ABA signaling prevents phosphodegradation of the SR45 splicing factor to alleviate inhibition of early seedling development in Arabidopsis

Abstract

Serine/arginine-rich (SR) proteins are conserved splicing regulators that play important roles in plant stress responses, namely those mediated by the abscisic acid (ABA) hormone. The Arabidopsis thaliana SR-like protein SR45 is a described negative regulator of the ABA pathway during early seedling development. How the inhibition of growth by ABA signaling is counteracted to maintain plant development under stress conditions remains largely unknown. Here, we show that SR45 overexpression reduces Arabidopsis sensitivity to ABA during early seedling development. Biochemical and confocal microscopy analyses of transgenic plants expressing fluorescently tagged SR45 revealed that exposure to ABA dephosphorylates the protein at multiple amino acid residues and leads to its accumulation, due to SR45 stabilization via reduced ubiquitination and proteasomal degradation. Using phosphomutant and phosphomimetic transgenic Arabidopsis lines, we demonstrate the functional relevance of ABA-mediated dephosphorylation of a single SR45 residue, T264, in antagonizing SR45 ubiquitination and degradation to promote its function as a repressor of seedling ABA sensitivity. Our results reveal a mechanism that negatively autoregulates ABA signaling and allows early plant growth under stress via posttranslational control of the SR45 splicing factor.

Keywords: Arabidopsis thaliana; SR proteins; abscisic acid; alternative splicing; protein phosphorylation.

Figure 1

Effect of SR45 overexpression on ABA signaling during cotyledon development. (A) qRT-PCR analysis of SR45–GFP transcript levels in pUBQ10::SR45–GFP/sr45-1 (OX1 and OX2) and pSR45::SR45–GFP/sr45-1 (C1 and C2) transgenic seedlings, as well as in Col-0 wild-type and sr45-1 mutant seedlings grown for 7 days under control conditions, using PEX4 as a reference gene and primers annealing to either the SR45 or the GFP sequence (see Supplemental Figure 1). Transcript levels of either the Col-0 (SR45 primers) or the C2 transgenic line (GFP primers) were set to 1. Results represent means ± SE (n = 3), and different letters indicate statistically significant differences between genotypes for each set of primers (P < 0.05; Student’s t-test). (B) Protein gel blot analysis using α-GFP antibodies of the SR45–GFP fusion protein in 7-day-old seedlings of overexpression (OX1 and OX2) and complementation (C1 and C2) transgenic lines grown under control conditions. A total of 30 ng of protein was loaded per sample, and Ponceau staining is shown as a loading control. Results are representative of at least three independent experiments. (C) Representative images and quantification of cotyledon greening in 7-day-old seedlings of the Col-0 wild type, the sr45-1 mutant, the OX1 and OX2 overexpression lines, and the C1 and C2 complementation lines grown under control conditions or in the presence of 0.5 μM ABA (means ± SE, n = 3). Different letters indicate statistically significant differences between genotypes under each condition (P < 0.05; Student’s t-test). Scale bars, 2.5 mm (control) or 1 mm (ABA).

Figure 2

Effect of ABA on SR45 protein levels. (A) Protein gel blot analysis using α-GFP antibodies of the SR45–GFP fusion protein in 7-day-old seedlings of the C2 complementation transgenic line treated for 0, 30, 60, 90, or 180 min with 2 μM ABA. Control samples were treated with the equivalent volume of the solvent of the ABA solution (ethanol). A total of 40 ng of protein was loaded per sample. Bands were quantified and relative protein levels determined using the Ponceau loading control as a reference, with control conditions set to 1. Results represent means ± SE (n = 3), and different letters indicate statistically significant differences between treatments (P < 0.05; Student’s t-test). (B) qRT-PCR analysis of SR45–GFP transcript levels in 2-day-old seedlings of the C2 complementation line (SR45–GFP/sr45-1) treated for 180 min with 1 μM ABA, using PEX4 as a reference gene and primers annealing to the GFP sequence (see Supplemental Figure 1). Control samples (set to 1) were treated with the equivalent volume of the solvent of the ABA solution (ethanol). Results represent means ± SE (n = 3), with no statistically significant differences found between treatments (P > 0.05; Student’s t-test). (C) Sum Z projection images (top) and segmentation (bottom) of fast SR45–GFP accumulation in the primary root of 4-day-old seedlings of the C2 complementation transgenic line treated with 10 μM ABA observed by confocal microscopy. Scale bar, 100 μm. (D) Quantification of the nuclear intensity of the segmentation shown in (C). Line indicates mean values, with shaded region indicating the 95% confidence interval.

Figure 3

Effect of ABA on phosphorylation and amounts of the SR45 protein and dependence on SnRK2 function. Phos-tag (A) and protein (B) gel blot analyses using α-GFP antibodies of the SR45–GFP fusion protein in 7-day-old seedlings of the C2 complementation line (SR45–GFP/sr45-1) and a transgenic line expressing the pSR45:gSR45–GFP construct in the snrk2.2/3/6 mutant background (SR45–GFP/snrk2.2/3/6) treated for 180 min with 2 μM ABA. Control samples were treated with the equivalent volume of the solvent of the ABA solution (ethanol), and a total of 40 ng of protein was loaded per sample. In (A), results are representative of at least three independent experiments. In (B), bands were quantified and relative protein levels determined using the Ponceau loading control as a reference, with results representing means ± SE (n = 4), control conditions set to 1, and different letters indicating a statistically significant difference between treatments for each genotype (P < 0.05; Student’s t-test). In both (A) and (B), all samples were run in the same gel, but images were cropped to show the relevant genotypes alongside one another.

Figure 4

Effect of ABA on SR45 protein stability and ubiquitination. (A) Protein gel blot analysis, using α-GFP antibodies, of the SR45–GFP fusion protein in 7-day-old seedlings of the C2 complementation line pretreated with MG132 and subjected to a 180-min treatment with 2 μM ABA. Control samples (−MG132 or −ABA) were treated with the equivalent volume of the solvent of the MG132 or ABA solution (DMSO or ethanol, respectively). A total of 40 ng of protein was loaded per sample. Bands were quantified and relative protein levels determined using the Ponceau loading control as a reference, with control conditions set to 1. Results represent means ± SE (n = 4), and different letters indicate statistically significant differences between treatments (P < 0.05; Student’s t-test). (B) Protein gel blot analysis of the SR45–GFP fusion protein immunoprecipitated from extracts of 7-day-old seedlings of the C2 complementation line treated for 180 min with 2 μM ABA using α-GFP (immunoprecipitation; IP) or α-UBQ11 (coIP) antibodies. Control samples were treated with the equivalent volume of the solvent of the ABA solution (ethanol). Ponceau staining is shown as a loading control for the input fraction. Signals were quantified and the UBQ/SR45–GFP ratio determined, with control conditions set to 1. Results represent means ± SE (n = 3), and different letters indicate statistically significant differences between treatments (P < 0.05; Student’s t-test).

Figure 5

Effect of T264 phosphorylation on SR45 ubiquitination and degradation. (A) Protein gel blot analysis of the SR45–GFP fusion protein immunoprecipitated from extracts of 7-day-old seedlings of the OX1 overexpression (pUBQ10:SR45–GFP/sr45-1), PMut1 phosphomutant (pUBQ10:SR45–GFP_T264A/sr45-1), and PMim1 phosphomimetic (pUBQ10:SR45–GFP_T264D/sr45-1) transgenic lines grown under control conditions using α-GFP (immunoprecipitation; IP) or α-UBQ11 (coIP) antibodies. Ponceau staining is shown as a loading control for the input fraction. Signals were quantified and the UBQ/SR45–GFP ratio determined, with control conditions set to 1. Results represent means ± SE (n = 3), and different letters indicate statistically significant differences between genotypes (P < 0.05; Student’s t-test). (B) Protein gel blot analysis, using α-GFP antibodies, of the SR45–GFP fusion protein in 7-day-old seedlings of the OX1 overexpression, PMut1 phosphomutant, and PMim1 phosphomimetic transgenic lines supplemented or not with MG132 and left at room temperature for 0, 15, 30, or 60 min. Control samples (−MG132) were treated with the equivalent volume of the solvent of the MG132 solution (DMSO), and a total of 20 ng of protein was loaded per sample. Bands were quantified and relative protein levels determined using the Ponceau loading control as a reference, with time 0 set to 1. Results are representative of at least three independent experiments.

Figure 6

Effect of T264 phosphorylation on ABA-dependent SR45 protein accumulation and cotyledon development. (A) Protein gel blot analysis, using α-GFP antibodies, of the SR45–GFP fusion protein in 7-day-old seedlings of the OX1 overexpression (pUBQ10:SR45–GFP/sr45-1), PMut1 phosphomutant (pUBQ10:SR45–GFP_T264A/sr45-1), and PMim1 phosphomimetic (pUBQ10:SR45–GFP_T264D/sr45-1) transgenic lines treated for 180 min with 2 μM ABA. Control samples were treated with the equivalent volume of the solvent of the ABA solution (ethanol), and a total of 20 ng of protein was loaded per sample. Bands were quantified and relative protein levels determined using the Ponceau loading control as a reference, with results representing means ± SE (n = 3), control conditions set to 1, and different letters indicating statistically significant differences between treatments for each genotype (P < 0.05; Student’s t-test). All samples were run in the same gel, but images were cropped to show the relevant genotypes alongside one another. (B) Cotyledon greening percentages of 7-day-old seedlings of the OX1 overexpression, PMut1 phosphomutant, and PMim1 phosphomimetic transgenic lines grown under control conditions or in the presence of 0.5 μM ABA, with representative images of ABA conditions (scale bar, 0.2 cm), and qRT-PCR analysis of SR45–GFP transcript levels in the same seedlings (control conditions), using PEX4 as a reference gene and primers annealing to the GFP sequence (see Supplemental Figure 1). Results represent means ± SE (n = 3), and different letters indicate statistically significant differences between genotypes (P < 0.05; Student’s t-test).

Figure 7

Model of ABA-mediated SR45 regulation of early seedling development. Under non-stress conditions, basal ABA levels, which are insufficient to activate stress signaling, allow normal plant growth, with phosphorylation of SR45 triggering its ubiquitination and subsequent proteasomal degradation. In response to stress, ABA accumulates in the cell, activating SnRK2s and thus downstream signaling. This reduces SR45 phosphorylation levels via a yet unknown mechanism, leading to reduced ubiquitination and stabilization of the protein. Gradual accumulation of SR45 results in negative regulation of ABA signaling, alleviating its inhibition of early seedling development and allowing plant growth to some extent under stress conditions.