Characterization of an Arabidopsis thaliana Mutant ...

Plant CellPhysiol. 39(8): 853-858 (1998) JSPP © 1998

Characterization of an Arabidopsis thaliana Mutant that Has a Defect in ABA Accumulation: ABA-Dependent and ABA-Independent Accumulation of Free Amino Acids during Dehydration Eiji N a m b a r a 1 - 2 , Hiroshi K a w a i d e 2 , Yuji K a m i y a 2 and Satoshi N a i t o 1 2

li3

Department of Applied Bioscience, Faculty of Agriculture, Hokkaido University, Sapporo, 060-8589 Japan Laboratory for Plant-Hormone Function, Frontier Research Program, The Institute of Physical and Chemical Research (RIKEN), Wako, Saitama, 351-0198 Japan

Key words: ABA — Arabidopsis — Branched-chain amino acids — Dehydration.

The phytohormone ABA is a sesquiterpenoid and is synthesized from xanthophylls (Zeevaart and Creelman 1988). ABA plays prominent roles in various physiological processes including induction of seed dormancy and adaptive responses to environmental stresses. ABA accumulates during seed development as well as during dehydration in vegetative tissues and decreases rapidly upon subsequent

seed germination or rehydration. ABA-deficient mutants have been reported in a number of plant species and have been used in elucidating the biosynthetic pathway of ABA in higher plants (Zeevaart and Creelman 1988, Taylor 1991). However, how the ABA signal is transduced into a physiological response is still unclear. One difficulty to understand the molecular basis of ABA signal transduction may be due to the complexity of the responses mediated by this hormone. For example, gene expression during seed maturation and stress responses in vegetative tissues have been shown to be regulated in both an ABA-dependent and an ABA-independent manner (Yamaguchi-Shinozaki and Shinozaki 1994, Keith et al. 1994, West et al. 1994, Ishitani et al. 1997). In plants, the levels of free amino acids are tightly regulated by metabolic and developmental factors, and are maintained at specific concentrations within the cell. However, some of the amino acids over-accumulate when a plant is subject to osmostress (Delauney and Verma 1993, Rhodes and Hanson 1993). Proline is one of the well characterized amino acids that accumulate during osmostress and is thought to play a role as an osmoprotectant or as an energy source (Delauney and Verma 1993). Proline accumulation is achieved by both activation of biosynthesis and inhibition of its degradation (Yoshiba et al. 1995, Kiyosue et al. 1996). In A. thaliana, ABA is involved in activation of the gene encoding A '-pyrroline-5-carboxylate synthase, the key enzyme of proline biosynthesis (Yoshiba et al. 1995). A notable feature of such a metabolic response to osmostress is that ABA dependence of proline accumulation is different among plant species (Stewart and Voetberg 1987). Moreover, some plant species accumulate different amino acids in combination with proline during osmotic stress (Hanson and Hitz 1982, Rhodes and Hanson 1993). We previously reported the isolation of ABA-insensitive mutants of A. thaliana which germinate in the presence of a gibberellin (GA) biosynthetic inhibitor (Nambara et al. 1992, 1994). Here, we report the isolation and characterization of an ABA-deficient mutant and demonstrate the accumulation of free amino acids by ABA-dependent and ABA-independent regulation during dehydration.

Abbreviations: DAI, days after imbibition; GA, gibberellin. To whom correspondence should be addressed, e-mail: naito® abs.agr.hokudai.ac.jp

3

853

Downloaded from http://pcp.oxfordjournals.org/ by guest on June 10, 2012

In an attempt to elucidate the physiological role of ABA in seed dormancy and the adaptive response to dehydration, we isolated an ABA-deficient mutant of Arabidopsis thaliana (L.) Heynh. which germinated in the presence of a gibberellin biosynthetic inhibitor. Genetic analysis showed this mutation is a new allele of a recently reported locus aba2, and therefore has been designated aba2-2. The levels of endogenous ABA in fresh and dehydrated tissues of the aba2-2 mutant were highly reduced compared to those of wild-type plants. As a consequence, aba2-2 plants wilt and produce seeds with reduced dormancy. Dark germinated seedlings of the aba2-2 mutant showed true leaves, which were not observed in those of the wild type, indicating that abal-2 embryos grew precociously during seed maturation. In the dehydrated tissues of the wildtype plants, the levels of free proline, isoleucine and leucine were elevated to a content approximately 100-fold higher than those in fresh tissues. In contrast to the wild-type plants, dehydration-induced accumulation of proline was highly suppressed in the aba2-2 mutant plants while that of leucine and isoleucine accumulated. Furthermore, exogenous application of ABA to wild-type plants promoted accumulation of free proline, but not leucine nor isoleucine. These results suggest that dehydration-induced accumulation of free leucine and isoleucine is achieved independent of ABA.

Physiological role of ABA in Arabidopsis'

854

Materials and Methods

Phenotypes of the aba2-2 mutant—The aba2-2 mutant plants produced seeds that are phenotypically similar to known ABA biosynthetic mutants in that the seeds have reduced dormancy and adult plants wilt under desiccating conditions. After bolting, the wilty phenotype became evident in the shoot apex, cauline leaves and immature siliques of this mutant. This phenotype was partially restored by application of 10 pM ABA to rosette leaves (data not shown). The wilty phenotype of the aba2-2 mutant plants was quite similar to that of the aba2-l mutant (Fig. 1). The aba2-2 mutant has smaller rosette leaves and mature plants are not yellowish as has been reported for abal mutants

•s

Results Isolation and genetic analysis of EN7—ABA-deficient and ABA-insensitive mutants can germinate in the presence of uniconazole, a GA biosynthetic inhibitor, and these conditions have been successfully used to identify new mutants (Nambara et al. 1992,1994). One strain, EN7, showed a wilty phenotype in adult tissues. EN7 seeds germinate in the presence of uniconazole but not in the presence of 2//M of exogenous ABA, indicating that EN7 is responsive to ABA. F] plants generated by a cross between EN7 and wild-type plants were not wilty (data not shown) and among 109 F 2 seeds, 28 germinated in the presence of uniconazole. Moreover, all uniconazole-insensitive seeds co-

40-

20 2

4

Dehydration (h)

Fig. 1 Loss of fresh weight during dehydration. The aerial parts of 30 DAI plants were detached and dehydrated on paper towels for various periods as indicated. The fresh weight after dehydration (% of initial FW) was calculated by (weight after dehydration/weight before dehydration) x 100 (%). Averages±SD of 4 independent plants are shown, - o - , Wild type, - • - , aba2-l mutant, - • - , aba2-2 mutant.

Downloaded from http://pcp.oxfordjournals.org/ by guest on June 10, 2012

Plant materials and growth conditions—Arabidopsis thaliana (L.) Heynh. ecotype Columbia carrying a gll mutation (Col-g/7) and M 2 seeds of Col-gll mutagenized by fast neutron irradiation were purchased from Lehle Seeds (Tucson, AZ, U.S.A.)- The mutant strain EN7 was isolated from the M2 seeds of Col-gll in this study and carries an aba2-2 mutation (see below). Plant growth conditions have been described previously (Nambara et al. 1995). Genetic analysis of the EN7 mutant—To map the mutation, EN7 was crossed with Landsberg erecta (Ler), and 34 wilty plants were obtained in the F 2 generation. PCR-based mapping (Konieczny and Ausubel 1993, Bell and Ecker 1994) was performed in the 34 F 2 lines. PCR primers were purchased from Research Genetics Inc. (U.K.). Physiological analyses—Dehydration treatments for the transpiration rate assay and free amino acid quantification were performed as follows. Plants at 30 d after imbibition (DAI) were detached, and the aerial parts were dehydrated on paper towels for various periods of time under ambient humidity. Plants were weighed before and after the dehydration treatment and the tissues were used for subsequent analysis. The procedure for dark germination analysis has been described previously (Nambara et al. 1995). Exogenous ABA at 100//M was added to the nutrient medium. Care was taken not to water the leaves with ABA-containing medium. Extraction, purification and quantification of endogenous ABA—These procedures were essentially the same as those for GA analysis (Gawronska et al. 1995). In short, 50 ng of [6-2H3]ABA was added to the extracts. Authentic ABA was used to determine the elution position of ABA in a separate injection. Samples were methylated with etherial diazomethane for subsequent gas chromatography-selected ion monitoring. The oven temperature was kept at 80°C for 1 min, and then increased to 200GC at a rate of 30°C min" 1 followed by a further increment to 230°C at 5°C min" 1 . Three prominent ion pairs (m/z= 134/137, m / z = 162/165 and m / z = 190/193) were monitored with dwell time of 50 ms. The endogenous contents of ABA were calculated from the peak area ratios of m/z=190 (endogenous ABA) and 193 (trideuterated ABA). Extraction and quantification of free amino acids—The procedures for the extraction and quantification of free amino acids have been described previously (Inaba et al. 1994). Values for amino acid content are shown relative to the FW of plants prior to dehydration.

segregated a wilty phenotype. These results are consistent with the interpretation that both wiltiness and uniconazole insensitivity in EN7 are caused by the same recessive mutation. EN7 was crossed with Ler, and F2 lines were used for PCR-based mapping (Konieczny and Ausubel 1993, Bell and Ecker 1994). The mutation carried by EN7 was linked to the markers on chromosome 1, nga280 and GAPB (Konieczny and Ausubel 1993, Bell and Ecker 1994). In A. thaliana, three loci (abal, aba2 and aba3) which confer ABA deficiency have been reported (Koornneef et al. 1982, Leon-Kloosterziel et al. 1996, Schwartz et al. 1997), and of these the aba2 mutation is genetically linked to GAPB. To determine if EN7 carries a mutation in a new locus, the mutant was reciprocally crossed with abal-1 and aba2-l mutants, and the F, plants were used for allelism tests by measuring the plant weights after dehydration. The results showed that the abal-1 mutant, but not the aba2-l mutant, complemented the wilty phenotype of EN7 (data not shown). Based on these results, we concluded that EN7 is a new aba2 allele which we designated aba2-2.

Physiological role of ABA in Arabidopsis

855

Table 1 Quantification of ABA in fresh and dehydrated tissues Genotype Wild type aba2-l aba2-2

ABA content (ng (g FW)~ Fresh tissue Dehydrated tissue Exp. 1" Exp. 2 ' Exp. 1* Exp. 2l 5.4 ND