The PIF1-miR408-PLANTACYANIN repression cascade regulates light-dependent seed germination

Abstract

Light-dependent seed germination is a vital process for many seed plants. A decisive event in light-induced germination is degradation of the central repressor PHYTOCHROME INTERACTING FACTOR 1 (PIF1). The balance between gibberellic acid (GA) and abscisic acid (ABA) helps to control germination. However, the cellular mechanisms linking PIF1 turnover to hormonal balancing remain elusive. Here, employing far-red light-induced Arabidopsis thaliana seed germination as the experimental system, we identified PLANTACYANIN (PCY) as an inhibitor of germination. It is a blue copper protein associated with the vacuole that is both highly expressed in mature seeds and rapidly silenced during germination. Molecular analyses showed that PIF1 binds to the miR408 promoter and represses miR408 accumulation. This in turn posttranscriptionally modulates PCY abundance, forming the PIF1-miR408-PCY repression cascade for translating PIF1 turnover to PCY turnover during early germination. Genetic analysis, RNA-sequencing, and hormone quantification revealed that PCY is necessary and sufficient to maintain the PIF1-mediated seed transcriptome and the low-GA-high-ABA state. Furthermore, we found that PCY domain organization and regulation by miR408 are conserved features in seed plants. These results revealed a cellular mechanism whereby PIF1-relayed external light signals are converted through PCY turnover to internal hormonal profiles for controlling seed germination.

Figure 1.

PCY is induced during seed development and rapidly silenced during germination. (A) Global expression profile of 31 phytocyanin genes in the seed based on data in the Arabidopsis eFP Browser. Expression levels are indicated by the color scale at the bottom. (B) Comparison of PCY expression patterns in the seed and other organs. Data are mean ± SD of three individual experiments obtained from the eFP Browser. (C) Diagram illustrating the phyA_OFF and phyA_ON regimes. Imbibed seeds were sequentially treated with the indicated light conditions. Arrows indicate when time points in various assays were analyzed. (D) Relative PCY transcript levels during the time course of phyA_OFF and phyA_ON described in (C), as determined by RT-qPCR analysis. Data are mean ± SD from three independent qPCR reactions performed on the same pool of reverse-transcribed cDNA. Different letters at the same time point denote significant differences (one-way ANOVA, P < 0.001, Supplemental Data Set 3).

Figure 2.

PCY is vacuole associated and removed upon far-red irradiation. (A) Detection of PCY–GFP in PCY_pro:PCY-GFP plants using an anti-GFP antibody. Size markers are indicated on the right. RPT5 was used as the loading control. (B) Subcellular localization of PCY. PCY–GFP and COPT5-mCherry were transiently expressed in the same onion epidermal cells and examined by confocal fluorescence microscopy. Bar, 10 μm. (C) Confirmation of vacuole association of PCY by immuno-gold labeling. Cotyledon cells of imbibed PCY_pro:PCY-GFP seed in phyA_OFF were analyzed by immuno-electron microscopy using an antibody for GFP. Arrows indicate the size-defined electron-dense immunogold particles labeling PCY-GFP. Bar, 0.2 μm. (D–E) Co-localization of GFP fluorescence with vacuole autofluorescence in cotyledon cells of imbibed PCY_pro:PCY-GFP seed in phyA_OFF (D) and phyA_ON (E). Bar, 10 μm.

Figure 3.

PCY negatively regulates germination. (A) Representative plates showing the germination states of wild type and pcy seeds in phyA_OFF (top) and phyA_ON (bottom). (B–C) Quantification of the germination rate over the time course of phyA_OFF (B) and phyA_ON (C). Data are mean ± SD from three individual experiments. Different letters represent genotypes with significant differences at 120 h (one-way ANOVA, P < 0.01, Supplemental Data Set 3). (D) RT-qPCR analysis of relative PCY transcript level in the indicated genotypes without and with the application of β-estradiol. Data are mean ± SD from three replicates performed on the same cDNA. Different letters denote genotypes with significant differences (one-way ANOVA, P < 0.05, Supplemental Data Set 3). (E) Representative plates showing the germination states of the wild type and iPCY-OX seeds in phyA_OFF and phyA_ON under the indicated treatments. (F) Quantification of germination rates of the wild type and iPCY-OX seeds. Data are mean ± SD from three independent experiments. Different letters denote genotypes with significant differences (one-way ANOVA, P < 0.05, Supplemental Data Set 3).

Figure 4.

PCY negatively regulates seedling greening. (A) Representative wild type and pcy seedlings. Seedlings were grown in the dark for 96 h and then exposed to continuous white light for 24 h. Bar, 1 mm. (B) Quantified greening rate as described in Methods. Data are mean ± SD from three individual experiments. Different letters denote significant differences (P < 0.05 by Student’s t- test, Supplemental Data Set 3). (C–D) Comparison of pigment profiles in pcy versus wild-type seedlings. Protochlorophyllide and chlorophylls were assayed by spectral analysis in 4-day-old etiolated seedlings and etiolated seedlings exposed to white light for 24 h, respectively. Shown on the left are results from a representative experiment. Shown on the right is quantification of the protochlorophyllide fluorescence at 634 nm and chlorophyll fluorescence at 678 nm. Data are mean ± SD from three individual experiments. Different letters denote significant differences (Student’s t-test, P < 0.05 for C, P < 0.01 for D). (E) RT-qPCR analysis of the relative transcript levels of pigment biosynthetic genes in 4-day-old etiolated seedlings. Data are mean ± SD from three replicates performed on the same cDNA. For each gene, different letters denote significant differences (P < 0.05 by Student’s t-test, Supplemental Data Set 3).

Figure 5.

miR408 represses PCY expression during germination. (A) Confirmation of miR408 targeting to PCY in the seed by RNA-ligation-based amplification of cDNA ends. The structure of PCY (top) and base pairing between miR408 and PCY (bottom) is shown. Arrows mark the detected transcript ends along with frequency of the corresponding clones. (B) Degradome sequencing data supporting miR408-guided cleavage of the PCY transcript. The frequencies of the sequenced 5′ ends are plotted against nucleotide position in the PCY transcript. The red dot indicates the position of the miR408 recognition site. (C) RT-qPCR analysis of relative miR408 levels over the time course of phyA_OFF and phyA_ON. Data are means ± SD from three replicates performed on the same cDNA. (D) Relative PCY transcript levels in seeds of the indicated genotypes under phyA_OFF and phyA_ON. Data are means ± SD from three replicates performed on the same cDNA. Different letters denote genotypes with significant differences (one-way ANOVA, P < 0.01, Supplemental Data Set 3).

Figure 6.

miR408 promotes seed germination and seedling greening. (A) Representative plates showing the germination states of the indicated genotypes under phyA_OFF (left) and phyA_ON (right). (B–C) Quantification of germination rate of the indicated genotypes over the time course of phyA_OFF (B) and phyA_ON (C). Data are mean ± SD from three individual experiments. Different letters denote genotypes with significant differences at 120 h (one-way ANOVA, P < 0.05 for B, P < 0.001 for C, Supplemental Data Set 3). (D) Representative etiolated seedlings of the indicated genotypes that were exposed to white light for 24 h. Bar, 1 mm. (E) Quantified greening rates of miR408-OX and amiR408 in comparison to the wild type. Data are mean ± SD from three independent experiments. Different letters represent significant differences (one-way ANOVA, P < 0.05, Supplemental Data Set 3). (F–I) Protochlorophyllide and chlorophylls were assayed in 4-day-old etiolated seedlings and etiolated seedlings exposed to white light for 24 h, respectively. Results from a representative spectral analysis are shown in F and H. The protochlorophyllide fluorescence at 634 nm (G) and chlorophyll fluorescence at 678 nm (I) were quantified. Data are mean ± SD from three individual experiments. Different letters represent significant differences (one-way ANOVA, P < 0.001, Supplemental Data Set 3).

Figure 7.

PIF1 suppresses miR408 expression by binding to the miR408 promoter. (A) PIF1 occupancy profile at the miR408 locus. Significantly enriched PIF1 ChIP-sequencing reads were obtained from Pfeiffer et al. (2014) and mapped onto the Arabidopsis genome coordinates. Loci are represented by grey and black arrows. The blue circle marks the G box (CACGTG) in the miR408 promoter (horizontal line). (B) ChIP-qPCR confirming PIF1 binding to the miR408 promoter. An anti-MYC antibody was used to precipitate chromatin from PIF1-OX and wild-type seeds. Enrichment of PIF1 binding was determined by qPCR analysis. Data are mean ± SD from three qPCR performed on the same DNA. Different letters denote significant differences (P < 0.05 by Student’s t-test, Supplemental Data Set 3). (C) Transient dual luciferase assay showing PIF1 repression of miR408. The miR408_pro:LUC reporter concatenated to 35S_pro:REN was used to transform tobacco protoplasts with either the empty vector (−PIF1) or a PIF1-expressing construct (+ PIF1). (D) Quantification of the LUC/REN luminescence ratio. Data are mean ± SD from three independent transfections. Different letters denote significant differences (P < 0.05 by Student’s t-test, Supplemental Data Set 3). (E) Histochemical staining for GUS activity from miR408_pro:GUS expressed in the wild type or pif1 seeds in phyA_OFF and phyA_ON. Bar, 0.5 mm. (F) Quantification of GUS concentration by ELISA. Data are mean ± SD from three individual experiments. Statistical analysis was performed for data from phyA_OFF and phyA_ON separately and different letters denote significant differences (P < 0.05 by Student’s t-test, Supplemental Data Set 3). DW, dry weight. (G–H) RT-qPCR analysis of relative miR408 (G) and PCY (H) transcript levels. Data are mean ± SD from three replicates performed on the same cDNA. Statistical analysis was performed for data from phyA_OFF and phyA_ON separately and different letters denote genotypes with significant differences (one-way ANOVA, P < 0.05, Supplemental Data Set 3).

Figure 8.

Genetic analysis of the PIF1-miR408-PCY pathway. (A) The amiR408 line was crossed with pif1 to generate the pif1 amiR408 double mutant. Seeds from these lines and the wild type were assayed for germination rates over the time course of phyA_OFF (left) and phyA_ON (right). (B) The miR408-OX line was crossed with PIF1-OX to generate the PIF1-OX miR408-OX double over-expression line. Seeds from these lines and wild type plants were assayed for germination rates in phyA_OFF and phyA_ON. (C) Comparison of germination rates of the PIF1-OX, pcy, and PIF1-OX pcy seeds. All data are all means ± SD from three individual experiments. Different letters denote genotypes with significant differences at 120 h (one-way ANOVA, P < 0.05, Supplemental Data Set 3).

Figure 9.

Transcriptomic analysis of the PIF1-miR408-PCY pathway. (A) Venn diagram showing the relationships of PIF1-, miR408-, and PCY-regulated genes. Differentially expressed genes were identified from RNA-sequencing analyses of pif1, miR408-OX, and iPCY-OX seeds versus the respective controls. (B) Scatter plots showing pairwise correlation of the relative expression levels of the three sets of coregulated genes in pif1, miR408-OX, and iPCY-OX versus the respective controls. R, Pearson correlation coefficient. (C) Hierarchical clustering of the 2,290 genes that were differentially expressed in pif1, miR408-OX, and iPCY-OX vs. the respective controls. Colors represent the Log₂-transformed fold changes. (D) Clustering analysis of the 218 genes associated with the GO term “seed germination” (GO:0009845). The genes were divided into three groups based on their relative expression levels in pif1 versus in the wild type. Group I, repressed in pif1; Group II, not differentially expressed; Group III, induced in pif1. (E) Expression pattern of representative Group I (SOM and RVE2) and Group III (JMJ22 and MAN7) genes in the indicated RNA-sequencing data sets.

Figure 10.

The PIF1-miR408-PCY pathway differentially regulates GA and ABA metabolic genes. (A) Diagram of a simplified GA biosynthesis pathway illustrating genes influenced by the PIF1-miR408-PCY pathway. Genes associated with the individual biosynthesis steps are shown above the arrows. From left to right, colored boxes indicate relative expression levels of the corresponding gene in pif1, miR408-OX, and iPCY-OX against the respective controls. (B) RT-qPCR analysis of relative transcript levels of GA3ox1 and GA2ox2 in the indicated seeds under phyA_OFF. Data are mean ± SD from three replicates performed on the same cDNA. Different letters denote genotypes with significant differences (one-way ANOVA, P < 0.05 for GA3ox1, P < 0.01 for GA2ox2, Supplemental Data Set 3). (C) Diagram of a simplified ABA biosynthesis pathway illustrating genes influenced by the PIF1-miR408-PCY pathway. Colored boxes indicate relative expression levels of the corresponding gene in pif1, miR408-OX, and iPCY-OX against the respective controls. (D) RT-qPCR analysis of the relative transcript levels of ABA metabolic genes ABA1, NCED6, and NCED9 in the indicated seeds. Data are mean ± SD from three replicates performed on the same cDNA. Different letters denote genotypes with significant differences (one-way ANOVA, p < 0.05, Supplemental Data Set 3).

Figure 11.

PIF1-miR408-PCY controls germination by modulating the GA/ABA ratio. (A–B) Quantification of endogenous GA₄ (A) and ABA (B) levels in imbibed seeds of the indicated genotypes. Data are mean ± SD from three individual experiments. Different letters denote genotypes with significant differences (one-way ANOVA, p < 0.001 for A, P < 0.05 for B, Supplemental Data Set 3). FW, fresh weight. (C–D) Calculated GA/ABA ratios in the indicated genotypes. Data are mean ± SD from three individual experiments. Different letters denote genotypes with significant differences (one-way ANOVA, P < 0.001 for C, P < 0.01 for D, Supplemental Data Set 3). (E) Germination rates of the indicated seeds in phyA_OFF and phyA_ON with different chemical treatments. Mock, no chemical treatment; GA₃, 10 μM GA₃; Paclobutrazol, 100 μM paclobutrazol; ABA, 5 μM ABA. Data are mean ± SD from three individual experiments.

Figure 12.

PCY exhibits features conserved in seed plants. Comparison of PCY and related blue copper proteins in representative land plants. Shown on the left is a species tree showing the reconstructed order of divergence of the indicated plant species using information obtained from the TimeTree database at timetree.org. The branch length reflects evolutionary divergence time in millions of years. Species with identified PCYs are shaded in green. PCY domains are shown with different colors on the right. Scale represents accumulative number of amino acid residues. The alignment of the PCY related proteins is provided as a text file in Supplemental File 1.

Figure 13.

Working model of far-red light-dependent seed germination mediated by PCY. The PIF1-miR408 module is critical for regulating PCY abundance in far-red light-induced seed germination. In darkness, due to the absence of active phyA, PIF1 is stabilized and binds to the G box of the miR408 promoter. This leads to transcriptional repression of miR408 and allows PCY to accumulate. Upon irradiation with far-red light, phyA-mediated PIF1 degradation leads to transcriptional derepression of miR408, which in turn silences PCY. Removal of PCY from storage vacuoles facilitates an increase in the GA/ABA ratio, and this ultimately sets germination in motion.