DELLA proteins are conserved growth regulators that play a central role in plant development in response to internal and external signals. As transcriptional regulators, DELLAs bind to transcription factors (TFs) and histone H2A via their GRAS domains and are recruited to act on promoters. Recent studies have shown that DELLA stability is regulated post-translationally by two mechanisms: polyubiquitination induced by the plant hormone gibberellin, which leads to their rapid degradation, and conjugation with small ubiquitin-like modifier (SUMO), which increases their accumulation. Moreover, DELLA activity is dynamically regulated by two distinct glycosylation mechanisms: O-fucosylation enhances DELLA-TF interaction, whereas O-linked N-acetylglucosamine (O-GlcNAc) modification inhibits DELLA-TF interaction. However, the role of DELLA phosphorylation is unclear as previous studies have shown conflicting results, with some suggesting that phosphorylation promotes or represses DELLA degradation and others suggesting that phosphorylation does not affect their stability. Here, we identify phosphorylation sites in the GA1-3 repressor (RGA), AtDELLA, purified from Arabidopsis thaliana by mass spectrometry and show that phosphorylation of two RGA peptides in the PolyS and PolyS/T regions enhances RGA activity by promoting H2A binding and RGA association with target promoters. Notably, phosphorylation did not affect RGA-TF interactions or RGA stability. Our study reveals a molecular mechanism by which phosphorylation induces DELLA activity.
Our mass spectrometric analysis revealed that both Pep1 and Pep2 were highly phosphorylated in RGA in the GA-deficient Ga1 background. In addition to this study, phosphoproteomic studies have also revealed Pep1 phosphorylation in RGA, although its role has not yet been studied53,54,55. In contrast, Pep2 phosphorylation has not been previously described since this peptide could only be detected using the RGAGKG transgene. Although the m1A mutation, which abolished Pep1 phosphorylation, only slightly reduced RGA activity in planta, it had an additive effect when combined with m2A in reducing RGA activity (Supplementary Figure 6). Importantly, Pep1 phosphorylation was significantly reduced in the GA-enhanced sly1 mutant compared to ga1, suggesting that GA promotes RGA dephosphorylation, reducing its activity. The mechanism by which GA suppresses RGA phosphorylation requires further investigation. One possibility is that this is achieved through the regulation of an unidentified protein kinase. Although studies have shown that expression of the CK1 protein kinase EL1 is downregulated by GA in rice41, our results indicate that higher-order mutations of the Arabidopsis EL1 homologue (AEL1-4) do not reduce RGA phosphorylation. In agreement with our results, a recent phosphoproteomic study using Arabidopsis AEL overexpressing lines and an ael triple mutant did not identify any DELLA proteins as substrates of these kinases56. When we prepared the manuscript, it was reported that GSK3, the gene encoding a GSK3/SHAGGY-like kinase in wheat (Triticum aestivum), can phosphorylate DELLA (Rht-B1b)57, although phosphorylation of Rht-B1b by GSK3 has not been confirmed in planta. In vitro enzymatic reactions in the presence of GSK3 followed by mass spectrometry analysis revealed three phosphorylation sites located between the DELLA and GRAS domains of Rht-B1b (Supplementary Fig. 3). Serine to alanine substitutions at all three phosphorylation sites resulted in decreased Rht-B1b activity in transgenic wheat, consistent with our findings that alanine substitutions in Pep2 RGA decreased RGA activity. However, in vitro protein degradation assays further demonstrated that phosphorylation can also stabilize Rht-B1b57. This is in contrast to our results showing that alanine substitutions in Pep2 RGA do not alter its stability in planta. GSK3 in wheat is an ortholog of brassinosteroid-insensitive protein 2 (BIN2) in Arabidopsis 57. BIN2 is a negative regulator of BR signaling, and BR activates its signaling pathway by causing BIN2 degradation 58. We showed that BR treatment does not reduce RGA stability 59 or phosphorylation levels in Arabidopsis (Supplementary Figure 2), suggesting that RGA is unlikely to be phosphorylated by BIN2.
All quantitative data were statistically analyzed using Excel, and significant differences were determined using Student’s t-test. No statistical methods were used to pre-determine sample size. No data were excluded from the analysis; the experiment was not randomized; and the researchers were aware of the allocation during the experiment and outcome assessment. Sample sizes are provided in the figure legends and in the raw data files.
Post time: Apr-15-2025