Endogenous level of abscisic acid down-regulated by brassinosteroids signaling via BZR1 to control the growth of Arabidopsis thaliana

ABSTRACT

The increased level of endogenous abscisic acid (ABA) in brassinosteroid (BR)-deficient mutants, such as det2 and cyp85a1 × cyp85a2, suggests that ABA synthesis is inhibited by endogenous BRs in Arabidopsis thaliana. Expression of the ABA biosynthesis gene ABA-deficient 2 (ABA2) was negatively regulated by exogenously applied BR but up-regulated by the application of brassinazole and in det2 and cyp85a1 × cyp85a2. In addition, ABA2 expression decreased in bzr1-1D, showing that ABA biosynthesis is inhibited by BR signaling via BZR1, intermediated by ABA2, in Arabidopsis. Four cis-element sequences (E-boxes 1–4) in the putative promoter region of ABA2 were identified as BZR1 binding sites. The electrophoretic mobility shift assay and chromatin immune precipitation analysis demonstrated that BZR1 directly binds to overlapped E-boxes (E-box 3/4) in the promoter region of ABA2. The level of endogenous ABA was decreased in bzr1-1D compared to wild-type, indicating that binding of BZR1 to the ABA2 promoter inhibits ABA synthesis in Arabidopsis. Compared to wild-type, aba2-1 exhibited severely reduced growth and development. The abnormalities in aba2-1 were rescued by the application of ABA, suggesting that ABA2 expression and ABA synthesis are necessary for the normal growth and development of A. thaliana. Finally, bzr1-KO × aba2-1 exhibited inhibitory growth of primary roots compared to bzr1-KO, verifying that ABA2 is a downstream target of BZR1 in the plant. Taken together, the level of endogenous ABA is down-regulated by BR signaling via BZR1, controlling the growth of A. thaliana.

Introduction

Brassinosteroids (BRs) are steroidal plant hormones involved in a wide range of growth-regulating activities associated with germination, cell division and expansion, vascular differentiation, photomorphogenesis, pathogen resistance, senescence, and the stress response in plants.^1–3 The signal transduction systems of BRs modify the expression of target genes and promote hormonal activity.⁴ Biologically active BRs, such as castasterone and brassinolide (BL), are recognized by the membrane-localized receptor kinase, brassinosteroid-insensitive 1 (BRI1), promoting dimerization between BRI1 and its co-receptor BRI1-associated receptor kinase 1. One fully activated, BRI1 phosphorylates brassinosteroid-signaling kinases and BRI1 suppressor 1, in turn, suppressing the kinase activity of brassinosteroid-insensitive 2 (BIN2), a negative regulator in BR signaling. The inactivation of BIN2 leads to the accumulation of BR transcription factors, such as brassinazole-resistant 1 (BZR1) and BRI1 EMS suppressor 1 (BES1) in the nucleus, which ultimately control transcription of BR-responsive genes to regulate the growth and development of plants.⁵

Abscisic acid (ABA) also plays an important role in controlling many aspects of plant growth and development, such as vegetative growth, seed development and maturation, desiccation tolerance and dormancy of seeds, stomatal opening/closing, tolerance to environmental stresses and pathogen infection, and the inhibition of flowering.⁶ In many plants, ABA synthesis occurs mainly through an indirect pathway.⁷ In Arabidopsis, zeaxanthin and antheraxanthin are converted into violaxanthin by zeaxanthin epoxidase, encoded by ABA-deficient 1 (ABA1) in plastids. After structural modification, violaxanthin is converted into 9’-cis-neoxanthin and/or 9-cis-violaxanthin. The epoxycarotenoids are then oxidized by 9-cis-epoxycarotenoid dioxygenase to generate the C₁₅ intermediary, xanthoxin. Xanthoxin is transported to the cytosol and converted to abscisic aldehyde by a short-chain dehydrogenase/reductase-1 encoded by ABA-deficient 2 (ABA2). Abscisic aldehyde oxidase 3 (AAO3), which needs a molybdenum cofactor sulfurase encoded by ABA-deficient 3, catalyzes the last step in the conversion of abscisic aldehyde into ABA.⁸

In some ABA physiologies, BRs show the opposite regulatory activities to ABA, implying that they are antagonistically interactive in plants. From the view that endogenous levels of gibberellin (GA) and the production of ethylene are controlled by BR signaling in Arabidopsis,⁹ BRs may also regulate the levels of endogenous ABA to show antagonistic activities in the plant. However, whether and how BRs control the level of endogenous ABA to regulate the growth and development of plants remain to be elucidated. This prompted us to investigate the mechanism of action for the BR-induced regulation of ABA biosynthesis in Arabidopsis thaliana in this study.

Results and discussion

To evaluate the effects of BRs in the biosynthesis of ABA in A. thaliana, the level of endogenous ABA in BR-deficient mutants was quantified by an enzyme-linked immunosorbent assay (ELISA). As shown in (Table 1), the level of ABA was increased in det2 and cyp85a1 × cyp85a2 compared to the wild-type. Thus, the presence of BRs seems to be essential for synthesizing a steady-state level of ABA in A. thaliana.

Table 1.

Endogenous level of ABA measured by ELISA in wild type (Col-0) and BR related mutants of A. thaliana. Light-grown 20-day-old seedlings (3 g) were used for quantitative analysis. Numbers in parentheses indicate standard deviation (S.D.)

Plant	ABA content (pM/g fresh weight)
	1st experiment	2nd experiment	3rd experiment	Average
Col-0	624	601	702	642 (± 52.9)
det2	1135	1091	1329	1185 (± 126.6)
cyp85a1 × cyp85a2	1300	1386	1221	1302 (± 82.5)
bzr1-1D	492	475	389	452 (± 55.2)

Among ABA biosynthetic genes in Arabidopsis, ABA2 has been identified as a BR-down-regulated gene in microarray analysis of BR target genes,^10,11 suggesting that the inhibition of ABA biosynthesis by BRs may occur via ABA2 in the plant. To verify this possibility, alternation of ABA2 expression was examined by applying exogenous BRs and using BR-related mutants. As shown in (Figure 1a), the expression of ABA2 was inhibited by BL in a concentration-dependent manner. The application of brassinazole (Brz), a biosynthetic inhibitor of BRs, increased ABA2 expression, but this effect was weakened by the application of BL (Figure 1b). Similarly, the enhanced ABA2 expression in det2 and cyp85a1 × cyp85a2 was inhibited by BL application. Consistent with semi-quantitative RT-PCR analysis, β-glucuronidase (GUS) activity was decreased and increased in transgenic Arabidopsis expressing the GUS gene driven by the ABA2 promoter (727-bp) by application of BL and Brz, respectively (Figures 1c and 1d). Therefore, we demonstrated that the inhibition of ABA synthesis by BRs is mediated by the down-regulation of ABA2 in A. thaliana.

Figure 1.

Expression of ABA2 in BR-related mutants. (a) Effect of exogenously applied BL on the expression of ABA2 in five DAG seedlings. (b) Relative expression of ABA2 in BR-related mutants (det2 and cyp85a1 × cyp85a2) compared to wild-type (Col-0). BL (10^–8 M) and/or Brz (10^–7 M) were applied to wild-type and BR-deficient mutants. (c and d) GUS expression level in five DAG seedlings of pABA2:GUS transgenic plants after application of BL (10^–8 M) and Brz (10^–7 M). Semi-quantitative RT-PCR was performed on total RNA. UBQ5 was used to normalize the expression level. Data from three independent experiments are presented as the mean ± standard error (n > 120). The asterisks indicate a significant difference between the control and chemical-treated plants at *P < .05 and **P < .01 by the Student’s t-test

The expression of ABA2 decreased in bzr1-1D (a gain-of-function mutant of the BZR1 gene) compared to the wild-type (Figure 2a). Additionally, the level of endogenous ABA was lower in bzr1-1D than in the wild-type (Table 1). These results suggest that the down-regulation of ABA2 for reducing the ABA level seems to occur by BR signaling via the BR transcription factor, BZR1. In the 727-bp region upstream of the ABA2 start codon, a putative promoter region of ABA2, four cis-elements (E-boxes, CANNTG) for BES1/BZR1 were present (Figure 2b). The four E-boxes denoted E-box 1–4 were, respectively, farthest to and nearest from the start codon of ABA2. Maltose-binding protein (MBP)–BZR1 was obtained from Escherichia coli using a construct in which MBP was fused to the N-terminal of full-length BZR1, and an electrophoretic mobility shift assay (EMSA) was performed. As shown in (Figure 2c), an overlapped probe in E-box 3/4 was strongly bound to MBP–BZR1. The binding of MBP–BZR1 was diminished by the mutation of E-box 3/4 (CANNTG to AAAAAA). A chromatin immune precipitation (ChIP) assay was also carried out using BZR1–YFP transgenic plants. As shown in (Figure 2d), BZR1–YFP successfully bound to the overlapped E-box 3/4 in Arabidopsis plants. These results demonstrate that BZR1 binds to the E-boxes in the promoter region of ABA2 both in vitro and in vivo.

Figure 2.

BZR1-mediated ABA2 expression in the regulation of growth in Arabidopsis. (a) Expression of ABA2 in BR-signaling mutants (bzr1-1D and bes1-D). Data from three independent experiments are presented as mean ± standard error (n > 20). The asterisks indicate a significant difference between wild-type and mutant plants at *P < .05 and **P < .01 by the Student’s t-test. (b) Schematic diagram of the ABA2 promoter region. White rectangles indicate E-boxes (CANNTG) in the intergenic region. E-box 3/4 contains overlapped E-boxes. The location of each E-box is marked with a negative number. (c) EMSA result for binding of BZR1 protein to the ABA2 promoter in vitro. [³²P]-labeled ABA2 probes were incubated with or without MBP–BZR1 protein. The arrow indicates complexes in which DNA and protein are bound. Mutated probes (MT probes) successfully inhibited the binding of MBP–BZR1 and ABA2 promoter. (d) ChIP assay result for binding of BZR1 protein to ABA2 promoter in vivo. ChIP with anti-YFP antibody prepared from 10-day-old light-grown 35S:YFP (negative control) and BZR1:YFP transgenic seedlings. ChIP-PCR was performed with primers to the E-box 3/4 position of the ABA2 promoter, SAUR-AC1 (positive control), and UBQ5 (negative control). Each assay was repeated twice. (e) Growth phenotypes of five DAG seedlings of aba2-1, bzr1-KO, and bzr1-KO × aba2-1. (f) The phenotypes of intact plants of aba2-1, bzr1-KO, and bzr1-KO × aba2-1 grown in soil for 3 weeks. (g) Transcript levels of ABA2 in aba2-1, bzr1-KO, and bzr1-KO × aba2-1. (h) The proposed scheme for BR-regulated ABA biosynthesis in the growth of roots of A. thaliana.

The ABA2 knock-out (KO) mutant aba2-1 (CS156, Supplementary Figure S1) was obtained from the Arabidopsis Biological Resource Center. In terms of seedling growth, the primary roots of aba2-1 (Figure 2e) and bzr1-KO (a loss-of-function mutant of the BZR gene) were shorter and longer, respectively, compared to the wild-type. A double-mutant for bzr1-KO and aba2-1, bzr1-KO × aba2-1, exhibited reduced primary root growth compared to bzr1-KO. As shown in (Figure 2f), a similar tendency of phenotypic alternations was found in intact plants of aba2-1, bzr1-KO, and bzr1-KO × aba2-1 compared to the wild-type plants, Semi-quantitative RT-PCR analysis revealed that the growth phenotypes in aba2-1, bzr1-KO, and bzr1-KO × aba2-1 were correlated with the expression of ABA2 in the mutants (Figure 2g). Therefore, ABA2 is considered to be epistatic to BZR1 in Arabidopsis growth. Consequently, ABA2 is a direct downstream target for the BR transcription factor BZR1, and the down-regulation of ABA2 by the binding of BZR1 to the ABA2 promoter decreases the level of endogenous ABA to inhibit the growth of A. thaliana (Figure 2h).

Like the previously reported phenotype for Arabidopsis aba2¹¹, the aba2-1 used in this study showed inhibited primary root growth, which was rescued by the application of exogenous ABA (Supplementary Figure S2). In intact plants, aba2-1 also showed severely retarded growth compared to the wild-type, such as smaller dark-green unexpanded leaves, shorter inflorescent stem, and fewer flowers/siliques. The abnormalities in aba2-1 were fairly well restored to the wild-type phenotype by the application of ABA, indicating that the expression of ABA2 is necessary for the normal growth of A. thaliana. Besides ABA2, other ABA-deficient mutants, such as aba1 and aao3, showed similar inhibitory growth that could be rescued by the application of ABA.¹² Therefore, ABA is likely to be a positive regulator, controlling the growth and development of Arabidopsis. Contrariwise, aba2-1 showed larger seeds than the wild-type; application of ABA decreased the size of aba2-1 seeds.¹³ Therefore, ABA seems to function as a negatively regulating hormone in the size determination of Arabidopsis seeds. Consequently, ABA can play both positive and negative roles in different growth stages and the development of plant organs.

Initially, the E-box and BR regulatory elements (BRRE, CGTGTC) in the promoter region of BR target genes were characterized as specific cis-elements for binding BES1 and BZR1, respectively.² Later, Sun et al. reported that both BES1 and BZR1 could bind to both the E-box and BRRE sequences in the promoter region of BR target genes.² No BRRE sequences are present in the intergenic region of ABA2 (Figure 2b). However, BZR1 directly binds to the E-box sequences in the promoter region of ABA2 to regulate the expression of ABA2, providing evidence that the binding of BZR1 to the E-box sequence is needed to regulate the expression of BR target genes in plants. Similar to bzr1-1D, bes1-D showed down-regulated expression of ABA2 (Figure 2a). Coupled with the presence of E-box sequences in the promoter region of ABA2, this suggests that the potential binding of BES1 to E-boxes regulates ABA2 expression in Arabidopsis. If so, BES1-mediated ABA2 expression may show different physiological functions from those induced by BZR1-mediated ABA2 expression in the regulation of growth and development of Arabidopsis. To examine this, the binding of BES1 to the promoter of ABA2, and its physiological role, are currently under investigation.

In both Arabidopsis and rice, BRs change the expression of GA metabolic genes, such as GA₂₀ oxidase/GA₃ oxidase for activation and GA₂ oxidase for inactivation, by triggering the binding of BES1/BZR1 to cis-elements, such as E-box/G-box, BRRE, or non-E-box motifs, in the promoter regions of the genes, which ultimately controls the endogenous levels of active GAs in plants.¹⁴ We previously demonstrated that BES1 or BZR1 binds directly to the E-box or BRRE sequences in the promoter region of ACC oxidases in ethylene biosynthesis, which alters ethylene production in A. thaliana.^15–17 Recently, we also found that BZR1 directly binds to BRRE sequences in the promoter of the isochorismate synthase 1 (ICS1) gene to increase the level of endogenous salicylic acid in response to pathogenic infections in Arabidopsis (data will be published elsewhere). Coupled with that, BZR1 directly binds to the E-box sequences in the promoter region of ABA2 to reduce endogenous ABA levels, as demonstrated in this study. These findings suggest that BRs are likely to be master regulators in the homeostasis of, at least, GA, ethylene, salicylic acid, and ABA, and possibly other plant hormones in A. thaliana.

Methods and materials

Plant materials and growth conditions

Arabidopsis thaliana Columbia-0 (Col-0) was used as the wild-type ecotype background for all mutants and transgenic plants, including aba2-1 obtained from SALK (CS156), det2, cyp85a1 × cyp85a2, ProABA2::GUS, bzr1-1D, bzr1-KO × aba2-1, and BZR1-YFP, in this study. Seeds were surface-sterilized with ethanol (EtOH)–H₂O (70:30, v/v) for 5 min, rinsed in distilled H₂O, and stratified at 4°C for 2 days. The seeds were planted on 0.5X Murashige–Skoog (MS) medium containing 1% (w/v) sucrose and 0.8% agar medium (Phytagel; Sigma, St. Louis, MO, USA). Plants were grown under light conditions at 22°C for 16 h and dark conditions at 20°C for 8 h in a growth chamber (Vision Scientific, Seoul, Korea).

Quantification of ABA (ELISA)

On the 20^th day after germination, Arabidopsis seedlings (3 g) were ground with a pestle and extracted three times with 30 mL of 80% methanol (MeOH). These extracts were concentrated on a rotary evaporator, solvent partitioned with ethyl acetate and Pi buffer (pH 2.5), and the ethyl acetate layer was collected, concentrated, and dried. The obtained ethyl acetate fraction was dissolved in 50% MeOH and loaded onto a Sep-Pak C₁₈ cartridge (Waters) column. After conditioning the column alternately with H₂O and MeOH several times, equilibrium was maintained with 50% MeOH. ABA was eluted by adding 5 mL of 50% MeOH to the column, and the collected fractions were concentrated and dried. The ABA concentrations in the fractions obtained from the Sep-Pak C₁₈ cartridge column were measured by ELISA using a Phytodetek ABA Test Kit (Agida, PDK09347/0096). First, 100 µL of a standard solution or each sample (1/1,000) was added to a given test well, and the same amount of ABA-tracer was added. The test well was covered with a plate sealer and placed in a humid environment (a box containing water) to react at 4°C for 3 h. When the reaction was over, all solutions in the test well were discarded and washed with 1X PBST. Then, 200 µL of the substrate solution was added to the washed test well, which was then sealed and placed in a humid environment (a box containing water) to react at 37°C for 30–60 min. After the reaction, the ELISA absorbance was measured at 405 nm, and the ABA concentration was calculated according to the given calculation method.

Total RNA isolation and qRT-PCR

Five days after germination, Arabidopsis seedlings were frozen in liquid nitrogen and homogenized using a hand grinder. About 0.1 g of powder was dissolved in 1 mL of QIAzol lysis reagent (QIAGEN Sciences, Germantown, MD, USA) and incubated at room temperature for 5 min. Then, 0.2 mL chloroform was added, and the solution was vortexed vigorously for 15 s, then left at room temperature for 10 min. After centrifugation at 12,000 × g and 4°C for 15 min, the supernatant was transferred to 0.5 mL isopropanol. The mixture was inverted and incubated at room temperature for 5 min, followed by centrifugation at 12,000 × g, 4°C, for 10 min. After discarding the supernatant, 1 mL of 70% EtOH was added to wash the pellet and re-centrifuged. The supernatant was discarded, and the resultant pellet was dried and dissolved in 20 µL of DEPC-treated H₂O. The extracted total RNA (5 µg) was reverse-transcribed by M-MLV Reverse Transcriptase (Promega, Madison, WI, USA). PCR was performed using iQ SYBR Green SuperMix and an iCycler iQ Real-time PCR Detection System (Bio-Rad). PCR conditions consisted of denaturation at 95°C for 3 min, followed by 40 cycles of denaturation at 95°C for 15 s, annealing at 58°C for 15 s, and elongation at 72°C for 30 s. The mRNA levels were quantified using the 2^−ΔΔCT method (Livak and Schmittgen, 2001). Expression of the target gene was normalized against protein phosphatase 2A (PP2A) mRNA expression as the housekeeping gene. (Supplementary Table S1) describes the gene-specific primers.

Transgenic plants harboring ProABA2::GUS constructs and histochemical analysis of GUS

Genomic DNA fragments from 727-bp upstream from the translation start codon of ABA2 (At1g52340) were introduced in the pBI121 binary vector containing the GUS (UidA) reporter gene. Constructs were transformed into Col-0 of Arabidopsis using the floral dipping method, and transgenic plants were selected by kanamycin (50 mg mL⁻¹). For GUS histochemical analysis, transgenic Arabidopsis seedlings of proABA2::GUS were fixed with 90% (v/v) acetone at 4°C and incubated in a staining solution containing 2 mM 5-bromo-4-chloro-3-indolyl-β-D-GlcUA (X-glcA; Duchefa Biochemie) in 50 mM Na₂HPO₄ buffer (pH 7.2), 2 mM of potassium ferrocyanide, 2 mM of potassium ferricyanide, and 0.2% Triton X-100 at 37°C overnight after vacuum infiltration on ice for 20 min. The samples were washed with EtOH–distilled H₂O (70:30, v/v) and observed under a dissecting microscope (Olympus SZ-PT) and a light microscope (Olympus CX21).

EMSA

To prepare the samples for EMSA, first, the N-terminal of full-length BZR1 was fused to MBP (New England Biolabs, Ipswich, MA, USA) using a pMALc2X vector (New England Biolabs). The recombinant protein was then expressed in E. coli and affinity-purified (BL21-CodonPlus [DE3]-RIL) using amylose resin. MBP served as a negative control in the assays. (Supplementary Table S2) describes the probes for ProABA2 used for the EMSA. These probes were labeled with γ-32P-dATP (10 μCi μL⁻¹) using the DNA 5ʹ-end labeling system (Promega), according to the manufacturer’s instructions. The protein was mixed with 5X binding buffer (50 mM Tris–HCl [pH 7.5], 250 mM NaCl, 2.5 mM EDTA, 2.5 mM DTT, 5 mM MgCl₂, and 10% glycerol), and incubated at room temperature for 10 min. Radioisotope-labeled probes were added to this mixture and incubated at room temperature for 10 min to allow the binding reaction to occur. The reactions were resolved using 4% native polyacrylamide gels with a 0.5X TBE buffer (5.4 g L⁻¹ of Tris base, 2.75 g L⁻¹ of boric acid, 1 mM EDTA [pH 8.0]) and exposed to a phospho-imager.

ChIP assay

Wild-type Arabidopsis (Col-0) and transgenic plants expressing BZR1–YFP were used for the ChIP assay, as described previously.¹⁸ The chromatin–protein complex was isolated from the plant tissue by incubation in a crosslinking extraction buffer (0.4 M sucrose, 10 mM Tris–HCl pH 8.0, 1 mM PMSF, 1 mM EDTA, and 1% formaldehyde) under a vacuum. Tissues were ground to fine powders with liquid nitrogen and resuspended in cold nuclei isolation buffer (0.25 M sucrose, 15 mM PIPES pH 6.8, 5 mM MgCl₂, 60 mM KCl, 15 mM NaCl, 1 mM CaCl₂, 0.9% Triton X-100, 2 µg mL⁻¹ pepstatin A, and 2 µg mL⁻¹ aprotinin). The homogenized slurry was filtered through several layers of Miracloth and centrifuged at 11,000 × g and 4°C for 20 min. The supernatant was discarded, and the pellet was resuspended in cold nuclei lysis buffer (50 mM HEPES pH 7.5, 150 mM NaCl, 1 mM EDTA, 1% SDS, 0.1% of sodium deoxycholate, 1% Triton X-100, 1 µg mL⁻¹ pepstatin A, and 1 µg mL⁻¹ aprotinin). The resuspended samples (DNA) were sheared into ~500-bp fragments by ultrasonication. ChIP-grade YFP antibody (Abcam, Cambridge, UK) was added to the samples and kept at 4°C for at least 5 h under gentle rotation. The immunocomplexes were precipitated with pre-equilibrated protein A agarose beads (Santa Cruz Biotechnology, Dallas, TX, USA). The eluted DNA was purified using a PCR purification kit (LaboPass, Seoul, Korea) and used as a template for PCR. (Supplementary Table S3) describes the region-specific primer sequences used in ChIP-PCR.

Funding Statement

This work was supported by the National Research Foundation of Korea (NRF) grant to S.-K. Kim (grant No. 2021R1A2C1007516), funded by the Korean government (Ministry of Science, ICT and Future Planning).

Disclosure statement

No potential conflicts of interest were disclosed.

Supplementary material

Supplemental data for this article can be accessed on the publisher’s website.