Phytochromes A and B Mediate Light Stabilization of BIN2 to Regulate Brassinosteroid Signaling and Photomorphogenesis in Arabidopsis

Phytochromes A and B (phyA and phyB) are the far-red and red lights photoreceptors mediating many light responses in Arabidopsis thaliana. Brassinosteroid (BR) is a pivotal phytohormone regulating a variety of plant developmental processes including photomorphogenesis. It is known that phyB interacts with BES1 to inhibit its DNA-binding activity and repress BR signaling. Here, we show that far-red and red lights modulate BR signaling through phyA and phyB regulation of the stability of BIN2, a glycogen synthase kinase 3 (GSK3)-like kinase that phosphorylates BES1/BZR1 to inhibit BR signaling. The BIN2 gain-of-function mutant bin2-1 displays an enhanced photomorphogenic phenotype in both far-red and red lights. phyA-enhanced accumulation of BIN2 promotes the phosphorylation of BES1 in far-red light. BIN2 acts genetically downstream from PHYA to regulate photomorphogenesis under far-red light. Both phyA and phyB interact directly with BIN2, which may promote the interaction of BIN2 with BES1 and induce the phosphorylation of BES1. Our results suggest that far-red and red lights inhibit BR signaling through phyA and phyB stabilization of BIN2 and promotion of BES1 phosphorylation, which defines a new layer of the regulatory mechanism that allows plants to coordinate light and BR signaling pathways to optimize photomorphogenesis.

There are five phytochromes in Arabidopsis designated phyA to phyE. Among them, phyA mainly acts as the far-red light photoreceptor under far-red light to promote photomorphogenesis, while phyB is the primary red light photoreceptor that mediates red light-induced photomorphogenesis under red light (Quail, 2002). Phytochromes perceive light with the covalently attached linear tetrapyrrole chromophore and exist in two different conformations: the red light-absorbing Pr form and the far red light-absorbing Pfr form (Rockwell et al., 2006). Phytochromes are in the inactive Pr form in the cytoplasm when synthesized, and then translocate into the nucleus as the active Pfr form upon light irradiation (Kircher et al., 1999;Li et al., 2011). Once activated, they interact with a group of PHYTOCHROME-INTERACTING-FACTORS (PIFs), which are basic helix-loop-helix transcription factors and act as the key negative regulators in the phytochrome signaling cascade (Ni et al., 1998;Leivar and Quail, 2011). These interactions lead to the phosphorylation, ubiquitination and degradation of PIFs (Al-Sady et al., 2006;Shen et al., 2007Shen et al., , 2008, and eventually the activation or repression of their downstream target genes (Leivar and Quail, 2011). The RING-finger E3 ubiquitin ligase COP1 is the master negative regulator of light signaling, which interacts with and targets the degradation of many transcription factors in light signaling, such as HY5 and CONSTANS (CO), to control photomorphogenesis and flowering, respectively (Deng et al., 1992;Osterlund et al., 2000;Jang et al., 2008;Liu et al., 2008). Phytochromes interact with COP1, and its enhancer SPAs (SPA1-4), to repress the activity of COP1 and promote the photomorphogenesis (Jang et al., 2010;Lu et al., 2015;Sheerin et al., 2015). Cryptochromes and UVR8 regulate blue and UV-B light signaling by interacting with COP1/SPA1, respectively (Favory et al., 2009;Lian et al., 2011;Liu et al., 2011;Zuo et al., 2011). Moreover, cryptochromes mediate responses to low intensity blue light or high temperature by interacting with PIFs (Ma et al., 2016;Pedmale et al., 2016).
Brassinosteroid is involved in inhibiting photomorphogenesis, as both BR biosynthesis and signaling deficient mutants display dwarf hypocotyl phenotype during photomorphogenesis (Chory et al., 1991;Li and Chory, 1997), and the gain-of-function mutant of BIN2, bin2-1, also displays a shortened hypocotyl phenotype in both darkness and light (Li et al., 2001). BIN2 integrates multiple signaling pathways by phosphorylating a range of proteins including the components in light signaling. For example, PIF3, PIF4, and PIF5 can be phosphorylated by BIN2 and then degraded via the 26S proteasome pathway to control transcription and hypocotyl growth (Bernardo-García et al., 2014;Ling et al., 2017). Meanwhile, light regulates the kinase activity of BIN2 through the COP1/SPA complex and HY5 (Ling et al., 2017;Li et al., 2020). Specifically, the COP1/SPA complex represses the activity of BIN2 and interferes with BIN2-PIF3 interaction in the dark to inhibit BIN2-mediated PIF3 destabilization (Ling et al., 2017), whereas HY5 enhances the activity of BIN2 via direct interaction in the light . It is known that BR induceddegradation of BIN2 is dependent on the F-box protein KIB1 , and that the repressor of BR signaling, the Nuclear Factor YCs (NF-YCs), interact with BIN2 to stabilize BIN2 (Zhang et al., 2021). However, whether light affects the stability of BIN2 to repress BR signaling is still unknown.
Previous studies have demonstrated that light represses BR signaling and promote photomorphogenesis through different mechanisms. For examples, both CRY1 and phyB interact with the dephosphorylated form of BES1 to inhibit its DNA-binding activity and repress BR signaling Wu et al., 2019). In addition, CRY1 interacts with both BIN2 and BZR1 to bring them together and regulate hypocotyl elongation (He et al., 2019). Furthermore, phyB is shown to promote the amount of the phosphorylated BZR1 (Kim et al., 2014). In this study, we show that phyA and phyB repress BR signaling by inhibiting BR-induced degradation of BIN2 and promoting the phosphorylation of BES1. We demonstrate by protein-protein interaction studies that phyA and phyB interact directly with BIN2, which may promote the accumulation of BIN2 protein, as well as the interaction between BIN2 and BES1, leading to the phosphorylation of BES1. Our results provide a new insight into the molecular mechanism by which phyA and phyB regulate BR signaling and photomorphogenesis.
After the imbibed seeds were keep at 4 • C for 3 days, they were grown on Murashige-Skoog (MS) nutrient medium plus 2% sucrose with 0.8% agar at 22 • C under white light. The experiments involving red and far-red light illuminations were described previously (Wei et al., 2021). Light spectra and intensity were measured with a Hand Held spectroradiometer (ASD) and a Li250 quantum photometer (Li-Cor) (Wei et al., 2021).

BIN2 Protein Degradation Assay in Arabidopsis
For the assays of the influence of phyA and phyB on BIN2 protein degradation in far-red and red lights, WT, phyA and phyB seedlings were grown in darkness, far-red or red light (1 µmol/m 2 /s or 50 µmol/m 2 /s) for 5 d. For the assays of the influence of far-red or red light exposure time on BIN2 protein degradation, WT and phyA or phyB seedlings were grown on MS plates supplemented with 2 µM BRZ (Sigma-Aldrich, USA) in darkness for 7 d, and then exposed to far-red (10 µmol/m 2 /s) or red light (100 µmol/m 2 /s) for 0, 5, 10, and 20 min, respectively. For the assay of the effects of far-red light intensity on the degradation of BIN2 protein, WT and phyA seedlings were grown in far-red light at the fluence rates of 0, 1, 3, and 10 µmol/m 2 /s for 5 d, respectively. For the assay of the effects of red light intensity on the degradation of BIN2 protein, WT and phyB seedlings were grown on MS plates supplemented with 2 µM BRZ in darkness for 7 d, and then exposed to red light at the fluence rates of 0, 50, 100, and 200 µmol/m 2 /s for 30 min, respectively. For the assays of the effects of phyA or phyB on brassinolide (BL)induced degradation of BIN2, WT and phyA or phyB seedlings were grown on MS plates supplemented with 2 µM BRZ in darkness for 7 d, and then transferred into liquid MS medium containing 1 µM eBL (Sigma-Aldrich, USA) and exposed to farred (10 µmol/m 2 /s) or red light (100 µmol/m 2 /s) for 30, 60, 90, and 120 min, respectively.
Lysis buffer containing 1 mM Pefabloc, cocktail and 50 µM MG132 was used to extract total protein, and Bradford assay (Bio-Rad, United States) was used to determine the total protein concentration. The supernatant of total protein was mixed with 5 × SDS loading buffer and boiled for 5 min, and subjected to Western blot analysis with an antibody against Arabidopsis BIN2 (Jiang et al., 2019).

Measurements of Hypocotyl Length and Anthocyanin Content
The seeds used for hypocotyl length analysis were plated on MS medium with 2% sucrose and kept at 4 • C for at 3 days, and then transferred to far-red or red light. Five-day-old seedlings were photographed with a digital camera (Nicon, Japan) . The hypocotyl lengths were measured using Image J software 1 . Anthocyanin content was analyzed as described previously .

BES1 Protein Phosphorylation Assay
For the assays of the influence of far-red and red lights exposure time on the phosphorylation of BES1 protein, BES1-Flag-OX/WT seedlings were grown in continuous darkness for 5 days, and then the dark-adapted seedlings were exposed to far-red or red light (10 µmol/m 2 /s or 50 µmol/m 2 /s) for different lengths of time (0, 1, 3, and 6 h). For the assay of the influence of phyA on the phosphorylation of BES1, BES1-Flag-OX/WT and BES1-Flag-OX/phyA seedlings were grown in darkness for 5 days, and then the dark-adapted seedlings were exposed to far-red light (10 µmol/m 2 /s) for different lengths of time (0, 1, 3, 6, 12, and 24 h). For the assays of the BIN2-dependent far-red and red lights regulation of the phosphorylation of BES1, BES1-Flag-OX/WT seedlings were grown on MS plates in darkness for 5 days, and then transferred into liquid MS supplemented with 10 mM LiCl, and kept in darkness or exposed to far-red or red light (10 µmol/m 2 /s or 50 µmol/m 2 /s) for different lengths of time (0, 1, 3, and 6 h), or BES1-Flag-OX/WT seedlings were grown on MS plates in darkness for 5 days, and then transferred into liquid MS supplemented with or without 1 µM BL or 10 mM LiCl for 3 h. For the assay of the BIN2-dependent phyA regulation of the phosphorylation of BES1, BES1-Flag-OX/WT and BES1-Flag-OX/phyA seedlings were grown on MS plates in darkness for 5 days, and then transferred into liquid MS medium containing 10 mM LiCl and kept in darkness or exposed to far-red light (10 µmol/m 2 /s) for different lengths of time (0, 1, 3, 6, 12, and 24 h).
Lysis buffer containing 1 mM Pefabloc, cocktail and 50 µM MG132 was used to extract total protein, and Bradford assay (Bio-Rad, United States) was used to determine the total protein concentration. The supernatant of total protein was mixed with 5 × SDS loading buffer and boiled for 5 min. An anti-Flag antibody (Sigma, F3165) was used to detect BES1-Flag protein.
To determine the effects of phyA and phyB on the interaction of BIN2 with BES1 via split-LUC assays, the cDNA fragments encoding phyA and GUS were cloned into vector pHB (Mao et al., 2005) to generate Myc-phyA and Myc-GUS. Cells harboring the constructs expressing cLUC-BES1 or BIN2-nLUC and Myc-phyA, Myc-phyB (Wu et al., 2019) or Myc-GUS (negative control) were mixed at a ratio of 1:1:1, and infiltrated into tobacco WT, phyA and phyB mutant seedlings were grown on MS plates supplemented with 2 µM BRZ in darkness for 7 d, and then exposed to far-red light (10 µmol/m 2 /s) (E) or red light (100 µmol/m 2 /s) (F) for the indicated lengths of time. (G, H) Western blot assays showing the effects of far-red or red light intensity on the degradation of BIN2 protein. WT and phyA mutant seedlings were grown on MS plates in far-red light at the fluence rates of 0, 1, 3, and 10 µmol/m 2 /s for 5 d, respectively (G); WT and phyB mutant seedlings were grown on MS plates supplemented with 2 µM BRZ in darkness for 7 d, and then exposed to the indicated light intensities of red light (H) for 30 min. (I, J) Western blot assays showing the effects of phyA or phyB on the BL-induced degradation of BIN2 protein. WT, phyA and phyB mutant seedlings were grown on MS plates supplemented with 2 mM BRZ in darkness for 7 d, and then treated with 1 mM BL, and then exposed to far-red (10 µmol/m 2 /s) (I) or red light (100 µmol/m 2 /s) (J) for the indicated lengths of time. Asterisks shown in (E), (F), (H), (I), and (J) denote the possible modified BIN2 protein upon BRZ treatment.
Frontiers in Plant Science | www.frontiersin.org leaves, and then the LUC signal in tobacco leaves was collected by a luminescent imaging workstation after 2 days' incubation with 1 mM D-luciferin sodium salt substrate.

Semi-in vivo Pull-Down Assays
For the assays of the effects of far-red and red light on phyA-BIN2 and phyB-BIN2 interactions, GST-BIN2 bait protein and MagneGST TM Glutathione Particles were incubated together to make proteins-bound MagneGST TM Glutathione Particles, which were washed with lysis buffer (50 mM Tris-HCl, pH 7.5, 150 mM NaCl, 0.2% Trition-X-100) for three times. The prey protein phyA-YFP-containing extract was prepared from phyA-YFP-OX seedlings grown on white light (100 µmol/m 2 /s) for 4-5 days, and then adapted in darkness for 3 days, and finally remained in darkness for 30 min or exposed to far-red light (10 µmol/m 2 /s) for 30 min, while the prey protein phyB-YFP-containing extract was prepared from Myc-phyB-OX seedlings adapted to darkness for 5 days, and then remained in darkness for 30 min or exposed to red light (50 µmol/m 2 /s) for 30 min.
For the assays of the effects of phyA and phyB on BIN2-BES1 interaction, the prey protein extracts were prepared from phyA, phyB, and phyA-YFP-OX or Myc-phyB-OX seedlings grown on white light (100 µmol m −2 s −1 ) for 4-5 days, and then adapted in darkness for 3 days, and finally exposed to red (50 µmol/m 2 /s) or far-red light (10 µmol/m 2 /s) for 30 min. Lysis buffer containing 1 mM Pefabloc cocktail and 50 µM MG132 was used to homogenize seedlings. The extracts were centrifugated, and the supernatant was incubated with bait proteins-bound MagneGST TM Glutathione Particles for 1 h and washed for 4-5 times with 1 mL lysis buffer for each time, and then the precipitates were eluted into 25 µL SDS 1× loading buffer and subjected to Western blot analysis. Anti-Myc (Millipore, 05-724) or anti-GFP (Abmart, M20004H) antibody was used to detect Myc-phyB or phyA-YFP.

Accession Numbers
Gene accession numbers used in this paper as follows:

phyA and phyB Are Involved in Mediating Far-Red and Red Lights Inhibition of Brassinolide-Induced Degradation of BIN2 Protein, Respectively
The GSK3-like kinase BIN2 is the key negative regulator of BR signaling, and the regulation of BIN2 stability is important for BR signaling (Peng et al., 2008). Given our previous demonstrations that photoreceptors CRY1 and phyB inhibits auxin signaling by stabilizing the key auxin signaling repressors AUX/IAA proteins , and that CRY1 inhibits GA signaling by stabilizing the key GA signaling repressors DELLA proteins , we explored whether phyA and phyB might affect the stability of BIN2 to regulate BR signaling. To this end, we firstly performed immunoblot assays using an anti-BIN2 antibody to detect BIN2 protein level in WT, phyA and phyB mutant seedlings grown in continuous darkness or far-red or red light, respectively. The results showed that, in the WT background, much more BIN2 protein accumulated in far-red or red light than in the dark, whereas in the phyA or phyB mutant background, basically similar very low level of BIN2 was detected in the dark and far-red or red light (Figures 1A,B; Supplementary Figure 4). RT-qPCR analysis demonstrated that BIN2 expression was not increased, but decreased to a varied extent in the WT or phyA or phyB mutant seedlings exposed to far-red or red light (Figures 1C,D). These results indicated that far-red and red lights induce the accumulation of BIN2 protein and that phyA and phyB might be responsible for mediating this process in far-red and red lights, respectively. As the BR biosynthesis inhibitor BRZ promotes BIN2 protein content in the etiolated seedlings (Peng et al., 2008), we then analyzed the accumulation of BIN2 protein using the BRZ-treated WT, phyA and phyB mutant seedlings grown in darkness for 7 d, and then exposed to different lengths of time of far-red and red lights, respectively. We found that BIN2, accumulated faster in WT than in phyA and phyB mutant backgrounds within 30 min far-red and red lights irradiation, respectively (Figures 1E,F; Supplementary Figure 4). We further analyzed BIN2 levels in the etiolated WT, phyA and phyB mutant seedlings exposed to different fluence rates offar-red and red lights, respectively. As WT and phyB mutants have low BIN2 protein levels when growing under continuous dark or red light, we treated WT and phyB mutant seedlings with 2 µM BRZ in darkness for 7 d, and then exposed to different fluence rates of red light for 30 min; While WT and phyA seedlings were grown in far-red light at the fluence rates of 0, 1, 3, and 10 µmol/m 2 /s for 5 d, respectively. The results showed that, in the WT background, BIN2 protein level increased as the fluence rate of far-red or red light increased (Figures 1G,H;  Supplementary Figure 4), but hardly increased in phyA or phyB mutant background (Figures 1G,H; Supplementary Figure 4). Taken together, these results demonstrate that phyA and phyB mediate far-red and red light-induced accumulation of BIN2 protein, respectively.
Given the previous demonstration that the exogenous application of an active BR, brassinolide (BL) induces BIN2 degradation (Peng et al., 2008), we further examined whether phyA and phyB would inhibit the BL-induced degradation of BIN2 protein under far-red and red lights. To minimize the potential endogenous BRs content difference in WT and phyA or phyB mutant with different treatments, we applied BRZ in the subsequent assays. We performed Western blotting assay using the BRZ treated WT, phyA and phyB mutant seedlings grown in darkness for 5 d, and then transferred to liquid medium containing 1 µM eBL exposed to different lengths of time of far-red and red lights. As shown in Figures 1I,J and  Supplementary Figure 4, the effects of red and far-red lights were stronger than those of BL within two hours of BL treatment, making BIN2 protein slightly increase in WT background as the light exposure time increased. Moreover, BIN2 protein was degraded much faster in phyA and phyB mutant seedlings than in WT seedlings when exposed to far-red and red lights. These results indicate that phyA and phyB are able to inhibit the BL-induced degradation of BIN2 protein under far-red and red lights.

BIN2 Acts to Promote Photomorphogenesis in Far-Red and Red Lights
To explore whether BIN2 might regulate photomorphogenic development in far-red and red lights, we examined the hypocotyl phenotype of the gain-of-function mutant of BIN2, bin2-1 (Li et al., 2001), under red and far-red lights, respectively. The results showed that bin2-1 seedlings developed significantly shorter hypocotyls than WT in far-red and red lights (Figures 2A-D). We then analyzed anthocyanin contents in bin2-1 mutant grown in far-red and red lights, and found that bin2-1 seedlings accumulated significantly more anthocyanin than WT in far-red or red light (Figures 2E,F). These results suggest that BIN2 is involved in promoting photomorphogenesis in far-red and red lights, respectively.

phyA Promotes the Accumulation of the Phosphorylated BES1 in a BIN2-Dependent Manner Under Far-Red Light
It is known that BES1 and BZR1 are the important transcription factors of BR signaling, and the ratios of phosphorylated BES1/BZR1 (pBES1/pBZR1) to dephosphorylated BES1/BZR1 (dBES1/dBZR1) is the read-out of the BR signaling status . BIN2 phosphorylates BES1 and BZR1 to inhibit their DNA-binding activities, thus blocking the transduction of the BR signaling (Vert and Chory, 2006). Based on our demonstration that phyA mediates far-red light inhibition of the degradation of BIN2, we asked whether phyA might mediate far-red light modulation of the phosphorylation of BES1 to regulate BR signaling. To test this possibility, we firstly used the etiolated transgenic seedlings over-expressing BES1 fused to Flag in the WT background (BES1-Flag-OX/WT)  exposed to different lengths of time of far-red light, and examined the phosphorylation status of BES1. The results showed that the ratio of pBES1/dBES1 was increased as the far-red light exposure time increased (Figure 3A and Supplementary Figures 3, 4). We  and BES1-Flag-OX/phyA (D) seedlings were grown in darkness for 5 days, and then exposed to the far-red light (10 µmol/m 2 /s) for the indicated lengths of time. (E,F) Western blot assays showing the effects of BIN2 on far-red and red lights promotion of the phosphorylation of BES1-Flag protein. BES1-Flag-OX/WT seedlings were grown on MS plates in darkness for 5 days, and then treated with 10 mM LiCl and exposed to far-red light (10 µmol/m 2 /s) (E) or red light (50 µmol/m 2 /s) (F) or adapted in the dark for the indicated lengths of time. (G,H) Western blot assays showing the effects of BIN2 on phyA-mediated far-red light promotion of the phosphorylation of BES1-Flag protein. BES1-Flag-OX/WT (G) and BES1-Flag-OX/phyA (H) seedlings were grown in darkness for 5 days, and then treated with 10 mM LiCl and exposed to far-red light (10 µmol/m 2 /s) for the indicated lengths of time.
also found that red light is able to promote the accumulation of the phosphorylated of BES1, and the ratio of pBES1/dBES1 was also increased as the red light exposure time increased ( Figure 3B and Supplementary Figures 3, 4). We then generated the transgenic seedlings overexpressing BES1-Flag in the phyA mutant background (BES1-Flag-OX/phyA), and analyzed the phosphorylation status of BES1 protein in BES1-Flag-OX/WT and BES1-Flag-OX/phyA seedlings adapted in the dark and then exposed to far-red light for different lengths of time, respectively. The results showed that the ratio of pBES1/dBES1 was increased faster in the WT background than in the phyA mutant background as the far-red light exposure time increased (Figures 3C,D and Supplementary Figures 3, 4). These results indicate that phyA mediates far-red light-induced accumulation of the phosphorylated BES1.
It is shown that LiCl inhibits the GSK3-like kinase activity of BIN2 and its homologs, leading to the accumulation of dephosphorylated BZR1/BES1 in vivo . To further determine whether phyA would affect BES1 phosphorylation in a BIN2-dependent manner, we first performed immunoblotting analyses using BES1-Flag-OX/WT seedlings grown in darkness for 5 days, and then transferred to the liquid medium containing 10 mM LiCl adapted in darkness or exposed to far-red or red light for different lengths of time. The results showed that, in the presence of LiCl, the phosphorylated BES1 failed to accumulate as the far-red or red light exposure time increased, and the ratio of pBES1/dBES1 remained at a similar low level ( Figures 3E,F and Supplementary Figures 3, 4). Moreover, in the presence of both LiCl and BL, the phosphorylated BES1 also hardly accumulated upon far-red or red light irradiation, and the ratio of pBES1/dBES1 was lower than in the presence of BL only or in the absent of both LiCl and BL (Supplementary Figures 1, 5). We then perform the same assays to evaluate the phosphorylation status of BES1 in the WT and phyA mutant backgrounds, respectively. The results showed that the treatment of LiCl reduced the accumulation of phosphorylated BES1 in both WT and phyA mutant backgrounds within 3 h (Figures 3G,H and  Supplementary Figures 3-5). After 6 h treatment, the effects of LiCl became weak, and the ratio of pBES1/dBES1 increased faster in the WT than in the phyA mutant (Figures 3G,H and  Supplementary Figures 3-5). After 24 h treatment, LiCl barely had effects, probably because the treatment was too long, which may trigger other processes affecting the phosphorylation of BES1 (Figures 3G,H and Supplementary Figures 3-5). Taken together, these results indicate that phyA-mediated far-red light promotion of the accumulation of the phosphorylated BES1 is dependent on BIN2. . Data in (B) is mean ± SD (n = 30). Data in (C) is mean ± SD (n = 2). Statistically significant differences are determined by a one-way analysis of variance (ANOVA), followed by Tukey's least significant difference (LSD) test (P < 0.05), which are marked by letters "a" to "d." Frontiers in Plant Science | www.frontiersin.org served as a bait. The protein extracts prepared from phyA-YFP-OX and Myc-phyB-OX seedlings served as preys. phyA-YFP-OX and Myc-phyB-OX seedlings were grown in dark for 5 days, and then exposed to far-red light (FR, 10 µmol/m 2 /s) and red light (R, 50 µmol/m 2 /s) for 30 min, respectively.

BIN2 Acts Genetically Downstream of PHYA to Regulate Photomorphogenesis Under Far-Red Light
To explore the genetic relationships between BIN2 and PHYA, we introgressed bin2-1 into phyA mutant background by genetic crossing and generated the phyA bin2-1 double mutant. Photomorphogenic phenotype analysis showed that the phyA bin2-1 double mutant developed significantly shorter hypocotyls than phyA mutant in far-red light (Figures 4A,B), and accumulated more anthocyanin than phyA mutant under farred light ( Figure 4C). However, the phyA bin2-1 double mutant developed a little longer hypocotyls, but produced much lower levels of anthocyanins than the bin2-1 mutant. These results indicate that BIN2 may act as an important downstream component of phyA signaling to regulate hypocotyl elongation in far-red light, but play a minor role in mediating phyA signaling to regulate anthocyanin accumulation.

Both phyA and phyB Physically Interact With BIN2
Given our previous demonstrations that CRY1 or phyB inhibits auxin and GA signaling by interacting with AUX/IAA and DELLA proteins to stabilize them, respectively , we explored whether phyA and phyB might affect the stability of BIN2 by interacting with BIN2. To verify this assumption, we performed pull-down experiments with the recombinant His-TF-phyA-N (phyA N-terminus), His-TF-phyA-C (phyA C-terminus), His-TF-phyB-N (phyB N-terminus) and His-TF-phyB-C (phyB C-terminus) fusion proteins as preys, and GST-BIN2 as a bait. As shown in Figures 5A,B and Supplementary Figure 5, His-TF-phyA-N, His-TF-phyA-C, His-TF-phyB-N and His-TF-phyB-C were all pulled down by GST-BIN2, but not by the His-TF control, indicating that both the N and C termini of phyA and phyB interact with BIN2 protein in vitro.
To further prove whether phyA and phyB interact with BIN2 in vivo, we performed split luciferase complementation (split-LUC) assays in Nicotiana benthamiana leaves. phyA or phyB and BIN2 were fused to the N-and C-terminal halves of firefly luciferase (cLUC-phyA or cLUC-phyB and BIN2-nLUC), respectively. We found that the luciferase activity was reconstituted when cLUC-phyA or cLUC-phyB was co-expressed with BIN2-nLUC, but not when cLUC-phyA or cLUC-phyB was co-expressed with nLUC or when BIN2-nLUC was co-expressed with cLUC (Figures 5C,D). These results indicated that both phyA and phyB interact with BIN2 in vivo.
To investigate whether light quality may affect the interactions of phyA and phyB with BIN2, we performed semi-in vivo pulldown experiments. The recombinant GST-BIN2 protein promotion of the interactions of BIN2 with BES1. GST-BIN2 served as bait. His-BES1 served as prey. The protein extracts prepared from the dark-adapted phyA and phyA-YFP-OX/WT or phyB and Myc-phyB-OX/WT seedlings exposed to far-red light (10 µmol/m 2 /s) or red light (50 µmol/m 2 /s) for 30 min, respectively. was incubated with the protein extracts prepared from transgenic Arabidopsis seedlings overexpressing phyA-YFP-OX or Myc-phyB-OX that were dark-adapted before exposure to farred or red light. As shown in Figures 5E,F and Supplementary Figure 5, BIN2 pulled down phyA and phyB from the extracts prepared from seedlings exposed to far-red and red lights, respectively, but not from those prepared from dark-adapted seedlings. These results suggest that phyA and phyB may interact with BIN2 in a far-red and red light-dependent manner, respectively.

phyA and phyB May Promote the Interaction of BIN2 With BES1
Based on our demonstrations that phyA and phyB interact with BIN2 ( Figure 5) and that phyA and phyB promote BIN2 accumulation and BES1 phosphorylation in far-red and red lights (Figures 1, 3), we explored whether phyA and phyB might promote the interaction of BIN2 with BES1. To do this, we first performed split-LUC assay in N. benthamiana leaves and confirmed the interaction between BIN2 with BES1 ( Supplementary Figure 2A). We then detected the interaction strength between BIN2 and BES1 in the presence of Myc-phyA, Myc-phyB, and the control protein Myc-GUS, respectively. The results showed that the interaction of BIN2 with BES1 was enhanced significantly by phyA and phyB, but not by GUS (Figures 6A,B,D,E). Western blot assays showed that Myc-GUS was expressed at higher or similar levels compared with Myc-phyA and Myc-phyB (Figures 6C,F and Supplementary  Figure 5). BIN2-nLUC and cLUC-BES1 were expressed at only slightly lower levels when co-expressed with Myc-GUS than when co-expressed with Myc-phyA (Supplementary  Figures 2B, 5) or Myc-phyB (Supplementary Figures 2C, 5).
These results indicate that phyA and phyB may promote the interaction of BIN2 with BES1.
We then performed cell-free pull-down assays with recombinant GST-BIN2, His-BES1 proteins and the protein extracts prepared from phyA, phyB, and phyA-YFP-OX or FIGURE 7 | A model illustrating how phyA and phyB may inhibit BL-induced degradation of BIN2 protein and BR signaling. (A,B) In darkness and in the presence of BR, phyA and phyB are inactive in the cytoplasm and unable to interact with BIN2 and BES1 proteins. BIN2 is ubiquitinated by KIB1 and degraded through the 26S proteasome, thus BES1 is dephosphorylated, and able to bind to its target genes to promote their expression and skotomorphogenesis (A); upon light irradiation, phyA and phyB are activated, and then translocated into the nucleus. We speculate that the interactions of phyA and phyB with BIN2 in light may promote the dissociation of BIN2 from its E3 ligase, such as KIB1, to stabilize BIN2 and enhance the interaction of BIN2 with BES1. Consequently, the phosphorylation of BES1 is promoted and its binding to its target genes is repressed, thus BR signaling is inhibited and photomorphogenesis is promoted (B).
Myc-phyB-OX seedlings adapted in darkness and exposed to red and far-red light, respectively. We first tested this system by confirming the interaction between BIN2 and BES1, and found that the strength of the interaction between BIN2 and BES1 was enhanced as the amounts of His-BES1 increased (Supplementary Figures 2D, 5), which is consistent with the previous study (Zhao et al., 2002). We then added the protein extracts prepared from phyA, phyB, and phyA-YFP-OX or Myc-phyB-OX seedlings, and found that much more His-BES1 was pulled down by GST-BIN2 in the presence than in the absence of the activated phyA or phyB (Figures 6G,H and Supplementary  Figure 5). These results indicate that phyA and phyB may promote the interaction of BIN2 with BES1 in a far-red and red light-dependent manner, respectively.

DISCUSSION
Given that the BR-deficient mutant det2 (de-etiolated 2) was initially identified in the genetic screening for mutants showing constitutive photomorphogenic phenotype in darkness (Chory et al., 1991), it is likely that light and BR signaling are connected in some way. Through 30 years' study, some impressive progresses have been made in dissecting the molecular mechanism of the crosstalk between light and BR signaling. It is known that BES1/BZR1 and BIN2 function at the heart of BR signaling networks, as BES1/BZR1 can directly regulate the expression of thousands of the downstream BR-responsive genes, and BIN2 can phosphorylate and destabilize BES1/BZR1 to repress BR signaling (Li et al., 2001;Wang et al., 2002;Yin et al., 2002;Vert and Chory, 2006). The multiple photoreceptors and the negative regulators of light signaling can mediate light repression of BR signaling by regulating the protein stability and the DNA-binding ability of BES1 and BZR1. Specifically, UVR8, phyB and CRY1 interact with dephosphorylated BES1 and BZR1 specifically to inhibit the DNA-binding ability of BES1 and repress the expression of their target genes (Liang et al., 2018;Wang et al., 2018;Wu et al., 2019). COP1 degrades the phosphorylated BZR1 in darkness, while SINAT E3 ligases are involved in degrading dephosphorylated BES1 protein in the light (Kim et al., 2014;Yang et al., 2017). It is shown that light regulates the kinase activity of BIN2 through the COP1/SPA1 complex and HY5 (Ling et al., 2017;Li et al., 2020). However, whether light regulates BR signaling through modulating the BIN2 protein stability is not known.
In this study, we report that phyA and phyB mediate light regulation of BR signaling and photomorphogenesis by modulating the stability of BIN2 protein. we show by protein expression assays that phyA-and phyB-mediated far-red and red lights signaling inhibits BL-induced degradation of BIN2 protein (Figure 1), and that the accumulated BIN2 promotes the phosphorylation of BES1 (Figure 3). Analysis of the phenotype of bin2-1 mutant in monochromatic light conditions demonstrates that BIN2 acts to promote photomorphogenesis in far-red and red lights, respectively (Figure 2). Genetic interaction analysis indicates that BIN2 acts partially downstream of phyA to regulate photomorphogenesis under far-red light (Figure 4). We show by protein-protein interaction studies that phyA and phyB interact directly with BIN2 (Figure 5), and that phyA and phyB may promote the interaction between BIN2 and BES1 (Figure 6), which may in turn enhance the phosphorylation of BES1 by BIN2. These results, combined with the previous studies, suggest that phytochromes may mediate light inhibition of BR signaling through modulation of the activity of both BES1 and BIN2. On the one hand, they interact with the dephosphorylated BES1 to inhibit its DNA-binding ability (Wu et al., 2019); On the other hand, they stabilize BIN2 to promote the accumulation of the phosphorylated BES1, thus inhibiting its DNA-binding ability to promote photomorphogenesis (this report).
How do phyA and phyB regulate BIN2 protein stability? Previous studies have shown that phyB mediates red light induction of the degradation of EIN3 by enhancing the binding of the EBF1/EBF2 E3 ligase to EIN3 and regulate ethylene response (Shi et al., 2016), and that blue light-triggered interactions of CRY1 with AUX/IAA proteins or DELLA proteins inhibit the associations of SCF TIR1 with AUX/IAA proteins or the interactions of the GA receptor GID1 and SCF SLY1 with DELLA proteins to repress auxin or GA signaling . Given these demonstrations and our results showing that phyA and phyB interact with BIN2 directly (Figure 5), we speculate that the interactions of phyA and phyB with BIN2 may promote the dissociation of BIN2 from its E3 ligase, such as KIB1 , to inhibit the ubiquitination and degradation of BIN2. Since it is known that NF-YCs promotes the autophosphorylation of BIN2 to protect BIN2 degradation, while the phosphorylated BSU1 dephosphorylates BIN2 to trigger the degradation of BIN2 via the 26S proteasome (Kim et al., 2009;Ryu et al., 2010;Zhang et al., 2021), it is also possible that the interactions of phyA and phyB with BIN2 may result in the inhibition of the dephosphorylation and degradation of BIN2. These possibilities will be worth exploring in future studies. In sum, we propose that, in darkness and in the presence of BR, as phyA and phyB are at their inactive Pr form, and localized in the cytoplasm, they cannot interact with BIN2 and BES1. BIN2 is ubiquitinated by KIB1 and degraded through the 26S proteasome, thus BES1 is dephosphorylated, and able to bind to its target genes to promote their expression and skotomorphogenesis ( Figure 7A); Upon light irradiation, phyA and phyB are activated and translocate from the cytoplasm to the nucleus, where they may inhibit the association of BIN2 with KIB1 to stabilize BIN2 and enhance the interaction of BIN2 with BES1. As a result, BIN2 accumulates and the phosphorylation of BES1 is enhanced, and BES1 binding to its target genes is inhibited, leading to the inhibition of BR signaling and the eventual promotion of photomorphogenesis ( Figure 7B).

DATA AVAILABILITY STATEMENT
The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding author/s.

AUTHOR CONTRIBUTIONS
WW and H-QY conceived the project and wrote the manuscript. JZ, GY, and WW designed the research plan. JZ and GY carried out the most of the experiments. All authors contributed the article and approved the submitted version. BES1-Flag-OX/WT seedlings were grown on MS plates in darkness for 5 days, and then treated with or without 10 mM LiCl and 1 µM BL, and exposed to far-red light (FR, 10 µmol/m 2 /s) (A) or red light (R, 50 µmol/m 2 /s) (B) for 3 h.