IPA1 functions as a downstream transcription factor repressed by D53 in strigolactone signaling in rice

Abstract

Strigolactones (SLs), a group of carotenoid derived terpenoid lactones, are root-to-shoot phytohormones suppressing shoot branching by inhibiting the outgrowth of axillary buds. DWARF 53 (D53), the key repressor of the SL signaling pathway, is speculated to regulate the downstream transcriptional network of the SL response. However, no downstream transcription factor targeted by D53 has yet been reported. Here we report that Ideal Plant Architecture 1 (IPA1), a key regulator of the plant architecture in rice, functions as a direct downstream component of D53 in regulating tiller number and SL-induced gene expression. We showed that D53 interacts with IPA1 in vivo and in vitro and suppresses the transcriptional activation activity of IPA1. We further showed that IPA1 could directly bind to the D53 promoter and plays a critical role in the feedback regulation of SL-induced D53 expression. These findings reveal that IPA1 is likely one of the long-speculated transcription factors that act with D53 to mediate the SL-regulated tiller development in rice.

Figure 1

Phenotypic characterization of IPA1 loss-of-function mutants. (A) Gross morphologies of wild-type (WT), ipa1-10, ipa1-11, ipa1-3D, and ipa1-4D plants before the heading stage. Bar = 10 cm. (B) Statistical analysis of tiller number in (A). Values are means ± sem (n = 11). Different letters at top of each column indicate a significant difference at P < 0.05 determined by Tukey's HSD test. (C) IPA1 expression levels in WT, ipa1-10, and ipa1-3D. Values are means ± SD (n = 3). The asterisks represent significant differences determined by Student's t test. ^*P < 0.05, ^**P < 0.001. (D) Gross morphologies of wild-type (WT), ipa1-10, ipa1-3D, d27, and d53 seedlings with or without rac-GR24 treatment. Seedlings were treated with 1.0 μM rac-GR24 (+) or mock (–). Bar = 5 cm. (E) Statistical analysis of tiller number in (D). Values are means ± SD (n = 5). The asterisks represent significant differences determined by Student's t test. ^**P < 0.01, ^**P < 0.001. ns, no significant difference.

Figure 2

Interaction of IPA1 with D53 in vitro and in vivo. (A) Interaction between IPA1 and D53 revealed by yeast two-hybrid. IPA1 was fused with the GAL4 binding domain (BD) and D53 with the GAL4 activation domain (AD). The yeast clones were grown on the SD medium without leucine, tryptophan, histone and adenine (SD-L-W-H-A) with dilutions to 10⁻¹ and 10⁻². Yeast grown on the SD medium without leucine and tryptophan (SD-L-W) were used as a loading control. (B) Interaction between IPA1 and D53 revealed by the BiFC assay in rice protoplasts. IPA1 was fused with the N-terminal of CFP and D53 with the C-terminal of CFP. (C) Interaction between IPA1 and D53 revealed by GST pull-down with GST-IPA1 purified from bacteria and D53 extracted from rice calli. D53 was detected by rabbit polyclonal antibodies anti-D53 and GST by mouse monoclonal antibody anti-GST. (D) In vivo interaction between 7mIPA1-GFP and D53 revealed by the Co-IP assay in rice protoplasts. Proteins were extracted from ProIPA1:7mIPA1-GFP or ProUB:GFP. D53 was detected by rabbit polyclonal antibodies anti-D53 and GFP by mouse monoclonal antibody anti-GFP. (E) Interaction between IPA1 and D53 revealed by the GST pull-down assay with HisTrx-D53 and GST-IPA1 purified from bacteria. D53 was detected by rabbit polyclonal antibodies anti-D53 and GST by Ponceau S staining.

Figure 3

D53 represses the transcriptional activation activity of IPA1. (A) Transcriptional activity assay in tobacco leaves, showing that IPA1 promotes the expression of the reporter gene Luciferase (LUC) driven by the GTAC-containing promoter. A. tumefaciens transformed with Pro35S:IPA1-MYC, Pro35S:GFP, Pro35S:GUS, and ProGTAC:LUC were mixed and injected into tobacco leaves. D-luciferin was used as the substrate of LUC. (B) Statistical analysis of (A). Values are means ± SD (n = 3). The asterisks represent significant difference determined by Student's t test. ^**P < 0.01. (C) Transcriptional activity assay in tobacco, showing that D53 represses the transcriptional activation activity of IPA1. A. tumefaciens transformed with Pro35S:IPA1-MYC, Pro35S:GFP, ProGTAC:LUC, Pro35S:GUS, and Pro35S:FLAG-D53 were mixed and injected into tobacco leaves. D-luciferin was used as the substrate of LUC. (D) Statistical analysis of (C). Values are means ± SD (n = 3). Different letters at top of each column indicate a significant difference at P < 0.05 determined by Tukey's HSD test. (E) Protein levels in different infiltration combinations in (C). D53 was detected by rabbit polyclonal antibodies anti-D53, GFP by mouse monoclonal antibody anti-GFP, and IPA1 was detected by rabbit polyclonal antibodies anti-IPA1. Ponceau S staining was used as loading control. See also Supplementary information, Figure S5.

Figure 4

Interaction of IPA1 and D53 inhibits IPA1 transcriptional activation activity. (A) Interaction between the N-terminal, the SBP domain or the C-terminal of IPA1 and D53 revealed by the BiFC assay in rice protoplasts. The N-terminal, SBP domain or C-terminal of IPA1 were fused with cCFP and D53 was fused with nCFP. (B) The C-terminal of IPA1 was sufficient for the transcriptional activation activity of IPA1. Values are means ± sem (n = 3). The double asterisks represent significant difference determined by Student's t test at P < 0.01. (C) Transcriptional activity assay in rice protoplasts, showing that the N-terminal and the SBP domain are necessary for the D53-mediated repression of IPA1 transcriptional activation activity. Values are means ± SD (n = 3). The asterisk represents significant differences determined by Student's t test at P < 0.05. ns, no significant difference. GUS was used as a control. The transcriptional activation activity of full length IPA1 or IPA1-ΔN, which could bind to D53, was repressed by added D53 compared with the control. However, the activity of IPA1-ΔN-ΔSBP, which could not bind to D53, was not influenced by added D53 compared with control. (D) EMSA assay, showing that D53 does not affect the DNA binding activity of IPA1. The shift bands indicated the binding of GST-IPA1 to the probe containing GTAC element, and the super shift bands indicated the binding of GST-IPA1 together with HisTrx-D53, which was enhanced by adding HisTrx-D53, but not by MBP.

Figure 5

IPA1 binds to the D53 promoter and regulates D53 expression. (A) IPA1 binding profile in the promoter of D53. The solid arrowhead refers to the GTAC around the peak summit, and red vertical line to peak summit. Primer pairs D53-ProF and D53-ProR (Supplementary information, Table S1) were used for ChIP-qPCR. The probe was used in EMSA. TSS, transcription start site. (B) Validation of IPA1 direct binding sites in the D53 promoter by ChIP-qPCR analysis. Values are means ± sem (n = 3). The double asterisk represents significant difference determined by the Student's t test at P < 0.01; ns, no significant difference. (C) Direct binding of IPA1 to the D53 promoter in the EMSA assay. The 20- and 50-fold excess non-labeled probes were used for competition. The D53 pro-m is a mutated version of D53 pro probe with the SBP binding motif GTAC changing to ATAC. (D) Transcriptional activity assay in tobacco leaf, showing that IPA1 could enhance the expression of the D53 promoter-drived LUC reporter. A. tumefaciens transformed with Pro35S:IPA1-MYC, Pro35S:GFP, and ProD53:LUC were mixed and injected into tobacco leaves. D-luciferin was used as the substrate of Luciferase. (E) Statistical analysis of (D). Values are means ± SD (n = 3). The asterisk represents significant difference determined by Student's t test. ^**P < 0.01.

Figure 6

D53 and IPA1 form a feedback regulation loop in SL signaling. (A) Transcriptional activity assay in tobacco, showing that D53 represses the IPA1-drived activation of the D53 promoter. A. tumefaciens transformed with Pro35S:IPA1-MYC, Pro35S:GFP, ProD53:LUC, and Pro35S:FLAG-D53 were mixed and injected into tobacco leaves. D-luciferin was used as the substrate of LUC. (B) Statistical analysis of (A). Values are means ± SD (n = 3). Different letters at top of each column indicate a significant difference at P < 0.05 determined by Tukey's HSD test. (C) D53 transcript levels in WT, d53 and ipa1 mutants. Values are means ± sem (n = 3). Different letters at top of each column indicate a significant difference at P < 0.05 determined by Tukey's HSD test. (D) D53 protein levels in WT, d53 and ipa1 mutants. D53 was detected by rabbit polyclonal antibodies anti-D53. Actin was used as the loading control. (E) Mutations in IPA1 disrupt SL-induced D53 transcription after rac-GR24 treatment. Values are means ± sem (n = 3). Statistical differences between mock and treatment at same time points were determined by Student's t test. ^*P < 0.05, ^**P < 0.01; ns, no significant difference.

Figure 7

Overexpression of miRNA156 compromises SL response. (A) Gross morphologies of wild-type and miRNA156 overexpression (miR156OE) plants with or without rac-GR24 treatment. Seedlings were treated with 1 μM rac-GR24 (+) or mock (–). Bar = 5 cm. (B) Statistical analysis of tiller number in (A). Values are means ± sem (n = 5). The asterisks represent significant difference determined by Student's t test. ^***P < 0.001; ns, no significant difference. (C) Overexpression of miRNA156 disrupts SL-induced D53 transcription after rac-GR24 treatment. Values are means ± sem (n = 3). Statistical differences between mock and treatment at the same time points were determined by Student's t test. ^**P < 0.01; ns, no significant difference. (D) Gross morphologies of d53, ipa1-1D, and d53 ipa1-1D double mutant plants. Bar = 20 cm. (E) Statistical analysis of (D). Values are means ± sem (n = 8). Different letters at top of each column indicate a significant difference at P < 0.05 determined by Tukey's HSD test. (F) A proposed model of the IPA1-mediated SL signaling pathway. In the absence of SLs, the D53 protein binds to IPA1, and together with TPL/TPR proteins represses the transcriptional activity of IPA1. In the presence of SLs, perception of SL leads to degradation of D53 by the proteasome system, which in turn releases the repression of IPA1-regulated gene expression and leads to SL response.