Light Dependent Degradation of PIF3 by SCF^EBF1/2 Promotes the Photomorphogenic Response in Arabidopsis

SUMMARY

Plant seedlings emerging from darkness into the light environment undergo photomorphogenesis, which enables autotrophic growth with optimized morphology and physiology. During this transition, plants must rapidly remove photomorphogenic repressors accumulated in the dark. Among them is PHYTOCHROME-INTERACTING FACTOR 3 (PIF3), a key transcription factor promoting hypocotyl growth. Here we report that, in response to light activation of phytochrome photoreceptors, EIN3-BINDING F BOX PROTEINs (EBFs), EBF1 and EBF2, mediate PIF3 protein degradation in a manner dependent on light-induced phosphorylation of PIF3. While PIF3 binds EBFs independently of light, the recruitment of PIF3-EBFs to the core SCF scaffold is facilitated by light signals or PIF3 phosphorylation. We also found that previously identified LIGHT-RESPONSE BRIC-A-BRACK/TRAMTRACK/BROAD (LRB) E3 ubiquitin ligases target phytochrome B (phyB) and PIF3 primarily under high light conditions, whereas EBF1/2 vigorously target PIF3 degradation under wide ranges of light intensity without affecting the abundance of phyB. Both genetic and molecular data support that SCF^EBF1/2 function as the photomorphogenic E3s during seedling development.

INTRODUCTION

Sunlight provides the ultimate energy source for nearly all living organisms on earth, largely through photosynthesis by green plants. To utilize sunlight efficiently, plants rely on their photoreceptors to sense light quality, quantity, direction, and period, and modulate their growth pattern correspondingly [1]. Phytochromes are the major photoreceptors that germinating plants use to determine whether to undergo photomorphogenesis (in light environments) or skotomorphogenesis (in dark environments). Phytochromes can sense red (R) or far-red (FR) light, and reversibly switch between the active (Pfr) and inactive (Pr) forms [2]. The general light activation scheme in plants starts with the light-induced conformational change of phytochromes into the active (Pfr) forms, which subsequently translocate into the nucleus and interact with, among others, a group of basic helix-loop-helix (bHLH) photomorphogenic repressors called PHYTOCHROME-INTERACTING FACTORs (PIFs) [3–5].

Among these PIFs, PIF3 is a critically important factor that regulates hypocotyl elongation, cotyledon expansion of seedlings, and chloroplast development. It works as a transcriptional factor by binding to the promoters of the target genes [3, 6–8]. In the dark, PIF3 accumulates to high levels to maintain the state of skotomorphogenesis in plants, and its stability is determined by CONSTITUTIVE PHOTOMORPHOGENIC 1 (COP1) and DE-ETIOLATED 1 (DET1), two pleiotropic suppressors of photomorphogenesis [9, 10]. While the mechanism whereby DET1 stabilizes PIF3 is still unclear, COP1 stabilizes PIF3 by inhibiting BRASSINOSTEROID-INSENSITIVE 2 (BIN2)-mediated PIF3 phosphorylation and degradation in the dark [11]. During the dark to light transition, PIF3 directly interacts with photo-activated phyA and phyB, and is rapidly phosphorylated on multiple residues, which are necessary for its subsequent degradation through the ubiquitin-proteasome pathway [12, 13]. Recent studies showed that phytochrome and PHOTOREGULATORY PROTEIN KINASEs (PPKs) may directly phosphorylate PIF3 [14, 15]. The sharp decline in PIF3 concentration permits robust expression of light-induced genes such as photosynthetic genes, which is a necessary step for plants to establish photomorphogenesis [12–16].

A key player in this light activation scheme is the E3 ubiquitin ligase that mediates removal of PIF3, the main light repressor [3, 6]. LIGHT-RESPONSE BRIC-A-BRACK/TRAMTRACK/BROAD (LRB) E3 ligases have been shown to mediate PIF3 ubiquitination, but LRBs simultaneously target both phyB and PIF3 for ubiquitination and concurrent degradation [17]. Based on the genetic evidence that lrb double and triple mutants are hypersensitive to red light due to over-accumulation of phyB, LRBs have been postulated to function as negative regulators of photomorphogenesis, possibly through PIF3-dependent phyB protein degradation [17–19].

In searching for the PIF3 E3 ligases that serve as positive regulators of photomorphogenesis, we report in this study that EBF1 and EBF2 F-box proteins interact with PIF3 and mediate light-induced PIF3 degradation through the ubiquitin-proteasome pathway. EBF1/2 were originally identified as inhibitors of the ethylene pathway by targeting the transcription factor EIN3 for degradation [20, 21]. This event is critical during de-etiolation or greening when etiolated seedlings grow out of the dark soil and are exposed to light [22–24]. Here we show that, upon light activation, EBF1/2 target PIF3 for ubiquitination without affecting phyB stability. Moreover, SCF^EBF1/2 mediated PIF3 ubiquitination is modulated, not at the level of substrate recognition by the F-box receptors of the SCF, but by a novel mechanism that involves substrate phosphorylation-dependent assembly of SCF^EBF1/2. Our study found that SCF^EBF1/2 function as the photomorphogenic E3 ligases targeting PIF3 for degradation, thus the mechanism reported here fulfills a long-standing gap in the plant light activation scheme.

RESULTS

EBF1 and EBF2 Potentiate the Light Response to Inhibit Hypocotyl Growth by Restricting the Activity of PIF3

As light inhibits the hypocotyl elongation rate of plants, hypocotyl length has been used as a physiological indicator of light responses. Seedlings grown under light (red light or white light) contain low levels of PIF3, while forced overexpression of PIF3 promoted hypocotyl elongation and decreased the output of the light signaling pathway (Figure 1A and Figure S1A). Importantly, the activity of PIF3 was notably suppressed by overexpression of EBF1 or EBF2, as indicated by the reduced hypocotyl lengths of PIF3-Myc/EBF-TAP seedlings in the light (Figure 1A and Figure S1A). Likewise, in loss-of-function mutants, light-grown ebf1–3 ebf2-2 seedlings exhibited longer hypocotyls, while the phenotype was completely suppressed by the pif3 mutation (Figure 1B and Figure S1B). In the dark, altering the levels of PIF3 alone did not affect hypocotyl elongation, as previously reported (6; Figure S2). Therefore, the hypocotyl assay to study the genetic relationship of PIF3 and EBF1/2 was effective only under light conditions. These genetic data imply that PIF3 plays key roles downstream of EBF1/2 to inhibit light stimulated hypocotyl response, and that EBF1/2 promote photomorphogenesis at least in part by counteracting PIF3 activity. Given that EBF1/2 are F-box proteins, which are substrate receptors for SCF ubiquitin E3 ligases, we hypothesized that EBF1/2 may target PIF3 for degradation in response to phytochrome activation.

Figure 1. EBF1/2 inhibit hypocotyl elongation by reducing PIF3 activity in red light.

(A) Overexpression of EBF1 or EBF2 suppressed PIF3-Myc induced hypocotyl elongation in red light. Seedlings grown under 10 µmol·m⁻²·s⁻¹ red light for around 4 days were measured for hypocotyl length. Representative seedlings are shown on the left and the mean hypocotyl lengths (mean±SEM) are shown as bar graphs on the right. Statistical significance was calculated by Student’s t test. n.s.: p>0.05; *: p<0.05; ***: p<0.001.

(B) The ebf1 ebf2 double mutant exhibited longer hypocotyls in red light compared to Col, and the phenotype could be suppressed by the pif3 mutation. Seedling growth and data analyses were performed as in (A).

See also Figure S1 & S2.

PIF3 Interacts with EBF1 and EBF2

We first tested whether EBF1 and EBF2 could interact with PIF3 in yeast-2-hybrid assays (Figure 2A). Normally, EBF1 and EBF2 bind the substrates via their leucine-rich repeat (LRR) domains and assemble into the SCF complexes via their F-box domains [20]. Our data showed that PIF3 specifically interacted with the LRR domains, and not the F box domains of EBF1 and EBF2 (Figure 2A), while EBF1 and EBF2 interacted with at least two regions of PIF3 in yeast (Figure S3). Pull-down assays using in vitro translated proteins showed that EBF1 and EBF2 could directly interact with PIF3 (Figure 2B). In vivo co-immunoprecipitation (co-IP) assays showed that EBF1 and EBF2 proteins fused with green fluorescence protein (GFP) could co-immunoprecipitate PIF3-Myc proteins in etiolated seedlings (Figure 2C). In addition, PIF3 could interact with EBF1 and EBF2 in a luciferase complementation imaging (LCI) assay (Figure 2D). These data showed that PIF3 could interact with EBF1 and EBF2, which would be consistent with PIF3 being a substrate of EBF1/2.

Figure 2. PIF3 interacts with EBF1 and EBF2.

(A) PIF3 interacts with both full length and LRR domains of EBF1 and EBF2 in yeast. Full length or partial EBF1 and EBF2 were fused with the binding domain (BD) as baits. Full length PIF3 was fused with the activation domain (AD) as prey. Empty vectors were used as negative controls.

(B) PIF3 interacts with EBF1 and EBF2 in vitro. The proteins of EBF1-flag, EBF2-flag and PIF3-myc were synthesized using in vitro translation system. flag antibody was used for in vitro pull down and pellets were analyzed by western blots using antibodies to flag and Myc.

(C) EBF1 and EBF2 co-immunoprecipitate PIF3 in Arabidopsis. Total proteins were extracted from 3-day-old dark-grown PIF3-Myc or EBF-GFP/PIF3-Myc seedlings. GFP antibody was used for immunoprecipitation and pellets were analyzed by western blots using antibodies to GFP and Myc.

(D) PIF3 interacts with EBF1 and EBF2 in LCI assays. PIF3 was fused to nLUC, and EBF1/2 without F box domain (EBF1/2 ΔF) were fused to cLUC, which were co-expressed in tobacco. Empty vectors were used as negative controls.

See also Figure S3.

EBF1 and EBF2 Promote PIF3 Degradation via the Ubiquitin-Proteasome System after Light Exposure

To determine whether EBF1 and EBF2 have roles in light-induced degradation of PIF3, we monitored PIF3 protein levels following exposure to a red light pulse in the dark-grown mutants or lines overexpressing EBFs. Since ebf1 ebf2 double mutants showed severely reduced fertility, fertile triple mutants ein3 ebf1 ebf2 were used to study the effects of EBF mutations, with ein3 as the corresponding control. Steady-state levels of PIF3 transcripts and proteins in dark-grown seedlings of all the above lines are shown in Figure S4. After light exposure, PIF3 protein levels dropped immediately, and deficiency of EBFs decreased the degradation rate of both endogenous PIF3 (Figure 3A) and PIF3-GFP driven by a 35S promoter (Figure 3B), the latter of which rules out transcriptional regulation of PIF3. In addition, overexpression of EBF (EBF1-TAP and EBF2-TAP) increased the degradation rates of both endogenous PIF3 and PIF3-Myc driven by the 35S–promoter after light exposure (Figure 3C and 3D). Further, the proteasomal inhibitor MG132 efficiently inhibited degradation of both endogenous and ectopically expressed PIF3 (Figure S5). These results showed that EBF1/2 are rate-limiting factors in light-induced degradation of PIF3 via the ubiquitin-proteasomal pathway.

Figure 3. EBF1/2 promote PIF3 degradation after light exposure.

(A and B) Mutation of EBF1 and EBF2 slowed down light induced degradation rate of endogenous PIF3 protein (A) and ectopic PIF3-GFP protein (B). Dark-grown seedlings (D) were treated with a red light pulse (Rp) and kept in the dark for the indicated times before protein extraction and western blotting with antibodies to PIF3 and RPN6. (A), Upper panel shows the representative western blot result; Lower panel shows the quantification results of PIF3 degradation kinetics from 3 independent biological repeats. PIF3 protein level was normalized to RPN6 loading control, and the dark PIF3 level was set as 100%. Data were shown in mean±SEM. (B), S.exp: short exposure time; L.exp: long exposure time.

(C and D) Overexpression of EBF1-TAP and EBF2-TAP accelerated light-induced degradation of endogenous PIF3 (C) or ectopic PIF3-Myc protein (D). Dark-grown seedlings (D) were treated with either pulse (Rp) or continuous (Rc) red light. Western blots were performed with antibodies to PIF3, RPN6, Myc, or RPT5.

See also Figure S4.

We next examined how polyubiquitination of PIF3 is affected by EBF1/2 using tandem ubiquitin binding entities (TUBE2). The total amounts of polyubiquitinated proteins pulled down by TUBE2 were similar in dark-grown and red-light treated plants, and were un-affected by expression of EBFs-TAP (Figure 4A). In contrast, PIF3-Myc specific polyubiquitination, indicated by the anti-Myc blots (Figure 4B and 4C), was strongly promoted by EBF1-TAP or EBF2-TAP specifically after light exposure. Consistently, deficiency of EBF1 and EBF2 dramatically reduced the light-induced increase of endogenous PIF3 polyubiquitination (Figure 4D). These results showed that EBF1/2 specifically promoted light-induced polyubiquitination of PIF3.

Figure 4. EBF1/2 promote light-induced PIF3 ubiquitination.

Total polyubiquitinated proteins were precipitated using tandem ubiquitin binding entities 2 (TUBE2) from 3-day dark-grown seedlings without/with R light pulse (D/Rp) treatment, and then were analyzed by western blots using antibodies to Ub (A), Myc (B and C) or PIF3 (D).

See also Figure S5.

EBF1/2-promoted PIF3 Degradation is Dependent on Light Induced PIF3 Phosphorylation

To investigate the light control of EBF-mediated degradation of PIF3, we first determined the role of phytochromes. We crossed EBF1-TAP and EBF2-TAP transgenes into phyA-211 phyB-9 double mutants and found that overexpression of either EBF1 or EBF2 could promote light-induced PIF3 degradation only in wild type but not in phyA-211 phyB-9 mutants (Figure 5A and 5B), confirming that phyA and phyB are necessary for EBF-mediated light-induced PIF3 degradation. Moreover, the transgenic PIF3 mutant, PIF3 mAPAmAPB (containing E31A, G37A, F203A and F209A) that was unable to interact with phytochromes [13], was more stable than wild type PIF3 after exposure to light, and was resistant to EBF1- or EBF2-TAP stimulated degradation (Figure 5C and 5D). Thus, not only are phytochromes necessary for light-induced EBF mediated PIF3 degradation, but the direct interaction of phytochrome with PIF3 is also required.

Figure 5. EBF1- and EBF2-mediated PIF3 degradation is dependent on phytochrome-induced PIF3 phosphorylation.

(A and B) EBF1/2-promoted PIF3 degradation upon light exposure is dependent on phyA and phyB. Dark-grown seedlings (D) were treated with a red light pulse (Rp) and kept in the dark for the indicated times before total proteins were extracted and analyzed by western blots using antibodies to PIF3 or RPN6.

(C and D) The interaction between PIF3 and phytochromes is necessary for EBF1/2-promoted PIF3 degradation. The GFP fused PIF3 transgenic lines PIF3 WT (wild type PIF3) and PIF3 mAPAmAPB (PIF3 with mutations in APA and APB domains) in wild type (Col) or EBF1/2-TAP backgrounds were used. Dark-grown seedlings (D) were treated with continuous red light (Rc) for the indicated times before western blotting using antibodies to GFP or RPN6.

(E and F) EBF1/2-promoted PIF3 degradation is dependent on light-induced PIF3 phosphorylation. The Myc fused PIF3 transgenic lines PIF3 WT (wild type PIF3) and PIF3 A20 (PIF3 protein with 20 mutations of light-induced phosphorylation sites) in wild type (Col) or EBF1/2-TAP backgrounds were used. Protein preparation was as described in (C) and western blots were performed with antibodies to Myc or RPN6.

Interaction of PIF3 with active phytochrome leads to phosphorylation of PIF3, which has been shown to be necessary for its subsequent degradation upon exposure to light [13]. To address the significance of PIF3 phosphorylation in conjunction with EBFs, a phospho-dead PIF3 (PIF3 A20), in which 20 light-induced phosphorylation sites were mutated to alanine [16], was expressed in EBF1-TAP and EBF2-TAP plants. Clearly, the mutation of PIF3 phosphorylation sites blocked light-induced as well as EBF1- or EBF2-TAP promoted degradation of PIF3 (Figure 5E and 5F). These results show that phosphorylation of PIF3 is critical for its degradation through EBF1/2.

Light signals promote recruitment of PIF3 to the core SCF scaffold through EBF1/2

We next asked how light or light-induced phosphorylation of PIF3 triggers its degradation via EBFs. F-box proteins typically function as substrate receptors that recruit specific substrates to the SCF type of Cullin-RING ubiquitin E3 ligase complexes for polyubiquitination [25, 26]. Usually it is the binding of the substrate to its substrate receptor that defines the specificity and signal-dependence of substrate degradation, especially for those substrates that contain phosphodegrons [27]. Accordingly, we examined the possibility that light may enhance the binding of EBFs to PIF3 as a result of PIF3 phosphorylation, analogous to LRB-mediated ubiquitination of PIF3 and phyB [17]. However, the result from the co-immunoprecipitation experiment showed that EBFs bound to PIF3 both in the dark and after light exposure with similar affinity (Figure 6A). In contrast to LRBs, which exhibited higher binding affinity to phosphorylated PIF3, EBF1/2 interacted with both unphosphorylated and phosphorylated PIF3 with similar binding affinity in yeast 2-hybrid (Figure 6B) as well as pull down assays (Figure S6). These data showed that EBF1 and EBF2 bind to PIF3 regardless of light signals and independently of the phosphorylation status. This signal-independent binding cannot explain why EBFs only stimulate ubiquitination and degradation of PIF3 after light exposure (Figure 3 and 4).

Figure 6. PIF3 phosphorylation promotes the assembly of PIF3-EBF1/2 with CUL1.

(A) EBF interacts with PIF3 under both dark and light conditions in Arabidopsis. Dark-grown (D) seedlings (3-day-old) were treated with a red light pulse (Rp) and then kept in darkness for 10 min and then total proteins were extracted. IgG sepharose was used for immunoprecipitation and pellets were analyzed by western blot using a Myc antibody.

(B) EBF1/2 interact with wild type and phospho-site mutants of PIF3 in yeast-2-hybrid assays. Full length EBF1, EBF2 and LRB2 were fused to DNA-binding domains (BD) as baits. Phospho-dead (A6 and A20), wild type and phospho-mimic (D6 and D19) PIF3s were fused to activation domains (AD) as preys. Empty vectors were used as negative controls.

(C) Light induces the association of CUL1 and PIF3 in Arabidopsis. Experiment conditions were the same as in (A). GFP antibody was used for immunoprecipitation (IP) and the pellets were analyzed by western blots using antibodies to GFP or CUL1. CUL1 was co-precipitated with YFP-PIF3 after light (Rp) but not in the dark (D).

(D) EBF1 and EBF2 are essential for light-induced association of CUL1 and PIF3 in Arabidopsis. Experiment conditions were the same as in (A). GFP antibody was used for IP and the pellets were analyzed by western blots using antibodies to CUL1 and PIF3.

(E) PIF3 phosphorylation can enhance its association with CUL1 without light activation. Dark-grown seedlings (3-day-old) were used for IP with Myc antibody. The pellets were analyzed by western blots using antibodies to Myc or CUL1. Endogenous CUL1 was pulled down by proteins of PIF3 phospho-mimic (PIF3-D6) but not PIF3 phospho-dead (PIF3-A20) mutant.

(F) Phospho-mimic PIF3 promotes EBF1 and CUL1 association. Experimental procedures were same as in (E). IgG sepharose was used for IP of EBF1-TAP, and the pellets were analyzed by western blots using antibodies to Myc or CUL1. CUL1 was strongly pulled down with EBF1-TAP in the presence of PIF3 phospho-mimic transgenic products, compared to those in PIF3 wild type, PIF3 phospho-dead, or no PIF3 transgene. The relative amount of CUL1 protein to EBF1-TAP protein in pellets were shown in mean±SD from 3 independent experiments, in which the ratio of EBF1-TAP transgenic plants was set as 1.0.

(G) ASK1 is necessary for the association of EBF1 and CUL1 promoted by phospho-mimic PIF3 in vitro. EBF1-flag, CUL1-His, PIF3 A20-Myc and PIF3 D6-Myc proteins were expressed using an in vitro translation system. GST and GST-ASK1 proteins were expressed and purified from E. coli. Anti-flag antibody was used to immunoprecipitate EBF1-flag and then the pellets were analyzed by western blot using antibodies against flag, His, Myc and GST.

See also Figure S6.

Given that EBF1/2 bind their PIF3 substrate constitutively, we next hypothesized that the signal-dependent step may take place at the recruitment of PIF3-EBF1/2 to the core SCF scaffold. We tested co-immunoprecipitation of PIF3 with CULLIN1 (CUL1), the scaffold component of the SCF complex, with plant extracts. As shown in Figure 6C, the association between YFP-PIF3 and CUL1 was only detected after light treatment. Importantly, light-enriched association between PIF3-GFP and CUL1 could be found in ein3 eil1 mutants, but not in ein3 eil1 ebf1 ebf2 mutants, which indicated that the light-enriched association between PIF3 substrate and CUL1 scaffold was dependent on EBF1/2 (Figure 6D). Thus the data unequivocally showed that light stimulates association of PIF3 and CUL1 through the F-box substrate receptors EBF1/2. Together, these data show that, while EBF1/2 bind PIF3 regardless of light signals, the assembly of PIF3-EBF1/2 with the core SCF scaffold to form active SCF^EBF1/2 E3 complexes is stimulated by light signals in plants.

PIF3 Phosphorylation enhances the assembly of PIF3-EBF1/2 to the core SCF scaffold in the absence of activated phytochromes

To delineate the specific role of phytochromes in SCF-mediated PIF3 ubiquitination, we asked whether PIF3 phosphorylation is sufficient to induce its association with CUL1 in the absence of light irradiation. We utilized phospho-mimic PIF3-myc (PIF3 D6) and phospho-dead PIF3-myc (PIF3 A6, or PIF3 A20) mutants that carry point-mutations in light-induced phosphorylation sites on PIF3 to mimic or block light-induced phosphorylation, respectively [17]. These phosphorylation mutants of PIF3 exhibited similar affinity to EBF1 in the in vitro pull-down experiment (Figure S6B). However as shown in Figure 6E, phospho-mimic PIF3 (PIF3 D6) exhibited higher association affinity with CUL1 than phospho-dead PIF3 (PIF3 A6) in etiolated seedlings, indicating that phosphorylation of PIF3 is sufficient to facilitate its recruitment to the SCF complex for ubiquitination. To address whether substrate phosphorylation status may affect association of F-box substrate receptor with CUL1, we examined by co-immunoprecipitation the association of EBFs with CUL1 in the presence of different PIF3 phosphorylation mutants. Remarkably, higher amounts of EBF1 were associated with CUL1 in the presence of phospho-mimic PIF3 D6 substrate compared to phospho-dead A20 PIF3 or wild type PIF3 in the dark (Figure 6F). Furthermore, the adaptor protein ASK1 was proven to be necessary for the association of EBF1 and CUL1 stimulated by phospho-mimic PIF3 D6 in an in vitro pull down assay (Figure 6G). These results indicate that substrate phosphorylation can stimulate recruitment of the F-box protein modular complex to CUL1. According to these data, once PIF3 is phosphorylated in response to phytochrome activation, it is able to propel its own assembly, through EBF1/2 and ASK1, into the SCF E3 complex for ubiquitination, in the absence of phytochromes.

EBF and LRB E3 Ligases Have Distinct Functions

LRB E3 ligases have previously been shown to mediate degradation of both PIF3 and phyB proteins [17]. With the discovery that EBF1/2 also act as PIF3 E3 ligases, we wanted to define functional distinctions of the two groups of E3 ligases to understand why plants require both types to target PIF3 degradation. While LRB E3 ligases were shown to target PIF3 as well as phyB for degradation [17], EBF1/2 E3 ligases had no effect on phyB stability (Figure 7A). Upon light-induced restraint of hypocotyl growth, lrb1 lrb2 lrb3 triple mutants exhibited significantly shorter hypocotyls than wild type under continuous red light, consistent with previous reports [17]; while ebf mutants had longer hypocotyls in the light (Figure 7B and Figure S7). These genetic phenotypes imply that lrb1 lrb2 lrb3 mutants are photo-hypersensitive, while ebf mutants are photo-hyposensitive.

Figure 7. EBF and LRB E3 ligases have distinct functions.

(A) LRBs regulate light-induced degradation of phyB whereas EBFs do not. 3-day-old dark-grown (D) seedlings were transferred to 60 µmol/m²/s red light for 72 h (Rc) and then total proteins were analyzed by western blots using antibodies to phyB or RPN6. Light-induced decline in phyB was detected only in lrb mutants but not in the other lines tested.

(B) lrb mutants show light hypersensitivity while ebf mutants show hyposensitivity with regard to the hypocotyl response. Hypocotyl lengths of 3-day-old seedlings growing under dark (Dc) or 5 µmol/m²/s red light (Rc) were measured (mean±SEM). Statistical significance was calculated by Student’s t test. n.s., p>0.05; ***, p<0.001. The lrb mutants displayed shorter hypocotyls while ebf mutants displayed longer hypocotyls under Rc.

(C) ebf mutants showed defects in PIF3 degradation under both strong and weak light conditions, while lrb mutants showed defects in PIF3 degradation predominantly under strong light condition. 3-day-old dark-grown seedlings were treated with either 3000 µmol/m² or 100 µmol/m² and then kept in darkness for 15 min before total proteins were extracted and analyzed by western blot using antibodies to PIF3 and RPN6.

(D) EBFs promote PIF3 degradation under both strong and weak light conditions while LRB’s effect on PIF3 is detected predominantly in strong light. 3-day-old dark-grown seedlings were treated with either 3000 µmol/m² (upper) or 100 µmol/m² (lower) red light pulses and kept in the dark for the indicated times before western blotting using antibodies to PIF3 or RPN6.

(E) A model illustrating the distinct functions of EBF1/2 and LRBs in plant light signaling pathway through their control of PIF3. In the dark, PIF3 accumulates to promote skotomorphogenesis. PIF3 binds EBF1/2 but the complex is not recruited to SCF core, and PIF3 does not bind LRBs. When plants sense weak light at the beginning of light exposure, PIF3 undergoes multisite phosphorylation (represented by the circles with “-” inside). The phospho-PIF3-EBF complex and core SCF scaffold assembles to form the SCF^EBF1/2-PIF3 complexes, resulting in the ubiquitination and degradation of PIF3. This PIF3 degradation pathway promotes photomorphogenesis. When plants perceive strong light, another PIF3 degradation pathway is triggered in addition to SCF^EBF1/2-PIF3, in which phyB-PIF3 complex is targeted by CUL3^LRBs E3 ligases. The LRB pathway results in the degradation of both phyB and PIF3 to attenuate plant light responses.

See also Figure S7.

Since the genetic functions of the LRBs are to attenuate light signaling, we speculated that LRB E3 ligases might manifest their roles predominantly under excessive light conditions. To test this, PIF3 degradation rates were examined in plants growing under different light intensities. We found that PIF3 degradation in lrb123 mutants was weakened predominantly under strong red light irradiation (3000 µmol/m²) while only marginally under weak red light irradiation (100 µmol/m²) (Figure 7C). In contrast, PIF3 degradation in ein3ebf1ebf2 mutants was severely impaired compared to ein3 under both light conditions (Figure 7C). In agreement, we also found that YFP-LRB2 could facilitate PIF3 degradation more evidently under strong red light irradiation (3000 µmol/m²), whereas EBF1-TAP stimulated PIF3 degradation in response to both strong (3000 µmol/m²) and weak red light (100 µmol/m²) (Figure 7D). Together we demonstrate that EBF E3 ligases function as photomorphogenic E3s to promote light signaling in response to a wide range of light intensities, while LRB E3 ligases function primarily as a photoreceptor E3 to down-regulate light signaling in response to excessive light (Figure 7E).

DISCUSSION

Light Induced Degradation of PIF3 Marks an Important Event in the Transition from Skotomorphogenesis to Photomorphogenesis

In adaptation to terrestrial life, seed plants have developed the skotomorphogenic developmental program, or etiolation, to cope with subterranean darkness during the early stage of their life [28]. Skotomorphogenesis is an active biological process that is essential for plants to survive the transition from the dark to light environments, and it is maintained by multiple hormonal signaling pathways and cellular factors [29–32]. Among them, the PIF3 transcription factor plays a critical role in skotomorphogenesis [6, 13], and its mutant is severely defective in surviving de-etiolation [7, 22]. PIF3 has been shown to directly bind 1064 sites in the Arabidopsis genome, and many of these target genes are involved in cellular processes that sustain skotomorphogenic development and repressing light responses [8, 33, 34]. In addition to its transcription activity, PIF3 has also been shown to promote the degradation of phyB to reduce the sensitivity of plant cells to red light [17, 18, 35].

As a photomorphogenic repressor, PIF3 accumulates to high levels in dark-grown seedlings [9]. When seedlings emerge from darkness to sunlight, it is important for plants to rapidly down-regulate PIF3 in order to switch to photomorphogenic growth [12, 36]. Naturally, since light-triggered PIF3 degradation is a critical step in light-mediated plant development, the identity of the PIF3 E3 ligase has been intensively sought after. The LRBs have been identified as PIF3 E3 ligases, but LRBs simultaneously target the phyB photoreceptor in the same process to attenuate light responses [17]. Indeed, we and others have shown that LRBs are negative regulators of light signaling, rather than the photomorphogenic E3 ligases in the light activation scheme (Figure 7; 17, 19). In contrast, EBF1/2 genetically promote light responses in hypocotyl regulation (Figures 1 and 7), and are rate-limiting factors in light-induced degradation of PIF3 (Figure 3). Both genetic and molecular data demonstrate that EBF1/2 are the PIF3 ubiquitin E3 ligases whose physiological function is to promote photomorphogenesis.

Light-dependent Targeting of PIF3 by SCF^EBF1/2 Revealed a Novel Mechanism of SCF

The SCF complexes are ubiquitin E3 ligases forming a highly-conserved family in eukaryotes. In SCF E3s, recognition of a substrate by its F-box substrate receptor usually determines the fate of the substrate, including phosphorylation-dependent substrate binding by the F-box protein [25–27]. Indeed, in light-induced degradation of EIN3 via SCF^EBF1/2, activated phyB is shown to act as a cofactor to enhance and bridge the binding of EIN3 to the EBF1/2 F-box receptor [24]. However, light-induced degradation of PIF3 by the same E3 is regulated by a different mechanism. In SCF^EBF1/2 mediated targeting of PIF3, the commitment of substrate ubiquitination occurs, not upon the binding of PIF3 to EBF1/2, but upon the recruitment of the phospho-PIF3-EBF1/2 complex to the core SCF scaffold (Figures 5 and 6). To our knowledge, this is the first time that the activation step in SCF mediated ubiquitination is reported to take place at the assembly of the substrate-F-Box complex to the CULLIN core, rather than upon binding of a substrate by the substrate receptor. Our findings reveal a novel activation mechanism by which SCF-type ubiquitin E3 complexes is regulated according to substrate status.

SCF^EBF1/2 Ubiquitin E3 Ligases converge light and ethylene signals to regulate early seedling development

EBF1/2 were originally identified as negative regulators of the ethylene pathway, in which EIN3/EIL1 are responsible for the activation of ethylene signaling, and EBF1/2 are the E3s targeting EIN3 for degradation [19, 21, 37]. Among the diverse processes it regulates in plants, ethylene plays a critical role in reinforcing skotomorphogenic development of dark-grown seedlings in response to the mechanical pressure of the soil cover [22]. One of the target genes of the EIN3 ethylene transcriptional factor is PIF3 [38]. Consequently, soil-induced stimulation of ethylene signaling in dark-grown seedlings would result in increased expression of PIF3, which in part accounts for the ethylene-enhanced skotomorphogenic program [22, 38]. When seedlings grow out of the dark soil and are exposed to light, light decreases ethylene signaling by stimulating EIN3 degradation through SCF^EBF1/2, which is necessary for adequate de-etiolation and greening of seedlings [24]. In line with its function in light-induced degradation of EIN3, our finding that under the same physiological circumstances, EBF1/2 directly target PIF3, a key canonical inhibitor of photomorphogenesis, solidifies the role of EBF1/2, or SCF^EBF1/2, as the bona fide photomorphogenic ubiquitin E3 ligases. The role of SCF^EBF1/2 therefore fulfills a long-standing gap in the plant light activation cascade (Figure 7E).

Since EIN3 can activate PIF3 gene expression by directly binding to its promoter [38], EBF1/2 is able to reduce the abundance of PIF3 by directly targeting the protein for ubiquitination, as well as by decreasing PIF3 expression via mediating EIN3 degradation after light exposure. EIN3 can be targeted and degraded by EBF without light, while light further promotes EBF-mediated EIN3 degradation [20, 21, 23, 24]. In contrast, PIF3 cannot be degraded by EBF in the dark, and light is essential to trigger EBF mediated PIF3 degradation (Figure 3–6, Figure S4). Based on the expression patterns of EIN3 and PIF3 [8, 38, 39, 40], as well as their closely related functions in etiolated seedlings [22], EIN3 and PIF3 are likely co-expressed in response to ethylene at least in some cells where they function coordinately to maintain skotomorphogenic development. Emerging from under the soil, seedlings are exposed to both light and the open air. On cues of simultaneous phytochrome activation and the plunge in ethylene concentration, EBF1/2 are utilized to rapidly and simultaneously remove both the PIF3 light repressor and the EIN3 ethylene activator, allowing the plant to switch to the photomorphogenic program. This mechanism highlights the extensive cross-talk and coordination between light and ethylene signaling.

STAR METHODS

CONTACT FOR REAGENT AND RESOURCE SHARING

Further information and requests for resources and reagents should be directed to and will be fulfilled by the Lead Contact, Haodong Chen (chenhaodong@pku.edu.cn).

EXPERIMENTAL MODEL AND SUBJECT DETAILS

The wild-type Arabidopsis ecotype used in this study was Columbia-0 (Col). All the mutants and transgenic plants were in the Col background. Adult plants were grown in soil under long day condition (16 h light/8 h night) at 22 °C.

METHOD DETAILS

Growth conditions of the seedlings

Seeds were first surface-sterilized with 15% NaClO for 5 min and then washed 4 times with sterilized ddH₂O before sowing on MS plates (4.4 g/L Murashige-Skoog powder, 8 g/L Agar, pH 5.7). After stratification for 3 days at 4 °C, seeds were treated with white light for 3 h followed by 5 min illumination with far-red light. For detection of endogenous PIF3 protein levels, 3-day-old dark-grown seedlings were treated with 60 µmol·m⁻²·s⁻¹ red light for 2 min (Rp) and then put back into darkness for the indicated times at 21 °C unless otherwise indicated. For detection of ectopic PIF3-Myc and PIF3-GFP protein levels, 3-day-old dark-grown seedlings were treated with continuous 10 µmol·m⁻²·s⁻¹ red light for the indicated times at 21 °C. For co-immunoprecipitation, seedlings were grown in darkness for 3 days at 21 °C before Rp treatment as indicated.

Yeast two-hybrid assays

The cDNAs corresponding to full length, F box domain, and LRR domain of EBF1 were amplified from the cDNAs of Col using primers pLexA-EBF1-F/pLexA-EBF1-R, pLexA-EBF1-F/BamHI-1Fbox-R and EcoRI-1LRR-F/pLexA-EBF1-R, and cloned into the EcoRI and BamHI restriction sites of the vector pLexA, to construct pLexA-EBF1, pLexA-EBF1 Fbox, and pLexA-EBF1 LRR respectively. The cDNAs corresponding to full length, F box domain, and LRR domain of EBF2 were amplified from the cDNAs of Col using primers pLexA-EBF2-F/pLexA-EBF2-R, pLexA-EBF2-F/BamHI-2Fbox-R, and EcoRI-2LRR-F/pLexA-EBF2-R, and cloned into the EcoRI and BamHI restriction sites of the vector pLexA, to construct pLexA-EBF2, pLexA-EBF2 Fbox, and pLexA-EBF2 LRR respectively. The cDNA of LRB2 was amplified from the cDNAs of Col using primers SalI-LRB2-F and SalI-LRB2-R and then cloned into the pLexA plasmid using SalI restriction site, to construct pLexA-LRB2.

The cDNAs of PIF3-A6, A20, WT, D6 and D19 were amplified from pT7CFE1 (CHis) PIF3-A6, A20, WT, D6 and D19 [17] using primers EcoRI-PIF3-F/XhoI-PIF3-R, and then cloned into the EcoRI and XhoI restriction sites of the vector pB42AD, to construct pB42AD-PIF3 A6, A20, WT, D6 and D19. The cDNAs of PIF3 NT338, bHLH and 181–524 were amplified from pB42AD-PIF3 WT using primers EcoRI-PIF3-F/XhoI-PIF3(338)-R, EcoRI-PIF3(bHLH)-F/XhoI-PIF3(bHLH)-R and EcoRI-PIF3 (181–524)-F/XhoI-PIF3-R, and then cloned into the EcoRI and XhoI restriction sites of the vector pB42AD, to construct pB42AD-PIF3 NT 338, bHLH and 181–524. The pB42AD-PIF3 NT120 and delAPA were from Dong et al., 2014 [10] and the pB42AD-PIF3 NT180, pB42AD-PIF3 181–338, pB42AD-PIF3 339–524 were from Ling et al., 2017 [11].

BD- and AD-fused plasmids were co-transformed into yeast strain EGY48 containing p8opLacZ according to the instructions provided with the Matchmaker LexA two-hybrid system (Clontech). Yeast transformants were then plated on minimal SD Agar Base (Clontech) plates supplemented with DO Supplement –His/– Trp/–Ura (Clontech) and further incubated for 4 days at 30 °C. Finally, well-grown colonies were plated onto minimal SD Agar Base/Gal/Raf (Clontech) plates supplemented with DO Supplement –His/–Trp/–Ura (Clontech) and 80 mg/L X-gal (Amresco) for β-gal activity assays.

Generation of plant materials

For generating 35S:EBF1-GFP/PIF3-Myc and 35S:EBF2-GFP/PIF3-Myc plants, the full length cDNA of EBF1 and EBF2 were amplified from pLexA-EBF1 or pLexA-EBF2 using primers XbaI-EBF1-F/StuI-EBF1-R and XbaI-EBF2-F/XhoI-EBF2-R, and then cloned into the XbaI/StuI and XbaI/XhoI sites of pJim19-GFP (bar) [40] respectively, to construct pJim19(bar)-EBF1-GFP and pJim19(bar)-EBF2-GFP. Then, pJim19(bar)-EBF1-GFP and pJim19(bar)-EBF2-GFP were transformed into 35S:PIF3-Myc [10] plants.

For generating 35S:PIF3(WT)-GFP/EBF-TAP and 35S:PIF3(mAPAmAPB)-GFP/EBF-TAP, the full length cDNAs of PIF3(WT), PIF3(mAPAmAPB) and GFP were amplified from pB42AD-PIF3 WT, eYFP: PIF3 mAPAmAPB [13] and pJim19-GFP (bar) [40] respectively, using primers SacI-PIF3-F/KpnI-PIF3-R or KpnI-GFP-F/SalI-GFP-R, and then cloned into pCAMBIA1300 using SacI/KpnI sites for PIF3 (WT or mAPAmAPB) and KpnI/SalI sites for GFP. Then, the obtained constructs pCAMBIA1300-PIF3(WT)-GFP and pCAMBIA1300-PIF3(mAPAmAPB)-GFP were transformed into 35S:EBF1/2-TAP [41] plants respectively, to generate 35S:PIF3(WT)-GFP/EBF1-TAP, 35S:PIF3(WT)-GFP/EBF2-TAP, 35S:PIF3(mAPAmAPB)-GFP/EBF1-TAP and 35S:PIF3(mAPAmAPB)-GFP/EBF2-TAP.

For generating 35S:PIF3–GFP/ein3eil and 35S:PIF3-GFP/ein3eil1ebf1ebf2, pCAMBIA1300-PIF3(WT)-GFP was transformed into ein3eil1 [41] or ein3eil1ebf1ebf2 [41] mutant plants, respectively. Agrobacterium strain GV3101 was used for all the Arabidopsis transformations.

For generating 35S:PIF3-Myc/EBF1-TAP and 35S:PIF3-Myc/EBF2-TAP, 35S: PIF3-Myc [10] was crossed with 35S: EBF1-TAP [41] and 35S: EBF2-TAP [41] respectively. For generating 35S:PIF3 A20-Myc/EBF1-TAP and 35S:PIF3 A20-Myc/EBF2-TAP, 35S: PIF3 A20-Myc [16] was crossed with 35S: EBF1-TAP [41] and 35S: EBF2-TAP [41] respectively. For generating 35S:PIF3 D6-Myc/EBF1-TAP, 35S: PIF3 D6-Myc [16] was crossed with 35S: EBF1-TAP [41]. For generating pif3-3ebf1–3ebf2-2, pif3-3 [7] was crossed with ebf1–3ebf2-2 [21] and pif3-3ebf1–3ebf2-2 homozygous plants were isolated from the progenies of pif3-3ebf1–3ebf2-2(+/−) plants. For generating 35S:EBF1-TAP/phyA-211phyB-9 and 35S:EBF2-TAP/phyA-211phyB-9, phyA-211phyB-9 [44] was crossed with 35S: EBF1-TAP [41] and 35S: EBF2-TAP [41] respectively.

Plant total protein extraction for western blot

Seedlings were ground into powder in liquid nitrogen and then total proteins were extracted into 100 µL denaturing buffer (8 M urea, 100 mM NaH₂PO₄, 100 mM Tris-HCl, pH 8.0) containing 1 mM PMSF and 1×protease inhibitor (Roche) and cleared by centrifugation (20000 g, 10 min) at 4 °C. Protein concentration was determined using bradford protein assay (Bio-Rad). 100 µL supernatant was then mixed with 25 µL 5×SDS loading buffer (250 mM Tris-HCl pH 6.8, 25 % glycerol, 10% SDS, 0.01% Bromo Phenol Blue, 10 mM DTT, 5% β-mercaptoethanol), boiled at 95 °C for 10 min and analyzed by western blot.

Co-immunoprecipitation (Co-IP) assays

3-day-old dark-grown seedlings were ground into powder in liquid nitrogen and total proteins were extracted using IP buffer (50 mM Tris-HCl pH7.5, 150 mM NaCl, 10 mM MgCl₂, 10% glycerol, 0.1% NP40) containing 1 mM PMSF, 1×protease inhibitor (Roche), 1×PhosStop (Roche) and 50 µM MG132. After centrifugation (20000 g, 10 min) twice at 4 °C, 10 µL of the anti-GFP mAb-agarose (MBL) or anti-Myc mAb-agarose (Thermo Fisher) or human IgG sepharose (GE Healthcare) was added into the supernatant and the tubes were rotated at 4 °C for 2 h. Then, the pellet was washed 5 times with IP buffer, and proteins were eluted by adding 20 µL 2×SDS loading buffer (diluted from 5×SDS loading buffer), boiled at 95 °C for 10 min and and analyzed by western blot.

In vitro pull down assays

For constructing pT7CFE1-EBF1-CFtag and pT7CFE1-EBF2-CFtag, full length cDNAs of EBF1 and EBF2 were amplified from pLexA-EBF1 and pLexA-EBF2 using primers NdeI-EBF1-F/EcoRI-EBF1-R and NdeI-EBF2-F/EcoRI-EBF2-R respectively, and then cloned into the NdeI and EcoRI restriction sites of pT7CFE1-CFtag (Pierce). For constructing pT7CFE1-PIF3-CMyc, full length cDNA of PIF3 was amplified from pB42AD-PIF3 WT using primers NdeI-PIF3-F/SalI-PIF3-R and then cloned into the NdeI and SalI restrictions of pT7CFE1-CMyc (Pierce). For constructing pT7CFE1-EBF1-CMyc, full-length cDNA of EBF1 was amplified from pLexA-EBF1 using primers NdeI-EBF1-F/EcoRI-EBF1-R, and then cloned into the NdeI and EcoRI restriction sites of pT7CFE1-CMyc (Pierce).

For constructing pT7CFE1-PIF3 A20-CMyc and pT7CFE1-PIF3 D6-CMyc, full length cDNAs of PIF3 A20 and PIF3 D6 were digested from pT7CFE1-PIF3 A20-CHis and pT7CFE1-PIF3 D6-Chis [17], and then inserted into the NdeI and SalI restriction sites of pT7CFE1-CMyc (Pierce) respectively. For constructing pT7CFE1-CUL1-Chis, full length cDNA of CUL1 was amplified from the cDNAs of Col using primers EcoRI-CUL1-F/XhoI-CUL1-R and then cloned into the EcoRI and XhoI restriction sites of pT7CFE1-CHis. For constructing pGEX-4T-1-ASK1, Full length cDNA of ASK1 was amplified from Col cDNAs using primers BamHI-ASK1-F/EcoRI-ASK1-R, and then cloned into the BamHI and EcoRI restriction sites of pGEX-4T-1.

All in vitro translated (IVT) proteins were expressed using 1-step human-coupled IVT kit (Pierce). GST and GST-ASK1 proteins were expressed in E. coli and then purified with Glutathione Sepharose 4B (GE Healthcare). Anti-flag (Sigma) or EZview Red c-Myc-Agarose (Sigma) was used to immunoprecipitate bait protein for 1h and then washed with IP buffer (the same as the buffer used in Co-IP). Then, pellets were eluted with 20 µL 2×SDS loading buffer (diluted from 5×SDS loading buffer), boiled at 95 °C for 10 min and analyzed by western blots.

Western blot analyses

Protein samples were separated on home-made polyacrylamide gels and transferred to a PVDF film (Immobilon-P; Millipore) using a semi-dry method. After being blocked for 1 h at room temperature with blocking solution containing 5% milk in PBST (2.56 g/L Na₂HPO₄·7H₂O, 8 g/L NaCl, 0.2 g/L KCl, 0.2 g/L KH₂PO₄, 0.1% Tween-20, pH 7.4), the film was incubated with primary antibody in PBST containing 3% bovine serum albumin (BSA; Sigma) overnight at 4 °C and then washed 3 times with PBST followed by incubation with 1:10000 dilution of Goat Anti-Mouse IgG (whole molecule)-Peroxidase antibody (Sigma) or Goat Anti-Rabbit IgG (whole molecule)-Peroxidase antibody (Sigma) in blocking solution for 1h at room temperature. After being washed 3 times with PBST, the PVDF film was developed with ECL prime (GE Healthcare) and visualized on X-ray film (GE Healthcare). The primary antibodies were anti-GFP (1:1000; Abmart), anti-Myc (1:1000; Sigma), anti-flag (1:2500; Sigma), anti-His (1:500; Sigma), anti-GST (1:1000; Cell Signaling Technology), anti-Ub (1:1000; Cell Signaling Technology) anti-CUL1 (1:1000) [42], anti-PIF3 (1:500) [10], anti-RPN6 (1:1000) [43] and anti-RPT5 (1:1000) [43].

Luciferase complementation imaging assays

The cluc-EBF1ΔF and cluc-EBF2ΔF plasmids were gifts from Dr. Hongwei Guo’s lab (unpublished). For generating PIF3-nluc plasmid, PIF3 cDNA was amplified from pB42AD-PIF3 WT using primers nluc-PIF3-F/nluc-PIF3-R, and then inserted into the KpnI and SalI restriction sites of pCAMBIA1300-nluc vector [45], to construct pCAMBIA1300-nluc PIF3. Then the cluc- and nluc-fused plasmids were transformed into Agrobacterium strain GV3101, and the transformants were infiltrated into tobacco leaves. After 3 days growth, the tobacco plants were transferred to darkness for 12 h, and then the luciferase activity was measured using LB985 NightShade (Berthold Techonologies).

Precipitation of polyubiquitinated proteins

3-day-old dark-grown seedlings were irradiated with saturating red light for 5 min and then transferred to darkness for another 5 min. Total proteins were extracted using MOPS buffer (100 mM MOPS, pH 7.4, 150mM NaCl, 0.1% Nonidet P-40, 1% Triton X-100) containing 1 mM PMSF, 1× protease inhibitor (Roche), 1×PhosStop (Roche), 50 µM MG132 and 20 µM PR-619 (LifeSensors) and then cleared by centrifugation (20000 g, 10 min) twice at 4 °C. 30 µL of tandem ubiquitin-binding entities 2 (TUBE2; Lifesensors) were added into the supernatant and incubated at 4 °C for 4 h. Then the pellet was washed 4 times with MOPS buffer and proteins were eluted by adding 20 µL 1×SDS loading buffer (diluted from 5×SDS loading buffer), boiled at 95 °C for 10 min and analyzed by western blots.

RNA extraction and quantitative RT-PCR

Total RNA was extracted from 4-day-old dark-grown seedlings according to the instruction of RNeasy Plant Mini Kit (Qiagen). Then first strand cDNA was synthesized from 1 µg total RNA using ReverTra Ace qPCR RT Master Mix (TOYOBO). The cDNA was then diluted 10 times and 1 µL was used for qPCR in each replicate. Quantitative PCR was performed using SYBR Premix Ex Taq (Takara) in an ABI 7500 fast real-time instrument. The quantitative PCR procedure was as follows: 95 °C for 30 seconds followed by 40 cycles of 95 °C for 5 seconds and 60 °C for 34 seconds. The relative expression levels of PIF3 were normalized to internal control PP2A. Materials for qRT-PCR were collected from three biological replicates, and three technical replicates were performed in each experiment. Primers used for qRT-PCR assays are qPIF3F, qPIF3R, qPP2AF and qPP2AR.

QUANTIFICATION AND STATISTICAL ANALYSIS

For quantification of the phenotypes, the hypocotyl lengths were measured using Image J software and the data were shown in mean±SEM. n indicates the number of the seedlings used for hypocotyl length measurements and here are the details. Figure 1A: Col (n=28), EBF1-TAP (n=27), EBF2-TAP (n=24), PIF3-Myc (n=27), PIF3-Myc/EBF1-TAP (n=28), PIF3-Myc/EBF2-TAP (n=28). Figure 1B: Col (n=22), pif3-3 (n=21), ebf1–3ebf2-2 (n=12), pif3-3ebf1–3ebf2-2 (n=13). Figure 7B, Dc: Col (n=25), lrb1lrb2lrb3 (n=27), ein3eil1 (n=33), ein3eil1ebf1ebf2 (n=31); Rc: Col (n=28), lrb1lrb2lrb3 (n=23), ein3eil1 (n=28), ein3eil1ebf1ebf2 (n=24). Figure S1A: Col (n=21), EBF1-TAP (n=23), EBF2-TAP (n=22), PIF3-Myc (n=19), PIF3-Myc/EBF1-TAP (n=19), PIF3-Myc/EBF1-TAP (n=19). Figure S1B: Col (n=29), pif3-3 (n=35), ebf1–3ebf2-2 (n=33), pif3-3ebf1–3ebf2-2 (n=17). Figure S2A: Col (n=26), EBF1-TAP (n=24), EBF2-TAP (n=22), PIF3-Myc (n=26), PIF3-Myc/EBF1-TAP (n=24), PIF3-Myc/EBF2-TAP (n=26). Figure S2B: Col (n=22), pif3-3 (n=23), ebf1–3ebf2-2 (n=11), pif3-3ebf1–3ebf2-2 (n=15). Statistical analysis was calculated by Student’s t test, with p values more than 0.05 being considered no significant (n.s.) while p values less than 0.05 being considered significant for the analyzed data: *, p<0.05; **, p<0.01; ***, p<0.001.

Highlights.

• SCF^EBF1/2 mediate light-induced PIF3 degradation to promote photomorphogenesis

• PIF3 phosphorylation triggers the recruitment of PIF3-EBF1/2 to core SCF scaffold

• EBF and LRB E3 ligases have distinct physiological roles in plant light responses

In Brief.

Light triggers the degradation of a group of repressors to initiate photomorphogenesis in plants. Dong et al. show that SCF^EBF1/2 work as the photomorphogenic E3 ligases that mediate light-induced PIF3 degradation, fulfilling a critical gap in the pathway of de-etiolation. A novel mechanism of signal-dependent SCF activation is revealed.