Plant, Cell and Environment (2008) 31, 227–234
doi: 10.1111/j.1365-3040.2007.01759.x
The ABA1 gene and carotenoid biosynthesis are required for late skotomorphogenic growth in Arabidopsis thaliana JOSÉ MARÍA BARRERO1*, PEDRO L. RODRÍGUEZ2*, VÍCTOR QUESADA1*, DAVID ALABADÍ2,3, MIGUEL A. BLÁZQUEZ2, JEAN-PIERRE BOUTIN4, ANNIE MARION-POLL4, MARÍA ROSA PONCE1 & JOSÉ LUIS MICOL1 1
División de Genética and Instituto de Bioingeniería, Universidad Miguel Hernández, Campus de Elche, 03202 Elche, Alicante, Spain, 2Instituto de Biología Molecular y Celular de Plantas, Universidad Politécnica de Valencia-Consejo Superior de Investigaciones Científicas, 46022 Valencia, Spain, 3Fundación de la Comunidad Valenciana para la Investigación Agroalimentaria (Agroalimed), 46113 Moncada, Valencia, Spain and 4Seed Biology Laboratory, UMR 204 INRA-AgroParisTech, Jean-Pierre Bourgin Institute, 78026 Versailles Cedex, France
ABSTRACT
INTRODUCTION
Several plant hormones, including auxin, brassinosteroids and gibberellins, are required for skotomorphogenesis, which is the etiolated growth that seedlings undergo in the absence of light. To examine the growth of abscisic acid (ABA)-deficient mutants in the dark, we analysed several aba1 loss-of-function alleles, which are deficient in zeaxanthin epoxidase. The aba1 mutants displayed a partially de-etiolated phenotype, including reduced hypocotyl growth, cotyledon expansion and the development of true leaves, during late skotomorphogenic growth. In contrast, only small differences in hypocotyl growth were found between wild-type seedlings and ABA-deficient mutants impaired in subsequent steps of the pathway, namely nced3, aba2, aba3 and aao3. Interestingly, phenocopies of the partially de-etiolated phenotype of the aba1 mutants were obtained when wild-type seedlings were dark-grown on medium supplemented with fluridone, an inhibitor of phytoene desaturase, and hence, of carotenoid biosynthesis. ABA supplementation did not restore the normal skotomorphogenic growth of aba1 mutants or fluridone-treated wild-type plants, suggesting a direct inhibitory effect of fluridone on carotenoid biosynthesis. In addition, aba1 mutants showed impaired production of the b-carotenederived xanthophylls, neoxanthin, violaxanthin and antheraxanthin. Because fluridone treatment of wild-type plants phenocopied the phenotype of dark-grown aba1 mutants, impaired carotenoid biosynthesis in aba1 mutants is probably responsible for the observed skotomorphogenic phenotype. Thus, ABA1 is required for skotomorphogenic growth, and b-carotene-derived xanthophylls are putative regulators of skotomorphogenesis.
The control that light exerts upon a plant ranges from its use as an energy source to its effects on specific developmental programs (Chen, Chory & Fankhauser 2004). One of the earliest events after germination involves the establishment of a morphogenic developmental plan. In the absence of light, seedlings grow in an etiolated form in a process known as skotomorphogenesis, which is characterized by enhanced cell expansion resulting in longer hypocotyls, and the maintenance of an apical hook with folded cotyledons. In contrast, light triggers the switch to photomorphogenesis, which is accompanied by a cessation of cell expansion, the formation of new organs and massive changes in gene expression, including the induction of genes involved in photosynthesis and chloroplast function (Casal et al. 2004). In Arabidopsis thaliana, photomorphogenesis is repressed in the dark by a mechanism dependent on the activity of COP1 (CONSTITUTIVE PHOTOMORPHOGENESIS 1), an E3 ubiquitin ligase that triggers the degradation of a battery of transcription factors necessary for the expression of light-regulated genes (Hoecker 2005).The repression of photomorphogenesis and the promotion of skotomorphogenesis are also under hormonal control (Kim, Kim & von Arnim 2002). For instance, brassinosteroid (BR)-deficient and BR-insensitive mutants display a de-etiolated phenotype in the dark, that is, shorter hypocotyls, cotyledon unfolding, and derepression of lightregulated genes such as CAB2 (CHLOROPHYLL A/B– BINDING PROTEIN2) and RbcS (RubisCO smallsubunit) (Clouse & Sasse 1998; Schumacher & Chory 2000). Several lines of evidence implicate auxins in the regulation of photomorphogenesis; for example, mutations in the auxin-regulated gene FIN219 (FAR-RED INSENSITIVE 219) suppress the de-etiolated phenotype of cop1 in the dark (Hsieh et al. 2000), and phytochromes have been reported to phosphorylate auxin-signalling elements (Colón-Carmona et al. 2000). The involvement of auxin in photomorphogenesis is in consonance with a large overlap in the regulation of gene expression by auxin and BRs
Key-words: ABA; aba1 mutants; skotomorphogenesis.
Correspondence: J. L. Micol. Fax: +34 96 665 85 11; e-mail:
[email protected] *These authors contributed equally to this work. © 2007 The Authors Journal compilation © 2007 Blackwell Publishing Ltd
227
228 J. M. Barrero et al. (Nemhauser, Mockler & Chory 2004). Gibberellins (GAs) are also implicated as repressors at photomorphogenic development. As one recent example, in the dark, impairment of GA biosynthesis and the activation of negative GA-signalling elements provoked de-etiolation that could be partially reverted by the exogenous application of BRs (Alabadí et al. 2004). Carotenoids are a large class of isoprenoid-derived compounds that have critical roles in all photosynthetic organisms (Hirschberg 2001). Xanthophylls, which are oxygenated carotenes, are accessory pigments in the lightharvesting antennae of the chloroplast, and function to
transfer photon energy to chlorophylls. Two branches of xanthophylls are generated from a- and b-carotene (Fig. 1). In particular, aba1 mutants are impaired in the epoxidation of zeaxanthin, a xanthophyll derived from b-carotene, to antheraxanthin, and to all-trans-violaxanthin. Despite the high concentration of carotenoids in etioplasts, and evidence showing that carotenoid biosynthesis affects chloroplast biogenesis (Park et al. 2002), little is known about the actual role of carotenoids in etiolated growth. We examined the phenotype of carotenoid and abscisic acid (ABA) biosynthetic mutants grown in the dark and found that the loss of function of the ABA1 gene, which
Phytoene
PDS
Fluridone
z− Carotene Neurosporene
LUT2
Lycopene g -Carotene
d -Carotene
b-Carotene
a-Carotene
OH
H 3C H H3C
C H33 CH
C H33 CH
CH33
Zeinoxanthin CH C H33
ABA1 CH C H33
H33C
CH C H33
H 3C
C H33 CH
C H33 CH
CH C H33
OH OH
O
O CH C H33
CH C H33 HO
CH C H33
Antheraxanthin
ABA1 H 3C
OH
O
CH C H33 HO
H H3C C
C H33 CH
C H33 CH
C H33 CH
CH C H33
Zeaxanthin
Lutein H H3C C
H33C
CH C H33
CH C H33 HO
H33C
CH33
CH C H33
All-trans-violaxanthin CH 3
H3C
CH3
H 3C H 3C
CH 3
O O
HO
H3C
CH3
OH
CH 3 H 3C H 3C CH3
9′-cis-neoxanthin
9-cis-violaxanthin
H 3C CH3 H 3C
CH3
O
H 3C
CH 3 H 3C
HO OH H 3C
OH
NCED3 CH CH33
C H33 CH
H 3C
O CHO H HO
C H33 CH
Xanthoxin
ABA2 C H33 CH
C H33 CH
H 3C
OH CHO O
CH C H33
Abscisic aldehyde
AAO3, ABA3 H3 C H 3
C H3 3
C H3 3
OH OH COOH O
C H3 3
ABA
Figure 1. Carotenoid and abscisic acid (ABA) biosynthetic pathways. The inhibitory effect of fluridone on phytoene desaturase (PDS) is indicated. LUT2 is a e-cyclase. The steps in the ABA biosynthetic pathway catalysed by ABA1, NCED3, ABA2, AAO3 and ABA3 are indicated.
© 2007 The Authors Journal compilation © 2007 Blackwell Publishing Ltd, Plant, Cell and Environment, 31, 227–234
De-etiolation in aba1 mutants 229 encodes zeaxanthin epoxidase (ABA1), partially removes the growth restraint imposed by the absence of light. This leads to the promotion of cotyledon expansion and to organogenesis. This phenotypic defect was not reverted by exogenous ABA, indicating that xanthophylls are part of the machinery that blocks plant growth in the absence of external stimuli.
MATERIALS AND METHODS Plant material and growth conditions Wild-type (Ler, Col-0 and Ws-2) and mutant Arabidopsis thaliana (L.) Heynh. plants were grown under sterile conditions on 150 mm Petri dishes containing 100 mL of agar medium at 20 ⫾ 1 °C, 60–70% relative humidity, and continuous illumination at 7000 lx (98 mmol m-2 s-1), as described in Ponce, Quesada & Micol (1998). The nced3-1 and lut2-1 mutants were kindly provided by H. Koiwa (Texas A&M University, College Station, TX, USA) and by P. Jahns (Heinrich Heine University, Düsseldorf, Germany), respectively. The ABA-deficient mutants aba1, aba2, aba3 and aao3, as well as the aba1-101::pBIN19-ABA1 transgenic line, were described previously (Gonzalez-Guzman et al. 2002, 2004; Barrero et al. 2005, 2006).
Pigment analysis and quantification Extractions were carried out on 6 mm leaf discs taken from the rosette leaves of 3-week-old plants grown in soil in a glasshouse (~22 °C, minimum 13 h photoperiod, maximum light intensity of 500 mmol m-2 s-1), and in separate experiments, on 10- or on 21-day-old etiolated seedlings grown in the dark. The pigments were extracted in acetone and separated by HPLC, as described in North et al. (2005). The carotenoids and chlorophylls were separated by reversephase high-performance liquid chromatography (HPLC) on a System Gold HPLC (Beckman-Coulter France, Villepinte, France) using two Adsorbosphere HS C18 3 mm columns (Alltech, Carquefou, France) in series (100 ¥ 4.6 mm; 150 ¥ 4.6 mm). The pigments were detected using a photodiode-array detector (Beckman-Coulter France) and identified by the comparison of retention times and absorption spectra with published values (Britton 1995) or commercially available standards [zeaxanthin, lutein, b-carotene (ExtraSynthèse, Genay, France), and chlorophyll a and b (Fluka, Sigma-Aldrich Chimie, St. Quentin Fallavier, France)]. The pigments were quantified by the integration of their peak areas.
RESULTS Hypocotyl growth is impaired in dark-grown ABA-deficient mutants
Approximately 50–100 seeds of each genotype were sown on Petri dishes containing non-supplemented medium or medium supplemented with 50, 100 or 500 nm ABA (SigmaAldrich A1049, Sigma-Aldrich, St. Louis, MO, USA), or 5, 10 or 20 mm fluridone (Sigma-Aldrich 45511), or 50 nm ABA and 10 or 20 mm fluridone. To measure hypocotyl growth, 20 etiolated seedlings of each genotype grown on Petri dishes for 10 d in the dark were photographed and analysed using the Image-J program (http://rsb.info.nih.gov/ij/docs/menus/ file.html). For the de-etiolation assays, the phenotype of 50–100 seedlings grown on Petri dishes for 21 d in the dark was scored and photographed.
To further substantiate the proposed role of ABA in cell expansion (Sharp et al. 2000; Sharp & LeNoble 2002; Barrero et al. 2005), we analysed hypocotyl elongation in dark-grown ABA-deficient and wild-type plants. ABAdeficient mutants showed reduced hypocotyl growth in the dark compared to wild-type seedlings (Fig. 2). The exogenous application of ABA promoted hypocotyl growth in ABA-deficient mutants, whereas it had no significant effect on wild-type Columbia (Col-0) plants (Fig. 2). Conversely, the application of fluridone, an inhibitor of carotenoid and ABA biosynthesis, dramatically reduced hypocotyl growth both in wild-type and ABA-deficient plants (Fig. 2).
Northern blots Total RNA was extracted from frozen, whole 7-day-old dark-grown seedlings using the RNeasy Plant Mini Kit (Qiagen, Hilden, Germany). Two micrograms of total RNA was used for Northern analysis with 32P-labelled CAB2 and RbcS probes, synthesized as described in Alabadí et al. (2004). All procedures were carried out as previously described (Alabadí et al. 2004).
Hypocotyl length (mm)
ABA and fluridone treatments
20
0
500 nM ABA
5
10
20 mM fluridone
16 12 8 4 0 Col-0
aba3-101 aba2-14
aao3-2
aba1-101
Cotyledon angle measurements
Figure 2. Effects of exogenous abscisic acid (ABA) and of
Seedlings were placed on an acetate sheet and scanned at a resolution of 800 dots per inch. The angle between the cotyledons was measured from the images obtained (Alabadí et al. 2004).
fluridone on hypocotyl length in 10-day-old dark-grown wild-type and ABA-deficient seedlings. The histogram shows the mean (n = 20) for seedlings grown on medium supplemented with the indicated concentration of ABA or fluridone. All seedlings were homozygous for the indicated mutations.
© 2007 The Authors Journal compilation © 2007 Blackwell Publishing Ltd, Plant, Cell and Environment, 31, 227–234
230 J. M. Barrero et al. Because the aba1-101 mutant showed the greatest reduction in hypocotyl growth, we decided to analyse this phenotypic defect in greater detail in several aba1 alleles, which exist in different genetic backgrounds. In all alleles examined, the aba1 mutants displayed reduced hypocotyls length compared to their corresponding wild-type plants (Fig. 3a). This phenotypic effect was particularly noticeable in aba1101/aba1-101 and aba1-102/aba1-102 seedlings, which are in the Col-0 and Ws-2 genetic backgrounds, respectively.
(a)
otyl length (mm) Hypoco
(b)
25 0
20
500 nM ABA
15 10 5 0
(c)
Hy ypocotyl length (mm))
0
5
10
20 mM fluridone
15 10 5
Figure 3. Effects of abscisic acid (ABA) and fluridone on hypocotyl length in 10-day-old dark-grown wild-type and aba1 mutant seedlings. (a) Representative seedlings are shown. (b,c) Histograms of the mean (n = 20) for seedlings grown on medium supplemented with the indicated concentration of ABA or fluridone. All seedlings were homozygous for the mutations indicated. The seedlings were scored 10 d after sowing. Scale bar = 1 mm.
Supplementing the growth medium with exogenous ABA increased the hypocotyls length of all of the aba1 mutants examined (Fig. 3b). Conversely, fluridone inhibited hypocotyl elongation in all of the aba1 mutants and their corresponding wild-type plants (Fig. 3c).
The aba1 mutants exhibit a de-etiolated phenotype The short hypocotyl of dark-grown ABA-deficient mutants could be the result of a defect in either ABA-induced cell elongation, as mentioned earlier, or ABA regulation of the transition between skotomorphogenesis and photomorphogenesis, as shown for BRs and GAs (Li et al. 1996; Szekeres et al. 1996;Alabadí et al. 2004).To distinguish between these two alternatives, we analysed a number of classical photomorphogenesis markers, including the angle between cotyledons, and the expression of CAB2 and RbcS genes. For this, 7-day-old dark-grown ABA-deficient seedlings were assayed. The angle between the cotyledons in all of the ABA-deficient mutants was similar to that of the wild type, and no change was observed in terms of sensitivity to paclobutrazol (PAC), an inhibitor of GA biosynthesis (Fig. 4a). Similarly, CAB2 and RbcS expression was unaffected in the ABA-deficient mutant seedlings (Fig. 4b), in contrast to the reported effect of PAC on GA biosynthesis (Alabadí et al. 2004). CAB2 and RbcS expression was also unaffected in fluridone-treated wild-type seedlings (Fig. 4c). These data collectively lead us to suggest that ABA itself is not the major player in the transition from skotomorphogenesis to photomorphogenesis. We next found a requirement for ABA1 to repress photomorphogenesis in the dark. The aba1 mutants were found to be partially de-etiolated when grown in the dark for 21 d. They exhibited open cotyledons and developed true leaves (Fig. 5a and Table 1). Unexpanded cotyledons were shown in >89% of the dark-grown wild-type seedlings, but only in
