Direct receptor competition gates RGL2 proteolysis for seed germination timing in Arabidopsis
Kaili Nie, Juntao Jiang, Changgen Xie, Hongyun Zhao, Yuan Zheng
TL;DR
The paper explains how plant seeds decide when to germinate by showing how two hormones, ABA and GA, compete to control a key protein called RGL2.
Contribution
The study reveals a new mechanism where ABA and GA receptors directly compete for RGL2, controlling its degradation and germination timing.
Findings
ABA receptors (PYLs) stabilize RGL2 by binding to it and sequestering DWA1.
GA receptors (GID1s) outcompete PYLs, allowing DWA1 to degrade RGL2.
This competition is regulated by ABA and GA concentrations, influencing germination timing.
Abstract
Seed germination is orchestrated by antagonistic gibberellin (GA) and abscisic acid (ABA) signals converging on the master germination repressor RGL2. Here, we unveil a receptor-competition paradigm where ABA receptors (PYLs) stabilize RGL2, both through direct physical interaction and through functional sequestration of DWA1, the CUL4–DDB1 E3 ligase substrate adapter mediating RGL2 ubiquitination. GA receptors (GID1s) counteract this stabilization by competitively displacing PYLs from RGL2, leveraging their superior binding capacity to license DWA1-mediated degradation. Crucially, this competition is defined by the concentration of abscisic acid and gibberellin as they regulate PYL and GID1 expression. Genetic epistasis confirms that PYLs act upstream of DWA1, competing directly with GID1 at RGL2. This receptor-occupied switch converts environmental fluctuations into proteolytic…
Genes, proteins, chemicals, diseases, species, mutations and cell lines named across the full text — each resolved to its canonical identifier and authoritative record.
Click any figure to enlarge with its caption.
Figure 10
Figure 11
Figure 12
Figure 1
Figure 2
Figure 3
Figure 4
Figure 5
Figure 6
Figure 7
Figure 8
Figure 9
Figure 13
Figure 14- —http://dx.doi.org/10.13039/501100001809MOST | National Natural Science Foundation of China (NSFC)
- —State Key Laboratory of Plant Environmental Resilience
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsSeed Germination and Physiology · Plant Molecular Biology Research · Plant Parasitism and Resistance
Introduction
The successful transition from seed dormancy to germination is a decisive ecological adaptation, orchestrated by the counteracting forces of gibberellin (GA) and abscisic acid (ABA). GA mobilizes metabolic reserves and drives embryonic expansion, while ABA maintains desiccation tolerance and suppresses premature activation (Finch-Savage and Leubner-Metzger, 2006; Shu et al, 2016a; Vishwakarma et al, 2017). This hormonal antagonism operates through dynamic concentration shifts: ABA dominates during maturation to enforce dormancy, whereas GA peaks upon imbibition to initiate germination (Liu and Hou, 2018; Chen et al, 2020). Crucially, the levels of the two hormones are regulated in opposite ways (Razem et al, 2006; Seo et al, 2006; Oh et al, 2007). For instance, ABA-activated ABI4 suppresses ABA metabolic genes (CYP707A1 and CYP707A2) while inducing GA catabolism (GA2ox7) and ABA synthesis (NCED6), establishing a self-reinforcing dormancy loop (Shu et al, 2013, 2016b).
Exogenous GA application can significantly enhance germination rates, whereas the GA biosynthesis inhibitor paclobutrazol (PAC) effectively suppresses germination by reducing GA levels. Notably, the GA-deficient mutant ga1 is unable to germinate without exogenous GA supplementation (Sun and Kamiya, 1994; Xu et al, 2014). GA signals are mainly activated by the degradation of key inhibitors, DELLA. DELLA proteins negatively regulate GA signals by inhibiting the activity of bound transcription factors. There are five DELLA proteins in Arabidopsis thaliana: RGA, GAI, RGL1, RGL2, and RGL3. Among them, RGL2 accumulates in dormant seeds and acts as a master brake on germination (Claeys et al, 2016; Thomas et al, 2016; Bao et al, 2020). Excessive RGL2 accumulation was observed in ga1 mutant seeds, and the rgl2 knockout fully rescues the germination defect of the ga1 mutant and confers PAC and ABA insensitivity (Lee et al, 2002; Tyler, 2004; Liu et al, 2016). Crucially, RGL2’s degradation serves as a GA signaling permissive switch. GA binding to GID1 receptors promotes their interaction with RGL2. More importantly, this interaction enhances the subsequent binding between RGL2 and SLY1/GID2, thereby promoting RGL2’s recruitment to the SCF^SLY1/GID2^ complex for ubiquitination and degradation (Murase et al, 2008; Sun, 2011; Blanco-Touriñán et al, 2020; Dahal et al, 2025; Islam et al, 2025). In addition, the MYB transcription factor RVE1, which is involved in light signal regulation, protects RGL2 by competitively binding to it with SLY1 (Yang et al, 2019).
Intriguingly, RGL2 also functions as a molecular amplifier of ABA signaling. ABA not only transcriptionally upregulates RGL2 through ABI4 but also post-translationally stabilizes RGL2 protein—a stark contrast to GA’s destabilizing effect (Piskurewicz et al, 2008; Xian et al, 2024). Mechanistically, RGL2 regulates the core transcription factor ABI5 in the ABA signaling pathway, which is the ultimate downstream inhibitor of the GA–ABA signal balance in seed germination (Piskurewicz et al, 2008; Skubacz et al, 2016; Zhao et al, 2022, 2025). First, RGL2 activates XERICO expression, elevating endogenous ABA levels by promoting NCED-mediated biosynthesis, enhancing ABI5 expression and protein content (Ko et al, 2006). Second, RGL2 physically interacts with ICE1 and NF-YC transcription factors to potentiate ABI5 activity (Liu et al, 2016; Hu et al, 2019). Thus, RGL2 functions as a bidirectional switch: GA reduces RGL2 levels to permit germination, whereas ABA elevates RGL2 to repress germination.
ABA signaling is mediated by PYR1/PYL/RCAR (referred to as PYL) receptors. ABA promotes their binding to PP2C phosphatases, thereby releasing SnRK2 kinases to phosphorylate transcription factors such as ABI5/ABFs and induce ABA-responsive genes (Park et al, 2009; Umezawa et al, 2010; Chen et al, 2020). PYLs also participate in cross-talk between ABA and other signaling pathways by direct interactions. For example, PYLs interact with PIF transcription factors to regulate seed responses to ABA in the dark (Qi et al, 2020) and with MYB30 to mediate the antagonistic regulation of seed dormancy by nitric oxide and ABA (Zhao et al, 2024). However, how ABA receptors perceive hormonal cues to influence RGL2 stability is unknown.
To decode the enigmatic ABA-mediated RGL2 stabilization, we performed systematic interactor screening that unexpectedly identified ABA receptor PYLs as direct binding partners of RGL2. This finding revealed a receptor competition paradigm: PYLs stabilize RGL2 through dual mechanisms, whereas GA receptors (GID1s) forcibly displace PYLs via superior binding capacity, thereby licensing degradation through the DWA1–E3 ligase module. Crucially, elevated PYL expression reciprocally attenuates GID1 access to RGL2, establishing a bidirectional proteolytic switch. This work redefines hormone cross-talk as direct receptor competition for a shared integrator, thereby optimizing developmental decisions in fluctuating environments.
Results
Direct interaction between PYL receptors and RGL2 mediates hormone-responsive stabilization
Our exploration of RGL2 regulatory networks unexpectedly revealed the ABA receptor PYLs as its direct binding partners. Bimolecular fluorescence complementation (BiFC) assays revealed that all 14 PYLs formed complexes with RGL2 in Nicotiana benthamiana leaf cells, with distinct fluorescence signals observed (Fig. 1A; Appendix Fig. S1). In vitro pull-down experiments using purified proteins further established the direct physical binding between MBP-RGL2 and GST-tagged PYR1/PYL2/4/5/8 (Fig. 1B). Notably, these interactions exhibited dynamic hormone responsiveness. ABA treatment enhanced the binding intensity, particularly for PYL11, whereas GA application consistently suppressed complex formation in luciferase complementation imaging (LCI) assays (Fig. 1C; Appendix Fig. S2).Figure 1PYL receptors directly stabilize RGL2 through hormone-modulated interactions.(A) Bimolecular fluorescence complementation (BiFC) visualization of PYL–RGL2 interactions. YFP signals from Nicotiana benthamiana co-expressing YNE-RGL2 and PYLs-YCE were visualized, scale bar = 50 µm. (B) In vitro pull-down of PYL–RGL2 complexes. GST-PYLs were incubated with MBP-RGL2, captured with glutathione resin, and immunoblotted. (C) Hormonal regulation of PYL–RGL2 interactions. Luciferase complementation imaging (LCI) quantification in N. benthamiana expressing nLUC-PYLs and cLUC-RGL2 treated with 10 µM ABA or GA₃. Data are means ± SD (n = 3 independent biological replicates). Significant differences were determined by two-way ANOVA. (D) PYL-mediated suppression of GA-induced RGL2 degradation. Five-day-old seedlings (genotypes indicated) expressing MYC-RGL2 were pretreated with 10 µM paclobutrazol (PAC) for 60 min, then challenged with 10 µM GA. Right: densitometric quantification (Data represent means from three independent biological repeats, two-way ANOVA followed by Tukey’s test). Source data are available online for this figure.
We then investigated whether PYLs are involved in GA-induced RGL2 degradation. For this purpose, 35S:MYC-RGL2 was expressed in wild-type (WT), pyr1 pyl1 pyl2 pyl4 quadruple mutant, 35S:PYL4-GFP and 35S:Flag-PYL11 seedlings (Zhao et al, 2020, 2024). Our results showed that PYL4 and PYL11 overexpression significantly reduced GA-induced RGL2 degradation in seedlings, whereas the pyr1 pyl1 pyl2 pyl4 quadruple mutant showed accelerated RGL2 turnover (Fig. 1D; Appendix Fig. S3), identifying a stabilization role of PYLs. This indicates that PYLs act as direct molecular stabilizers of the central germination repressor through hormone-modulated interactions.
Competitive receptor occupancy on RGL2 implements a bidirectional switch
The convergence of GA and ABA signaling receptors on RGL2 prompted us to investigate potential competitive interactions. Quantitative assessment revealed a striking binding hierarchy. LCI assays showed the GID1–RGL2 interaction intensity was dramatically stronger than PYR1/PYL4/11–RGL2 under respective hormone treatments (Fig. 2A; Appendix Fig. S4). This was confirmed by in vitro pull-down quantification, which showed that GID1a/b captured dramatically more RGL2 than PYR1, PYL4, or PYL11 at similar concentrations (Fig. 2B). Furthermore, our microscale thermophoresis (MST) assays revealed that in the presence of GA, both GID1a and GID1b recruited fluorescently labeled RGL2 with markedly higher affinity than any ABA receptor tested. Under identical conditions, PYR1, PYL4, and PYL11 consistently produced weaker binding signals, with PYL11 showing the closest but still clearly reduced interaction (Fig. EV1). Thus, the data establish GID1 as the preferential RGL2 partner, providing a quantitative basis for receptor competition.Figure 2. Bidirectional receptor competition governs RGL2 occupancy.(A) Quantitative receptor–RGL2 binding. N. benthamiana leaves co-expressing nLUC-receptors and cLUC-RGL2 were treated with 10 µM ABA (PYLs) or GA₃ (GID1s). Right: relative LUC signal (means ± SD, n = 3 independent biological replicates; different letters denote P < 0.05, one-way ANOVA). (B) In vitro binding affinity assessment. GST-tagged receptors incubated with MBP-RGL2 with corresponding hormones, pulled down with anti-GST agarose. (C) GID1 disrupts PYL–RGL2 complexes. BiFC in N. benthamiana co-expressing YNE-RGL2, PYR1/PYL4/11-YCE, and mCherry or GID1a/b-mCherry. Scale bar = 25 µm. (D) PYL interferes with GID1–RGL2 association. BiFC in N. benthamiana co-expressing YNE-RGL2, GID1a/b-YCE, and mCherry/PYR1/PYL4/11-mCherry. Scale bar = 25 µm. (E) Fluorescence quantification of (C, D) (YFP/mCherry ratio). Data are means ± SD; n = 3 independent biological replicates, two-way ANOVA followed by Tukey’s test. (F) Yeast three-hybrid competition assay (PYL vs GID1). Left: AD-RGL2 + BD-PYL-GID1 grown on Met-containing medium (Met + ); GID1 is not induced, and the PYL–RGL2 interaction is recorded. Middle: identical strain on Met-free medium (Met–); GID1 is expressed but, without GA, remains inactive, so PYL–RGL2 signal is unchanged. Right: Met-free medium plus GA (Met–/GA + ); GA-activated GID1 now binds RGL2 and outcompetes PYL, causing almost abolishing the PYL–RGL2 reporter growth. All plates were supplemented with 10 µM ABA to keep intracellular ABA constant; growth on SD/–Trp/–Leu/–His/–Ade is shown after 4 days. (G) Yeast three-hybrid competition assay (GID1 vs PYL). Left (Met + ): PYL is not induced; basal GID1–RGL2 interaction is detected. Middle (Met–): PYL expression is switched on; PYL binds RGL2 and reduces the GID1–RGL2 signal even without ABA. Right (Met–/ABA + ): 10 µM ABA strengthens PYL–RGL2 association, further leading the reporter signal to drop. All plates were supplemented with 10 µM GA to keep intracellular GA constant. Source data are available online for this figure.
Our competition assays employed complementary systems to decipher regulatory hierarchies. BiFC revealed that receptor abundance alone is sufficient to drive competition. GID1 overexpression abolished approximately 80% PYR1/PYL4/11–RGL2 fluorescence without exogenous GA, whereas PYR1/PYL4/11 overexpression reduced GID1a/b–RGL2 binding by ~40% without exogenous ABA (Fig. 2C–E; Appendix Fig. S5), indicating that intrinsic binding capacity differences enable basal competition.
Furthermore, our yeast three-hybrid (Y3H) system revealed hormones as essential enablers of receptor competition, establishing a tripartite regulatory paradigm. Crucially, GID1 could not interact with RGL2 without GA supplementation in yeast (Fig. EV2), demonstrating that GA binding is the fundamental prerequisite for its competitive activity. Under GA-saturated conditions, induced GID1 expression nearly blocked the PYR1/PYL4/11–RGL2 complexes formation (Fig. 2F; Appendix Fig. S6), confirming that GA licenses GID1’s competitive superiority. Conversely, suppression assays required dual-hormone conditions—GA to maintain GID1–RGL2 binding and ABA to activate PYLs. Induced PYR1/PYL4/11 expression inhibited the formation of GID1–RGL2 complexes, and ABA enhanced this suppression (Fig. 2G; Appendix Fig. S6), revealing a role for ABA in optimizing PYL competitive efficacy against GA-bound complexes. This hormone-gated system operates through three integrated layers: GA binding enables GID1’s competitive capacity, ABA signaling enhances PYL’s inhibitory potential, and intrinsic binding capacity differences determine the baseline displacement efficiency. The resulting activation and binding capacity-tuned switch precisely converts hormone availability into binding outcomes, thus modulating germination timing under environmental fluctuations.
Genetic architecture of the receptor competition module
Genetic epistasis confirmed hierarchical organization. The pyr1 pyl1 pyl2 pyl4 quadruple mutant and rgl2 knockout mutant exhibited strong insensitivity to both PAC-induced GA deficiency and ABA-mediated germination suppression. PYL4 overexpression conferred hypersensitivity to both compounds, while GID1b overexpression promoted significant resistance. Moreover, PYL4 overexpression only partially inhibited GID1b-overexpression effects, showing a similar germination rate to wild-type when co-expressed, whereas rgl2 mutation in PYL4 overexpression abolished all phenotypic modifications (Fig. 3A,B). This receptor antagonism creates a tunable decision threshold. PYLs delay germination for stress assessment during transient ABA spikes, while GID1’s superior binding capacity ensures timely emergence when conditions improve, optimizing fitness in fluctuating environments.Figure 3. Genetic architecture of the receptor competition threshold.(A) Genetic hierarchy of germination. Seeds germinated on ½ MS ± 1 µM PAC or ABA, scale bar = 0.2 mm. (B) Quantification of hormone sensitivity. Germination rates after 2 days are analyzed. Data are means ± SD, n = 3 independent biological replicates; different letters indicate P < 0.05 by two-way ANOVA followed by Tukey’s test. Source data are available online for this figure.
DWA1 functions as a tunable degradation conduit
Considering that ABA only moderately regulates the content of RGL2, while RGL2 accumulates excessively when SLY1 function is inhibited, we explored new components regulated by ABA and involved in controlling RGL2 stability regulation (Ariizumi and Steber, 2007; Piskurewicz et al, 2008). This led to the identification of DWA1, a WD40-repeat protein associated with ubiquitination pathways, as a direct interactor of RGL2. Co-immunoprecipitation (Co-IP) assays showed that RGL2-GFP co-precipitated with MYC-DWA1 when co-expressed in N. benthamiana leaf cells (Fig. 4A), and BiFC assays showed clear YFP fluorescence only when YNE-RGL2 and DWA1-YCE were co-expressed in N. benthamiana (Fig. 4B).Figure 4DWA1 executes GA-dependent RGL2 ubiquitination.(A) Co-immunoprecipitation (Co-IP) of DWA1–RGL2 complexes. Total proteins from N. benthamiana leaves co-expressing MYC-DWA1 and RGL2-GFP were immunoprecipitated with anti-MYC agarose and immunoblotted with the indicated antibodies. IP immunoprecipitation. (B) BiFC validation of DWA1–RGL2 interaction. YFP signals in N. benthamiana epidermal cells co-expressing YNE-RGL2 and DWA1-YCE were visualized (scale bar = 50 µm). (C) Time-course of GA-induced RGL2 degradation. Five-day-old MYC-RGL2 seedlings in the indicated backgrounds were pretreated with 10 µM PAC for 60 min before 10 µM GA₃ exposure. RGL2 levels were quantified by anti-MYC immunoblotting (right: densitometric analysis normalized to ACTIN; results represent the means ± SD, two-way ANOVA, n = 3). (D) DWA1-mediated RGL2 ubiquitination in vivo. RGL2-GFP and Flag-UBQ10 were expressed in N. benthamiana with/without MYC-DWA1. Tissues were treated with 50 µM MG132 (12 h), followed by anti-GFP immunoprecipitation and anti-Flag immunoblotting. (E) GA enhancement of DWA1-dependent RGL2 ubiquitination. Five-day-old seedlings (genotypes indicated) were treated with 50 µM MG132 for 6 h (-GA), and then treated with 10 µM GA₃ + 50 µM MG132 for 3 h (+GA). Ubiquitinated RGL2 was detected by anti-ubiquitin immunoblotting after anti-MYC IP. (F) Endogenous RGL2 abundance during germination. Seeds were imbibed in 10 µM PAC at 4 °C for 48 h, washed to remove the inhibitor, and allowed to germinate for an additional 48 h on PAC-free medium; ACTIN was used as loading control. (G) Ubiquitination of endogenous RGL2 after germination. After the same imbibition, indicated seeds were germinated for 12 h in the presence of 50 µM MG132. RGL2 was immunoprecipitated with anti-RGL2 antibody and probed with anti-ubiquitin antibody. Source data are available online for this figure.
Given DWA1’s known association with the CUL4–DDB1 E3 ubiquitin ligase complex (Lee et al, 2010), we investigated its role in GA-mediated RGL2 degradation. Seedlings with MYC-RGL2 expressed in WT, dwa1-1 (a DWA1 null mutant), and DWA1-overexpression (35S:Flag-DWA1) backgrounds were used for analysis (Appendix Fig. S7). In dwa1-1, GA-induced RGL2 degradation was impaired, with MYC-RGL2 protein levels remaining at 48% after 60-minute GA treatment, compared to 24% in wild-type controls. Conversely, DWA1 overexpression accelerated RGL2 turnover, reducing MYC-RGL2 to 44% compared to 66% in wild-type at 30 min (Fig. 4C; Appendix Fig. S3).
In vivo ubiquitination assays in N. benthamiana demonstrated that co-expression of MYC-DWA1 strongly increased the RGL2-GFP ubiquitination levels (Fig. 4D). This phenomenon was also observed in 35S:Flag-DWA1, and this enhancement was further amplified when GA was supplied (Fig. 4E). To examine the genetic requirement for GA, we monitored RGL2 dynamics in the GA-deficient mutant ga1-t. In this background, RGL2 remained stable throughout germination and showed virtually no ubiquitination (Fig. 4F,G). Expressing 35S:Flag-DWA1 in ga1-t barely enhanced RGL2 turnover or its ubiquitination, demonstrating that DWA1-mediated RGL2 ubiquitination strictly depends on bioactive GA (Fig. 4F,G). These findings establish DWA1 as a critical E3 ligase substrate receptor executing GA-triggered RGL2 degradation.
Receptor-mediated regulation of the degradation machinery
Our analysis revealed that DWA1 activity was modulated by both receptor systems. ABA receptors sequester DWA1 through direct binding. BiFC assays detected clear YFP fluorescence when all 14 tested PYL receptors were paired with DWA1 (Fig. 5A; Appendix Fig. S1), indicating universal binding capacity. In vitro pull-down assays confirmed the physical interaction of DWA1 with PYR1/PYL2/4/5/8 (Fig. 5B). Hormonal modulation studies revealed that ABA significantly stabilized PYL–DWA1 complexes, with LCI assays quantifying luminescence increases for key PYLs (PYR1, PYL2, PYL4) following ABA treatment (Fig. 5C; Appendix Fig. S8).Figure 5PYLs directly interact with DWA1 in an ABA-dependent manner.(A) BiFC visualization of PYL–DWA1 interactions. YFP signals in N. benthamiana co-expressing YNE-DWA1 and YCE-PYLs were visualized, scale bar = 25 µm. (B) In vitro pull-down of PYL–DWA1 complexes. GST-tagged PYLs were incubated with MBP-DWA1, pulled down with anti-GST agarose, and immunoblotted. GST alone served as control. (C) ABA enhancement of PYL–DWA1 interactions. LCI assays in N. benthamiana expressing nLUC-PYLs and cLUC-DWA1. Leaves were injected with 10 µM ABA 3 h pre-assay. Data are means ± SD (n = 3 independent biological repeats). Significant differences were determined by two-way ANOVA. Source data are available online for this figure.
Functional consequence analysis showed that the co-expression of PYL4 dramatically inhibited DWA1-mediated RGL2-GFP ubiquitination (Fig. 6A). Moreover, the RGL2 level in 35S:PYL4-GFP 35S:Flag-DWA1 after GA treatment was similar to that in 35S:PYL4-GFP, which was higher than that in wild type (Fig. 6B). Co-IP revealed that PYL4 competed with RGL2 for DWA1 binding. DWA1 still interacted with both proteins, but the RGL2 signal in the anti-MYC precipitate dropped 4-fold when PYL4 was present (Fig. 6C). LCI and BiFC assays corroborated this competitive displacement. PYL4-mCherry halved the LUC signal and reduced the YFP fluorescence of the DWA1–RGL2 complex by ~70% compared with mCherry alone (Fig. 6D,E; Appendix Figs. S9A and S10A). The same PYL4-mCherry construct also weakened the interaction between DWA1 and the CUL4 adapter DDB1A/B (Fig. EV3; Appendix Fig. S10B), indicating that ABA-bound PYL receptors sequester DWA1 from the entire CUL4 complex and thus stabilize RGL2.Figure 6. Receptors gate the DWA1–RGL2 interaction.(A) PYL4 restrains DWA1-catalyzed ubiquitination. N. benthamiana leaves co-expressing RGL2-GFP, Flag-UBQ10, and MYC-DWA1 were infiltrated ± PYL4-mCherry and 50 µM MG132 was added 12 h before harvest. (B) Cell-free degradation of recombinant MBP-RGL2. Purified MBP-RGL2 was incubated with extracts from 5-day-old WT, 35S:Flag-DWA1, 35S:PYL4-GFP or 35S:Flag-DWA1 35S:PYL4-GFP seedlings in the presence of 10 µM GA₃ for the indicated times. Rubisco (Ponceau S) is a loading control. (C) Co-IP confirmation of PYL4 interference. MYC-DWA1 and RGL2-GFP were co-expressed ± PYL4-mCherry; anti-MYC IP, immunoblot as indicated. (D) PYL4 disrupts the DWA1–RGL2 association (LCI). N. benthamiana expressing nLUC-RGL2 + cLUC-DWA1 + mCherry/PYL4-mCherry. Data are means ± SD, n = 3 independent biological replicates. Significant differences were determined by one-way ANOVA followed by Tukey’s test. (E) Same interaction monitored by BiFC. YNE-RGL2 + DWA1-YCE + mCherry/PYL4-mCherry; scale bar = 25 µm. (F) GA-dependent enhancement of DWA1–RGL2 interaction. nLUC-RGL2 + cLUC-DWA1 expressed for 2 days with 50 µM MG132, then treated ± 10 µM GA₃ for 3 h. Data are means ± SD, n = 3 independent biological repeats. Significant differences were determined by unpaired *t-*test (P = 0.00012). (G) GID1 requirement for GA-induced complex assembly. Co-IP in wild-type and gid1b/c protoplasts expressing MYC-DWA1 and RGL2-GFP ± 10 µM GA₃. gid1b/c: GID1-deficient mutant. (H) GID1 promotes DWA1-mediated RGL2 ubiquitination in vivo. MYC-DWA1, RGL2-GFP and Flag-UBQ10 expressed ± GID1a/b-mCherry. Tissues were treated with 50 µM MG132 and 10 µM GA₃ (12 h), followed by anti-GFP immunoprecipitation and anti-Flag immunoblotting. (I) GID1 dependence of GA-accelerated RGL2 turnover. MBP-RGL2 protein was incubated with extracts from the 5-day-old WT, 35S:Flag-DWA1, gid1b/c or 35S:Flag-DWA1 gid1b/c seedlings plus 10 µM GA_3_ for the indicated times. Source data are available online for this figure.
Conversely, GID1 proved essential for GA to accelerate DWA1–RGL2 interaction. LCI showed that GA tripled the luciferase output from nLUC-RGL2/cLUC-DWA1 pair (Fig. 6F; Appendix Fig. S9B). Co-IP reproduced this GA dependence, showing a 2.6-fold GA-stimulated increase in DWA1 association with RGL2-GFP, which was largely eliminated in the gid1a/b double mutant background (Fig. 6G). Furthermore, the presence of GID1a or GID1b greatly enhanced the DWA1-mediated RGL2 ubiquitination, and the promotion of RGL2 degradation by DWA1 was almost blocked in the gid1b/c double mutant (Fig. 6H,I). Taken together, these data demonstrate that PYL and GID1 act as antagonistic switches on DWA1, and DWA1 functions as a receptor-gated degradation channel that integrates the two hormone signals into a single output—RGL2 stability.
Hierarchical organization of the DWA1–RGL2 signaling network
Genetic epistasis defined the signaling hierarchy. Phenotypic characterization revealed that dwa1 mutants exhibited pronounced hypersensitivity to PAC and ABA-mediated germination inhibition, while 35S:Flag-DWA1 lines showed significant PAC and ABA resistance (Fig. 7A,B). Moreover, PYL4 overexpression completely suppressed the PAC/ABA resistance phenotypes conferred by DWA1 overexpression, positioning PYLs upstream in the genetic hierarchy (Fig. 7A,B). Crucially, genetic epistasis analysis demonstrated that rgl2 mutation completely suppressed the PAC/ABA-hypersensitive phenotype of dwa1 mutants (Fig. 7A,B), establishing RGL2 as the exclusive downstream effector. This integrative analysis establishes RGL2 as the terminal effector, with receptor competition determining its stability through regulation of the DWA1-mediated degradation pathway.Figure 7. Hierarchical organization of the DWA1–RGL2 signaling network.(A) Genetic hierarchy of germination. Seeds germinated on ½ MS ± 1 µM PAC or ABA, scale bar = 0.2 mm. (B) Quantification of hormone sensitivity. Germination rates after 2 days are analyzed. Data are means ± SD, n = 3 independent biological repeats; different letters indicate P < 0.05 by two-way ANOVA followed by Tukey’s test. Source data are available online for this figure.
Discussion
Seed germination is dynamically controlled by GA–ABA antagonism converging on RGL2 stability. Our study establishes a paradigm in which ABA receptors (PYLs) and GA receptors (GID1s) engage in direct bidirectional competition for physical occupancy of RGL2, the central germination repressor. This receptor-level confrontation operates as a molecular switch: PYL binding stabilizes RGL2 through dual mechanisms—direct interaction and functional sequestration of DWA1, the substrate receptor for E3 ubiquitin ligase. GID1s overcome this stabilization by competitively displacing PYLs, leveraging superior binding capacity to license DWA1-mediated degradation. Crucially, this receptor competition works alongside classical hormone concentration dynamics to determine germination outcomes (Fig. 8).Figure 8. Bidirectional receptor competition on RGL2 establishes a tunable degradation switch for germination control.ABA-activated PYL receptors stabilize RGL2 through dual locking mechanisms: (i) functional sequestration of DWA1 to disrupt CUL4–DDB1 E3 ligase assembly, and (ii) direct degron masking via physical binding. GA-bound GID1 receptors overcome PYL-mediated stabilization through competitive displacement, leveraging their superior binding capacity for RGL2 to liberate degradation sites. Crucially, this receptor competition synergizes with hormone concentration dynamics—PYLs attenuate GID1–RGL2 interaction under ABA dominance to impose a reversible pause state, while sustained GA enables GID1s to license DWA1-mediated polyubiquitination and proteasomal degradation. This cooperative mechanism converts fluctuating hormone ratios into a proteolytic threshold that optimizes germination timing in variable environments.
The competition mechanism integrates two synergistic tiers that confer dynamic responsiveness. First, basal receptor abundance sets the competition threshold, demonstrated in hormone-free systems where PYLs and GID1s compete for RGL2 occupancy without hormonal input (BiFC; Fig. 2C–E). Second, hormones modulate competition through asymmetric mechanisms: GA is essential for GID1 binding competence, whereas ABA enhances PYL’s inhibitory potency (Y3H; Fig. 2F,G). This receptor-centric layer cooperates with hormone accumulation pathways, in which GA-dependence prevents premature commitment, and ABA-amplification enables rapid stress responses.
Classical ABA/GA ratio models remain essential for interpreting bulk DELLA accumulation (Liu and Hou, 2018). However, receptor competition provides complementary explanatory power for context-specific paradoxes: gid1 mutants fail to germinate even under saturating GA conditions (Griffiths et al, 2006; Iuchi et al, 2007), and ABA treatment induces only moderate RGL2 accumulation compared to PAC-induced extremes (Ariizumi and Steber, 2007; Piskurewicz et al, 2008). Our data demonstrate that hormone concentration and receptor competition jointly determine functional outcomes. Crucially, the constitutive binding superiority of GID1 facilitates its displacement of PYLs even when ABA and GA are present at comparable concentrations. This unique advantage permits the generation of graded responses in scenarios where alterations in hormone concentration alone would prove insufficient for such modulation.
Having confirmed that DWA1 catalyzes RGL2 ubiquitination in a GA-enhanced manner, we asked how DWA1 relates to the canonical F-box protein SLY1. Our data position DWA1 as a context-dependent modulator that operates alongside, but is distinct from, SLY1. Immunoblots showed that germinating dwa1 seeds contained only trace amounts of RGL2, whereas sly1 seeds retained abundant RGL2 (Fig. EV4A) (McGinnis et al, 2003; Ariizumi and Steber, 2007), confirming that SLY1 is the principal germination-triggered degrader. Accordingly, DWA1 loss-of-function delayed RGL2 disappearance much less severely than SLY1 loss-of-function. Nevertheless, DWA1 overexpression can modestly accelerate RGL2 turnover even in the sly1 background (Fig. EV4B). Moreover, we found that the two proteins interacted in vivo without DWA1 altering SLY1 abundance (Fig. EV4C–E). These observations, together with DWA1’s WD40 propeller structure (Skowyra et al, 1997; Stirnimann et al, 2010), suggest that DWA1 does not replace SLY1 but acts as a scaffold that facilitates SLY1-mediated substrate recognition or ubiquitin transfer when environmental cues demand fine-tuned control of DELLA abundance. Thus, SLY1 provides the constitutive “on/off” switch for germination, whereas DWA1 supplies an adjustable rheostat that integrates hormone and stress signals to set the speed of RGL2 removal.
Previous work established DWA1 as a negative regulator of ABA signaling by accelerating ABI5 degradation (Lee et al, 2010). Our study extends this role to the DELLA protein RGL2, revealing a common biochemical logic: DWA1 assembles with CUL4–DDB1 to ubiquitinate both transcriptional repressors, thereby relieving ABA-imposed dormancy. Conversely, we show that all tested PYLs physically interact with DWA1, and that PYL4 competitively excludes DDB1 to prevent ligase activation. Consequently, PYL binding simultaneously protects RGL2 and, presumably, ABI5 from proteasomal removal, amplifying ABA responses. This dual control converts the PYL–DWA1 association into a rapid hormonal toggle: when GA levels rise, GID1 displaces PYLs from RGL2, permitting CUL4–DWA1 complex to polyubiquitinate RGL2; when ABA accumulates, elevated PYL expression sequesters DWA1 in an inactive complex, stabilizing both RGL2 and ABI5. Such reversible sequestration of a shared E3 adapter provides an elegant mechanism for seeds to calibrate the balance between stress tolerance and germination commitment.
The intricate regulation of receptor expression during seed germination reveals unexpected feedback mechanisms that refine signal perception. Paradoxically, while ABA promotes the expression of specific PYL isoforms (e.g., PYL7/9/11/12/13), it suppresses others (e.g., PYR1/PYL1/2/4/6/8) (Zhao et al, 2020), a functional dichotomy mirrored in GA signaling where GA represses GID1a/b/c transcription, whereas GA biosynthesis inhibitors (e.g., uniconazole) strongly induce their expression (Iuchi et al, 2007). This counterintuitive regulation suggests that negative feedback loops function to prevent signal oversaturation, thereby enabling precise hormonal gating. Crucially, these transcriptional dynamics align with our observed isoform-specific functional divergence: ABA enhances PYR1/PYL4/11–RGL2 binding and potentiates PYR1/PYL2/4’s DWA1 sequestration capacity, yet leaves PYL5/8 unaffected. This likely reflects physiological partitioning, whereby stress-inducible PYL11 (seed-specific) prioritizes germination control, while constitutively expressed PYL5 regulates unrelated processes such as stomatal closure (Dittrich et al, 2019; Zhao et al, 2020).
Environmentally, this system converts fluctuating conditions into adaptive germination decisions. Transient ABA spikes (e.g., during soil drying) enable PYLs to rapidly stabilize RGL2 through enhanced binding capacity, imposing a reversible “pause state”. Conversely, sustained GA elevation progressively activates pre-existing GID1 pools, cumulatively displacing PYLs to commit to germination. The PYL4-OE phenotype, delayed but not blocked germination (Fig. 3), validates a commitment threshold tuned by both hormone levels and receptor interactions.
In conclusion, our “Competitive Receptor Arbitration” model advances three key concepts. First, receptor competition and hormone concentration jointly fine-tune signal integration. Second, the two-tier mechanism (basal competition and hormonal modulation) enables micro-adjustments to fluctuating environments. Third, DWA1 functions as a tunable executor that translates competitive outcomes into proteolytic commitment. This cooperative framework illuminates how plants achieve precision control in noisy environments, suggesting engineering PYL–GID1 binding hierarchies could optimize crop resilience, e.g., modulating PYL11 induction kinetics might extend the “assessment window” in climate-variable regions.
Methods
Reagents and tools tableReagent/resourceReference or sourceIdentifier or catalog number Experimental models Arabidopsis: Col-0Widely distributedN/AArabidopsis: gid1b/cZhong et al, 2021N/AArabidopsis: 35S:Flag-GID1bZhong et al, 2021N/AArabidopsis: 35S:PYL4-GFPZhao et al, 2024N/AArabidopsis: 35S:Flag-PYL11Zhao et al, 2020N/AArabidopsis: pyr1 pyl1 pyl2 pyl4Park et al, 2009N/AArabidopsis: dwa1-1ABRCSALK_051022Arabidopsis: dwa1-2ABRCSALK_021789Arabidopsis: rgl2ABRCSALK_124231Arabidopsis: sly1ABRCSAIL_836_B01Arabidopsis: 35S:Flag-DWA1This paperN/AArabidopsis: dwa1-1 rgl2This paperN/AArabidopsis: 35S:MYC-RGL2This paperN/AArabidopsis: 35S:MYC-RGL2 dwa1-1This paperN/AArabidopsis: 35S:MYC-RGL2 35S:Flag-DWA1This paperN/AArabidopsis: 35S:MYC-RGL2 pyr1 pyl1 pyl2 pyl4This paperN/AArabidopsis: 35S:MYC-RGL2 35S:PYL4-GFPThis paperN/AArabidopsis: 35S:MYC-RGL2 35S:Flag-PYL11This paperN/AArabidopsis: 35S:Flag-DWA1 35S:PYL4-GFPThis paperN/AArabidopsis: 35S:PYL4-GFP dwa1-1This paperN/AArabidopsis: 35S:PYL4-GFP rgl2This paperN/AArabidopsis: 35S:PYL4-GFP 35S:Flag-GID1bThis paperN/AArabidopsis: 35S:Flag-DWA1 sly1This paperN/AArabidopsis: 35S:Flag-DWA1 gid1b/cThis paperN/A Recombinant DNA
pCAMBIA1300-35S-RGL2-GFPc This paperN/A pCAMBIA1307-35S-MYC-DWA1 This paperN/A pCAMBIA1307-35S-MYC-RGL2 This paperN/A pCAMBIA1300-35S-mCherry This paperN/A pCAMBIA1300-35S-GID1a-mCherry This paperN/A pCAMBIA1300-35S-GID1b-mCherry This paperN/A pCAMBIA1300-35S-PYR1-mCherry This paperN/A pCAMBIA1300-35S-PYL4-mCherry This paperN/A pCAMBIA1300-35S-PYL11-mCherry This paperN/A pSPYNE-RGL2 This paperN/A pSPYNE-DWA1 This paperN/A pSPYCE-DWA1 This paperN/A pSPYCE-PYR1 This paperN/A pSPYCE-PYL1 This paperN/A pSPYCE-PYL2 This paperN/A pSPYCE-PYL3 This paperN/A pSPYCE-PYL4 This paperN/A pSPYCE-PYL5 This paperN/A pSPYCE-PYL6 This paperN/A pSPYCE-PYL7 This paperN/A pSPYCE-PYL8 This paperN/A pSPYCE-PYL9 This paperN/A pSPYCE-PYL10 This paperN/A pSPYCE-PYL11 This paperN/A pSPYCE-PYL12 This paperN/A pSPYCE-PYL13 This paperN/A pCAMBIA1300-nLUC-PYR1 This paperN/A pCAMBIA1300-nLUC-PYL2 This paperN/A pCAMBIA1300-nLUC-PYL4 This paperN/A pCAMBIA1300-nLUC-PYL5 This paperN/A pCAMBIA1300-nLUC-PYL8 This paperN/A pCAMBIA1300-nLUC-PYL11 This paperN/A pCAMBIA1300-nLUC-GID1a This paperN/A pCAMBIA1300-nLUC-GID1b This paperN/A pCAMBIA1300-nLUC-RGL2 This paperN/A pCAMBIA1300-cLUC-DWA1 This paperN/A pCAMBIA1300-cLUC-RGL2 This paperN/A pGEX-4T-1-PYR1 This paperN/A pGEX-4T-1-PYL2 This paperN/A pGEX-4T-1-PYL4 This paperN/A pGEX-4T-1-PYL5 This paperN/A pGEX-4T-1-PYL8 This paperN/A pGEX-4T-1-PYL11 This paperN/A pGEX-4T-1-GID1a This paperN/A pGEX-4T-1-GID1b This paperN/A pET21a-MBP-DWA1 This paperN/A pET21a-MBP-RGL2 This paperN/A pET21a-sfGFP-RGL2 This paperN/A pGADT7-RGL2 This paperN/A pBridge-BD-GID1a-PYR1 This paperN/A pBridge-BD-GID1a-PYL4 This paperN/A pBridge-BD-GID1a-PYL11 This paperN/A pBridge-BD-GID1b-PYR1 This paperN/A pBridge-BD-GID1b-PYL4 This paperN/A pBridge-BD-GID1b-PYL11 This paperN/A pBridge-BD-PYR1-GID1a This paperN/A pBridge-BD-PYL4-GID1a This paperN/A pBridge-BD-PYL11-GID1a This paperN/A pBridge-BD-PYR1-GID1b This paperN/A pBridge-BD-PYL4-GID1b This paperN/A pBridge-BD-PYL11-GID1b This paperN/A Antibodies Anti c-Myc AntibodyCWBIOCat#: CW0229MAnti-ACTINCWBIOCat#: CW0264Anti-GFP antibodySangon BiotechCat#: D110008anti-FLAG antibodySangon BiotechCat#: D110005Anti-RGL2 antibodyAgriseraCat#: AS11 1803Anti-SLY1 antibodyPHYTOABCat#: PHY0883SAnti-luciferase antibodySigma-AldrichCat#: No. L0159Anti-cLuc antibodySigma-AldrichCat#: No. L2164anti-ubiquitin antibodySanta Cruz BiotechnologyCat#: c0519Anti-GST antibodyABclonalCat#: AE006Anti-mCherry antibodyproteintechCat#: 26765Anti-MBP antibodyproteintechCat#: 66003HRP-conjugated Goat Anti-Mouse IgG(H + L)ProteintechCat#: SA00001-1HRP-conjugated Goat Anti-Rabbit IgG(H + L)ProteintechCat#: SA00001-2 Chemicals, enzymes, and other reagents Anti-MYC Nanobody Magarose BeadsAlpalifeBioCat#: KTSM1336Anti-GFP Nanobody Magarose BeadsAlpalifeBioCat#: KTSM1334GST-Sefinose Resin 4FFSangon BiotechCat#: C600031Abscisic acidMCECat#: HY-N2549Gibberellin A3MCECat#: HY-N1964PaclobutrazolSigma-AldrichCat#: No. 46046MG-132MCECat#: HY-13259D-LuciferinMCECat#: HY-12591A Software ImageJNIHRRID: SCR_003070GraphPad Prism 9Informer Technologies, Inc.RRID: SCR_002798
Plant materials and growth conditions
The Arabidopsis thaliana ecotype Columbia-0 (Col-0) was used as the wild-type in this study. The dwa1-1 (SALK_051022), dwa1-2 (SALK_021789), gid1b/c, and 35S:Flag-GID1b were described previously (Lee et al, 2010; Zhong et al, 2021). The pyr1 pyl1 pyl2 pyl4 quadruple mutant and 35S:PYL4-GFP were described previously (Park et al, 2009; Zhao et al, 2024). The T-DNA insertion lines rgl2 (SALK_124231) and sly1 (SAIL_836_B01) were obtained from the Arabidopsis Biological Resource Center.
Arabidopsis plants were planted in a growth chamber under a 16-h-light/8-h-dark photoperiod at 100 μmol m^−2^ s^−1^ at 23 °C. To grow Nicotiana benthamiana, the seeds were first sown on soil (nutrient soil:vermiculite; [1:1; v/v]) for 10 days, and then the young seedlings were transplanted into culture boxes individually under a 16-h-light/8-h-dark photoperiod at 100 μmol m^−2^ s^−1^ at 23 °C.
Seed germination assays
For the seed germination assays, harvested seeds dried at room temperature for 4 weeks were used for analysis. Seeds were plated on 1/2 MS or 1/2 MS containing 0.5–1 μM PAC or ABA, as indicated. After stratification in the dark at 4 °C for 3 days, plates were transferred to the growth chamber. Seeds were identified as germinated when a radicle had emerged from the seed coat (Shu et al, 2013).
Construction of transgenic plants
The full-length coding sequences of RGL2, PYR1, PYL4, PYL11, GID1a, and GID1b were amplified and cloned into pCAMBIA1300 vectors to generate the 35S:GFP and 35S:mCherry constructs. Similarly, the entire coding sequences of RGL2 and DWA1 were amplified and cloned into pCAMBIA1307 vectors to generate the 35S:MYC and 35S:Flag constructs. The resultant plasmids were confirmed by sequencing and used for plant transformation. All transgenic plants were generated using the Agrobacterium-mediated flower-dip transformation method, and the homozygous transgenic plants from the T_4_ generation were used for phenotype analyses in this study. The 35S:MYC-RGL2 line was crossed to dwa1-1, 35S:Flag-DWA1 #1, pyr1 pyl1 pyl2 pyl4, 35S:PYL4-GFP and 35S:Flag-PYL11 to express MYC-RGL2 in respective genetic backgrounds.
Yeast two-hybrid (Y2H) and yeast three-hybrid (Y3H) assays
Full-length GID1a/b/c coding sequences were cloned into pGBKT7 (BD) as bait vectors. RGL2 was fused to the pGADT7 (AD) vector. The corresponding vectors were cotransformed into the yeast (Saccharomyces cerevisiae) strain AH109. Yeast cells were selected on medium lacking Trp and Leu (SD/-Trp-Leu) to select transformants or on medium lacking Trp, Leu, Ade, and His (SD/-Trp-Leu-Ade-His) with or without 10 µM GA to identify interactions.
For Y3H, PYR1/PYL4/11 and GID1a/b were cloned into pBridge-BD vector (Clontech) to generate BD-PYL-GID1 or BD-GID1-PYL fusion constructs. Selected transformants were transferred into medium lacking Trp, Leu, Ade, His, and Met (SD/-Trp-Leu-Ade-His-Met) containing 10 μM GA and ABA with or without 1 mM Met to assess interactions (Suzuki et al, 2009).
BiFC and LCI assays
The yellow fluorescent protein (YFP) was splited into the N-terminal fragment (YNE) and the C-terminal fragment (YCE). The full-length CDS of DWA1, RGL2, and DDB1A/B genes were cloned into the pSPYNE vector, and DWA1, 14 PYLs without stop codons were cloned into the pSPYCE vector. The resulting constructs were co-expressed in N. benthamiana leaves. YFP fluorescence signals were detected using a laser scanning confocal microscope (ZEISS LSM980; lasers: 488 nm, 15%; 560 nm, 15%; gains: 600; pinhole: 100 μm).
For the LCI assay, PYR1/PYL2/4/5/8/11, GID1a/b, and RGL2 coding sequences were fused with the N-terminal region of LUC (nLUC) to generate nLUC-PYL and nLUC-RGL2, and DWA1, RGL2 with the C-terminal region of LUC (cLUC) to generate cLUC-DWA1 and cLUC-RGL2. The resulting constructs were cotransformed into N. benthamiana by Agrobacterium-mediated transformation for 2 d. For GA and ABA treatments, 10 μM GA or ABA was injected into leaves 3 h before detection. In all, 100 μM D-luciferin was sprayed on the leaves before image collection by a CCD camera. The primers used are listed in Appendix Table S1. Total proteins were extracted for immunoblot analysis. nLUC and cLUC fusion proteins were detected with anti-luciferase antibody and anti-cLUC antibody, respectively (Guo et al, 2025).
Microscale thermophoresis (MST) assay
Microscale thermophoresis (MST) was performed with sfGFP-RGL2 (target; 0.5 µM final concentration) produced in E. coli and purified GST-GID1A/B or GST-PYR1/PYL4/PYL11 (ligands). The reaction buffer contained 20 mM HEPES pH 7.5, 250 mM NaCl, 0.5 mM TCEP, 0.01% (v/v) Tween-20, 5% (v/v) glycerol, and 100 µM GA₃ or ABA as indicated. Ligands were prepared in 16 two-fold serial dilutions in the same buffer, mixed 1:1 (10 µL each) with sfGFP-RGL2 solution, incubated for 10 min at 25 °C, and loaded into standard capillaries. Measurements were carried out on a Monolith NT.115 (blue channel; LED 20%; MST power: High) with three independent replicates per combination (Zhang et al, 2024).
Co-IP assays
For DWA1–RGL2 interaction identification, 35S:MYC-DWA1 was co-expressed with 35S:RGL2-GFP in N. benthamiana leaves. Total proteins were extracted using extraction buffer (10 mM Tris pH 7.5, 0.5% Nonidet P-40, 2 mM EDTA, 150 mM NaCl, 1 mM PMSF, and 1% [v/v] protease inhibitor cocktail). Proteins were purified using anti-MYC or anti-GFP agarose and detected using the corresponding antibody.
For GA treatment, RGL2-GFP and MYC-DWA1 were transiently expressed in WT or gid1b/c mutant protoplasts with 50 μM MG132 applied; 10 μM GA_3_ was added 3 h before protein extraction. For evaluating PYL4’s effect on DWA1–RGL2 interaction, RGL2-GFP and MYC-DWA1 were transiently expressed in N. benthamiana leaves with or without PYL4-mCherry.
In vitro pull-down assay
The coding sequences of PYR1, PYL2, PYL4, PYL5, PYL8, and PYL11 were cloned into pGEX-4T-1 vectors to obtain GST-PYL constructs, while DWA1 and RGL2 were cloned into pET21a-MBP vectors to obtain MBP-DWA1/RGL2 constructs. For in vitro pull-down assays, the indicated proteins were included in the pull-down buffer (50 mM Tris-HCl, pH 7.5, 100 mM NaCl, 0.2% [v/v] Glycerol, 0.6% [v/v] Triton X-100, 0.1 mM EDTA) for incubation, and then proteins were purified using GST antibody-conjugated agarose, and their interacting proteins were detected through immunoblot analysis using anti-MBP antibody.
In vivo ubiquitination assay
For the ubiquitination of RGL2, 35S:RGL2-GFP and 35S:Flag-UBQ10 were transiently expressed in N. benthamiana with or without 35S:MYC-DWA1. The leaves were infiltrated with 50 μM MG132 for 12 h before being sampled. Total proteins were extracted using extraction buffer and RGL2-GFP protein was purified by GFP antibody-conjugated agarose. Ubiquitinated forms of RGL2 were detected by the anti-Flag antibody. For the ubiquitination assay showing DWA1’s role in GA-mediated RGL2 ubiquitination, 5-day-old 35S:MYC-RGL2 and 35S:MYC-RGL2 35S:Flag-DWA1 seedlings were first treated with 50 μM MG132 for 6 h, then 10 µM GA_3_ and 50 μM MG132 were subsequently supplied for 3 h. Total proteins were extracted, and polyubiquitinated RGL2 protein was detected by anti-ubiquitin antibody.
In vivo protein degradation
For time-course of RGL2 degradation after GA application, 5-day-old seedlings of 35S:MYC-RGL2, 35S:MYC-RGL2 dwa1-1, 35S:MYC-RGL2 35S:Flag-DWA1, 35S:MYC-RGL2 35S:PYL4-GFP, 35S:MYC-RGL2 35S:Flag-PYL11, and 35S:MYC-RGL2 pyr1 pyl1 pyl2 pyl4 seedlings were treated with 10 µM PAC for 60 min followed by 10 µM GA_3_ for 30 or 60 min. RGL2 protein was detected using an anti-MYC antibody. ACTIN was used as a loading control.
Cell-free degradation
The cell-free degradation assay was performed as previously described (Wang et al, 2009). Total proteins of 5-day-old seedlings were extracted in degradation buffer (25 mM Tris-HCl, pH 7.5, 10 mM NaCl, 10 mM MgCl_2_, 4 mM PMSF, 5 mM DTT, and 10 μM ATP). The concentration of total protein extracts was adjusted to be equal to the degradation buffer (100 µL, 200 µg) in the presence of 10 µM GA₃. A total of 20 µg of the E. coli-purified MBP-RGL2 protein was added to the extracts for the individual assays and incubated in a growth chamber under continuous white light for the indicated times. Samples were taken to determine RGL2 protein abundance by immunoblot using anti-MBP antibody. Rubisco (Ponceau S) was used as the control of extract content.
Quantification and statistical analysis
Statistical parameters are reported in the figures and figure legends. Statistical analysis was performed using GraphPad Prism 9. The two-tailed Student’s t test was used to analyze the statistical significance between the two groups. One-way or two-way analysis of variance (ANOVA) followed by appropriate post-hoc tests was used to analyze the statistical significance between more than two groups. Different letters represent significant differences at P < 0.05. Values are represented as means ± SD.
Supplementary information
Appendix Peer Review File Source data Fig. 1 Source data Fig. 2 Source data Fig. 3 Source data Fig. 4 Source data Fig. 5 Source data Fig. 6 Source data Fig. 7 Appendix Fig S1 Source Data Expanded View Figures
The reference list from the paper itself. Each links out to its DOI / PubMed record.
- 1Guo P, Chong L, Jiao Z, Xu R, Niu Q, Zhu Y (2025) Salt stress activates the CDK 8-AHL 10-SUVH 2/9 module to dynamically regulate salt tolerance in Arabidopsis. Nat Commun 16:245410.1038/s 41467-025-57806-6PMC 1190395540074748 · doi ↗ · pubmed ↗
- 2Zhao H, Shin D, Zhu Y, Kim J (2025) Bridging the knowledge gap: Utilization of mediator subunits for crop improvement. Plant Cell Environ 48:213–22510.1111/pce.1514239254322 · doi ↗ · pubmed ↗