The SnRK2-APC/CTE regulatory module mediates the antagonistic action of gibberellic acid and abscisic acid pathways

Abscisic acid (ABA) and gibberellic acid (GA) antagonistically regulate many developmental processes and responses to biotic or abiotic stresses in higher plants. However, the molecular mechanism underlying this antagonism is still poorly understood. Here, we show that loss-of-function mutation in rice Tiller Enhancer (TE), an activator of the APC/CTE complex, causes hypersensitivity and hyposensitivity to ABA and GA, respectively. We find that TE physically interacts with ABA receptor OsPYL/RCARs and promotes their degradation by the proteasome. Genetic analysis also shows OsPYL/RCARs act downstream of TE in mediating ABA responses. Conversely, ABA inhibits APC/CTE activity by phosphorylating TE through activating the SNF1-related protein kinases (SnRK2s), which may interrupt the interaction between TE and OsPYL/RCARs and subsequently stabilize OsPYL/RCARs. In contrast, GA can reduce the level of SnRK2s and may promote APC/CTE-mediated degradation of OsPYL/RCARs. Thus, we propose that the SnRK2-APC/CTE regulatory module represents a regulatory hub underlying the antagonistic action of GA and ABA in plants.


Supplementary Figure 2
The te mutant exhibits reduced sensitivity to GA. (a) 1 µM GA 3 more effectively induces α-amylase secretion by OE17 and wild type (WT) than by te half seeds, as shown by the degradation of starches in the medium in these α-amylase assays. The half seeds were plated on medium for three days at 30 °C. (b-d) Treatment with 100 µM GA 3 for 12 hours effectively induces the expression of α-amylase gene RAmy1A in WT and OE17 plants but not in te plants. Values are means ± s.d. (n=3 replicates). Student's t-test analysis indicates a significant difference (compared with control, **P<0.01).

Supplementary Figure 3
Alignment of PYR/PYL/RCAR proteins. Alignment of 10 rice PYR/PYL/RCAR proteins and 3 Arabidopsis PYR/PYL/RCAR proteins shows the conserved three-dimensional structural motifs. In addition, the conserved D-box mainly located in Lid loop one (composed of β3 and β4) is shown.

Supplementary Figure 8
The quantification analysis of relative R10/HSP82 protein levels The quantification analysis of relative R10/HSP82 protein levels corresponding to Figure 4a (a) and Figure  4b (b). Note: in (a), the R10 protein level in OE17 seeds after 4h cycloheximide treatment (regardless of GA application) is notably less than that in OE17 seeds before treatment. Values are means ± s.d. (n=3 replicates). Asterisks mark significant differences according to Student's t test (**P < 0.01). N.S. =no significant.

Supplementary Figure 9
Interruption of the GA signaling pathway stabilizes rice SnRK2 proteins. (a) Western blot analysis shows the levels of SAPK10 protein 1, 3 or 6 hours after addition of 100 μM GA 3 or Paclobutrazol (a GA biosynthesis inhibitor). WT: wild type. (b-d) Western blot analysis shows the levels of SAPK10-GFP, SAPK8-GFP and SAPK9-GFP proteins 1, 3 or 6 hours after addition of 100 μM GA 3 or Paclobutrazol. The 'α-HSP82' signal shows that roughly equal amounts of total plant extracts were used.

Supplementary Figure 10
Effects of ABA or GA on rice SnRK2 mRNA expression. (a-c) qRT-PCR results show that ABA can promote the expression of SAPK10 (a) and SAPK8 (b) but not SAPK9 (c) in WT plants; but GA 3 does not substantially alter the expression of all three rice SnRK2 genes (a-c) in WT plants. Values are means ± s.d. (n=3 replicates). Asterisks mark significant differences according to Student's t test (*P < 0.05, **P < 0.01).