.58

1.02

0.03

1.00 6.66

0.39

0.11

0.75

0.1

0.01 100

(e)

(j)

0.50

-2 l-0 01 01 14 -1 o3 -1 2Co 3 a 1 a a a a ab ab ab

proved the specificity of each set of primers. According to previous data obtained by northern blot analysis, transcription of the ABA1, ABA2 and AAO3 genes is dramatically reduced in the aba1-101, aba2-14 (González-Guzmán et al. 2002) and aao3-2 (González-Guzmán et al. 2004) mutants, respectively. Consistent with this, in control conditions the

NaCl

7.13

24.14 NaCl

1.13 Control

Control

12.24 NaCl

Control

NaCl

Control

8.75

0.34

0.20

1.00 NaCl

Control

32.26

4.21

-2 l-0 01 01 14 o3 -1 -1 2Co a 3 1 a a a a ab ab ab

ABA

Control

22.94 ABA

Control

30.34 ABA

0.46

0.49

0.48 Control

16.00 ABA

ABA

Control

0.1

9.02

1

0.64

1.00

10

Control

NCED3 expression

(i)

6.10

9.27

12.38

1.03

5.29

1

0.87

1.00

10

(d)

0.03

0.001 100

AAO3 expression

Relative expression (2–DDCT)

(f)

2.94

0.65

0.35 2.24

0.30

0.04

3.96

1

0.20

1.00

(a)

2.18

ABA1 expression

NaCl treatments

ABA treatments

10

Figure 2. Quantitative reverse transcriptase-polymerase chain reaction analysis of the expression of genes involved in abscisic acid (ABA) biosynthesis. All quantifications were made in triplicate on RNA samples obtained from plants sprayed once with 100 µM ABA or exposed to 300 mM NaCl for 5 h (see Materials and Methods section). The expression levels shown are relative to those of Col-0 plants in control conditions, to which a value of 1 was given. Error bars indicate standard deviations.

transcription of ABA1, ABA3 and AAO3 genes was severely reduced in the aba1-101, aba3-101 and aao3-2 mutants, respectively, compared with the wild type (Fig. 2a,c,d), whereas ABA2 transcripts were not found in the aba2-14 mutant (Fig. 2b), as expected from its large deletion.


2004 J. M. Barrero et al. After treatment of 15-day-old whole plants with 100 µM ABA for 5 h (see Materials and Methods section), the NCED3, ABA1 and AAO3 genes were induced both in the wild-type and ABA-deficient mutants (Fig. 2). The ratio of expression after ABA treatment/control indicated that the highest level of induction by ABA was that observed for the NCED3 gene (Fig. 2e), which increased its expression 9-fold in Col-0, 25-fold in aba1-101, 63-fold in aba2-14, 46-fold in aba3-101 and 9-fold in aao3-2. The ABA1 gene was found to be induced in all cases (2-fold in Col-0, 5-fold in aba1-101, 13-fold in aba2-14, 6-fold in aba3-101 and 2-fold in aao3-2; Fig. 2a), as was AAO3 (5-fold in Col-0, 12-fold in aba1-101, 10-fold in aba2-14, 8-fold in aba3-101 and 3-fold in aao3-2; Fig. 2d). The results obtained for NCED3 are noticeably different from those of Xiong et al. (2002), who found no or limited induction of NCED3 after ABA treatment. In contrast, our results agree with those of Cheng et al. (2002), who observed the strong induction of NCED3 in roots after ABA treatment. Contrary to the discrepancy observed for NCED3, our results for ABA1 and AAO3 qualitatively agree with those obtained by Xiong et al. (2002). Treatment of 15-day-old whole plants of the wild-type Col-0 with 300 mM NaCl (Fig. 2f–j) induced all the studied genes, with the exception of ABA2. Just as for the ABA treatment, the NCED3 gene (Fig. 2j) was the most responsive to NaCl, as it was found 32-fold induced in Col-0. The AAO3 gene (Fig. 2i) was also markedly induced (6-fold), whereas the ABA1 (Fig. 2f) and ABA3 (Fig. 2h) genes were induced to a lesser extent (3- and 2-fold, respectively). Interestingly, induction of the ABA biosynthetic genes in ABA-deficient mutants was clearly observed after treatment with NaCl, although the absolute expression level reached was always lower than in the wild type. Moreover, a 14- to 43-fold increase in NCED3 expression, i.e. expression after NaCl treatment/control, was measured in severely ABA-deficient mutants after NaCl treatment, pointing to an ABA-independent pathway for the induction of NCED3 by NaCl. Under

non-stress conditions, the expression of NCED3 (Fig. 2e,j) was reduced in ABA-deficient mutants compared to wildtype plants. The downregulation of NCED3 in ABA-deficient mutants together with the strong induction of this gene in response to ABA support the hypothesis that NCED3 is a key element of a positive feedback mechanism regulated by ABA levels. Under the control conditions used for the ABA treatment, the ABA1 gene was also notably downregulated in all the mutants studied (Fig. 2a). This is consistent with the qualitative results of Xiong et al. (2002). However, Audran et al. (2001) found that ABA1 transcript levels were similar in aba2, aba3 and wild-type plants, which is consistent with the results that we obtained for these mutants when used as controls for NaCl treatment studies (Fig. 2f; these mutants were transferred from culture medium to filter paper saturated with distilled water). Our divergent results, as well as those of previous authors, are likely to be a consequence of the different manipulation procedures used, rather than a genuine effect of ABA or NaCl on ABA1 expression.

QRT-PCR analyses in roots and rosettes The results described previously were obtained upon RNA analyses from whole plants. As some differences were previously described regarding expression of ABA biosynthetic genes in different tissues (Audran et al. 2001; Cheng et al. 2002; Seo & Koshiba 2002; Xiong et al. 2002), particularly between roots and rosettes, we performed quantitative analyses similar to those described earlier in the same tissues. To this end, after ABA and NaCl treatments, roots and rosettes of the plants were collected separately and QRT-PCR-analysed (Table 1). After treatment with 100 µM ABA for 5 h, the NCED3, ABA1 and AAO3 genes, and to a lesser extent ABA3, were induced both in roots and rosettes of the wild-type and ABA-deficient mutants (Table 1). The highest level of induction by ABA was that of the NCED3 gene (Table 1), whose expression in roots

Table 1. Induction of gene expression after abscisic acid (ABA) or NaCl treatments 100 µM ABA treatment ABA1

Col-0 aba1-101 aba2-14 aba3-101 aao3-2

ABA2

300 mM NaCl treatment ABA3

AAO3

NCED3

ABA1

ABA2

ABA3

AAO3

NCED3

A

R

A

R

A

R

A

R

A

R

A

R

A

R

A

R

A

R

A

R

1.9 3.0 4.0 4.0 1.6

2.9 2.6 5.0 7.3 4.5

1.0 1.1 nd 0.7 0.6

1.0 1.0 nd 1.6 2.2

1.7 4.3 4.0 nd 1.9

1.2 1.8 2.5 nd 1.4

3.9 10.7 7.7 12.3 nd

2.6 2.2 3.8 4.0 nd

8.8 4.1 13.0 8.0 15.6

14.0 12.4 12.2 14.0 21.3

4.5 2.1 2.0 1.2 3.2

3.3 6.4 2.1 3.5 3.3

1.2 0.8 nd 0.9 0.8

0.9 1.3 1.3 1.2 1.4

5.0 1.6 2.3 nd 1.0

1.6 1.2 1.4 nd 2.2

10.0 3.4 5.6 3.9 nd

2.4 2.2 2.6 2.5 nd

166.2 29.4 73.4 44.6 114.8

90.9 54.3 27.0 21.5 68.0

Numbers indicate the induction level of genes indicated, after treatment with ABA or NaCl, in the wild-type Col-0 and ABA-deficient mutants. Values are the ratio of expression level reached by each genotype after treatment relative to that of the same genotype in control conditions. All quantifications were made in triplicate on RNA samples obtained from plants sprayed once with ABA or exposed to NaCl for 5 h (see Materials and Methods section). A, aerial tissues (rosettes); R, roots; nd, not detected. © 2006 The Authors Journal compilation © 2006 Blackwell Publishing Ltd, Plant, Cell and Environment, 29, 2000–2008


Natural variability in the induction by NaCl and ABA of NCED3 Analysis of the inducibility of NCED3 by ABA in previous semi-quantitative studies rendered contrasting results. For instance, Xiong et al. (2002) reported that NCED3 was not induced by ABA in wild-type C24 plants; however, in contrast, it was clearly induced in wild-type Landsberg erecta (Ler) plants, although to lower levels than other ABA biosynthetic genes. Our results for NCED3 induction by ABA in Col-0 show a clear and predominant induction of this gene with respect to the other ABA biosynthetic genes studied (Fig. 2a–e & Table 1). In order to test whether NCED3 plays a pivotal role both in ABA- and NaCl-mediated inductions of ABA biosynthesis, we analysed the induction of NCED3 by ABA and NaCl in different A. thaliana wild-type races (accessions). They were selected on the basis of their different response to NaCl at germination, which was discovered in a previous work where we

(a)

Germination (%)

100 80 60 Ws2 Ler Col La-1 Ak-1 Bu-18 Hodja

40 20 0

0

50 100 150 200 250 300 350 400 450

[NaCl] (mM)

2.7

1.9

1.4

2.6

2.8

1.5

2.5

2.9

2.9

2.0

2.5

4.3

1

2.1

10

2.9

ABA1 expression

(b)

(c) 100 10 1

control NaCl 32.2 control ABA 9.0 control NaCl 37.8 control ABA 7.0 control NaCl 32.9 control ABA 18.5 control NaCl 23.7 control ABA 10.8 control NaCl 12.5 control ABA 5.0 control NaCl 31.4 control ABA 4.0 control NaCl 51.3 control ABA 3.2

NCED3 expression

was increased 14-fold in Col-0, 12-fold in aba1-101 and aba2-14, 14-fold in aba3-101 and 21-fold in aao3-2. In rosettes, NCED3 also showed the highest level of induction, although it was lower than in roots in all cases except in the that of the aba2-14 mutant (Table 1). The ABA1 gene was found to be induced in all cases in roots (2.9-, 2.6-, 5-, 7- and 4.5-fold in Col-0, aba1-101, aba2-14, aba3-101 and aao3-2, respectively; Table 1). Similar results were obtained for rosettes, except in the aao3-2 mutant, for which only a 1.6-fold induction was found. In roots, the AAO3 gene was induced 2.6-fold in Col-0, 2-fold in aba1-101 and 4-fold in aba2-14 and aba3-101 (Table 1). However, a higher level of induction was detected in rosettes both in the wild type (3.9-fold) and all the mutants studied: 10.7-, 7.7- and 12.3fold in aba1-101, aba2-14 and aba3-101, respectively. In summary, as previously observed in whole plant tissues, NCED3 induction in response to ABA was clearly predominant over AAO3 and ABA1. Treatment of wild-type plants with 300 mM NaCl (Table 1) induced the expression of all the studied genes at least 2-fold in roots and rosettes, with the exception of ABA2, or ABA3 in roots. In wild-type plants, higher induction ratios by NaCl were obtained in rosettes than in roots. The NCED3 gene (Table 1) was by far the most responsive, as it was found 91- and 166-fold induced in Col-0 roots and rosettes, respectively. The AAO3 gene (Table 1) was also markedly induced in wild-type rosettes (10-fold), whereas the ABA1 and ABA3 genes were induced to a lesser extent (4.5- and 5-fold, respectively). In particular, ABA3 expression was 5-fold induced by NaCl in rosettes, whereas induction in roots was below 2-fold. Of particular note was the fact that a 21- to 68-fold (in roots) and 29- to 114-fold (in rosettes) increase in NCED3 expression was measured in severely ABA-deficient mutants after NaCl-treatment (Table 1). As previously observed in whole plant tissues, these results confirm that an ABA-independent pathway must operate to induce NCED3 expression in severe ABAdeficient mutants.

Col-0

Ler

Ws-2

La-1

Ak-1 Bu-18 Hodja

Figure 3. (a) Salt dose response of the germination of several Arabidopsis thaliana accessions. (b–c) Quantitative reverse transcriptase-polymerase chain reaction analysis of the expression of ABA1 (b) and NCED3 (c) genes in accessions exposed to 300 mM NaCl for 5 h. Quantifications were performed as indicated in Fig. 2. The expression levels shown are relative to those found in control conditions for each of the accessions studied, to which a value of 1 was given. Error bars indicate standard deviations.

compared the ability of 102 accessions to germinate on 250 mM NaCl, finding a wide range of variation among them (Quesada et al. 2002). Figure 3a shows a dose– response curve of germination under different salt concentrations for these accessions. On the one hand Ak-1, La-1, Col-0 and Ler were more tolerant, while on the other hand, Ws-2, Bu-18 and Hodja were more sensitive (Fig. 3a). The NaCl concentration, which permitted half of the seeds to germinate (IC50), as calculated from the dose–response curve of germination (Fig. 3a), was around 175 mM for Bu-18 and Hodja, 250 mM for Ws-2, 340 mM for La-1, 375 mM for Ler and Col-0 and 400 mM for Ak-1. On the basis of the QRT-PCR data obtained earlier (Fig. 2), two NaCl- and ABA-induced genes of the ABA biosynthetic pathway, ABA1 and NCDE3, were selected for quantitative gene expression analysis in the different accessions. The rationale was to test whether the low- versus high-induced expression pattern of ABA1 and NCED3 in response to ABA and NaCl was (or not) a common trend among the accessions studied. Several conclusions can be


2006 J. M. Barrero et al. extracted from the data in Fig. 3. Firstly, NCED3 induction by ABA appears to be a general issue in the different accessions. However, clear natural variability was observed, ranging from 3-fold induction in Hodja to 18-fold in Ws-2. Secondly, the induction of NCED3 by ABA is usually several folds higher than that observed for ABA1. Only in the case of Hodja were both ratios similar. In particular, our results for Ler and C24 clearly differ from those of Xiong et al. (2002). We found that NCED3 was clearly induced by ABA in C24, whereas Xiong et al. (2002) did not detect any induction. In addition, ABA induction was stronger for NCED3 than for ABA1 in Ler in our working conditions, the opposite being reported by Xiong et al. (2002). Regarding the inducibility of NCED3 and ABA1 by NaCl, our results in different accessions clearly confirm the prevailing notion that NCED3 induction by osmotic stress is predominant over ABA1. However, it was not possible to establish a correlation between, for instance, enhanced salt resistance at the germination step (observed mainly in the Ak-1, La-1, Col-0 and Ler accessions) and lower expression of NCED3. Thus, although Ak-1 displayed the lowest induction levels of NCED3 by NaCl (12-fold), a NaClsensitive ecotype, Bu-18, showed an induction ratio of 31-fold, quite similar to the more NaCl-tolerant ecotypes (Col-0, Ler and Ws-2).

DISCUSSION The induction of ABA biosynthetic genes by stress has mostly been analysed qualitatively or semi-quantitatively (Audran et al. 1998, 2001; Iuchi et al. 2000, 2001; Xiong et al. 2001), allowing (Xiong et al. 2002) to put forward the model for the stress induction of ABA biosynthesis represented in Fig. 4a. In this work, according to QRT-PCR analyses, we revise the model and propose its modification, particularly regarding the ABA-mediated positive feedback on ABA biosynthesis (Fig. 4b). Nevertheless, our QRT-PCR analyses (Figs. 2 & 3) confirm that NCED3 is the ABA biosynthetic gene most responsive to NaCl treatment, and this feature is present in the different accessions examined (Fig. 3). These data are consistent with the well-established role of NCED3 as a key player in controlling osmotic stressinduced ABA biosynthesis (Qin & Zeevaart 1999; Iuchi et al. 2000; Xiong et al. 2002; Tan et al. 2003). In addition, significant induction by NaCl was also observed for AAO3, ABA1 and, to a low level, ABA3 gene expression in whole plant tissues. In order to discern the relative contribution of the ABAdependent pathway to the NaCl-mediated induction of ABA biosynthetic genes, both wild-type plants and severe ABA-deficient mutants were analysed. The expression level of NCED3, AAO3 and ABA1 after the NaCl treatment was lower than in the wild type (40–80% of the absolute wild-type levels). This suggests that positive feedback regulation by ABA is an important element in the induction process, as proposed by Xiong et al. (2002). However, both the remaining induction (at least 40% of the absolute wildtype levels) and high ratio of induction (expression after

(a)

Osmotic stress

NCED3

(b) ABA1

NaCl

NCED3

ABA ABA2 ABA3

AAO3, ABA3, ABA1

AAO3

ABA

ABA

Figure 4. Regulation of abscisic acid (ABA) biosynthesis by NaCl and ABA. (a) Model for stress induction of ABA biosynthesis proposed by Xiong et al. (2002). Solid lines indicate stimulation of gene expression (except arrows pointing to ABA, which indicate stimulation of the production of ABA). A dotted line indicates limited stimulation. The AAO3, ABA3 and ABA1 genes appear in the model in order of decreased inducibility. (b) Alternative model proposed in this work. Straight white arrows indicate steps in the ABA biosynthesis pathway. Solid curved arrows represent positive regulatory effects. The NaCl signal leads to enhanced expression of ABA biosynthetic genes; among them, NCED3 induction is quantitatively predominant. Increased expression of the indicated genes leads to increased ABA level, which exerts a positive feedback on them. In contrast to model (a), NCED3 induction is predominant over other ABA biosynthetic genes both in the NaCl-mediated as well as ABA-mediated induction pathways. Both pathways are required for full induction of the indicated genes, although in ABA-deficient mutants, a major induction of them can be attained through an ABA-independent pathway.

NaCl treatment/control) of these genes by NaCl in severe ABA-deficient mutants indicate the existence of a NaCldependent but ABA-independent pathway for the induction of NCED3, AAO3 and ABA1. These results are in agreement with the current picture of the regulatory network involved in stress-responsive transcription, where both ABA-independent and ABA-dependent pathways regulate transcription of stress-responsive genes (Shinozaki, Yamaguchi-Shinozaki & Seki 2003). A major task for the future will be the identification of the cis- and transacting factors that regulate transcription of ABA biosynthetic genes, particularly in response to stress constraints. With regard to the positive feedback regulation by ABA of the expression of ABA biosynthetic genes, the model emerging from our QRT-PCR analysis is not in complete agreement with the model of Xiong et al. (2002) (compare Fig. 4a versus Fig. 4b). Thus, under our experimental conditions, the most responsive gene to ABA treatment was NCED3, irrespective of considering whole plant tissues or separately root and rosette tissues. The strong induction of NCED3 by ABA has also been reported by Cheng et al. (2002) in root tissues of the Col-0 wild type. In contrast, Xiong et al. (2002) found that this gene was not induced by


QRT-PCR analysis of ABA biosynthesis genes 2007 ABA in the wild-type C24 accession and only to a low level (compared with other ABA biosynthetic genes) in Col-0 and Ler. In our hands, NCED3 expression was induced by ABA in a range of 3- to 18-fold in the different accessions hereby reported. In particular, in Col-0 and Ler ecotypes the ratio of induction by ABA of NCED3 was 9- and 7-fold, respectively, and clearly above that observed for ABA1 (2-fold in both cases). Additionally, we found that the AAO3 and, to a lesser extent, ABA1 and ABA3 genes were induced in response to ABA, both in the wild-type and ABA-deficient mutants. Therefore, we propose a model for the positive feedback regulation of the biosynthesis pathway by ABA, in which NCED3 is the most responsive gene to the newly synthesized ABA (Fig. 4b). In addition, AAO3, and to a lesser extent ABA1 and ABA3 genes are also responsive to this ABA-dependent pathway. The fact that ABA-deficient mutants still show a major induction of ABA biosynthetic genes (NCED3, AAO3 and ABA1) reveals the existence of an additional ABA-independent pathway. In summary, our QRT-PCR analysis depicts a scenario whereby the osmotic stress signal is mainly sensed by the NCED3 gene, whose expression increases dramatically, and also by AAO3 and, to a lesser extent, ABA1 and ABA3. The increased transcription of these genes might lead to an increase in ABA levels, which, in turn, would induce a positive feedback regulation of NCED3, AAO3 and ABA1 gene expressions (Fig. 4b; Xiong et al. 2002). A precise balance between ABA biosynthesis and catabolism can be attained through the complementary action of ABA catabolic enzymes, such as ABA 8′-hydroxylases (Kushiro et al. 2004). Finally, transcriptional regulatory mechanisms do not appear to be the only ones to regulate ABA biosynthesis. For instance, in the case of AAO3 expression, a regulatory post-transcriptional mechanism must operate, as enhanced transcript levels in dehydrated leaves did not correlate with enhanced levels of AAO3 protein (Seo et al. 2000a). However, in the case of NCED3, a correlation between enhanced expression of the transcript and enhanced levels of protein and ABA was clearly established (Qin & Zeevaart 1999).

ACKNOWLEDGMENTS The authors wish to thank P. Robles for his comments on the manuscript, and J.M. Serrano and V. García-Sempere for excellent technical assistance. This work was supported by a fellowship (to J.M.B.) and research grants (BIO20001082 to J.L.M. and BIO2005-01760 to P.L.R.) from the Ministerio de Educación y Ciencia of Spain and FEDER (to P.L.R.).

REFERENCES Addicott F.T., Lyon J.L., Ohkuma K., Thiessen W.E., Carns H.R., Smith O.E., Cornforth J.W., Milborrow B.V., Ryback G. & Wareing P.F. (1968) Abscisic acid: a new name for abscisin II (dormin). Science 159, 1493.

Audran C., Borel C., Frey A., Sotta B., Meyer C., Simonneau T. & Marion-Poll A. (1998) Expression studies of the zeaxanthin epoxidase gene in Nicotiana plumbaginifolia. Plant Physiology 118, 1021–1028. Audran C., Liotenberg S., Gonneau M., North H., Frey A., TapWaksman K., Vartanian N. & Marion-Poll A. (2001) Localization and expression of zeaxanthin epoxidase mRNA in Arabidopsis in response to drought stress and during seed development. Australian Journal of Plant Physiology 28, 1161– 1173. Barrero J.M., Piqueras P., González-Guzmán M., Serrano R., Rodríguez P.L., Ponce M.R. & Micol J.L. (2005) A mutational analysis of the ABA1 gene of Arabidopsis thaliana highlights the involvement of ABA in vegetative development. Journal of Experimental Botany 56, 2071–2083. Bittner F., Oreb M. & Mendel R.R. (2001) ABA3 is a molybdenum cofactor sulfurase required for activation of aldehyde oxidase and xanthine dehydrogenase in Arabidopsis thaliana. Journal of Biological Chemistry 276, 40381–40384. Cheng W.H., Endo A., Zhou L. et al (2002) A unique short-chain dehydrogenase/reductase in Arabidopsis glucose signaling and abscisic acid biosynthesis and functions. Plant Cell 14, 2723– 2743. Cnops G., Jover-Gil S., Peters J.L., Neyt P., De Block S., Robles P., Ponce M.R., Gerats T., Micol J.L. & Van Lijsebettens M. (2004) The rotunda2 mutants identify a role for the LEUNIG gene in vegetative leaf morphogenesis. Journal of Experimental Botany 55, 1529–1539. Cornforth J.W., Milborrow B.V., Ryback P.F. & Wareing P.F. (1965) Chemistry and physiology of ‘dormins’ in sycamore: Identity of sycamore ‘dormin’ with abscisin II. Nature 205, 1269–1270. Cutler A.J. & Krochko J.E. (1999) Formation and breakdown of ABA. Trends in Plant Science 4, 472–478. González-Guzmán M., Apostolova N., Bellés J.M., Barrero J.M., Piqueras P., Ponce M.R., Micol J.L., Serrano R. & Rodríguez P.L. (2002) The short-chain alcohol dehydrogenase ABA2 catalyzes the conversion of xanthoxin into abscisic aldehyde. Plant Cell 14, 1833–1846. González-Guzmán M., Abia D., Salinas J., Serrano R. & Rodríguez P.L. (2004) Two new alleles of the abscisic aldehyde oxidase 3 gene reveal its role in abscisic acid biosynthesis in seeds. Plant Physiology 135, 325–333. Iuchi S., Kobayashi M., Taji T., Naramoto M., Seki M., Kato T., Tabata S., Kakubari Y., Yamaguchi-Shinozaki K. & Shinozaki K. (2001) Regulation of drought tolerance by gene manipulation of 9-cis-epoxycarotenoid dioxygenase, a key enzyme in abscisic acid biosynthesis in Arabidopsis. Plant Journal 27, 325–333. Iuchi S., Kobayashi M., Yamaguchi-Shinozaki K. & Shinozaki K. (2000) A stress-inducible gene for 9-cis-epoxycarotenoid dioxygenase involved in abscisic acid biosynthesis under water stress in drought-tolerant cowpea. Plant Physiology 123, 553–562. Kushiro T., Okamoto M., Nakabayashi K., Yamagishi K., Kitamura S., Asami T., Hirai N., Koshiba T., Kamiya Y. & Nambara E. (2004) The Arabidopsis cytochrome P450 CYP707A encodes ABA 8′-hydroxylases: key enzymes in ABA catabolism. EMBO Journal 23, 1647–1656. Livak K.J. & Schmittgen T.D. (2001) Analysis of relative gene expression data using real-time quantitative PCR and the 2-∆∆CT method. Methods 25, 402–408. Marin E., Nussaume L., Quesada A., Gonneau M., Sotta B., Hugueney P., Frey A. & Marion-Poll A. (1996) Molecular identification of zeaxanthin epoxidase of Nicotiana plumbaginifolia, a gene involved in abscisic acid biosynthesis and corresponding to the ABA locus of Arabidopsis thaliana. EMBO Journal 15, 2331–2342. Milborrow B.V. (2001) The pathway of biosynthesis of abscisic acid


2008 J. M. Barrero et al. in vascular plants: a review of the present state of knowledge of ABA biosynthesis. Journal of Experimental Botany 52, 1145– 1164. Nambara E. & Marion-Poll A. (2005) Abscisic acid biosynthesis and catabolism. Annual Review of Plant Biology 56, 165–185. Ohkuma K., Lyon J.L., Addicott F.T. & Smith O.E. (1963) Abscisin II, an abscission-accelerating substance from young cotton fruit. Science 142, 1592–1593. Pérez-Pérez J.M., Ponce M.R. & Micol J.L. (2004) The ULTRACURVATA2 gene of Arabidopsis encodes an FK506binding protein involved in auxin and brassinosteroid signaling. Plant Physiology 134, 101–117. Ponce M.R., Quesada V. & Micol J.L. (1998) Rapid discrimination of sequences flanking and within T-DNA insertions in the Arabidopsis genome. Plant Journal 14, 497–501. Qin X. & Zeevaart J.A. (1999) The 9-cis-epoxycarotenoid cleavage reaction is the key regulatory step of abscisic acid biosynthesis in water-stressed bean. Proceedings of the National Academy of Sciences USA 96, 15354–15361. Qin X. & Zeevaart J.A. (2002) Overexpression of a 9-cisepoxycarotenoid dioxygenase gene in Nicotiana plumbaginifolia increases abscisic acid and phaseic acid levels and enhances drought tolerance. Plant Physiology 128, 544–551. Quesada V., Ponce M.R. & Micol J.L. (1999) OTC and AUL1, two convergent and overlapping genes in the nuclear genome of Arabidopsis thaliana. FEBS Letters 461, 101–106. Quesada V., Ponce M.R. & Micol J.L. (2000) Genetic analysis of salt-tolerant mutants in Arabidopsis thaliana. Genetics 154, 421– 436. Quesada V., García-Martínez S., Piqueras P., Ponce M.R. & Micol J.L. (2002) Genetic architecture of NaCl tolerance in Arabidopsis thaliana. Plant Physiology 130, 951–963. Schwartz S.H., Tan B.C., Gage D.A., Zeevaart J.A. & McCarty D.R. (1997) Specific oxidative cleavage of carotenoids by VP14 of maize. Science 276, 1872–1874. Seo M. & Koshiba T. (2002) Complex regulation of ABA biosynthesis in plants. Trends Plant Science 7, 41–48.

Seo M., Koiwai H., Akaba S., Komano T., Oritani T., Kamiya Y. & Koshiba T. (2000a) Abscisic aldehyde oxidase in leaves of Arabidopsis thaliana. Plant Journal 23, 481–488. Seo M., Peeters A.J., Koiwai H., Oritani T., Marion-Poll A., Zeevaart J.A., Koornneef M., Kamiya Y. & Koshiba T. (2000b) The Arabidopsis aldehyde oxidase 3 (AAO3) gene product catalyzes the final step in abscisic acid biosynthesis in leaves. Proceedings of the National Academy of Sciences USA 97, 12908– 12913. Shinozaki K., Yamaguchi-Shinozaki K. & Seki M. (2003) Regulatory network of gene expression in the drought and cold stress responses. Current Opinion in Plant Biology 6, 410–417. Tan B.C., Joseph L.M., Deng W.T., Liu L., Li Q.B., Cline K. & McCarty D.R. (2003) Molecular characterization of the Arabidopsis 9-cis epoxycarotenoid dioxygenase gene family. Plant Journal 35, 44–56. Taylor I.B., Burbidge A. & Thompson A.J. (2000) Control of abscisic acid synthesis. Journal of Experimental Botany 51, 1563– 1574. Xiong L. & Zhu J.K. (2003) Regulation of abscisic acid biosynthesis. Plant Physiology 133, 29–36. Xiong L., Ishitani M., Lee H. & Zhu J.K. (2001) The Arabidopsis LOS5/ABA3 locus encodes a molybdenum cofactor sulfurase and modulates cold stress – and osmotic stress – responsive gene expression. Plant Cell 13, 2063–2083. Xiong L., Lee H., Ishitani M. & Zhu J.K. (2002) Regulation of osmotic stress-responsive gene expression by the LOS6/ABA1 locus in Arabidopsis. Journal of Biological Chemistry 277, 8588– 8596. Zeevaart J.A. & Creelman R.A. (1988) Metabolism and physiology of abscisic acid. Annual Review of Plant Physiology and Plant Molecular Biology 39, 439–473. Zhu J.K. (2002) Salt and drought stress signal transduction in plants. Annual Review of Plant Biology 53, 247–273. Received 8 March 2006; received in revised form 8 June 2006; accepted for publication 25 June 2006
