Brassinosteroids promote etiolated apical structures in darkness by amplifying the ethylene response via the EBF-EIN3/PIF3 circuit

Abstract

Germinated plants grow in darkness until they emerge above the soil. To help the seedling penetrate the soil, most dicot seedlings develop an etiolated apical structure consisting of an apical hook and folded, unexpanded cotyledons atop a rapidly elongating hypocotyl. Brassinosteroids (BRs) are necessary for etiolated apical development, but their precise role and mechanisms remain unclear. Arabidopsis thaliana SMALL AUXIN UP RNA17 (SAUR17) is an apical-organ-specific regulator that promotes production of an apical hook and closed cotyledons. In darkness, ethylene and BRs stimulate SAUR17 expression by transcription factor complexes containing PHYTOCHROME-INTERACTING FACTORs (PIFs), ETHYLENE INSENSITIVE 3 (EIN3), and its homolog EIN3-LIKE 1 (EIL1), and BRASSINAZOLE RESISTANT1 (BZR1). BZR1 requires EIN3 and PIFs for enhanced DNA-binding and transcriptional activation of the SAUR17 promoter; while EIN3, PIF3, and PIF4 stability depends on BR signaling. BZR1 transcriptionally downregulates EIN3-BINDING F-BOX 1 and 2 (EBF1 and EBF2), which encode ubiquitin ligases mediating EIN3 and PIF3 protein degradation. By modulating the EBF-EIN3/PIF protein-stability circuit, BRs induce EIN3 and PIF3 accumulation, which underlies BR-responsive expression of SAUR17 and HOOKLESS1 and ultimately apical hook development. We suggest that in the etiolated development of apical structures, BRs primarily modulate plant sensitivity to darkness and ethylene.

Brassinosteroids repress EBF1 and EBF2, and the resulting increase in EIN3 and PIFs stimulates transcription of SAUR17 and HLS1 through the EIN3–PIFs–BZR1 complex, producing etiolated apical structures.

IN A NUTSHELL.

Background: Seeds germinate underneath the soil; to help the shoot tissues emerge from the soil, in plants such as Arabidopsis the shoot forms an apical hook and the cotyledons remain closed and unexpanded. This pointed apex helps the growing plant penetrate through the soil. This process is called etiolated development and is sensitive to environmental signals such as light and soil, but is also regulated by internal hormonal signals such as brassinosteroids (BRs), auxin, and more. The molecular nature of the crosstalk between those signals that determine the apical structures has remained unclear. SMALL AUXIN UP RNA17 (SAUR17), a gene that functions in controlling cell size, is expressed solely in apical organs of young dark-grown seedlings to maintain etiolated apical structures.

Question: What signals and factors drive the expression of SAUR17? What signaling inputs and interactions do plants use to maintain apical etiolation?

Findings: Arabidopsis SAUR17 is not regulated by auxin but is stimulated by darkness via PHYTOCHROME-INTERACTING FACTORs (PIFs), by ethylene (which responds to soil mechanical pressure) via ETHYLENE INSENSITIVE 3 (EIN3) and its homolog EIN3-LIKE 1 (EIL1), and by BRs via BRASSINAZOLE RESISTANT1 (BZR1). These transcription factors form protein complexes, which bind and activate the SAUR17 promoter primarily through EIN3 and PIFs. However, BR signaling is necessary for the stability of EIN3 and PIF3 proteins, in part by BZR1-mediated down-regulation of an ubiquitin ligase gene. By elevating EIN3 and PIF3 protein levels and enhancing their promoter-binding activities, BR stimulates the expression of SAUR17 and HOOKLESS1, and ultimately the formation of apical hook.

Next steps: This study suggests that BRs promote etiolated apical structure predominantly by enhancing the ethylene/darkness pathways that respond to environmental cues. SAUR17 expression also reflects the transient and developmental-stage-specific nature of etiolation, but the nature of this remains unknown.

Introduction

When germinating while buried under soil in darkness, seeds of plants such as Arabidopsis thaliana (Arabidopsis) undergo a form of skotomorphogenesis known as etiolation. They produce an apical hook, small and closed cotyledons with etioplasts, and a rapidly elongating hypocotyl (Darwin, 1881; Wei et al., 1994; Arsovski et al., 2012). The apical hook protects the apical meristem from mechanical damage caused by emergence through the soil, and the closed and unexpanded cotyledons keep the apex of the seedling in a pointed shape suitable for soil penetration (Goeschl et al., 1966; Darwin, 1881; Harpham et al., 1991; Ecker, 1995). The components and mechanisms regulating development of apical structures are not identical to those modulating hypocotyl elongation (de Wit et al., 2016; Sun et al., 2016; Shi et al., 2018). Apical hook development is precisely regulated by auxin: the auxin concentration gradient across the hook drives the asymmetric cell growth that results in the curvature of the apical hook (Beziat and Kleine-Vehn, 2018; Du et al., 2022). The differential distribution of auxin at the hook is associated with etiolation, but our understanding of how etiolation is established and maintained in plants to guide the development of apical organs remains superficial.

Lack of a light signal is the predominant determinant of etiolation in plant seedlings. The PHYTOCHROME-INTERACTING FACTORs (PIFs), which are basic helix–loop–helix transcription factors, play key roles in etiolated development of both the apical organs and the hypocotyl. The pifq quadruple mutant (pif1 pif3 pif4 pif5) of Arabidopsis lacks an apical hook and, when grown in the dark, exhibits opened cotyledons and a short hypocotyl (Ni et al., 1998; Leivar et al., 2008; Shin et al., 2009). PIFs accumulate to high levels in the dark and are degraded upon the transition to light (Al-Sady et al., 2006; Paik et al., 2017).

In darkness, the development of the seedling apex is profoundly influenced by ethylene (Ecker, 1995; Harpham et al., 1991; Raz and Ecker, 1999). Ethylene is produced in plants in response to the mechanical pressure of the overlying soil, and this induces an exaggerated apical hook along with a thickened short hypocotyl (Goeschl et al., 1966; Ecker, 1995; Zhong et al., 2014). Ethylene responses are mediated by the transcription factor ETHYLENE INSENSITIVE 3 (EIN3) and its homolog EIN3-LIKE 1 (EIL1) (Chao et al., 1997; Solano et al., 1998). In the absence of ethylene, EIN3 and EIL1 are targeted for degradation by EIN3-BINDING F-BOX 1 and 2 (EBF1 and EBF2) (Guo and Ecker, 2003; Potuschak et al., 2003; Gagne et al., 2004). The same ubiquitin E3 ligases also mediate light-induced degradation of PIF3 (Dong et al., 2017).

Brassinosteroid (BR) hormones also play important roles in seedling etiolation, as BR-deficient mutants grown in the dark display short hypocotyls and open cotyledons without apical hooks (Li et al., 1996). BR responses are mediated by the transcription factors BRASSINAZOLE RESISTANT1 (BZR1), BRI1-EMS-SUPPRESSOR1 (BES1), and their homologs (Wang et al., 2002; Vert and Chory, 2006). To regulate hypocotyl growth, BRs coordinate with light, temperature, and auxin signaling pathways through the BZR1-PIF4-ARF6 transcription module (Bai et al., 2012; Oh et al., 2012, 2014). However, in apical organs, BRs have been shown to collaborate with ethylene to modulate apical hook development (Smet et al., 2014; Zhao et al., 2021).

A key gene in apical hook development is HOOKLESS 1 (HLS1) (Lehman et al., 1996), which is thought to work at least in part by modulating auxin distribution and responses in the hook region (Lehman et al., 1996; Li et al., 2004), although its specific biochemical function has not been clearly defined. HLS1 can be activated by either EIN3/EIL1 or PIFs (Shi et al., 2018; Zhang et al., 2018) in cooperation with BZR1 (Zhao et al., 2021), but the molecular nature of the BR-ethylene relation remains unclear.

Downstream of the above-mentioned signaling pathways and transcription factors are target genes or effector genes that directly modulate cell growth. Among the effector genes involved in seedling etiolation are SMALL AUXIN UP RNA (SAUR) genes (Spartz et al., 2014; Ren and Gray, 2015; Stortenbeker and Bemer, 2019). Most SAUR proteins induce cell expansion by inhibiting D-clade type 2C protein phosphatases (PP2C-D), resulting in activation of the H⁺-ATPase proton pump (Spartz et al., 2014; Sun et al., 2016; Du et al., 2020; Yin et al., 2020). Consistent with this cellular function, proteins encoded by the SAUR6/12/14/16/50 subgroup, whose expression is induced by light in cotyledons, promote cotyledon separation and expansion following the dark-to-light transition (Dong et al., 2019; Wang et al., 2020). In contrast to other SAURs, SMALL AUXIN UP RNA17 (SAUR17) works to protect PP2C-D1 against SAUR50, thereby preventing cell expansion (Wang et al., 2020). In the dark, PP2C-D1 predominantly associates with SAUR17 in apical organs (Wang et al., 2020), where it inhibits cell expansion and promotes the structures of the apical hook and closed cotyledons (Wang et al., 2020; Rovira et al., 2021; Du et al., 2022). Upon the transition to light, the expression of SAUR17 declines, while the expression of SAUR6/12/14/16/50 increases in apical organs, causing inactivation of PP2C-D1 and presumably concomitant cell expansion in the hook and cotyledon regions (Wang et al., 2020). Thus, differential control of SAUR gene expression represents an important mechanism underlying the drastic morphogenic changes seen in the apical organs of seedlings during de-etiolation.

SAUR mRNAs tend to be short lived, and their expression can rapidly respond not only to auxin but also to light, BRs, gibberellin, abscisic acid, and ethylene, usually in a tissue-specific and developmental stage-dependent manner (Newman et al., 1993; Chae et al., 2012; Spartz et al., 2012; van Mourik et al., 2017). SAUR17 is expressed only in the apical organs of dark-grown seedlings, where it promotes hook development and closed cotyledons (Wang et al., 2020). In the current study, we aimed to gain insights into how plants establish and maintain the etiolation state in the apical organs by determining the signals and mechanisms governing SAUR17 expression. We found that SAUR17 expression is stimulated by darkness, ethylene, and BRs through transcription factor complexes composed of PIFs, EIN3, and BZR1. However, the roles of these transcription factors are distinct. BZR1 depends on EIN3 and PIFs for transcriptional activity and promoter binding, while EIN3 and PIF3 depend on BR signaling for protein stability. We report that BR induces protein accumulation of EIN3 and PIF3, and this largely accounts for BR-responsive induction of SAUR17, HLS1, and apical hook development. The increased EIN3 and PIF3 levels are caused in part by BZR1-mediated downregulation of EBF1/2 expression. We found that the EBF1/2-EIN3/PIF3 protein stability circuit plays a central role in the interdependent regulation of apical structures by BRs and ethylene.

Results

PIFs and EIN3/EIL1 induce SAUR17 expression to regulate etiolated development of apical organs

SAUR17 is expressed specifically in the apical organs of young dark-grown wild-type seedlings (Wang et al., 2020), but the transcripts are essentially absent in the pifq mutant, indicating that PIF transcription factors are necessary for SAUR17 expression (Figure 1A). In addition, SAUR17 is expressed transiently from approximately 1.5 to 3 days post germination and then declined and nearly disappeared following prolonged growth in the dark (Figure 1A). The timing of its peak expression largely overlaps with the peak abundance of endogenous PIF3 and PIF4 (Supplemental Figure S1, A and B). Overexpression of the PIFs strongly increased the level of expression of SAUR17 in the dark, confirming that PIFs serve as positive regulators of SAUR17, while light irradiation inhibited SAUR17 expression (Figure 1B and Supplemental Figure S1H). Likewise, cop1-6 and det1-1 mutants, which lack stable accumulation of PIF proteins (Bauer et al., 2004; Dong et al., 2014; Ling et al., 2017), showed nearly no expression of SAUR17 (Supplemental Figure S1C). However, HY5, a positive regulator of photomorphogenesis (Osterlund et al., 2000), did not affect SAUR17 expression (Supplemental Figure S1D). To evaluate the functional contribution of SAUR17 as an effector gene of the PIFs, we forced SAUR17 expression by adding the 35S enhancer upstream of the SAUR17 promoter to drive SAUR17-GFP expression (see “Materials and methods”). This transgene successfully expressed SAUR17 in pifq at levels moderately above the level of the wild type (Supplemental Figure S1, E and F). Remarkably, forced expression of SAUR17 in pifq partially rescued the cotyledon separation phenotype of pifq (Figure 1, C and D). These results show that SAUR17 requires PIFs for expression in the dark, and that it functions as a key downstream effector of PIFs in maintaining closed cotyledons during etiolated development.

Figure 1.

PIFs and EIN3/EIL1 mediate SAUR17 expression to maintain cotyledon closure and apical hook development in the dark. A, Expression levels of SAUR17 in Col and pifq seedlings grown in the dark for 1–6 days (D1–D6). RNA samples extracted from whole seedlings were analyzed by RT–qPCR. Statistical significance was calculated by two-way ANOVA along with Sidak’s post-test. ***P < 0.001; *P < 0.05; ns, no significant difference. B, SAUR17 transcript levels in PIF-overexpressing and wild-type seedlings. RT-qPCR assay of RNA samples from 3-day-old dark-grown plants exposed to white light for the indicated times. Two-way ANOVA was used to calculate significant differences. Tukey’s post-test was used, and different lowercase letters above the bars indicate significant differences at P < 0.001. C and D, Forced expression of SAUR17 suppressed the cotyledon-opening phenotype of pifq in the dark. pifq and three independent transgenic lines of Enh35S:ProSAUR17:SAUR17-GFP/pifq were grown in the dark for 3 days. The cotyledon opening angles are shown as the means ± sem (n = 52, 28, 21, 60, 79 for each genotype seedlings). Significance analysis was based on one-way ANOVA along with Bonferroni correction (P < 0.05). E, ACC enhanced the expression of SAUR17. Dark-grown 3-day-old wild-type seedlings were treated with 10 μM ACC or water (mock) for 1, 3, or 6 h, and RNA samples from the indicated tissues were analyzed by RT-qPCR. Statistical significance was calculated by two-way ANOVA along with Tukey’s post-test. ***P < 0.001, **P < 0.01. F, SAUR17 transcript levels were elevated in the ethylene-overproducing mutant eto1-2 but were decreased in the ethylene-insensitive mutant etr1-1. RT-qPCR assay of RNA samples from 3-day-old dark-grown plants. Two-way ANOVA was used to calculate significant differences. Tukey’s post-test was used and different lowercase letters above the bars indicate significant differences at P < 0.01. G, SAUR17 expression was abolished in the ein3/eil1 mutant and enhanced in EIN3 overexpression seedlings. RT-qPCR assay of RNA samples from 3-day-old dark-grown whole seedlings exposed to white light for the indicated times. Two-way ANOVA was used to calculate significant differences. Tukey’s post-test was used and different lowercase letters above the bars indicate significant differences at P < 0.001. H, Mechanical pressure-induced expression of SAUR17 was dependent on EIN3/EIL1. RT-qPCR assay of RNA samples from 3-day-old dark-grown seedlings on MS plates pressed with or without applied pressure of ∼150 Pa (using a cylindrical glass plate) for 6 h. Statistical significance was calculated by two-way ANOVA along with Sidak’s post-test. ***P < 0.001; ns, no significant difference. I and J, Forced expression of SAUR17 suppressed the hook-opening phenotype of ein3 eil1 in the dark. WT, ein3 eil1, and two independent transgenic lines of Enh35S::ProSAUR17:SAUR17-GFP/ein3 eil1 were grown in the dark for 3 days. The hook angles are shown as the means ± sem, n > 32 seedlings. Significance analysis was based on one-way ANOVA along with Tukey’s correction (P < 0.001). A, B, and E–H, All RNA analysis results are shown as the means ± sd from three biological replicates, with PP2A as the internal control. Results of all statistical analyses are presented in Supplemental Data Set S1.

Our previous study showed that the saur17 mutant displays a deficiency in ethylene-induced apical hook exaggeration (Wang et al., 2020), hinting that ethylene may regulate SAUR17. Therefore, 1-aminocyclopropane-1-carboxylic acid (ACC), the immediate precursor of ethylene, was applied to dark-grown seedlings, and SAUR17 transcript levels were drastically increased 3 h after the treatment (Figure 1E). In addition, the SAUR17 transcript level was higher in eto1-2, a mutant that oversynthesizes ethylene, and lower in etr1-1, an ethylene-insensitive mutant (Figure 1F), implying that ethylene signaling positively regulates SAUR17 expression. Similar to the case for PIFs, SAUR17 transcripts were almost undetectable in ein3 eil1 mutants, while they were significantly increased in Pro35S:EIN3 compared with the wild type (Figure 1G), indicating that EIN3 and EIL1 are critical mediators of SAUR17 expression.

In the natural environment, ethylene signals reflect mechanical pressures from the compact soil (Goeschl et al., 1966; Zhong et al., 2014). We directly tested whether mechanical pressure could induce SAUR17 expression in etiolated seedlings pressed by a cylindrical plate. In response to pressure, the transcript levels of SAUR17 were increased in the wild type and even more strongly increased in EIN3-GFP/ein3 eil1 and EIN3-Flag/ein3 eil1 transgenic plants, but not in ein3 eil1 mutants (Figure 1H). Thus, EIN3 and EIL1 are necessary for SAUR17 expression and for induction by mechanical pressure. In addition, forced expression of SAUR17 partially rescued the hook-opening phenotype of ein3 eil1 (Figure 1, I and J and Supplemental Figure S1G), indicating that SAUR17 plays an important role in EIN3-/EIL1-mediated hook development. Together, our data show that PIFs and EIN3/EIL1 are critical inducers of SAUR17, which functions in maintaining the apical hook and closed cotyledons.

SAUR17 does not respond to auxin

Because auxin is critical in apical hook formation and is known to activate many SAUR genes, we examined the role of auxin in the expression of SAUR17 (Ren and Gray, 2015; Beziat and Kleine-Vehn, 2018). Dark-grown wild-type seedlings were treated with various concentrations of indole-3-acetic acid (IAA). While the expression of SAUR15, a SAUR family member known to respond to auxin, was clearly induced by IAA (Supplemental Figure S2A), the levels of SAUR17 transcripts were almost unchanged (Supplemental Figure S2B). In another experiment, dark-grown ProSAUR17:GFP seedlings were treated with various concentrations of the auxin mimic picloram. Picloram clearly affected the seedling phenotypes, but it did not change either the amount of GFP fluorescence or the location of GFP expression (Supplemental Figure S2C). These results indicate that auxin does not play a significant role in regulating SAUR17 transcription.

BRs induce SAUR17 expression in the dark

BRs are necessary for etiolation, and mutants deficient in BR synthesis or signaling display lack of an apical hook, separated cotyledons and a short hypocotyl in the dark (Supplemental Figure S3). Exogenous supplementation of BRs using the analog 2,4-epibrassinolide (eBL) enhanced the apical hook in wild-type dark-grown seedlings by forming an obtuse and exaggerated hook (Figure 2, A and B). This exaggeration of the apical hook was generally compromised in saur17 mutants (Figure 2, A and B), indicating that SAUR17 functions to promote the BR-induced apical hook response. Similar to ethylene treatment, BR treatment induced the expression of SAUR17 by more than five-fold in approximately 3 h (Figure 2C and Supplemental Figure S4A), and the sensitivity to BRs persisted for at least 24 h (Figure 2C). Either brassinazole (Brz) or propiconazole (PPZ), which inhibit BR biosynthesis, caused a marked decline in SAUR17 levels (Supplemental Figure S4, B and C). In addition, mutants deficient in BR biosynthesis or signaling, such as cpd, det2-1, br6ox2, and bri1-116, showed lower SAUR17 transcript levels than the wild type (Figure 2D).

Figure 2.

BRs induce SAUR17 expression and is involved in the development of etiolated apex. A and B, The saur17 mutant exhibited a reduced BR response in hook curvature. A, Representative seedlings of saur17 mutants and the wild type were grown on MS plates containing 1 μM eBL or ethyl alcohol (mock) for 3 days in darkness. Bar = 1 mm. B, The hook angles of seedlings in (A) were determined. n = 50 seedlings. Two-way ANOVA with Tukey’s multiple comparisons test (P < 0.05) was used to analyze the significant differences. C, BR treatment induced SAUR17 expression in the dark. RT-qPCR analysis of RNA samples from 3-day-old dark-grown seedlings treated with liquid MS containing 1 μM eBL or ethyl alcohol (mock) for the indicated times. Statistical significance was calculated by two-way ANOVA along with Sidak’s post-test, P < 0.01. D, Mutants deficient in BR synthesis or signaling showed reduced expression levels of SAUR17. The levels of SAUR17 transcripts in 3-day-old dark-grown Col and the indicated mutants were analyzed by RT-qPCR. One-way ANOVA was used to calculate significant differences. Tukey’s post-test, P < 0.01. E, Mutated BZR1 subfamily members display a photomorphogenesis phenotype in the dark. Col and the bzr1 bes1 beh1 beh2 beh3 beh4 (+/−) (bzr^quintuple beh4 [+/−]) seedlings were grown in the dark for 3.5 days and photographed. Bar = 1 mm. F, The BZR-deficient mutant (bzr^quintuple beh4 [+/−]) exhibited reduced SAUR17 levels. For each repeat, 15 3-day-old dark-grown seedlings of each genotype were collected for RNA extraction followed by RT-qPCR. Student’s t test, ***P < 0.001. G, BR-induced SAUR17 expression in bzr1bes1 and BZR1 overexpression plants. Dark-grown 3-day-old seedlings of the indicated genotypes were treated with 1 μM eBL or equimolar ethyl alcohol (mock) for 3 or 6 h. RNA levels were analyzed by RT-qPCR. Two-way ANOVA was used to calculate significant differences. Tukey’s post-test was used and different lowercase letters above the bars indicate significant differences at P < 0.05. H, Light inhibited BZR1-mediated enhancement of SAUR17 expression. RT-qPCR analysis of RNA samples from 3-day-old dark-grown seedlings transferred to white light for the indicated times. Two-way ANOVA was used to calculate significant differences. Tukey’s post-test was used and different lowercase letters above the bars indicate significant differences at P < 0.001. In panels C, D, F–H, the data are shown as the means ± sd from three biological replicates (separate biological materials and extraction processes) with PP2A as the internal control. Results of all statistical analyses are presented in Supplemental Data Set S1.

BR-responsive gene expression is mediated by the BZR1 family of transcription factors, which contains six members in Arabidopsis (Sun et al., 2010; Chen et al., 2019). In an effort to determine the role of BZR1-family transcription factors in the expression of SAUR17, we attempted to create bzr1 bes1 beh1 beh2 beh3 beh4 sextuple mutants using the CRISPR/Cas9 gene editing system (Wang and Chen, 2020; Supplemental Figure S5). We managed to obtain a few seeds of bzr1 bes1 beh1 beh2 beh3 beh4 (+/−) (bzr^quintuple beh4 [+/−]), which were infertile. Dark-grown bzr^quintuple beh4 (+/−) seedlings displayed a reduced hook curvature and slightly separated cotyledons similar to those of BR-deficient mutants (Figure 2E and Supplemental Figure S3), and they had clearly decreased transcript levels of SAUR17 (Figure 2F). In fact, the bzr1bes1 double mutant already displayed decreased SAUR17 expression with or without BR stimulation (Figure 2G). BZR1 overexpression (Pro35S:BZR1-YFP) moderately elevated the levels of SAUR17 transcripts and significantly enhanced the BR-responsive increase in SAUR17 to above 12-fold that of the wild type (Figure 2, H and G), while light irradiation abolished SAUR17 expression despite BZR1 overexpression (Figure 2H). These results demonstrate that SAUR17 expression in the dark is strongly stimulated by BR signaling through BZR1 transcription factors.

PIFs, EIN3, and BZR1 directly target the SAUR17 promoter

Having demonstrated that PIFs, EIN3/EIL1, and BZR1 are each necessary for SAUR17 expression in the dark, we next studied whether these transcription factors associate with the SAUR17 promoter in vivo. The SAUR17 promoter region contains one PIF binding motif, the PBE-box (CACATG), and three putative EIN3 binding sites (EBSs), with EBS2 and EBS3 only 48 bp apart (Figure 3A). EBS2 (ATACAT) is the reverse complementary sequence of EBS1 (ATGTAT), which is a confirmed EBS in the ERF1 promoter (Song et al., 2015). The reverse complementary sequence of EBS3 (TTTGAAAT) has been extensively documented to be directly bound by EIN3 in the HLS1 promoter (An et al., 2012). The BZR1-binding motif BRRE (He et al., 2005) is not present in the SAUR17 promoter region, but BZR1 is known to work with PIFs and bind to the G-box and PBE-box (Sun et al., 2010; Oh et al., 2012).

Figure 3.

PIFs, EIN3, and BZR1 bind to the SAUR17 promoter. A, Schematic diagram of the SAUR17 promoter region showing the EIN3 and PIFs/BZR1 binding sites. Numbers indicate the distance from the ATG start codon. The lower part summarizes the results from panels B–G. Filled circles indicate binding at the corresponding sites; empty circles indicate lack of binding; half-filled circles indicate reduced binding. B and C, ChIP-qPCR assays showing PIF3 and PIF4 binding to the SAUR17 promoter in vivo. Seedlings were grown in the dark for 4 days and an anti-Myc antibody was used for ChIP. D, EIN3 bound to EBSs of the SAUR17 promoter. Plants were grown in the dark for 3 days and an anti-GFP antibody was used for ChIP. E, BZR1 bound to PBE and EBSs on the SAUR17 promoter. Plants were grown in the dark for 3 days and were collected for ChIP-qPCR assays using an anti-GFP antibody. F, BZR1 showed reduced binding to the SAUR17 promoter in the pifq mutant. Plants were grown in the dark for 3 days and were collected for ChIP-qPCR assays using an anti-GFP antibody. G, BZR1 could not bind to the EBSs of the SAUR17 promoter in the ein3 eil1 mutant. Plants were grown in the dark for 3 days and were collected for ChIP-qPCR assays using an anti-GFP antibody. For all ChIP assays, the data shown are the means ± sd (n = 3, three independent experiments were performed). 18S was used as a negative control. B–G, Two-way ANOVA was used to calculate significant differences with Tukey’s post-test. Different lowercase letters above the bars indicate significant differences at P < 0.001. ***P < 0.001, ns, no significant differences, Supplemental Data Set S1.

Chromatin immunoprecipitation (ChIP)-qPCR of Pro35S:PIF3-Myc and Pro35S:PIF4-Myc plants showed that PIF3 and PIF4 bind to the PBE-box as well as EBSs on the SAUR17 promoter (Figure 3, B and C). This result is consistent with published ChIP-seq data (Oh et al., 2012; Pfeiffer et al., 2014) on the SAUR17 promoter. ChIP–qPCR experiments using a Pro35S:EIN3-GFP/ein3 eil1 line showed that EIN3 can directly bind to EBS1, EBS2, and EBS3 (Figure 3D). The ChIP-qPCR results using BZR1-YFP transgenic plants showed that BZR1 is enriched not only on the PBE box but also on the EBSs of the SAUR17 promoter (Figure 3E). Taken together, the results indicate the PBE-box is bound by PIFs and BZR1, while the EBSs are bound by EIN3, PIFs, and BZR1 (Figure 3A).

EIN3 and EIL1 mediate the binding of BZR1 to the EBSs of the SAUR17 promoter

To investigate whether the binding of BZR1 to the SAUR17 promoter required EIN3/EIL1 and the PIFs, we performed BZR1-GFP ChIP experiments in the wild type, ein3 eil1, and pifq backgrounds (Pro35S:BZR1-GFP, Pro35S:BZR1-GFP/ein3 eil1, and Pro35S:BZR1-GFP/pifq transgenic plants, respectively). Our results showed that BZR1 could bind to the PBE box of the SAUR17 promoter in the absence of PIFs or EIN3/EIL1, but the amounts of enrichment were decreased (Figure 3, F and G). Strikingly, BZR1 binding to the EBSs was abolished in the ein3 eil1 background, while binding to the PBE-box was reduced (Figure 3G). These ChIP results indicate that EIN3 and EIL1 are responsible for recruiting BZR1 to the EBSs, while PIFs facilitate the binding of BZR1 to the PBE-box and EBSs. Notably, although BZR1 still bound to the SAUR17 promoter in the pifq and ein3 eil1 mutant backgrounds, SAUR17 was essentially not expressed in those mutants, suggesting that the binding of BZR1 to the promoter on its own, without PIFs and EIN3/EIL1, was insufficient to mediate SAUR17 transcription in the dark (Figure 1, A, G, and H).

BZR1/BES1 interacts with EIN3 and PIF3

Previous studies have reported interactions between PIF3 and EIN3 in the apical hook and for chloroplast development (Liu et al., 2017; Zhang et al., 2018) and between BZR1 and PIF4 in the context of hypocotyl elongation (Bai et al., 2012; Oh et al., 2012, 2014). In addition to PIF4, we found that BZR1 also interacted with PIF3, as BZR1-YFP coimmunoprecipitated endogenous PIF3 with or without eBL treatment in dark-grown seedlings (Figure 4A). Both the phosphorylated and dephosphorylated forms of BZR1 could be pulled down by PIF3 in vitro. However, the results indicated that the interaction between PIF3 and dephosphorylated BZR1 was stronger than that between PIF3 and phosphorylated BZR1 (Figure 4B).

Figure 4.

BZR1 interacts with PIF3 and EIN3. A, BZR1 interacted with PIF3 in vivo. Dark-grown 3-day-old WT and Pro35S:BZR1-YFP seedlings were treated with 1 μM eBL or buffer (mock) for 4 h. Co-IP was performed with anti-GFP mAb-agarose and analyzed by immunoblotting using antibodies against GFP and PIF3. RPN6 was used as a control. B, In vitro pulldown of phosphorylated and unphosphorylated BZR1 by MBP-PIF3. Amylose resin beads were used in the pull-down assays, and His-TrxA-BZR1 was examined by anti-His immunoblotting. C, BZR1 interacted with EIN3 in vivo. Plants grown in the dark for 3 days were treated with 1 μM eBL or equimolar ethyl alcohol (mock) for 6 h. GFP-Trap beads were used for precipitation. D, BES1 interacted with EIN3 in vivo. Plants grown in the dark for 3 days were treated with 1 μM eBL or equimolar ethyl alcohol (mock) for 6 h. IP was performed using GFP-Trap beads. An anti-BES1 antibody was used for immunoblotting.

We examined the potential interaction of EIN3 with BZR1 and BES1, the latter being a close homolog of BZR1 with 88% overall amino acid sequence identity. Similar to PIF3, BZR1-YFP in both the phosphorylated and dephosphorylated forms could coimmunoprecipitate EIN3-Flag in dark-grown seedlings (Figure 4C). In the opposite direction, EIN3-GFP coimmunoprecipitated endogenous BES1 in EIN3-GFP/ein3eil1 etiolated seedlings (Figure 4D). We noticed that EIN3-GFP accumulated at elevated steady-state levels after eBL treatment, and this might have contributed to the increased amount of coprecipitated BES1 in the eBL-treated sample (Figure 4D). Together, our data demonstrate that BZR1 and BES1 interact with EIN3 and that BZR1 also interacts with PIF3 in vivo, all of which occur regardless of the phosphorylation state of BZR1/BES1. Evidently, the EIN3–BZR1 interaction enables BZR1 to associate with the EBSs of the promoter DNA through piggybacking on EIN3 (Figure 3G).

BR induction of SAUR17 expression is dependent on PIFs and EIN3/EIL1

We further examined the roles of EIN3/EIL1 and PIFs in BR-directed stimulation of SAUR17 expression. While eBL application induced the expression of SAUR17 more than five-fold in dark-grown wild-type seedlings, the induction was nearly absent in pifq, ein3 eil1, or pifq ein3 eil1 mutants, and the overall transcription of SAUR17 was dramatically diminished in these mutants (Figure 5, A–C). Strikingly, ein3 eil1 showed a complete BR insensitivity to an extent similar to, if not more severe than, the BR-receptor null mutant bri1-116, in which the SAUR17 levels were well below that of the wild-type mock treatment (Figure 5, B and D). These results demonstrate that EIN3/EIL1 and PIFs play essential roles in BR-induced transcriptional activation of SAUR17. EIN3/EIL1 and the PIFs do not regulate BZR1 expression, since BZR1 transcript levels were not affected by pifq and ein3/eil1 mutations (Supplemental Figure S6F), suggesting EIN3/EIL1 and the PIFs are involved in other aspects of BR signaling.

Figure 5.

BR-mediated induction of SAUR17 is dependent on EIN3/EIL1 and PIFs, whose binding to the promoter is enhanced by BR. A–D, BR-induced increases in SAUR17 transcript levels were diminished in the pifq mutant (A), the ein3 eil1 mutant (B), the pifq ein3 eil1 mutant (C), and the BR receptor mutant bri1-116 (D). Seedlings grown in the dark for 3 days were treated with 1 μM eBL or equimolar ethyl alcohol (mock) for 3 or 6 h. Total RNA from dissected cotyledons with hooks (A and B) or whole seedlings (C and D) was used for RT-qPCR. The data are shown as the means ± sd from three biological replicates (separate biological materials and extraction processes). Two-way ANOVA was used to calculate significant differences with Tukey’s post-test. Different lowercase letters above the bars indicate significant differences at P < 0.05. E and F, The BZR1-mediated promotion of SAUR17 expression was enhanced by PIF4 overexpression and diminished by pifq or by ein3eil1. RNA from seedlings grown in the dark for 4 days was used for RT-qPCR analysis. The data are shown as the means ± sd from three biological replicates. Different letters indicate statistically significant differences (one-way ANOVA, Tukey’s multiple comparison, P < 0.05). G and H, ChIP-qPCR assays showing BR-induced PIF3 and EIN3 binding to the SAUR17 promoter in vivo. Seedlings grown in the dark for 3 days were treated with 0.5 μM eBL or equimolar ethyl alcohol (mock) for 6 h, and an anti-GFP antibody was used for ChIP. Two-way ANOVA was used to calculate significant differences with Tukey’s post-test. ***P < 0.001 and ns, no significant differences. Results of all statistical analyses are presented in Supplemental Data Set S1.

Overexpression of BZR1 (Pro35S:BZR1-GFP) achieved less than a two-fold increase in SAUR17 transcript levels, similar to the effect of bzr1-1D, the constitutively active mutant of BZR1. bzr1-1D worked additively with overexpression of PIF4 in the stimulation of SAUR17 (Figure 5E). However, SAUR17 transcript levels were sharply decreased in Pro35S:BZR1-GFP/pifq (Figure 5E and Supplemental Figure S6, A–D) and Pro35S:BZR1-GFP/ein3 eil1 (Figure 5F and Supplemental Figure S6E). These results show that BZR1-mediated SAUR17 expression strongly depends on PIFs and EIN3/EIL1.

Given the essential roles of PIFs and EIN3/EIL1 in BR-mediated induction of SAUR17 transcription (Figure 5, A, B, E, and F), we next investigated whether BRs affect PIFs and EIN3/EIL1 binding to the SAUR17 promoter by ChIP-qPCR. Our results showed that the BR significantly induced PIF3 binding to the EBS and PBE-box sites (Figure 5G) and EIN3 binding to the EBSs of the SAUR17 promoter (Figure 5H). These results further suggest that BRs increase SAUR17 transcript levels by increasing the promoter-binding activities of PIFs and EIN3.

BRs stabilize EIN3 and PIF3 and repress EBF1/2 transcription

In the protein interaction experiments, we noticed that BR treatment appeared to increase EIN3 and PIF3 protein levels (Figure 4, A, C, and D). This was further confirmed by immunoblotting of EIN3 using EIN3-GFP/ein3 eil1 and EIN3-Flag/ein3 eil1 transgenic seedlings, in which the EIN3 transgenes are driven by the cauliflower mosaic virus 35S promoter (Pro35S). As shown in Figure 6, A and B and Supplemental Figure S7, A and B, EIN3 protein steady-state levels were increased after treatment with eBL and decreased after treatment with Brz, an inhibitor of BR biosynthesis. Similarly, endogenous PIF3 levels increased in seedlings treated with eBL and declined in seedlings treated with Brz (Figure 6, C and D). BR also enhanced the PIF3 transcript level by over two-fold (Supplemental Figure S7C), but it hardly affected the EIN3 transcript level after 3 h of treatment and slightly decreased the transcript level after 6 h of treatment (Supplemental Figure S7D). Thus, EIN3 and PIF3 protein levels were increased by BRs, while they decreased under low-BR conditions. These results are consistent with the observation that the BR stimulated the promoter-binding activity of PIF3 and EIN3 (Figure 5, G and H).

Figure 6.

BR treatment elevates EIN3 and PIF3 protein levels and decreases EBF1/2 transcript levels. A and B, BR treatment increased EIN3 levels. Seedlings grown in the dark for 3 days were treated with 1 μM eBL or equimolar ethyl alcohol (mock) for 3 or 6 h (A), or the seedlings were grown on MS medium supplied with 0.5 µM BR-synthesis inhibitor Brz in the dark for 3.5 days (B). Total proteins were extracted and immunoblotted with antibodies against GFP and actin (the loading control). C and D, BR treatment increased PIF3 protein levels. Seedlings grown in the dark for 3 days were treated with 1 μM eBL (C) or 1 μM Brz (D) for 4 or 8 h. The total proteins were immunoblotted with antibodies against PIF3 and ACTIN (loading control). E, BR treatment did not affect EBF1 and EBF2 protein levels. EBF1-TAP and EBF2-TAP seedlings were grown in the dark for 3 days and then treated with 1 μM eBL or equimolar ethyl alcohol (mock) for 3 or 6 h. Proteins from dissected cotyledons and hooks were analyzed by immunoblotting with antibodies against Myc and ACTIN (the loading control). F, EBF1 and EBF2 were downregulated by BR treatment. Wild-type seedlings grown in the dark for 3 days were treated with 1 μM eBL or equimolar ethyl alcohol (mock) for 3 h. Total RNA from the whole seedlings was analyzed by RT-qPCR. G, RT-qPCR results showing elevated transcription levels of EBF1 and EBF2 in the BZR-deficient mutant (bzr^quintuple beh4 [+/−]). F and G, The data are shown as the means ± sd from three biological replicates. Statistical significance was determined using Student’s t test (***P < 0.001; **P < 0.01). H, Schematic illustration of the EBF1 and EBF2 promoter regions. The number indicates the distance from the ATG start codon. The G-box (CACGTG) and PBE-box (CACATG) are the confirmed BZR1 binding sites. I, ChIP-qPCR assay showing BZR1 binding to the promoters of EBF1 and EBF2. Seedlings were grown in the dark for 3 days and an anti-GFP antibody was used to enrich immunoprecipitated DNAs. The error bars represent the sd from three biological replicates. Two-way ANOVA was used to calculate significant differences with Tukey’s post-test. Different lowercase letters above the bars indicate significant differences at P < 0.001. Results of all statistical analyses are presented in Supplemental Data Set S1.

We next examined EBF1 and EBF2, the ubiquitin E3 components that mediate EIN3 and PIF3 degradation. The protein stability of EBF1 and EBF2 was not significantly affected by BR treatment, as indicated by immunoblot analyses of EBF1-TAP and EBF2-TAP (Figure 6E). However, the transcript levels of EBF1 and EBF2 moderately declined after eBL treatment of dark-grown seedlings (Figure 6F). Moreover, EBF1 and EBF2 transcript levels were elevated in the BR-response-deficient mutant bzr^quintuplebeh4 (+/−) (Figure 6G). Reexamination of previously published microarray data revealed that EBF1 and EBF2 were identified as BZR1-repressed genes (Sun et al., 2010). Recent RNA-Seq data in bzr1 bes1 beh1 beh2 beh3 beh4 sextuple mutants also showed upregulation of EBF2 (Chen et al., 2019), consistent with our observations. Together, these data suggest that the BR signaling pathway negatively regulates EBF1 and EBF2 expression.

In a search for putative BZR1 binding sites, we found a G-box (CACGTG) and two PBE-boxes (CACATG) in the EBF1 promoter and a PBE-box in the EBF2 promoter, while we did not find BR-response elements (BRRE, CGTG^T/_CG) in either promoter (Figure 6H). ChIP-qPCR assays using Pro35S:BZR1-YFP plants showed that BZR1 was enriched in the PBE boxes of EBF1 and EBF2 (Figure 6I). In agreement with this result, EBF1 and EBF2 were among the high-stringency BZR1 targets analyzed by a ChIP-chip study (Sun et al., 2010). Together, these data suggest that EBF1 and EBF2 are repressed in response to BR by BZR1-subfamily transcription factors, most likely through direct binding to the promoters.

BRs induce SAUR17 and apical hook exaggeration through EBF1/2-dependent increases in EIN3 and PIF protein abundance

Given the key role of EBFs as the ubiquitin E3 ligases of EIN3 and PIF3, we hypothesized that the BR-triggered decline in EBF expression, which should result in the accumulation of EIN3 and PIF3, may in turn cause increased expression of SAUR17. In that case, constitutive expression of EBFs might buffer the levels of EBFs from the BR-induced decreases in endogenous EBFs, thereby interfering with BR-responsive SAUR17 expression. To test this hypothesis, we first examined endogenous EIN3 and PIF3 levels in EBF1- and EBF2-overexpressing transgenic plants with or without BR treatment. The results showed that the BR-induced protein increase in EIN3 and PIF3 was indeed largely eliminated in EBF1- and EBF2-overexpressing lines, along with overall diminished levels of the proteins (Figure 7, A and B and Supplemental Figure S7, E and F). Because the protein accumulation of EIN3 and PIF3 was essentially prevented in these lines, we used them to evaluate the contributions of increased levels of EIN3 and PIF3 to BR-induced expression of SAUR17. The overall expression of SAUR17 as well as the induction by the BR were significantly diminished in EBF overexpression lines (Figure 7C). In particular, the EBF2-GFP line, which exhibited greater BR insensitivity and greater reductions in the levels of EIN3 and PIF3 than the EBF1-GFP line, showed a more drastic decrease in SAUR17 expression, with a severity comparable to that seen in the BR-insensitive mutant bri-116 (Figure 5D). Thus, EBFs are critical factors in BR-induced protein accumulation of EIN3 and PIF3 and consequently BR-responsive gene expression of SAUR17.

Figure 7.

EBF-dependent increases in EIN3 and PIF3 are necessary for BR-responsive expression of SAUR17 and HLS1 and for etiolated apical structure development. A and B, Overexpression of EBF1 and EBF2 diminished BR-induced PIF3 and EIN3 accumulation. Seedlings grown in the dark for 3 days were treated with 1 μM eBL or equimolar ethyl alcohol (mock) for 3 or 6 h. Total proteins were immunoblotted with antibodies against PIF3, EIN3, and Actin (loading control). C, Overexpression of EBF1 and EBF2 decreased SAUR17 expression levels and SAUR17 induction by BR. D, Overexpression of EBF2 abolished BR-induced HLS1 expression. In panels (C and D), seedlings grown in the dark for 3 days were treated with 1 μM eBL or equimolar ethyl alcohol (mock) for 3 or 6 h. Total RNA from whole seedlings was analyzed by RT-qPCR. The data are shown as the means ± sd from three biological replicates. Two-way ANOVA was used to calculate significant differences with Tukey’s post-test. Different lowercase letters above the bars indicate significant differences at P < 0.05. E, Overexpressing EBF2 abolished BR-induced exaggeration of apical hook angles. Seedlings were grown in the dark for 3 days with 1 μM eBL or equimolar ethyl alcohol (mock). On each box, the central line indicates the median, and the bottom and top edges of the box indicate the 25th and 75th percentiles, respectively. Two-way ANOVA with Tukey’s multiple comparisons test was used to analyze the significant differences (n = 51, 97, 42, 54, 37, 72 for each genotype seedlings). F, A model illustrating the mechanism of BR in promoting etiolated development of apical organs in darkness. BR treatment inhibited EBF1/2 transcription via BZR1, resulting in stabilization of the EIN3/EIL1 and PIF3 proteins. As a result, increasing amounts of EIN3 and PIF3 bound to the SAUR17 promoter, where they formed PIF–EIN3–BZR1 transcription factor complexes via protein–protein interactions. Solid lines indicate direct actions, dashed lines indicate multistep actions, and gray lines are based on reported studies. Results of all statistical analyses are presented in Supplemental Data Set S1.

To determine whether BR-induced increases in EIN3 and PIF3 might have broader implications beyond the regulation of SAUR17, we examined the expression of HLS1, which is also regulated by BR and ethylene (Zhao et al., 2021). HLS1 was induced by eBL under the same conditions, although not as strongly as SAUR17 (Figure 7, C and D). Strikingly, the BR-responsive increase in HLS1 was completely abolished in the EBF2 overexpression line, as the plants were completely insensitive to the BR with regard to HLS1 expression (Figure 7D). Furthermore, EBF2-overexpressing seedlings exhibited reduced hook angles and, more importantly, lacked BR-responsive changes in the apical hook phenotype (Figure 7E). Upon comparison of the wild-type, EBF1-GFP, and EBF2-GFP lines, we observed a correlation: compromise of the BR-induced increases in EIN3/PIF3 protein levels was associated with reduced BR responses in SAUR17 expression (Figure 7C), HLS1 expression (Figure 7D), and ultimately in the apical hook phenotype (Figure 7E). These results strongly suggest that BR activation of SAUR17 transcription, as well as the BR-induced apical hook phenotype, is mostly the result of BR-responsive increases in EIN3 and PIF3 levels (Figures 6 and 7).

Discussion

Apical structures such as the apical hook and small, closed cotyledons are transient growth forms of etiolated seedlings germinating underneath the soil, where the main environmental cues are darkness and soil pressure. This study elucidates a key mechanism of the phytohormones BRs in this event. We report that transcription of SAUR17, an etiolation effector, is mediated by an apical-organ-specific transcriptional module composed of PIF3/4, EIN3/EIL1, and BZR1 (Figure 7F). This differs from the reported PIF4-BZR1-ARF6 module that regulates hypocotyl elongation (Oh et al., 2014). The transcription module is regulated via the EBF-EIN3/PIF3 protein-stability circuit that converges the three-way crosstalk from the light/dark, ethylene (soil condition), and BR signaling pathways (Figure 7F). By impinging on this circuit, BRs elevate the protein levels of the ethylene transcription factor EIN3 and the light repressor PIF3, enhancing the promoter binding of BZR1–EIN3–PIF complexes and increasing transcription of SAUR17.

SAUR17 versus HLS1

HLS1 is a genetically well studied gene in ethylene-mediated apical hook formation (Lehman et al., 1996; Li et al., 2004). At the level of transcription, HLS1 appears to be regulated by a similar set of transcription factors as is SAUR17 (Zhao et al., 2021; Figure 7, C and D), but the functions of the two genes differ in several ways. First, SAUR17 influences apical hook formation as well as cotyledon opening. Forced expression of SAUR17 reversed the cotyledon-opening phenotype of pifq (Figure 1, C and D) and partially reversed the apical hook-opening phenotype of ein3 eil1 (Figure 1, I and J), while HLS1 was studied mostly for its role in apical hook development. Second, SAUR17 is a downstream effector gene that serves to restrict cell expansion in situ (Wang et al., 2020). In contrast, HLS1 is a more upstream regulator whose precise biochemical function has yet to be defined, but it works in part by regulating auxin distribution and signaling in the hook, and this in turn influences many genes involved in cell growth (Lehman et al., 1996; Li et al., 2004). Nonetheless, both genes are regulated by BRs and ethylene in an interdependent manner.

Interdependence of ethylene–BR signaling at the level of transcription factors

BRs collaborate extensively with ethylene to regulate apical hook development (Smet et al., 2014; Zhao et al., 2021). We show here that the dependence of BRs on ethylene manifests in several layers at the molecular level. First, BZR1 requires EIN3 to bind to the EBSs of the SAUR17 promoter. Second, it requires both EIN3/EIL1 and PIFs to become transcriptionally active. In the absence of EIN3/EIL1 (or PIFs), even though BZR1 weakly associated with the promoter (Figure 3, F and G), it barely activated transcription (Figure 5, E and F), indicating that BZR1 alone is insufficient to activate SAUR17 transcription on the promoter. Third, BR-induced protein accumulation of EIN3 and PIF3 was necessary for the robust transcriptional increase in SAUR17 as well as the morphologic change in the apical hook in response to BRs. When the increase in EIN3 and PIF3 was compromised by EBF2 overexpression, the BR responses in terms of expression of SAUR17 and HLS1 and exaggeration of the apical hook were consequently diminished (Figure 7, C–E).

Conversely, we found that the ethylene response depended on BR signaling for the stability of its transcription factors because low-BR conditions lead to destabilization of EIN3 (Figure 6B and Supplemental Figure S7B), PIF3 (Ling et al., 2017; Figure 6D), and PIF4 proteins (Bernardo-Garcia et al., 2014). BR signaling elevates EIN3 protein levels and increases EIN3 promoter-binding activities to achieve transcriptional activation. Through this mechanism, BRs invoke ethylene transcription factors (and PIFs) to mediate its responses in apical etiolation. However, given that BZR1 is recruited by EIN3 to the EBSs of the SAUR17 promoter, it seems more likely that EIN3 and PIF3 act together with BZR1 to stimulate SAUR17 transcription, although the specific contribution of BZR1 to the PIF–EIN3–BZR1 transcription module remains unclear.

Notably, the transcription of SAUR17 is extremely sensitive to light, and its enhancement by individual overexpression of PIFs, EIN3, or BZR1 is sharply inhibited by exposure to light (Figures 1, B and G and 2H). We believe this is because these transcription factors are required simultaneously for transcription. For example, light-triggered rapid inactivation of endogenous EIN3 (Shi et al., 2016) would still occur in PIF-overexpressing seedlings, which would deactivate SAUR17 expression despite of increased PIF levels (Figure 1B). Nonetheless, involvement of other unknown light-sensitive factors or mechanisms is also possible.

Mechanisms of BR-induced protein stabilization of PIF3 and EIN3

Our data suggest that the hub of the crosstalk between BRs and ethylene/darkness is the EBF-EIN3/PIF protein stability circuit (Figure 7F). EBF1/2 F-box proteins were originally identified as negative regulators of the ethylene pathway by mediating EIN3/EIL1 degradation (Gagne et al., 2004; Guo and Ecker, 2003; Potuschak et al., 2003; An et al., 2010). Later, EBF1/2 emerged as a key photomorphogenic component, as EBFs remove EIN3 and PIF3 in response to light (Shi et al., 2016; Dong et al., 2017). EBFs are primarily regulated by ethylene and light via posttranscriptional mechanisms (Olmedo et al., 2006; Li et al., 2015; Merchante et al., 2015; Shi et al., 2016; Hao et al., 2021). However, here we found that EBF1/2 can be transcriptionally repressed by BZR1, possibly as a mechanism of fine-tuning regulation. EBF1 can also be transcriptionally induced by MYC2, which reduces EIN3/EIL1 levels in response to jasmonate (Zhang et al., 2014). We suggest that the transcriptional downregulation of EBFs at least partially accounts for the BR-induced increase in the protein abundance of EIN3/EIL1 and PIF3. Nonetheless, other ways by which BR signaling downregulates EBFs and/or stabilizes EIN3 may also participate. For example, based on microarray data, BZR1 induces COP1 expression (Sun et al., 2010), which may cause destabilization of EBFs and consequently stabilization of PIF3 (Shi et al., 2016b; Ling et al., 2017). It is also conceivable that the process of EIN3 ubiquitination by SCF^EBFs is affected by BRs. By modulating the protein stability of EIN3 and PIF3/4, it seems that BRs promote the formation of etiolated apical structures to a large extent by reinforcing the sensitivity of plants to darkness and ethylene/soil conditions, the environmental signals most pertinent to the shapes of the apical organs of seedlings germinating under the soil.

What is the role of SAUR17 in auxin-directed apical hook development?

Auxin plays a central role in apical hook development, as its concentration gradient across the hook region dictates the differential cell growth at the opposite sides of the hook (Beziat and Kleine-Vehn, 2018; Du et al., 2022). Nonetheless, auxin had no effect on SAUR17 regarding either the expression level or localization (Supplemental Figure S2). The uniform spatial expression pattern of SAUR17 observed at the hook (Wang et al., 2020) did not match the distinctive auxin gradient distribution pattern previously described (Beziat and Kleine-Vehn, 2018; Du et al., 2022). On the other hand, SAUR12, SAUR16, SAUR50, and PP2C-D1, which exhibit differential enrichment on the concave side of the apical hook (Ren et al., 2018; Wang et al., 2020) that matches the auxin gradient, would be better candidates as targets of auxin regulation. Indeed, the expression of PP2C-D1 in the apical hook (Du et al., 2022) and SAUR50 in the hypocotyl (Sun et al., 2016) has been demonstrated to be regulated by auxin. Due to the structural complexity of apical organs, morphogenic changes in the apical hook and cotyledons likely involve coordinated actions of multiple different cell effectors at different locations, including SAUR50 and PP2C-D1 on the concave side, SAUR57 on the convex side (Wang et al., 2020), and SAUR17 uniformly on both sides. It is plausible that each of the effectors is regulated by different combinations of signal inputs.

The timing of SAUR17 expression roughly correlates with apical hook formation and maintenance when the lower hypocotyl cells undergo robust cell expansion in the dark (Wei et al., 1994; Raz and Ecker, 1999; Smet et al., 2014; Wang et al., 2020; Du et al., 2022). Considering that SAUR17 functions to prevent cell expansion, we speculate that the role of SAUR17 is to create a general growth-restricting environment in the apical region in seedlings. Against this background, auxin, through its gradient distribution, can induce or inhibit the expression of genes such as PP2C-D1, SAUR50, and SAUR57 in specific locations to trigger localized cell expansion. In other words, the role of SAUR17 is probably not to cause morphogenic changes but rather to provide an environment that enables morphogenic activities to manifest in specific locations in order to achieve changes in shape and structure. Our work shows that to establish and maintain an etiolated cellular environment in apical organs, seedlings express the etiolation effector SAUR17. SAUR17 expression is driven by darkness, ethylene, and BRs through transcription factor complexes containing PIFs, EIN3/EIL1, and BZR1, with BRs supporting and enhancing the sensitivity to darkness and ethylene (soil conditions) (Figure 7F).

Materials and methods

Plant materials and growth conditions

Arabidopsis thaliana hy5-q (N663420), dwf4-102 (N520761), cpd (N523532), det2-1 (N6159), br6ox2 (N678514), and bri1-5 (N6126) were purchased from The European Arabidopsis Stock Centre. Pro35S:BZR1-YFP/EIN3-Flag/ein3 eil1 was obtained by crossing Pro35S:BZR1-YFP to EIN3-Flag/ein3 eil1. In this study, all transgenic plants and mutants except for bri1-5 were in the A. thaliana Columbia-0 (Col-0) ecotype background, and bri1-5 was in the Wassilewskija (Ws) background. The seeds were surface-sterilized with 15% NaClO for 5 min, washed with sterile water five times, and then sown on MS medium (4.4 g/L Murashige and Skoog powder [Sigma, M5519], 1% [w/v] sucrose, 0.6% [w/v] agar, pH = 5.8). After stratification in darkness for 4 days, seeds were irradiated with white light (FHILIPS F17T8/TL841 17 WATT, 120 μmol·m⁻²·s⁻¹) for 6–12 h to promote germination and then transferred to darkness for the indicated days at 20°C. For chemical treatment, eBL (Sigma, E1641), the ethylene biosynthesis precursor 1-aminocyclopropanecarboxylic acid (ACC [Sigma, A3903]), PPZ (Sigma, 45642), Brz (Sigma, SML1406), or picloram (Sigma, P5575) were supplied in MS medium for phenotypic analysis, RT-PCR assays, or fluorescence observation, respectively. Seedlings for RNA extraction and co-immunoprecipitation (Co-IP) assays were submerged in MS liquid medium containing mock treatment or eBL.

Plasmid construction and generation of transgenic lines

To construct Enh35S:ProSAUR17:SAUR17-GFP/pifq and Enh35S:ProSAUR17:SAUR17-GFP/ein3 eil1 transgenic plants, four repeats of CaMV 35S enhancers from pEN-L4-4Enh-R1 (Tao et al., 2013) were amplified and cloned into the pJim19 (gentamicin^r) vector. Then, the SAUR17 genomic sequence fused with GFP was amplified from pJim19-SAUR17g-GFP (Basta) (Wang et al., 2020) and inserted into the pJim19-Enh35S vector using a Gibson assembly cloning kit. To generate Pro35S:BZR1-GFP in the wild-type (Col-0), ein3-1 eil1-1, and pifq backgrounds, GFP was first amplified from pJim19-GFP (Basta) with the primers GFP_SpeI F and GFP_SacI R and cloned into pJim19 (Gentamicin) using Spe I and Sac I to create pJim19 (gentamicin^r)-GFP. Then, the full-length coding sequence of BZR1 was cloned from wild-type complementary DNA (cDNA) with the primers BZR1 CDS_KpnI F and BZR1 CDS without SC_SpeI R and then inserted into pJim19 (gentamicin^r)-GFP to generate pJim19 (gentamicin^r)-BZR1-GFP. To generate Pro35S:PIF4-4×Myc and Pro35S:PIF5-4×Myc transgenic plants, the sequences of 4×Myc were synthesized and ligated into pJim19 (Basta) using a Gibson assembly cloning kit. PIF4 and PIF5 were cloned from wild-type cDNA using the primer pairs PIF4 CDS F/PIF4 CDS R and PIF5 CDS F/PIF5 CDS R and inserted into pJim19 (Basta)-4×Myc. To generate bzr1 bes1 beh1 beh2 beh3 beh4 (+/−) (bzr^quintuple beh4 [+/−]) mutants, the CRISPR/Cas9 gene editing system was used. Two sgRNAs were designed to target BEH1, BEH2, BEH3, and BEH4. The vector was transformed into the bzr1 bes1 background (Wang and Chen, 2020). The tandem method for the construction of multiple sgRNAs was as described previously (Wang and Chen, 2020). All the constructs were transformed into Arabidopsis by the floral dip method using Agrobacterium tumefaciens GV3101 (Lab stock).

Hook curvature and cotyledon opening angle measurements

Images of individual seedlings were acquired using a Leica stereoscope (M205FA), and both hook angles and cotyledon opening angles were measured by LAS 4.8 software as described previously (Wang et al., 2020).

RNA extraction and RT-qPCR

Total RNA was extracted from dark-grown whole seedlings or etiolated apical organs (cotyledons and hook) using a TaKaRa MiniBEST Plant RNA Extraction Kit (Takara, Cat# 9769). One microgram of total RNA was used to perform the reverse transcription assay. First-strand cDNA was synthesized using Hifair III 1st strand cDNA synthesis supermix for qPCR (YEASEN, Cat# 11141ES60). RT-qPCR was performed on a 7500 Fast Real-Time PCR System (Applied Biosystems) using TB Green Premix Ex Taq (Takara, Cat# RR420A). The RT-qPCR data were analyzed and are shown as the expression level relative to that of PP2A. Double normalization was performed: the expression level was first normalized to the PP2A control level and the relative expression level was then normalized to that in the first sample, which was set to a value of 1. The RT-qPCR primers are listed in Supplemental Table S1.

ChIP assay

A ChIP assay was performed as described previously (Sun et al., 2016). Briefly, 3-day-old dark-grown seedlings were crosslinked in 1% formaldehyde solution for 20 min under vacuum. The crosslinked seedlings were ground into a powder in liquid nitrogen and then the nuclei were extracted using sucrose gradient density centrifugation and resuspended in Nuclei Lysis Buffer (50 mM Tris–HCl pH = 8.0, 10 mM EDTA, 1% SDS, 0.1 mM PMSF, 1 mM phenylmethylsulfonyl fluoride [Amresco, 0754-5G], and 1× complete protease inhibitor cocktail [Roche, 04693159001]). After sonication, the supernatant containing chromatin was precleared with protein G beads (Invitrogen) and incubated at 4°C for 1 h. Then, the sample was divided into two equal parts. An anti-Myc antibody (Sigma, C3956, 1:100 dilution) or anti-GFP antibody (Invitrogen, G10362, 1:100 dilution) was added to one of two parallel samples, and the other sample received no antibody as a control. The antibody and sample were incubated at 4°C for 12 h. After that, protein G beads were added to the incubated mixture to conjugate with antibodies for 1 h. Then, the immune complex was washed, de-crosslinked, and subjected to DNA extraction. The precipitated DNA and input DNA were quantified using RT-qPCR, and the primers used for ChIP-qPCR are listed in Supplemental Table S1. 18S rDNA and PP2A were used as nonbinding controls.

Co-IP assay

All experimental procedures except for immunoblotting were carried out under dim green light in a dark room. Three-day-old dark-grown seedlings were submerged into liquid MS medium supplemented with 1 µM eBL or equimolar ethyl alcohol for 6 h. After treatment, the materials were harvested and ground to a powder in liquid nitrogen. The proteins were extracted in immunoprecipitation (IP) buffer (50 mM Tris–HCl pH 7.5, 150 mM NaCl, 10 mM MgCl₂, 10% glycerol, 0.1% NP40, 1 mM phenylmethylsulfonyl fluoride, 1× complete protease inhibitor cocktail, and 50 µM MG132 [Millipore, 474790]). IP was performed using GFP-Trap beads (ChromoTek, gta-20) or GFP mAb-Agarose beads (MBL, D153-8) for 2 h at 4°C. After washing five times with IP buffer, proteins were eluted from the beads by boiling with 2× SDS loading buffer. For immunoblotting, anti-GFP (Abmart, M20004, 1:1,000 dilution), anti-PIF3 (Lab stock, 1:1,000 dilution), anti-RPN6 (Lab stock, 1:5,000 dilution), anti-Flag (Sigma–Aldrich, F3165-2MG, 1:1,000 dilution), anti-BES1 (Abclonal, A9794, 1:1,000 dilution), and anti-Actin (EASYBIO, BE0027-100, 1:5,000 dilution) antibodies were used.

Protein extraction and immunoblotting

Similar to the Co-IP assay, protein extraction was performed under dim green light in a dark room. Proteins were extracted in 8 M urea, 100 mM NaH₂PO4, 1 mM PMSF, 1× protease inhibitor (Roche), and 100 mM Tris–HCl (pH 8.0). The extracts were centrifuged at 13,000 g at 4°C for 10 min and the supernatant was transferred into a new tube containing SDS loading buffer. The protein samples were separated by electrophoresis in 8% (w/v) SDS-polyacrylamide gels and transferred to a polyvinylidene fluoride film (Roche, 03010040001). The anti-GFP (Abmart, M20004, 1:1,000 dilution), anti-PIF3 (Lab stock, 1:1,000 dilution), anti-RPN6 (Lab stock, 1:5,000 dilution), anti-Flag (Sigma–Aldrich, F3165-2MG, 1:1,000 dilution), anti-Myc (Sigma–Aldrich, C3956-2MG, 1:2,000 dilution), anti-Actin (EASYBIO, BE0027-100, 1:5,000 dilution), and anti-HSP (Beijing Protein Innovation, AbM51099-31-PU, 1:5,000 dilution) antibodies were used as primary antibodies. Signals were detected using Amersham ECL Western Blotting Detection Reagents (GE Healthcare, RPN2106).

In vitro pulldown assay

His-TrxA-BZR1, His-BIN2, and MBP-PIF3 were transformed into the Escherichia coli Rosetta (DE3) strain, which was induced with 0.5 mM isopropyl-β-d-thiogalactopyranoside and incubated at 16°C for 18 h. After protein purification, His-TrxA-BZR1 proteins were first phosphorylated by His-BIN2 and then incubated with MBP-PIF3 for the pull-down assay. Two micrograms of His-TrxA-BZR1 was premixed with or without 0.4 µg of His-BIN2 into 50 µL of kinase buffer (20 mM Tris–HCl [pH 7.5], 100 mM NaCl, 12 mM MgCl₂, and 0.1 mM ATP) and the mixture was kept at 37°C for 1 h. Then, 2 μg of MBP or MBP-PIF3 protein was added to the reaction system, and the mixture was transferred to 1 mL of binding buffer (20 mM Tris–HCl [pH 7.5], 150 mM NaCl, and 0.1% Nonidet P-40) and incubated for 2 h at 4°C. After that, 20 μL of amylose resin beads were prewashed three times with binding buffer and used to pull down MBP and MBP-PIF3 for 1 h at 4°C. After incubation, the amylose resin beads were collected, washed five times with binding buffer, and eluted with 2× SDS loading buffer at 100°C for 10 min. The eluted protein samples were analyzed by immunoblotting using anti-His (Sigma–Aldrich, H1029-.2ML, 1:1,000 dilution) and anti-MBP (New England Biolabs, E8031S, 1:4,000 dilution) antibodies.

Primer sequences

Sequences of primers used in this study are listed in Supplemental Table S1.

Statistical analysis

Statistical data are provided in Supplemental Data Set S1.

Accession numbers

Sequence data from this article can be found in the EMBL/GenBank data libraries under the following accession numbers: SAUR17 (AT4G09530); SAUR15 (AT4G38850); PIF1 (AT2G20180); PIF3 (AT1G09530); PIF4 (AT2G43010); PIF5 (AT3G59060); HY5 (AT5G11260); COP1 (AT2G32950); DET1 (AT4G10180); EIN3 (AT3G20770); EIL1 (AT2G27050); EBF1 (AT2G25490); EBF2 (AT5G25350); ETO1 (AT3G51770); ETR1 (AT1G66340); HLS1 (AT4G37580); BZR1 (AT1G75080); BES1 (AT1G19350); BEH1 (AT3G50750); BEH2 (AT4G36780); BEH3 (AT4G18890); BEH4 (AT1G78700); DWF4 (AT3G50660); CPD (AT5G05690); DET2 (AT2G38050); BR6OX2 (AT3G30180); BRI1 (AT4G39400); BIN2 (AT4G18710); and PP2A (AT1G69960).

Supplemental data

The following materials are available in the online version of this article.

Supplemental Figure S1. SAUR17 expression is affected by COP1 and DET1 but not by HY5, and it acts downstream of PIFs and EIN3/EIL1.

Supplemental Figure S2. Auxin does not regulate the level or location of SAUR17 expression.

Supplemental Figure S3. BR mutants display abnormal photomorphogenic phenotypes with similarities to light signaling mutants.

Supplemental Figure S4. BR up-regulates SAUR17 expression.

Supplemental Figure S5. The mutation sites of bzr1 bes1 beh1 beh2 beh3 beh4 (+/−) mutants.

Supplemental Figure S6. Involvement of PIFs and EIN3/EIL1 in BR induction of SAUR17 expression.

Supplemental Figure S7. BRs increased EIN3 and PIF3 protein levels, while did not affect EBF1 and EBF2 protein levels.

Supplemental Table S1. Primers used in this research.

Supplemental Data Set S1. Statistical analysis tables.

 

J.W., H.C., X.W.D., and N.W. conceived the research. J.W., N.S., N.W., and H.C. designed the experiments. L.Z. and J.W. generated the bzr^quintuple beh4 (+/−) mutant and performed the in vitro pulldown assay. F.Z. and J.W. performed the BZR1-YFP Co-IP assay. M.X. tested the transcript levels of PIFs in PIF-overexpressing lines. J.W. performed all other experiments. J.W., H.C., and N.W. analyzed the data. J.W., X.W.D., and N.W. wrote the manuscript. J.W., X.W.D., H.C., and N.W. edited the article.

The author responsible for distribution of materials integral to the findings presented in this article in accordance with the policy described in the Instructions for Authors (https://academic.oup.com/plcell) is: Ning Wei (weining@swu.edu.cn).