EARLY BUD-BREAK 1 and EARLY BUD-BREAK 3 control resumption of poplar growth after winter dormancy

Abstract

Bud-break is an economically and environmentally important process in trees and shrubs from boreal and temperate latitudes, but its molecular mechanisms are poorly understood. Here, we show that two previously reported transcription factors, EARLY BUD BREAK 1 (EBB1) and SHORT VEGETATIVE PHASE-Like (SVL) directly interact to control bud-break. EBB1 is a positive regulator of bud-break, whereas SVL is a negative regulator of bud-break. EBB1 directly and negatively regulates SVL expression. We further report the identification and characterization of the EBB3 gene. EBB3 is a temperature-responsive, epigenetically-regulated, positive regulator of bud-break that provides a direct link to activation of the cell cycle during bud-break. EBB3 is an AP2/ERF transcription factor that positively and directly regulates CYCLIND3.1 gene. Our results reveal the architecture of a putative regulatory module that links temperature-mediated control of bud-break with activation of cell cycle.

Fig. 1. Isolation and molecular analysis of the early bud-break 3D (ebb3D) poplar mutant.

a ebb3D poplar mutant plants show early bud-break in the field during the start of the second growing season as compared to neighboring WT-717 control trees, arrows point to two ebb3D ramets that show early bud-break compared to the WT-717 plants and other neighboring activation tagging events. b Early bud-break in ebb3D mutant plant as compared to WT-717 under controlled growth chamber conditions (see Methods for more details). c Genome position of the activation tag insertion in the ebb3D mutant, 4X enhancers derived from the CaMV35S promoter. d The average number of days to bud-break in WT-717 and ebb3D mutant plants. e, f Potri.012G108400 and Potri.012G108500 genes are upregulated in the ebb3D mutant. Expression values are the average of three biological replicates ±SEM, normalized to the reference ACT7 gene. At least 10 plants of the ebb3D mutant and WT-717 genotypes were used in the bud-break analysis. Asterisks (*) and (**) indicate significant differences at P < 0.02 and P < 0.002 compared to WT-717 control plants by two-tailed paired t-tests. Source data underlying Fig. 1a–d are provided as a Source Data file.

Fig. 2. EBB3 over- and under-expressing lines show early and late bud-break phenotypes respectively.

a Early bud-break of a representative over-expressing line compared to WT-717 control plant. b EBB3 relative expression in EBB3-OE lines. c Time to bud-break in WT-717 plants and EBB3-OE lines. d Delayed bud-break in representative EBB3 under-expressing line compared to WT-T89 control plant. e EBB3 relative expression in EBB3-RNAi lines. f Time to bud-break relative to WT-T89 control plants in EBB3-RNAi lines. Expression values are average of three biological replicates ±SEM, normalized to the reference ACT7 for WT-717 and UBQ gene for WT-T89 clones, respectively. Asterisks (*) indicate significant and (**) indicate extremely significant differences compared to WT at P < 0.05 and P < 0.0001, respectively, and determined by two-tailed paired t-tests. Source data underlying Fig. 2a, c, and d are provided as a Source Data file.

Fig. 3. EBB3 is primarily expressed in the shoot apex and differentially regulated during the dormancy cycle and bud-break.

a Relative expression of EBB3 in different tissues of WT-717 plants. b Relative expression of EBB3 in vegetative buds of wild-growing aspen (Populus tremuloides) trees. c Relative expression of EBB3 in WT-T89 plants under manipulative dormancy treatments in controlled growth chamber conditions. Expression values are the average of three biological replicates ±SEM, normalized to the reference genes ACT7 for WT-717 clones, and UBQ for WT-T89 clones, respectively. 0 W: time point before inductive treatments, 10 WSD: 10 weeks under SD photoperiod, 2WC: 2 weeks under cold, 5WC: 5 weeks under cold, 2WLD: 2 weeks under LD photoperiod and warm temperature. d Average H3K27me3 abundance of three biological replicates in the EBB3 locus, including 2 Kb downstream and upstream regions, before dormancy inductive treatments (0 W) and during dormancy induction, (6WSD, 10WSD) and dormancy release (4WC). H3K27me3 abundance was normalized relative to H3 abundance. Shaded areas indicate ±standard deviation. TSS transcription start site, TTS transcription termination site, cds coding sequence, UTR untranslated region. Source data underlying Fig. 3c are provided as a Source Data file.

Fig. 4. EBB1 directly binds to SVL promoter in vivo in chromatin immunoprecipitation (ChIP) assays to regulate its expression.

a EBB1 and b SVL relative expression during dormancy cycle and bud-break under manipulative dormancy treatments in controlled growth chamber conditions; 0 W: time point before inductive treatments, 10WSD: 10 weeks under SD photoperiod, 5WC: 5 weeks under cold, 2WLD: 2 weeks under LD photoperiod and warm temperature. c Relative expression of SVL in EBB1 over-expressing (EBB1-OE) and underexpressed (EBB1-ami) plants. Expression values are the average of three biological replicates ±SEM, the expression values normalized to the reference ACT7 gene. d Enrichment of a DNA fragment in the SVL promoter containing a GCC-box and quantified by ChIP-qPCR. The green box is a schematic representation of the SVL promoter showing the position of the GCC-box (red). Red arrows delineate the position of DNA fragments containing the GCC-box and green arrows demarcate the position of DNA fragments with no GCC-box used as a negative control (NC) in ChIP-qPCR analysis. Chromatin from EBB1-GFP DNA transfected poplar protoplasts was isolated using anti-GFP antibody and IgG used as a control antibody. ChIP-purified DNA was used to perform ChIP-qPCR, expression values are represented as the percentage of input (% of input) DNA. Values are the average of three biological replicates ±SEM. Asterisks (*) and (**) indicate extreme significant differences compared to their respective control at P < 0.003 and P < 0.0005 respectively and determined using two-tailed paired multiple t-tests. Source data underlying Fig. 4d are provided as a Source Data file.

Fig. 5. EBB1, SVL, and ABA regulate the expression of EBB3.

a Relative expression of EBB3 in EBB1 over-expressing and under-expressing plants. Relative expression of EBB3 in SVL over-expressing (b) and under-expressing (c) plants. d Relative expression of EBB3 in ABA-treated apices of WT-717 plants. e EBB3 expression in apices of transgenic hybrid aspen with reduced ABA response (abi1 mutant); 0W: time point before inductive treatments, 10WSD: 10 weeks under SD photoperiod, 5WC: 5 weeks under cold, 2WLD: 2 weeks under LD photoperiod and warm temperature. Expression values are the average of three biological replicates ±SEM, normalized to the reference genes - UBQ for WT-T89 and ACT7 for WT-717 clones respectively. Asterisks (*) indicate significant and (**) indicate extremely significant differences compared to WT at P < 0.04 and P < 0.001 respectively and determined using two-tailed paired multiple t-tests.

Fig. 6. EBB3 binds directly to the CYCD3.1 promoter in vivo in chromatin immunoprecipitation assays to regulate its expression.

a Number of common and unique downregulated genes in EBB3 RNAi under-expressing plants in different dormancy stages. The Cyclin D3.1 (CYCD3.1) is the only downregulated gene in all dormancy stages. See Supplementary Data 4–7 for more details. b Relative expression of CYCD3.1 in EBB3-OE and WT-717 plants; 0W: time point before inductive treatments, 10WSD: 10 weeks under SD photoperiod, 5WC: 5 weeks under cold, 2WLD: 2 weeks under LD photoperiod and warm temperature, expression values are normalized to the ACT7 gene. Expression values are the average of three biological replicates ±SEM. c Enrichment of a DNA fragment in the CYCD3.1 promoter containing a GCC-box and quantified by ChIP-qPCR. The green box is a schematic representation of the CYCD3.1 promoter showing the position of the GCC-box (red). Red arrows delineate the position of DNA fragments containing the GCC-box and green arrows demarcate the position of DNA fragments with no GCC-box used as a negative control (NC) in ChIP-qPCR analysis. Chromatin from EBB3-GFP DNA transfected poplar protoplasts was isolated using anti-GFP antibody and IgG used as a control antibody. ChIP-purified DNA was used to perform ChIP-qPCR, expression values are represented as percentage input (% input) DNA. Values are the average of three biological replicates ±SEM. Asterisks (*) and (**) indicate significant differences compared to their respective control at P < 0.05 and P < 0.004 respectively and determined using two-tailed paired multiple t-tests. Source data underlying Fig. 6a and c are provided as a Source Data file.

Fig. 7. Hypothetical model of the roles EBB1, SVL and EBB3 play in control of bud-break in poplar.

EBB1 is positively regulated by low temperature, leading to suppression of SVL expression. Declining SVL expression breaks the SVL/ABA feedforward loop. SVL/ABA repression leads to the upregulation of EBB3 and consequently to activation of CYCD3.1, cell proliferation, and bud-break. Green arrows indicate positive regulation while red bars indicate negative regulation. Dash arrows indicate indirect regulation. See the text for additional description.