Evolutionarily conserved regulatory mechanisms of abscisic acid signaling in land plants: characterization of ABSCISIC ACID INSENSITIVE1-like type 2C protein phosphatase in the liverwort Marchantia polymorpha

Abstract

Abscisic acid (ABA) is postulated to be a ubiquitous hormone that plays a central role in seed development and responses to environmental stresses of vascular plants. However, in liverworts (Marchantiophyta), which represent the oldest extant lineage of land plants, the role of ABA has been least emphasized; thus, very little information is available on the molecular mechanisms underlying ABA responses. In this study, we isolated and characterized MpABI1, an ortholog of ABSCISIC ACID INSENSITIVE1 (ABI1), from the liverwort Marchantia polymorpha. The MpABI1 cDNA encoded a 568-amino acid protein consisting of the carboxy-terminal protein phosphatase 2C (PP2C) domain and a novel amino-terminal regulatory domain. The MpABI1 transcript was detected in the gametophyte, and its expression level was increased by exogenous ABA treatment in the gemma, whose growth was strongly inhibited by ABA. Experiments using green fluorescent protein fusion constructs indicated that MpABI1 was mainly localized in the nucleus and that its nuclear localization was directed by the amino-terminal domain. Transient overexpression of MpABI1 in M. polymorpha and Physcomitrella patens cells resulted in suppression of ABA-induced expression of the wheat Em promoter fused to the beta -glucuronidase gene. Transgenic P. patens expressing MpABI1 and its mutant construct, MpABI1-d2, lacking the amino-terminal domain, had reduced freezing and osmotic stress tolerance, and associated with reduced accumulation of ABA-induced late embryogenesis abundant-like boiling-soluble proteins. Furthermore, ABA-induced morphological changes leading to brood cells were not prominent in these transgenic plants. These results suggest that MpABI1 is a negative regulator of ABA signaling, providing unequivocal molecular evidence of PP2C-mediated ABA response mechanisms functioning in liverworts.

Figure 1.

Effect of ABA on growth of the M. polymorpha gametophyte. Gemmae of M. polymorpha gametophytes were grown on control M51C agar medium or medium containing 0.1, 0.2, 0.5, or 1.0 μm ABA. After photographs had been taken, the gametophyte tissues were used for measurement of fresh weight. The tissues were then dried at 90°C for 24 h for measurement of dry weight. A, Photographs of representative gametophytes grown for 4 weeks. Bars = 10 mm. B, Fresh weight and dry weight of gametophytes grown for 4 weeks (n = 5).

Figure 2.

Molecular characteristics of MpABI1. A, Sequence alignment of MpABI1 and Arabidopsis ABI1. The alignment was made by the T-COFFEE program. Identical amino acids are marked with asterisks, very similar ones are marked with colons, and weakly similar ones are marked with dots. Conserved motifs 1 to 11 of PP2Cs assigned by Bork et al. (1996) are indicated by overlines. Putative NLSs are underlined. B, Phylogenetic representations of MpABI1, PpABI1A, PpABI1B, and Arabidopsis ABI1-related PP2Cs. The phylogenetic tree was built on the alignment of amino acid sequences of the PP2C domains using the ClustalW program. The Arabidopsis Ap2C1 phosphatase sequence was used as the outgroup. The bar indicates 0.1 substitutions per site. C, Northern analysis of M. polymorpha mRNA for examination of MpABI1 expression. Total RNAs isolated from ABA-treated (+) or nontreated (−) gemma and thallus were electrophoresed, blotted onto a nylon membrane, and hybridized with a ³²P-labeled MpABI1 cDNA probe. Ethidium bromide-stained rRNA of each sample is shown to verify equal loading.

Figure 3.

Protein phosphatase assays of GST-MpABI1 fusion proteins. A, Mg²⁺- and Mn²⁺-dependent activities of GST-MpABI1 toward the ³²P-labeled MBP. B, Schematic representation of various GST-MpABI1 constructs used for protein phosphatase assays. C, SDS-PAGE of GST fusion proteins of MpABI1 and its mutant proteins. The proteins were stained with Coomassie Brilliant Blue after electrophoresis. Positions of the molecular mass markers are shown on the left. Lane 1, GST-MpABI1; lane 2, GST-MpABI1-d1; lane 3, GST-MpABI1-d2; lane 4, GST-MpABI1-G298D. D, Protein phosphatase activity of wild-type and mutant GST fusion proteins of MpABI1 toward the ³²P-labeled MBP (n = 3). Phosphatase reactions were carried out at 30°C for 10 min and terminated by the addition of TCA. Activity is represented as a percentage of [³²P]orthophosphate released by GST-MpABI1 in the presence of 5 mm Mg²⁺.

Figure 4.

Localization study using GFP fusion proteins. A, Schematic representation of GFP fusion constructs of MpABI1 and its mutant constructs used for analysis. B, Cellular localization of MpABI1-GFP. The DNA construct for the MpABI1-GFP fusion protein was bombarded into epidermal cells of onion, and the cells were observed after 1 d of incubation at 25°C in the dark. Bright-field, GFP fluorescence, and merged images are shown. Bars = 30 μ m. C, N-terminal domain-directed nuclear localization of MpABI1. Fluorescent images show onion epidermal cells bombarded with DNA constructs for the MpABI1 N-terminal domain (MpABI1-N; amino acids 1–225), MpABI1-N (Δ 7–45) lacking NLS, and the C-terminal PP2C domain (MpABI1-C; amino acids 213–568) fused to the upstream region of the GFP coding sequence. Bars = 50 μ m.

Figure 5.

Effect of transient overexpression of MpABI1 on ABA-induced gene expression. A, Constructs used for transient assays. The wheat Em promoter (Em-p) fused to the GUS reporter gene was used as a reporter construct for the detection of ABA-induced gene expression (Marcotte et al., 1989). The rice actin promoter (Actin-p) fused to MpABI1, MpABI-d1, MpABI1-d2, or MpABI1-G298D was used as an effector construct to see their effects on ABA-induced GUS gene expression. The rice ubiquitin promoter fused to the luciferase gene was used as a control construct (data not shown). Plasmid DNAs with these constructs were cobombarded into M. polymorpha or P. patens cells using the PDS-1000He particle delivery system, and the cells were then incubated with or without ABA (for details, see “Materials and Methods”). Protein fractions prepared from the cells were used for GUS and LUC assays, and the relative GUS activity-to-LUC activity ratios (GUS/LUC) are represented in B to D. B, Effect of different concentrations of ABA on Em-GUS expression in M. polymorpha cells. C, Effect of MpABI1 overexpression on 1- μm ABA-induced Em-GUS expression in M. polymorpha cells. D, Effect of overexpression of wild-type MpABI1 and deletion mutants MpABI1-d1, MpABI1-d2, and MpABI1-G298D on ABA-induced Em-GUS expression in P. patens.

Figure 6.

Northern-blot analysis of transgenic P. patens plants expressing MpABI1 and MpABI1-d2. Total RNAs extracted from nontransgenic (NT) and independent transgenic lines expressing intact MpABI1 (W-1 to W-6) and MpABI1-d2 (D2-1 to D2-4) were electrophoresed and blotted onto a nylon membrane. The blotted RNAs were hybridized with ³²P-labeled probes for either the N-terminal domain region (N) or the C-terminal domain region (C) of MpABI1 cDNA. Ethidium bromide-stained rRNA for each sample is also shown.

Figure 7.

Effect of ABA treatment on osmotic and freezing stress-induced damage of control and transgenic P. patens plants. Protonema cells of nontransgenic control (NT) and independent transgenic lines expressing wild-type MpABI1 (W-1 to W-6) and MpABI1-d2 (D2-1 to D2-4) were incubated for 1 d with or without 10 μm ABA and used for osmotic stress and freezing treatments. A, Representative results of growth of cells that had been either pretreated with 10 μm ABA (+) or nontreated (−) prior to the osmotic stress treatment by the indicated concentrations of mannitol or NaCl. After 15 min of osmotic stress treatment at room temperature, a piece of protonemata was planted on fresh BCD agar medium and grown at 25°C for 1 week. B, Representative results of growth after freezing of nontransgenic and transgenic plants that had been pretreated with 10 μm ABA for 1 d. All samples were ice inoculated at −1°C and subjected to equilibrium freezing to −5°C. After thawing, a piece of protonemata was planted on fresh BCD agar medium and grown at 25°C for 1 week.

Figure 8.

Freezing damage of nontransgenic (NT) and representative transgenic lines expressing different levels of transcripts of MpABI1 (W2, W3, and W4) and MpABI1-d2 (D2-1 and D2-2). Protonema cells were incubated for 1 d with or without 10 μm ABA and subjected to freezing. The cells were then ice inoculated at −1°C and subjected to equilibrium freezing at a rate of −2.4°C h⁻¹ to desired temperatures. After thawing, electrolyte leakage was measured and survival was estimated as described previously (Nagao et al., 2005). A and B, Comparison of electrolyte leakage (%) of unfrozen and frozen plants that had been pretreated with 10 μm ABA for 1 d prior to freezing. A, Comparison of nontransgenic, W2, W3, and W4 plants. B, Comparison of nontransgenic, D2-1, and D2-2 plants. C and D, Nontreated (C) or 10 μm ABA-treated (D) protonema cells of nontransgenic, W-2, and D2-1 plants were frozen to −3°C, −6°C, and −9°C, and survival (%) after thawing was estimated by electrolyte leakage measurement.

Figure 9.

Analysis of ABA-induced soluble proteins in control (nontransgenic [NT]) and transgenic plants, W-2 and D2-1. Total soluble proteins (left) and boiling-soluble proteins (right) were subjected to SDS-PAGE and stained with Coomassie Brilliant Blue. Positions of molecular mass markers are shown in kD. B, The boiling-soluble proteins were used for immunoblot analysis after electroblotting onto an Immobilon-P membrane. The blotted proteins were reacted with antiserum raised against the recombinant 17B9 protein. Signals were detected using an alkaline phosphatase-conjugated secondary antibody and nitroblue tetrazolium/5-bromo-4-chloro-3-indolyl phosphate color-developing reagent.

Figure 10.

ABA-induced morphological changes of protonema cells in control and transgenic plants. The protonemata of control, W-2, and D2-1 plants were incubated in normal BCD medium or the medium containing 10 μm ABA for 7 d. A, Nontreated control protonemata. B, ABA-treated control protonemata. C, ABA-treated W-2 protonemata. D, ABA-treated D2-1 protonemata. Bars = 100 μ m. [See online article for color version of this figure.]