1. Plant Biology
Download icon

Shade-induced nuclear localization of PIF7 is regulated by phosphorylation and 14-3-3 proteins in Arabidopsis

  1. Xu Huang
  2. Qian Zhang
  3. Yupei Jiang
  4. Chuanwei Yang
  5. Qianyue Wang
  6. Lin Li  Is a corresponding author
  1. Fudan University, China
Research Article
  • Cited 1
  • Views 2,378
  • Annotations
Cite this article as: eLife 2018;7:e31636 doi: 10.7554/eLife.31636

Abstract

Shade avoidance syndrome enables shaded plants to grow and compete effectively against their neighbors. In Arabidopsis, the shade-induced de-phosphorylation of the transcription factor PIF7 (PHYTOCHROME-INTERACTING FACTOR 7) is the key event linking light perception to stem elongation. However, the mechanism through which phosphorylation regulates the activity of PIF7 is unclear. Here, we show that shade light induces the de-phosphorylation and nuclear accumulation of PIF7. Phosphorylation-resistant site mutations in PIF7 result in increased nuclear localization and shade-induced gene expression, and consequently augment hypocotyl elongation. PIF7 interacts with 14-3-3 proteins. Blocking the interaction between PIF7 and 14-3-3 proteins or reducing the expression of 14-3-3 proteins accelerates shade-induced nuclear localization and de-phosphorylation of PIF7, and enhances the shade phenotype. By contrast, the 14-3-3 overexpressing line displays an attenuated shade phenotype. These studies demonstrate a phosphorylation-dependent translocation of PIF7 when plants are in shade and a novel mechanism involving 14-3-3 proteins, mediated by the retention of PIF7 in the cytoplasm that suppresses the shade response.

https://doi.org/10.7554/eLife.31636.001

Introduction

Because chlorophyll preferentially absorbs light in the red and blue ranges but not in the far-red range of the light spectrum, a perceived decrease in the ratio of red/far-red (R/FR) radiation, and thus in photosynthetically active radiation (PAR) of between 400 and 700 nm, provides a signal that shading by other plants is imminent. Shade-intolerant plants, such as Arabidopsis thaliana, sense this reduction and initiate the shade avoidance syndrome (SAS). In the SAS, energy resources are reallocated from storage organs to stem-like organs, including hypocotyls and petioles, thereby enabling plants to initiate a rapid growth response (Cole et al., 2011; Casal, 2012). Prolonged shade exposure leads to reduced branching, early flowering and seed set, and reduced yield (Ballaré, 1999; Franklin and Whitelam, 2005; Procko et al., 2014).

The R/FR-absorbing photoreceptor phytochrome B (phyB) plays the most dominant role during the SAS (Reed et al., 1993). In an open environment under sunlight (where the ratio of R/FR is about 1.2–1.5), most of the phyB is in the far-red-absorbing (Pfr) active form and moves to the nucleus, where it interacts with basic helix-loop-helix (bHLH) proteins known as PHYTOCHROME-INTERACTING FACTORS (PIFs) (Duek and Fankhauser, 2005; Leivar and Quail, 2011). The photoactivation of phyB induces the rapid phosphorylation of PIF1/3/4/5 prior to their degradation (Shen et al., 2005; Al-Sady et al., 2006; Shen et al., 2007), although the short half-lives of these PIFs impedes the tracing of the phosphorylated forms. When plants are in the shade, phyB is mostly in the inactive red-absorbing (Pr) cytosolic form, which facilitates the accumulation of PIFs and restricts the growth of the hypocotyl (Lorrain et al., 2008; Leivar and Quail, 2011).

PIF7 is a major regulator of the shade response, as shown by the severe shade-defective phenotype of pif7 mutants (Li et al., 2012; de Wit et al., 2015; Mizuno et al., 2015). PIF7 is less vulnerable than PIF1/3/4/5 to the rapid turnover of induced by the Pfr form of phyB (Leivar et al., 2008). Instead, the activity of PIF7 is controlled by rapid de-phosphorylation in response to shade, which leads to its binding to G-boxes in the promoters of auxin biosynthesis genes, causing an increase in auxin levels and a rapid growth response (Li et al., 2012).

The 14-3-3 proteins are highly conserved in all eukaryotes. Research in recent years has revealed several putative 14-3-3 targets in plants (Jaspert et al., 2011; Wang et al., 2011; Yoon and Kieber, 2013; Zhou et al., 2014). These studies have revealed that 14-3-3 proteins can interact with the phosphorylated forms of their client proteins in response to certain signals, and that this binding finalizes the signaling event by enabling a change in the subcellular localization, protein stability or intrinsic enzymatic activity of the client, or by promoting an association between the client and other proteins. The cellular 14-3-3 'pool' enables these proteins to react to altered signaling cues in an immediate and precise way through dynamic interactions with their clients.

Here, we demonstrate a shade induction of the nuclear localization of dephosphorylated PIF7 and a role for the 14-3-3 proteins in the cytoplasmic retention of PIF7 in Arabidopsis. Our work reveals a novel mechanism that rapidly switches PIF7 function in response to light conditions and the role of 14-3-3 proteins in SAS.

Results

Shade induces the rapid nuclear localization of PIF7

To monitor the cellular localization of PIF7, we generated 35S::GFP-PIF7 transgenic plants and analyzed the GFP signal in white-light-grown seedlings before and after shade treatment. Impressively, GFP-PIF7 rapidly accumulated in the nucleus when plants were placed in the shade, as observed in the cotyledon and the hypocotyl of transgenic lines. The extent of this shade response decreased gradually from the top to the bottom of the hypocotyls (Figure 1—figure supplement 1a). At the top of hypocotyls of two independent transgenic lines, the nuclear/cytoplasmic ratio of GFP-PIF7 increased within 5 min of moving the plants into shade and continued to increase for 45 min (Figure 1a,b). The localization of GFP, which was used as the control, was not affected by shade (Figure 1a,b; Figure 1—figure supplement 1a).

Figure 1 with 1 supplement see all
Shade induces the nuclear localization of PIF7.

(a) Subcellular localization of GFP-PIF7 at the top of the hypocotyls of two independent transgenic seedlings grown under white light at different time points after transfer to shade. Transgenic Arabidopsis expressing GFP-PIF7 or GFP was grown on 1/2 MS medium under white light for 5 days. Seedlings were treated with shade for 5, 15, or 25 min, and images of the GFP signal were obtained using confocal microscopy. White scale bar represents 25 μm. (b) Kinetics of the shade-induced nuclear accumulation of GFP-PIF7. GFP-PIF7 or GFP seedlings were treated as in (a). ImageJ was used to quantify the fluorescence intensities. Ratios of the nuclear and cytoplasmic signal intensities were calculated from 10 cells for each treatment. Error bars represent standard deviations. (c) Shade induces the nuclear localization of dephosphorylated PIF7. Immunoblot of the PIF7-Flash proteins using anti-Myc antibody in the total, nuclear and non-nuclear fractions from white-light- and shade-treated seedlings. Histone H3 is a nuclear marker, and the RuBisCO large subunit (RbcL), a chloroplast protein, is a non-nuclear fraction marker.

https://doi.org/10.7554/eLife.31636.002

Subcellular fractionation experiments using whole seedlings of 35S::PIF7-Flash (9xMyc-6xHis-3xFLAG) transgenic lines further demonstrated that PIF7-Flash was enriched in the nuclear fraction under shade conditions (Figure 1c, Figure 1—figure supplement 1b). Shade treatment resulted in an increase of PIF7 in the nucleus and a decrease of PIF7 in the non-nuclear fraction, indicating that the increased nuclear fraction of PIF7 was probably translocated from the cytoplasmic compartment.

PIF7 interacts with 14-3-3 proteins

To identify potential PIF7-binding proteins, we conducted a yeast two-hybrid (Y2H) screen using PIF7 as bait. Interestingly, this study identified the phosphopeptide-binding protein 14-3-3 κ as a binding partner of PIF7. As we have shown, 14-3-3 λ and 14-3-3 κ can interact with PIF7 when co-expressed in the yeast system (Figure 2a). A bimolecular fluorescence complementation (BiFC) assay in Nicotiana benthamiana cells also supported the interaction between PIF7 and 14-3-3 λ/κ (Figure 2b). In fact, there are at least six 14-3-3 proteins (14-3-3 λ, κ, χ, γ, μ and ε) that can interact with PIF7 in Y2H and BiFC assays (Figure 2a,b; Figure 2—figure supplement 1).

Figure 2 with 1 supplement see all
PIF7 interacts with 14-3-3 proteins.

(a) PIF7 interacts with 14-3-3 λ and 14-3-3 κ in a yeast two-hybrid assay. Each yeast clone containing pGADT7 (AD) or pGADT7-PIF7 (AD-PIF7), together with pGBKT7 (BD), pGBKT7-14-3-3 λ (BD-14-3-3 λ) or pGBKT7-14-3-3 κ (BD-14-3-3 κ), was grown on transformation selection (SD-L-T) or interaction selection (SD-L-T-H+3AT) plates. Dilution of the inoculation is shown at the top of the picture. Yeast growth on SD-L-T-H+3AT indicates a positive protein–protein interaction. (b) Interaction between PIF7 and 14-3-3 λ or 14-3-3 κ as detected by BiFC. The C-terminal half of yellow fluorescent protein (YFP) was fused to 14-3-3 λ or 14-3-3 κ and the N-terminal half of YFP was fused to PIF7. The constructs were co-transformed into tobacco leaf cells, and fluorescence images were obtained by confocal microscopy. White scale bar represents 75 μm. (c) Interaction between PIF7 and 14-3-3 λ or 14-3-3 κ as detected by semi-in vivo pull-down assay. 14-3-3 λ and 14-3-3 κ fused to GST were expressed and purified from Escherichia coli. Protein extracts from plants that overexpressed PIF7-Flash, grown under white light conditions or after 1 hr of shade, were used for the pull-down assay. Immunoblots of the PIF7-Flash proteins used anti-Myc antibody. CBB: Coomassie Brilliant Blue stain. (d) Interaction between PIF7 and 14-3-3s as detected by co-immunoprecipitation. Anti-FLAG M2 agarose beads were used to precipitate PIF7-Flash from PIF7 overexpression plants grown under white light or after 1 hr of shade. Western blots using anti-Myc and anti-14-3-3s antibodies were performed as indicated in the 'Materials and methods'.

https://doi.org/10.7554/eLife.31636.005

The 14-3-3 proteins are well known to bind phosphopeptides. When GST (glutathione S-transferase)−14-3-3 fusion proteins were used to pull down the protein lysate from the 35S::PIF7-Flash transgenic line grown under white-light and shade conditions, more PIF7 protein was enriched in the white-light-grown seedlings (Figure 2c). Moreover, a co-immunoprecipitation experiment further confirmed that 14-3-3 proteins are precipitated with PIF7 from white-light-grown transgenic seedlings (Figure 2d), probably because more PIF7 is phosphorylated in white-light-grown seedlings.

Phosphorylation sites of PIF7 mediate its binding to 14-3-3 proteins

There are two types of specific 14-3-3 binding motifs in mammalian and plant systems, mode I, R/KXXpSX, and mode II, R/KXXXpSXP (where X = any amino acid, R = arginine, K = lysine, pS = phosphoserine and p=proline) (Muslin et al., 1996; Muslin and Xing, 2000; Schoonheim et al., 2007). Although an obvious interaction occurred between PIF7 and the 14-3-3 proteins, no typical mode I or mode II motifs can be identified in PIF7. We reasoned that non-canonical motifs, such as RXXS, might mediate the interaction between PIF7 and 14-3-3 proteins, as observed in PHOT1 (Figure 3—figure supplement 1a) (Kinoshita et al., 2003).

In order to identify the phosphorylation sites of PIF7, we immunoprecipitated the PIF7 complex in 35S::PIF7-Flash transgenic plants and then performed a liquid chromatography-tandem mass spectrometry (LC-MS/MS) experiment. The results showed that PIF7 was phosphorylated at S139 and S141 in seedlings grown under while light and in seedlings treated with 5 min of shade, but not in seedlings treated with 1 hr of shade (Figure 3—figure supplement 1b). These phosphorylation sites constituted the putative 14-3-3 binding sequence (RSGSET). Because the LC-MS/MS experiment did not cover the entire protein, we also used the online software NetPhos 2.0 to predict the potential phosphorylation sites of PIF7, and the results showed the following high-score sites: S78, S80, S125, S139 and S141 (Figure 3—figure supplement 1c). On the basis of the 14-3-3-binding sequence in CDC25C (KTVSLC) (Chan et al., 2011), the sequence KDGSCS (75–80) of PIF7 may be another possible 14-3-3 binding motif (Figure 3—figure supplement 1a).

To determine whether the amino acids 138–143 (RSGSET) and 75–80 (KDGSCS) of PIF7 could mediate the binding of PIF7 to the 14-3-3 proteins, we mutated both S139 and S141 to alanine (to mimic the unphosphorylated state of these residues, resulting in PIF7[2A]: S139A S141A), deleted amino acids 138–141 (PIF7△), or mutated all five serine residues at amino acids 78, 80, 125, 139 and 141 (PIF7[5A]: S78A S80A S125A S139A S141A) of PIF7 to examine its interaction with 14-3-3 proteins. We also mutated S139 and S141 to aspartic acid residues to mimic phosphorylation (PIF7[2D]: S139D S141D) (PIF7[5D]: S78D S80D S125D S139D S141D) (Figure 3—figure supplement 1a).

In yeast, PIF7(2A) showed decreased interaction with 14-3-3λ/κ proteins and PIF7(2D) displayed a level of interaction that was similar to that of wildtype PIF7 (Figure 3a). In tobacco leaves, the weak BiFC signals from PIF7(2A) and PIF7△ mainly occurred in the nucleus (Figure 3b; Figure 3—figure supplement 2), which is probably due to the nuclear localization of PIF7(2A) and PIF7△ or to an attenuated interaction with 14-3-3 λ/κ proteins. When all five serine residues were mutated (PIF7[5A]), fluorescence was largely absent (Figure 3b, Figure 3—figure supplement 2), indicating that these serine residues are critical for binding to 14-3-3 proteins. However, no signal was observed from PIF7(2D) or PIF7(5D) in a BiFC assay (Figure 3—figure supplement 3a). Furthermore, PIF7(2D) extracted from a white-light-grown PIF7(2D)-Flash transgenic line was not able to pull down GST-14-3-3s (Figure 3—figure supplement 3c), probably because the interaction of the 14-3-3 protein with phosphorylated PIF7 cannot be mimicked by S-to-D substitution in some systems, as has also been reported for other clients of 14-3-3 proteins (de Chiara et al., 2009; Menon et al., 2012). In a semi-in vivo pull-down assay, PIF7(2A)-Flash from the white-light-grown 35S::PIF7(2A)-Flash transgenic line showed reduced levels of binding with GST-14-3-3 fusion proteins (Figure 3—figure supplement 3b). Finally, in an in vivo Co-IP assay, more 14-3-3 proteins were co-immunoprecipitated with PIF7(2D) than with wild type PIF7 or PIF7(2A) (Figure 3c), suggesting that the phosphorylation sites of PIF7 mediated its binding to 14-3-3 proteins.

Figure 3 with 3 supplements see all
Interactions between PIF7 derivatives and 14-3-3 proteins.

(a) Interactions between PIF7, PIF7(2A) or PIF7(2D) and 14-3-3 λ or 14-3-3 κ in a yeast two-hybrid assay. Each yeast clone containing the pGADT7 (AD), or AD-PIF7 (2A), or AD-PIF7 or AD-PIF7 (2D) together with pGBKT7 (BD) or BD-14-3-3 λ or BD-14-3-3 κ was grown on transformation selection (SD-L-T) or interaction selection (SD-L-T-H+3AT) plates. Dilution of the inoculation is shown at the top of the picture. Yeast growth on SD-L-T-H+3AT indicates a positive protein–protein interaction. (b) Interaction between PIF7 derivatives and 14-3-3 λ or 14-3-3 κ detected by BiFC. nYFP-PIF7, nYFP-PIF7(2A), nYFP-PIF7△ or nYFP-PIF7(5A) and 14-3-3 λ-cYFP or 14-3-3 κ-cYFP constructs were co-transformed into tobacco leaf cells. YFP fluorescence images were obtained using a confocal microscope. White scale bar represents 75 μm. (c) Interaction between PIF7, PIF7(2A) or PIF7(2D) and 14-3-3s as detected by co-immunoprecipitation. Anti-FLAG M2 agarose beads were used to precipitate PIF7-Flash, PIF7(2A)-Flash or PIF7(2D)-Flash from overexpression plants grown under white light. Western blots were performed using anti-Myc and anti-14-3-3s antibodies as indicated in the 'Materials and methods'.

https://doi.org/10.7554/eLife.31636.007

Phosphorylation sites of PIF7 are important for its localization and function

To investigate the roles of the phosphorylation sites on the cellular localization of PIF7, we generated 35S::GFP-PIF7 derivatives in which the serine residues were mutated to alanine residues (PIF7[2A] and PIF7[5A] or to aspartic acid residues (PIF7[2D] and PIF7[5D]). When over-expressed in tobacco leaf cells, the wildtype GFP-PIF7 localized in both the cytoplasm and the nucleus. The GFP-PIF7(2A), GFP-PIF7(5A) and GFP-PIF7△ mutants exhibited strong nuclear signals, whereas the GFP-PIF7(2D) and GFP-PIF7(5D) mutants showed stronger cytoplasmic signals than did GFP-PIF7 (Figure 4a; Figure 4—figure supplement 1). To further examine the effect of the phosphorylation sites on PIF7 localization in Arabidopsis, we generated transgenic plants expressing GFP-PIF7△ and GFP-PIF7(5A). Consistent with the findings in tobacco, GFP-PIF7△ and GFP-PIF7(5A) displayed stronger nuclear signals than did wildtype GFP-PIF7 (Figure 4b). The shade treatment caused substantial translocation of the wildtype GFP-PIF7, whereas its effects on GFP-PIF7△ and GFP-PIF7(5A) were minimal (Figure 4b).

Figure 4 with 1 supplement see all
Phosphorylation sites of PIF7 are important for the localization and function of this protein.

(a) Subcellular localization of GFP-PIF7, GFP-PIF7(2A), GFP-PIF7 (2D), GFP-PIF7△, GFP-PIF7(5A) and GFP-PIF7 (5D) in tobacco cells. White scale bar represents 25 μm. (b) Subcellular localization of the GFP-PIF7, GFP-PIF7△ and GFP-PIF7(5A) proteins at the top of the hypocotyl in transgenic Arabidopsis plants treated with white light or after 1 hr in shade. White scale bar represents 25 μm. (c) Expression of PIF7 in white-light-grown 35S::PIF7(2A)-Flash, 35S::PIF7-Flash and 35S::PIF7(2D)-Flash transgenic lines as determined using anti-Myc antibody. RbcL is used as the loading control. (d) Subcellular fractionation experiments using 35S::PIF7(2A)-Flash transgenic lines. Immunoblot of the PIF7(2A)-Flash proteins using anti-Myc antibody in the total, nuclear and non-nuclear fractions from white-light- and shade-treated transgenic seedlings. Histone H3 is a nuclear maker and RbcL is a non-nuclear fraction marker. (e) Subcellular fractionation experiments using 35S::PIF7(2D)-Flash transgenic lines. Immunoblot of the PIF7(2D)-Flash proteins using anti-Myc antibody in the total, nuclear and non-nuclear fractions from white-light- and shade-treated transgenic seedlings. Histone H3 is a nuclear marker and RbcL is a non-nuclear fraction marker. (f) Expression levels of IAA19 and YUCCA8 in Col-0, pif7-1 and transgenic lines harboring PIF7-Flash, PIF7(2A)-Flash and PIF7(2D)-Flash in the pif7-1 background. Mean ± SE from three independent biological replicates, after normalization to the internal control AT2G39960, are shown. Bars marked with different letters denote significant differences (p<0.05) in the mean expression levels. (g) Quantification of the hypocotyl lengths of transgenic lines harboring PIF7-Flash, PIF7(2A)-Flash and PIF7(2D)-Flash in the pif7-1 background. Seedlings were grown under white light for 4 days and maintained in white light or transferred to shade for the next 5 days, before hypocotyl lengths were measured. More than 20 seedlings were measured for each line. Bars marked with different letters denote significant differences (p<0.05) in the mean hypocotyl lengths.

https://doi.org/10.7554/eLife.31636.011

To determine the effect of the phosphorylation sites of PIF7 on the shade response in vivo, PIF7 with phosphorylation site mutations (35S::PIF7[2A]- Flash, 35S::PIF7[2D]-Flash) and wildtype (35S::PIF7-Flash) were expressed in the pif7-1 background, and the number of complementary transgenic lines was recorded on the basis of hypocotyl length under shade. A substantial proportion of the first generation (T1) of transgenic plants transformed with PIF7 (13/118: 11%) and PIF7(2A) (47/212: 22%) showed normal or enhanced shade responses. By contrast, all of the transgenic plants transformed with 35S:: PIF7(2D)-Flash (83 T1 lines) showed a shade-defective phenotype similar to that of pif7-1. To further confirm the phenotype, we obtained the third-generations of the 35S::PIF7(2A)-Flash, 35S::PIF7(2D)-Flash and 35S::PIF7-Flash transgenic lines and confirmed that their expression levels were similar to those of the equivalent T1 lines. As we have shown, the phosphorylated form of PIF7 was decreased in the PIF7(2A) transgenic lines (Figure 4c). In a subcellular fractionation experiment, PIF7(2A) was present in the nuclear fraction, but almost undetectable in the cytosol under both white light and shade (Figure 4d). By contrast, PIF7(2D) was mainly present in the cytoplasmic fraction (Figure 4e). Furthermore, when grown under white light, the PIF7 and PIF7(2A) transgenic plants showed greatly increased expression of the PIF7 downstream genes IAA19 and YUCCA8 (Figure 4f), resulting in longer hypocotyls (Figure 4g). By contrast, overexpression of PIF7(2D) in the pif7-1 background resulted in reduced shade-induction of both gene expression (Figure 4f) and hypocotyl elongation (Figure 4g). Notably, the effects of PIF7(2A) overexpression were greater than those of PIF7 overexpression, indicating the nuclear localization is critical for the function of PIF7. It is also noteworthy that the shade-induced effects on gene expression and hypocotyl elongation were not totally abolished in the PIF7(2A) transgenic lines (Figure 4f,g), implying that other phosphorylation sites or other mechanisms that regulate PIF7 may exist. For example, shade may regulate additional essential factors that interact with PIF7.

14-3-3 proteins regulate the localization and dephosphorylation of PIF7

As PIF7 translocates to the nucleus under shade conditions and is able to bind 14-3-3 proteins, we hypothesized that 14-3-3 proteins sequester phosphorylated PIF7 in the cytoplasm.

Previous studies have shown that the interactions of 14-3-3 proteins with their client proteins can be disrupted by the R18 peptide (Wang et al., 1999). We therefore took advantage of this peptide to discover that the shade-induced nuclear localization of GFP-PIF7 (Figure 5a,b) and de-phosphorylation of PIF7-Flash (Figure 5c; Figure 5—figure supplement 1) were accelerated after treatment with R18, but not after treatment with R18(Lys) (a non-functional control) (Figure 5—figure supplement 2). In addition, we crossed 35S::GFP-PIF7 and 35S::PIF7-Flash transgenic lines with the double mutant 14-3-3 λκ, in which the expression of 14-3-3s was reduced (Zhou et al., 2014). When compared with wildtype background, more GFP-PIF7 accumulated in the nucleus of the double mutant after 2 min of shade treatment (Figure 5d,e) and the phosphorylated PIF7 disappeared faster in the double mutant (Figure 5f; Figure 5—figure supplement 2), suggesting that 14-3-3 proteins mediate the cytoplasmic retention of phosphorylated PIF7 during the transition from white light to shade.

Figure 5 with 2 supplements see all
14-3-3 proteins delay shade-induced nuclear translocation and dephosphorylation of PIF7.

(a) The effect of R18 on the shade-induced nuclear localization of GFP-PIF7. GFP-PIF7 #14 transgenic plants grown under white light were treated with R18 or R18(Lys) for 3 hr after 10 min of vacuum, and were transferred to shade for indicated periods. White scale bar represents 25 μM. (b) Quantification of the shade-induced nuclear accumulation of GFP-PIF7. ImageJ was used to quantify the fluorescence intensities. Ratios of the nuclear and cytoplasmic signal intensities were calculated from 10 cells for each treatment. Error bars represent standard deviations. Significant differences between two treatments are shown as asterisks. *p<0.05 by Student’s t-test. (c) The effect of R18 on the shade-induced dephosphorylation of PIF7-Flash. Five-day-old 35S::PIF7-Flash transgenic seedlings were treated with R18 or R18(Lys) for 3 hr after 10 min of vacuum under white light, and transferred to shade for the indicated periods. Data for three biological replicates are presented. The level of PIF7-Flash was detected using anti-Myc antibody. RbcL was used as the loading control. (d) Shade-induced nuclear localization of GFP-PIF7 in 14-3-3 λκ (Salk-075219CxSalk-071097). A GFP-PIF7 #14 transgenic plant was crossed with 14-3-3 λκ. Five-day-old white-light-grown GFP-PIF7/14-3-3 λκ and GFP-PIF7/Col-0 seedlings were transferred to shade for the indicated periods. White scale bar represents 25 μM. (e) Quantification of the shade-induced nuclear accumulation of GFP-PIF7. ImageJ was used to quantify the fluorescence intensities. Ratios of the nuclear and cytoplasmic signal intensities were calculated from 10 cells for each treatment. Error bars represent standard deviations. Significant differences between14-3-3 λκ and Col-0 background are shown as asterisks. *p<0.05 by Student’s t-test. (f) Shade-induced dephosphorylation of PIF7-Flash in 14-3-3 λκ. A 35S::PIF7-Flash transgenic plant was crossed with 14-3-3 λκ. Five-day-old white-light-grown PIF7-Flash/14-3-3 λκ and PIF7-Flash/Col-0 seedlings were transferred to shade for the indicated periods. Three biological replicates were presented. The level of PIF7-Flash was detected using anti-Myc antibody. RbcL was used as the loading control.

https://doi.org/10.7554/eLife.31636.013

14-3-3 proteins are negative regulators of the shade response

Consistently, treatment with R18 significantly promoted shade-induced hypocotyl elongation (Figure 6a) and shade-induced gene (IAA19 and YUCCA8) expression (Figure 6b). However, this promotion is blunted in R18-treated pif7-1. We also determined whether a loss of 14-3-3 λ or 14-3-3 κ function would affect shade-induced hypocotyl elongation in Arabidopsis. Single mutants for each gene and the double mutant displayed enhanced shade responses, whereas the 14-3-3 λ transgenic lines (14-3-3 λ OE and 35S::FLAG-HA-14-3-3 λ) showed a reduced shade response (Figure 6c, Figure 6—figure supplement 1), as measured by hypocotyl elongation and levels of expression of IAA19 and YUCCA8 (Figure 6d). Moreover, with R18-treated pif7-1, the hypocotyl length of the double mutant of pif7-1 and 14-3-3 λ−2 was consistently more like that in pif7-1 (Figure 6c), suggesting that the function of 14-3-3 proteins is mediated by PIF7. Overall, the phenotype and gene expression analysis demonstrated that 14-3-3 λ and 14-3-3 κ negatively regulate the shade response through PIF7.

Figure 6 with 1 supplement see all
14-3-3 proteins negatively regulate shade-induced hypocotyl elongation and gene expression.

(a) Quantification of the hypocotyl length of Col-0 seedlings grown in plates containing 200 μg/ml R18 or R18(Lys) peptide under white light or shade conditions. Seedlings were grown under white light for 4 days and maintained in white light or transferred to shade for next 5 days before the measurement of hypocotyl length. More than 20 seedlings were measured. Significant differences between two treatments are shown as asterisks. ***p<0.001 by Student’s t-test. (b) IAA19 and YUCCA8 expression level in the Col-0 and pif7-1 seedlings treated with R18 and R18(Lys) under white light or shade. Seedlings were grown with 1/2 MS medium containing 200 μg/ml R18 or R18(Lys) under white light for 5 days. Then, the seedlings were kept in white light or transferred to shade for 1 hr. Mean ± SE from three independent biological replicates, after normalization to the internal control AT2G39960, are shown. Significant differences between two treatments are indicated by asterisks. ***p<0.001, by Student’s t-test. (c) Quantification of the hypocotyl length of Col-0, pif7-1, 14-3-3 mutants and the overexpression line grown under white light or shade. More than 20 seedlings were measured. Bars marked with different letters denote significant differences (p<0.05) of the means of hypocotyl length. (d) Shade induction of IAA19 and YUCCA8 in Col-0, pif7-1, 14-3-3 mutants and 14-3-3 λ OE. The seedlings were grown under white light for 5 days. Then, the seedlings were kept in white light or transferred to shade for 1 hr. The expression levels were normalized to a reference gene (AT2G39960) and then normalized to the expression under white light condition. The relative shade inductions were shown. Significant differences between mutants and Col-0 are indicated by asterisks. *p<0.05, **p<0.01,***p<0.001 by Student’s t-test.

https://doi.org/10.7554/eLife.31636.018

Discussion

In the current study, we improved the model of shade signal transduction by demonstrating a shade-sensitive subcellular localization of PIF7, which is conferred by interactions with 14-3-3 proteins (Figure 7). Under white-light conditions, phosphorylation of PIF7 results in the cytoplasmic location of this protein and enables its binding to 14-3-3 proteins. When plants sense shade conditions, unknown phosphatases remove the phosphorylation of PIF7. De-phosphorylated PIF7 does not interact with 14-3-3 proteins and translocates to the nucleus, where it promotes the expression of downstream genes, leading to shade-induced phenotypic changes. 14-3-3 proteins retain phosphorylated PIF7 in the cytoplasm to regulate shade-induced hypocotyl elongation negatively.

A molecular model illustrating the role of 14-3-3 proteins in PIF7-mediated SAS.

In response to shade light, de-phosphorylated PIF7 accumulates in the nucleus. 14-3-3 proteins retain phosphorylated PIF7 in the cytoplasm and hence regulate shade-induced hypocotyl elongation negatively.

https://doi.org/10.7554/eLife.31636.020

Current and previous studies have provided several lines of evidence to strongly support the importance of the phosphorylation of PIFs for the transcriptional activity of these proteins. Several phosphorylation sites have been identified in PIF1/3/4 (Bu et al., 2011; Ni et al., 2013; Bernardo-García et al., 2014). These phosphorylation events lead to apparent ubiquitylation and degradation through the ubiquitin-proteasome system (Al-Sady et al., 2006). In sustained light, PIF1/3/4/5 are maintained at a relatively low steady-state level. After subsequent exposure to shade light, new protein synthesis is required for the accumulation of these PIF proteins (Lorrain et al., 2008). PIF7, however, is a light-stable bHLH factor (Leivar et al., 2008; Li et al., 2012).

In the current study, at least two phosphorylation sites (S139 and S141) in PIF7 are found to be critical for its activity and for hypocotyl elongation. These sites also mediate the binding of PIF7 with 14-3-3 λ and κ. Shade-induced nuclear accumulation of PIF7 and the constitutive nuclear localization of PIF7(2A) mutants in transgenic plants suggest that the interaction of 14-3-3 proteins and PIF7 is involved in determining the subcellular distribution of PIF7. Although there was a lack of interaction in the BiFC and pull-down assays, S-to-D substitution does enhance the interaction between 14-3-3 proteins and PIF7 in vivo. Moreover, PIF7(2D)-Flash was mainly localized in cytoplasm and was unable to complement the shade-defective gene expression and phenotype of pif7-1, which functionally reflects the phosphorylation defects in the in vivo system. The phosphorylation state of PIF7 determines its localization and function, and also affects its ability to bind to 14-3-3 proteins.

14-3-3 proteins have been reported to regulate transcription factors by sequestering them in the cytoplasm; for example, BZR1 and RSG are regulated by the binding of 14-3-3 λ, ω or μ (Igarashi et al., 2001; Gampala et al., 2007). Although functional redundancy and different combinations of 14-3-3 isoforms bring difficulties in clarifying the specific roles of 14-3-3 proteins (Jaspert et al., 2011), the involvement of 14-3-3 proteins in light signaling has been illustrated by the elongated hypocotyls (relative to those of Col-0 seedlings) of 14-3-3 κ, ν and χ mutants grown in red light (Mayfield et al., 2007; Adams et al., 2014). In our work, there was no significant effect of R18 treatment and 14-3-3 mutations on PIF7’s localization and phosphorylation state under white light, which could be due to the strong light radiance, the inhibitory potency of R18 or the redundancy of the 13 14-3-3 proteins in Arabidopsis. By contrast, 14-3-3s significantly delay the shade-induced translocation and de-phosphorylation of PIF7 (Figure 5), and consequently enhance shade-induced hypocotyl elongation which is dependent on PIF7 (Figure 6). The weak shade phenotype might be caused by functional redundancy of 14-3-3 proteins. It is also possible that a compensatory increase in other isoforms occurs in 14-3-3 λκ. It is possible that 14-3-3 proteins sequester phosphorylated PIF7 in cytoplasm by protecting the PIF7 proteins from phosphatases during the transition from white light to shade.

PIF7 is a major controller for shade-induced hypocotyl elongation, as demonstrated by the severe shade-defective phenotype of pif7 mutants (Li et al., 2012). It is known that Arabidopsis grows rapidly in response to the shade stimulus, with an induction of PIL1 transcript levels detectable after only 8 min of low R/FR and growth measurable after just 30 or 45 min of exposure to shade (Salter et al., 2003; Cole et al., 2011). A quick shade regulatory mechanism is required to achieve this rapid response. Phosphorylation-dependent translocation of PIF7 is such a quick mechanism that can give rise to efficient photomorphogenesis. Moreover, several negative regulators of PIF7 have been shown to reduce the transcriptional activity of light-responsive genes and to prevent exaggerated shade responses (Hornitschek et al., 2009; Galstyan et al., 2011; Hao et al., 2012; Li et al., 2014). In our current study, the binding of 14-3-3 proteins delays the de-phosphorylation and nuclear import of PIF7 in response to shading, forming another layer of regulation to determine the appropriate SAS.

A conserved 14-3-3-binding motif has been identified in PIF3 (RNPSPP), and PIF3 has been found to be a 14-3-3 interaction partner in an affinity-purification assay using His-tagged 14-3-3-coated beads (Adams et al., 2014). However, the disruption of putative phosphorylation sites on the 14-3-3-binding motifs of PIF3 did not prevent 14-3-3 from binding to PIF3 or disturb the nuclear localization of PIF3 (Adams et al., 2014). Although the 14-3-3-binding sites of PIF7 are not typical of those of the other PIFs, the functional analysis of mutations of these sites (from Ser to Ala) illustrated their critical roles in PIF7 function.

The expression level and localization of 14-3-3 λ protein remain stable after shading is introduced (Figure 6—figure supplement 1), suggesting that 14-3-3 proteins probably exert their cargo function constantly. The activity of PIF7 is mostly determined by its phosphorylation status, which is controlled by a kinase and a phosphatase that are light-dependent. To date, CK2 (Bu et al., 2011), PPKs (Ni et al., 2017) and BIN2 (Bernardo-García et al., 2014) have been reported to be the kinases of PIF1, PIF3 and PIF4, and TOPP4 has been reported to be the phosphatase of PIF5 (Yue et al., 2016); however, no specific kinase or phosphatase of PIF7 has yet been identified. The shade-induced localization response of PIF7 varied in the different tissues (Figure 1—figure supplement 1), implying that the phosphorylation of PIF7 is probably regulated by upstream signals that have tissue and/or developmental specificity. One important goal for the future is to identify these upstream kinases, as well as the phosphatase(s) responsible for the de-phosphorylation of Ser 139 and 140, and hence for the disassociation of 14-3-3 proteins in response to shading.

Our findings offer novel insights into the mechanism through which a key transcription factor is activated by light. We already knew that a high R/FR light ration promotes the nuclear import of phyB, which positively regulates photomorphogenesis. Here, we propose that shade promotes the nuclear import of PIF7, which promotes SAS. Subcellular translocations contribute to the antagonistic action of phyB and PIF7. More detailed cooperation on light-mediated development will be necessary in the future.

Materials and methods

Plant materials and growth conditions

All of the Arabidopsis thaliana plants used in this study were of the Columbia-0 ecotype. The mutants of the 14-3-3 λ T-DNA lines (Salk_075219C and CS482153) and the 14-3-3 κ T-DNA lines (Salk_148929C and Salk_071097) were obtained from ABRC. The 14-3-3 λ and 14-3-3 κ double mutant (Salk_071097X Salk_075219C), the 35S::FLAG-HA-14-3-3 λ transgenic plants (14-3-3 λ OE) (Zhou et al., 2014) and the 35S::PIF7-Flash plants (Li et al., 2012) have been described previously. For the phenotypic analysis, seeds were germinated on 1/2 Murashige and Skoog (MS) medium (Duchefa Biochemie, Haarlem, The Netherlands) plates with 1% agar (Sangon, Shanghai, China) and without sucrose. After stratification, the plates were incubated in growth chambers under continuous white light for 4 days at 22°C, and then the plates were either left in white light or transferred to canopy shade for 5 days before hypocotyl measurements were performed. HR-350 (HiPoint, Taiwan) was used to measure the light conditions. The shade conditions were as described for previous studies (Tao et al., 2008; Li et al., 2012). Nicotiana benthamiana plants were grown at 26°C under 16-hr-light long-day conditions.

Identification of PIF7-interacting proteins by LC-MS/MS

The different light-treated 35S::PIF7-Flash seedlings were ground to a fine powder in liquid nitrogen and solubilized with lysis buffer (50 mM Tris-HCl [pH 8.0], 150 mM NaCl, 2 mM DTT, 1% NP40, 10% glycerol and protease inhibitor cocktail) (Roche, USA). The extracts were incubated at 4°C for 1 hr on a rotating wheel, and the insoluble material was removed by centrifugation at 20,000 x g at 4 °C for 15 min three times until the supernatant was clear. The supernatant was incubated with a prewashed anti-FLAG M2 agarose gel (Cat# A2220 RRID: AB_10063035, Sigma-Aldrich, USA) at 4 °C for 3 hr in the rotating wheel. The beads were recovered by centrifugation at 800 rpm at 4 °C for 2–5 min. After six washes with lysis buffer, SDS-loading buffer was added to the pellet fraction. The samples were boiled for 5 min, centrifuged at maximum speed for 15 min, and then loaded onto an SDS-PAGE gel. After running the gel halfway, the gel was cut into 1 mm3 cubes and sent for MS analysis. The trypsin-digested peptides were concentrated and analyzed using a Finniqan LTQ mass spectrometer (Thermoquest, San Jose, USA) coupled with a surveyor HPLC system.

Isolation of the nuclear fraction

For nuclear fractionation, the different light-treated seedlings were ground to a fine powder and lysed with a buffer (20 mM Tris-HCl [pH 7.5], 20 mM KCl, 2 mM EDTA, 2.5 mM MgCl2, 250 mM sucrose, 5 mM DTT, 25% glycerol and protease inhibitor cocktail) and filtered through Miracloth (Calbiochem, San Diego, USA). The obtained solution was centrifuged at 1500 x g for 10 min. The pellet was washed five times with a re-suspension buffer (NRBT: 20 mM Tris-HCl [pH 7.5], 2.5 mM MgCl2, 0.2% Triton X-100, 25% glycerol and protease inhibitor cocktail). After washing, the pellet was suspended in NRB2 (20 mM Tris-HCl [pH 7.5], 250 mM sucrose, 10 mM MgCl2, 0.5% Triton X-100, 5 mM β-mercaptoethanol and protease inhibitor cocktail) and slowly added on top of the same volume of NRB3 (20 mM Tris-HCl [pH 7.5], 1.7 M sucrose, 10 mM MgCl2, 0.5% Triton X-100, 5 mM β-mercaptoethanol and protease inhibitor cocktail). After centrifugation at 16,000 x g for 45 min at 4°C, the pellet was suspended in lysis buffer as the nuclear fraction.

Confocal microscopy and quantitation of the fluorescent protein signal

The fluorescence images of GFP and YFP expression were obtained with a Leica confocal microscope (Leica SP8) at 488 and 514 nm. GFP-PIF7 transgenic lines were grown in white light and then transferred to the shade for 0, 5 15, 25, and 45 min. The fluorescence at each time point was recorded using a 40 × 1.3 objective lens. ImageJ (http://rsb.info.nih.gov/ij/) was used to quantify the fluorescence intensities. The images were converted to an 8-bit format, and the fluorescence intensity was integrated from all pixels in the selected area. To measure the ratio between the nuclear and cytoplasmic signals for each cell, the entire cellular and nuclear area was selected for quantification of fluorescence intensity. The cytoplasmic intensity was calculated by subtracting the value for the nuclear area from that for the whole cell. The ratio between the nuclear and cytoplasmic signals was calculated for 10 cells, and three repeated measurements were performed in each condition.

Yeast two-hybrid screens and assay

A Matchmaker Gold Yeast Two-Hybrid system was used. The CDS of PIF7 was cloned into a pGBKT7 vector and used as bait to identify interacting proteins from a cDNA library, which was prepared by Oebiotech (China) with RNA from Col-0 seedlings grown under white-light conditions for 4 days. The cDNA synthesized from this material was cloned into the bait vector pGADT7. The interactions were tested on SD medium without Leu, Trp, and His but with 5 mM 3-amino-1,2,4-triazole (3AT, Sigma), using the yeast strain AH109 according to the manufacturer’s manual (Clontech).

To confirm the interaction between PIF7 and 14-3-3 proteins, the CDS of PIF7, PIF7(2A) and PIF7(2D) was cloned into pGADT7 and 14-3-3s were cloned into pGBKT7. Interactions were tested on SD medium without Leu, Trp and His but with 5 mM 3AT.

Semi-in vivo pull-down assay

Plant materials were ground with liquid nitrogen and re-suspended in extraction buffer (100 mM Tris-HCl [pH 7.5], 300 mM NaCl, 2 mM EDTA, 1% Trion X-100, 10% glycerol, and protease inhibitor cocktail). Protein extracts were centrifuged at 20,000 x g for 10 min, and the resulting supernatant was incubated with pretreated GST-14-3-3 beads for 2 hr. GST was used as a negative control. Beads were re-suspended with SDS-PAGE loading buffer and analyzed by SDS-PAGE and immunoblotting.

Generation of transgenic plants

For the overexpression of PIF7 fused with GFP, the full-length CDS of PIF7 and mutated PIF7 (△, 5A) were cloned into pMDC43 and transformed into a Col-0 background. To generate transgenic plants that overexpressed mutated PIF7 in the pif7-1 background, the mutated PIF7(2A) and PIF7(2D) were created in a plasmid of 35S::PIF7-Flash using site-directed mutagenesis. All the constructs were transformed into Agrobacterium GV3101. The primers are listed in Supplementary file 1.

Hypocotyl measurements

Quantitative measurements of hypocotyls were performed on scanned images of seedlings using ImageJ software. For measurements of mutants and stable transgenic lines, at least 20 seedlings were used per treatment or genotype. For hypocotyl analysis of T1 transgenic lines under shade, the numbers of seedlings with elongated hypocotyls were counted.

Transient transformation

Agrobacterium cells (GV3101) containing 35S::GFP-PIF7, 35S::GFP-PIF7(5A), 35S::GFP-PIF7(2A), 35S::GFP-PIF7(2D), 35S::GFP-PIF7(5D), 35S::GFP-PIF7△ or BiFC expression vectors were re-suspended in the induction medium (10 mM MES buffer [pH 5.6], 10 mM MgCl2 and 200 μM acetosyringone) and were infiltrated into young leaves of 4-week-old tobacco plants. The expression of various fluorescent proteins was analyzed using confocal microscopy or western blotting 36 hr after infiltration.

Gene expression analysis by quantitative real-time RT-PCR

Total RNA was extracted using an RNApre Plant Kit (TIANGEN, China), and the first-strand cDNA was synthesized using a FastQuant RT kit (with gDNase) (TIANGEN, China). Real-time PCR was performed with a Biorad CFX Connect system. All of the oligonucleotide primers were listed in Supplementary file 1.

Antibodies

Anti-Myc (Cat# M4439 RRID: AB_439694) and anti-FLAG (Cat# F3165 RRID: AB_259529) were purchased from Sigma-Aldrich (USA). Anti-H3 (Cat# AS10 710 RRID: AB_10750790), anti-RbcL (Cat# AS03 037–200 RRID: AB_2175288) and anti-14-3-3 (Cat# AS12 2119 RRID: AB_2619715) were from Agrisera (Sweden). Anti-GFP (Cat# MMS-118P-200 RRID: AB_10063778) was from Covance (USA).

References

  1. 1
  2. 2
  3. 3
  4. 4
  5. 5
  6. 6
  7. 7
  8. 8
  9. 9
  10. 10
  11. 11
  12. 12
  13. 13
  14. 14
  15. 15
  16. 16
  17. 17
  18. 18
  19. 19
  20. 20
  21. 21
  22. 22
  23. 23
  24. 24
  25. 25
  26. 26
  27. 27
  28. 28
  29. 29
  30. 30
  31. 31
  32. 32
  33. 33
  34. 34
  35. 35
  36. 36
  37. 37
  38. 38
  39. 39
  40. 40
  41. 41
  42. 42
  43. 43

Decision letter

  1. Zhi-Yong Wang
    Reviewing Editor; Carnegie Institution for Science, United States

In the interests of transparency, eLife includes the editorial decision letter and accompanying author responses. A lightly edited version of the letter sent to the authors after peer review is shown, indicating the most substantive concerns; minor comments are not usually included.

Thank you for submitting your article "Shade-induced nuclear localization of PIF7 is regulated by phosphorylation and 14-3-3 proteins in Arabidopsis" for consideration by eLife. Your article has been favorably evaluated by Christian Hardtke (Senior Editor) and three reviewers, one of whom, Zhi-Yong Wang (Reviewer #1), is a member of our Board of Reviewing Editors.

The reviewers have discussed the reviews with one another and the Reviewing Editor has drafted this decision to help you prepare a revised submission.

Summary:

Overall, the reviewers found your manuscript interesting and potentially a good contribution to eLife. However, they also raised important issues that must be addressed. Some of the issues require additional experimental evidence. Particularly, the role of 14-3-3s in controlling PIF7 nuclear localization needs more rigorous tests.

Essential revisions:

1) PIF7 is known to mediate shade avoidance syndrome (SAS) in Arabidopsis. Unlike other PIFs, PIF7 is not degraded rapidly in response to phytochrome-induce phosphorylation. How phosphorylation affects PIF7 activity has remained an outstanding question. Therefore, your findings that 14-3-3 proteins retain phospho-PIF7 in cytoplasm and shade-induced dephosphorylation causes PIF7 nuclear localization answer an important question in plant biology. While the results are interpreted clearly, the biological implication was not fully elaborated. For example, how such mechanism may contribute to the rapid response to shade, compared to the other light-labile PIFs that would require new protein synthesis to regain activity upon shade inactivation of phytochrome? We suggest you add a discussion of the possible implication of this nuclear localization mechanism in the contexts of the degradation mechanisms of other PIFs as well as your earlier finding that pif7 lacks an early shade avoidance response (Li et al., 2012).

2) Two previous studies on GFP/CFP-tagged PIF7 (Leivar et al., 2008 and Kidokoro et al., Plant Physiol., 2009) did not report such cytoplasmic localization. Instead, GFP-PIF7 was shown to be nuclear and in speckles in light-treated tissues. The possible reasons for the discrepancy must be discussed, and ideally the possible reasons should be tested experimentally.

3) Why are BiFC signals stronger in the nucleus than in cytoplasm? If the interaction causes cytoplasmic retention, BiFC signals should be weaker or absent in the nucleus. Co-expression of a PIF7-RFP would help determine whether the tethering to 14-3-3 by BiFC affects PIF7 localization. Comparing the localization patterns of PIF7/14-3-3 BiFC signal with PIF7-GFP after short time shade treatment will provide further evidence as to whether change of 14-3-3 binding is required for shade-induced nuclear localization.

4) The R18 peptide treatment and 14-3-3 mutant experiments did not support your model. Unlike mutations of 14-3-3 binding site in PIF7 (PIF7(2A) and PIF7(5A)), both R18 treatment and 14-3-3 mutations had no effects on the PIF7 localization or target gene expression under white light. They only slightly increased the N/C ration after shade treatment. This needs to be resolved. Perhaps a positive control (another protein whose subcellular localization affected by 14-3-3 binding) is required to make sure the R18 treatment is effective, and higher order 14-3-3 mutants might show more dramatic effects. Similarly, the effects of R18 and 14-3-3 mutations on PIF7 phosphorylation status are weak and variable (Figure 4C and 4F), whereas PIF7(2A) had a strong effect on dephosphorylation under white light conditions (Figure 3C). The quantitative change of phosphorylation of PIF7(2A) (Figure 3C) suggests an indirect effect of the mutations on phosphorylation of additional sites, which may contribute to the change of nuclear localization and phenotypes.

5) The 14-3-3 binding to PIF7(2D) and PIF7(5D) needs to be tested quantitatively in vitro and in vivo, and the phenotypes of plants expressing PIF7(2D) and PIF7(5D) should be analyzed. It has been reported that 14-3-3 binding requires phosphorylation and cannot be mimicked by S-to-D substitution. If PIF7(5D) cannot bind to 14-3-3s, it would suggest that cytoplasmic localization is independent of 14-3-3, though dependent on phosphorylation (negative charge). In this case, it would be important to test whether PIF7(2D) and PIF7(5D) rescue the pif7 mutant and change localization in response to shade, and you may want to test whether 14-3-3 inhibits PIF7 through other mechanisms such as DNA-binding or transcription activity.

6) Overexpression from the 35S promoter was used for both subcellular localization and in vivo interaction experiments. Reviewers are concerned that overexpression may contribute to artifacts on both cytoplasmic localization and interaction with 14-3-3 proteins. It is important to repeat these experiments using PIF7 expressed from its native promoter at its normal level.

7) The interaction between PIF7 and 14-3-3 should be analyzed by co-immunoprecipitation, or quantitative IP-MS (Figure 2—figure supplement 1 was not quantitative), in Arabidopsis under light and shade conditions.

[Editors' note: further revisions were requested prior to acceptance, as described below.]

Thank you for resubmitting your work entitled "Shade-induced nuclear localization of PIF7 is regulated by phosphorylation and 14-3-3 proteins in Arabidopsis" for further consideration at eLife. Your revised article has been favorably evaluated by Christian Hardke (Senior Editor) and three reviewers, one of whom is a member of our Board of Reviewing Editors.

The manuscript has been improved but there are some remaining issues that need to be addressed before acceptance, as outlined below. Ordinarily we do not encourage multiple rounds of review. Given the remaining substantial concerns, it will be necessary to limit this to just a single additional review before we make a binding final decision on your manuscript.

The main issues are raised by your new data showing that the PIF7(2D) mutation abolished the binding to 14-3-3 λ and 14-3-3 κ, but unexpectedly lead to cytoplasmic retention of PIF7.

1) The Discussion of the role of 14-3-3s must be consistent with the results. There are overstatements about this in the current manuscript.

2) Stronger evidence is needed to clarify the role of 14-3-3s in regulating PIF7. As the effects of R18 treatment and 14-3-3 mutations are subtle, the results need to be quantified with at least three biological replicates. Further, treatment of the 14-3-3 mutants with R18 and R18(Lys) should be tested for possibly stronger effects on the dephosphorylation and nuclear localization of PIF7.

3) Is it possible that PIF7(2D) binds to some other 14-3-3 family members in vivo? One test of this possibility is to immunoprecipitate PIF7 and PIF7(2D) followed by immunoblotting using the anti-14-3-3 antibodies, which may detect other isoforms. Another test is IP-MS as explained below.

4) The IP-MS experiment (Figure 2—figure supplement Figure 1) is invalid due to the lack of a negative control. The IP-MS experiment needs to be re-performed with a negative control using quantitative mass spectrometry to show specificity. If PIF7(2D) is included and compared to wild type PIF7 under the shade condition or compared to PIF7(2A), the protein that mediates cytoplasmic retention of PIF7(2D) can potentially be identified.

Reviewer #1:

The revision has improved the manuscript. Particularly, the co-IP experiment (Figure 2DFigure) provides convincing evidence for in vivo interaction that is light/shade-dependent.

However, the new data showing the lack of 14-3-3 binding to PIF7(2D) suggests that binding by 14-3-3 is not essential for cytoplasmic retention and inactivation of PIF7, and additional 14-3-3-independent mechanisms may exist. This should be clearly discussed. The statement "Here we demonstrate… an essential role of the 14-3-3 proteins in cytoplasmic-retention of PIF7 in Arabidopsis" is no longer supported by the new data.

The mass spectrometry data (Figure 2—figure supplement 1) lacks a proper negative control and thus is invalid. Such IP-MS often detects many non-specific background proteins, and a negative control is essential to get meaningful results. I suggest this data be removed. As the interaction has been confirmed by Y2H, BiFC, co-IP, and site-mutagenesis, removing the MS data does not affect the conclusion while avoiding below-standard results.

Reviewer #2:

The authors have addressed most of my comments experimentally and provided explanation for discrepancies.

The only concern that I have is the lack of proof that the epitope-tagged lines of PIF7 used in the study is functional (i.e. if they can rescue pif7 mutants under native conditions). Though, as the authors have argued with other supporting data and time constrain, this may be addressed in future studies.

Reviewer #3:

The revised manuscript by Huang et al. addressed some issues raised by reviewers, but the functional roles of 14-3-3 proteins in the regulation of PIF7 phosphorylation and localization are still not clear and not supported by experimental evidence. One of major molecular functions of 14-3-3 proteins is to retain their target proteins in the cytoplasm, thereby preventing them from translocating to the nucleus. However, several experimental evidences (see below) suggest that the 14-3-3 proteins are not directly involved in the PIF7 localization although they seem to directly bind to the phosphorylated PIF7.

1) The 14-3-3 proteins interact with PIF7 specifically under white light (not shade).

2) Neither R18 treatment nor 14-3-3 mutations affected the PIF7 localization under white light.

3) PIF7(2D) does not interact with the 14-3-3 proteins, but it is mainly localized in the cytoplasm.

Since the cytoplasmic localization of PIF7 is not dependent on the 14-3-3 proteins (as the authors also suggested in "Responses to reviewers' comments"), the manuscript (especially Discussion part) should be re-written to clearly define the role of 14-3-3 proteins. The title should be changed; the 14-3-3 proteins only slightly affect the PIF7 localization in the very limited condition.

In addition, although authors claim the 14-3-3 proteins somehow affect the shade-induced PIF7 localization and dephosphorylation, the effects of R18 and 14-3-3 mutations on the PIF7 localization are very subtle (Figure 4A, D) and their effects on the dephosphorylation of PIF7 are not obvious (Figure 4C, F). Since these are the only effects of 14-3-3 proteins on PIF7 that the authors showed in this manuscript, they should provide more convincing data.

The 14-3-3 proteins interact with PIF7 only under white light, but they appear to regulate the PIF7 localization only during the transition from white to shade. Possible explanations for this inconsistency should be provided in Discussion part.

https://doi.org/10.7554/eLife.31636.027

Author response

Essential revisions:

1) PIF7 is known to mediate shade avoidance syndrome (SAS) in Arabidopsis. Unlike other PIFs, PIF7 is not degraded rapidly in response to phytochrome-induce phosphorylation. How phosphorylation affects PIF7 activity has remained an outstanding question. Therefore, your findings that 14-3-3 proteins retain phospho-PIF7 in cytoplasm and shade-induced dephosphorylation causes PIF7 nuclear localization answer an important question in plant biology. While the results are interpreted clearly, the biological implication was not fully elaborated. For example, how such mechanism may contribute to the rapid response to shade, compared to the other light-labile PIFs that would require new protein synthesis to regain activity upon shade inactivation of phytochrome? We suggest you add a discussion of the possible implication of this nuclear localization mechanism in the contexts of the degradation mechanisms of other PIFs as well as your earlier finding that pif7 lacks an early shade avoidance response (Li et al., 2012).

We have followed the reviewer’s suggestion by including the related discussion in our revised manuscript (Discussion, second and fifth paragraphs).

2) Two previous studies on GFP/CFP-tagged PIF7 (Leivar et al., 2008 and Kidokoro et al., Plant Physiol., 2009) did not report such cytoplasmic localization. Instead, GFP-PIF7 was shown to be nuclear and in speckles in light-treated tissues. The possible reasons for the discrepancy must be discussed, and ideally the possible reasons should be tested experimentally.

As the reviewer mentioned, GFP/CFP-tagged PIF7 has been published in two previous studies.

According to the phenotype of loss function of mutants (pif7-1, pif7-2) (Leivar et al., 2008, Li et al., 2012), PIF7 is a positive regulator for hypocotyl elongation. In our hand, overexpressing either PIF7-Flash (Li., et al., 2012 and current study) or N-terminated tagged GFP-PIF7 (current study) could rescue the shorter hypocotyl phenotype of pif7-1 or pif7-2 under shade. So we believe that the PIF7-Flash or GFP-PIF7 transgenic lines represent its normal function and probable localization. The line reported in literature, PIF7-CFP overexpression lines from Dr. Peter Quail’s lab show a somewhat short hypocotyl phenotype in white light and dark (Leivar et al., 2008), indicating C-terminate fused GFP protein might interfere the normal function of PIF7. It is a possible explanation of the discrepancy of these localization results.

3) Why are BiFC signals stronger in the nucleus than in cytoplasm? If the interaction causes cytoplasmic retention, BiFC signals should be weaker or absent in the nucleus. Co-expression of a PIF7-RFP would help determine whether the tethering to 14-3-3 by BiFC affects PIF7 localization. Comparing the localization patterns of PIF7/14-3-3 BiFC signal with PIF7-GFP after short time shade treatment will provide further evidence as to whether change of 14-3-3 binding is required for shade-induced nuclear localization.

The seemly stronger nuclear signals in BiFC assay is probably because these intensive signals gathered in a small area in tobacco leaf cells. In Arabidopsis, GFP-PIF7 localized in both nucleus and cytoplasm (Figure 1), but 14-3-3 λ mainly localized in cytoplasm (Figure 6—figure supplement 1). To further clarify this question, we quantitated the nuclear/cytoplasmic fluorescence signal ratio in tobacco leaf cells which co-expressed 14-3-3λ-cYFP and nYFP-PIF7 or only expressed GFP-PIF7. As shown in Author response image 1A, nuclear/cytoplasmic signal is significant lower in BiFC assay, indicating that the interaction of 14-3-3s and PIF7 occurs mainly in cytoplasm.

The experiment suggested by the reviewers is to figure out whether change of 14-3-3 binding is required for shade-induced nuclear localization of PIF7. However, the localization of GFP-PIF7 expressed in tobacco leaf cells doesn’t respond to shade treatment (Author response image 1B). The shade response in tobacco leaf cells may not share the same molecular mechanism as that in Arabidopsis. Regardless of light response, we could test whether co-expression of 14-3-3s affects the localization of PIF7 in tobacco leaf cells. When 14-3-3 λ was co-expressed with GFP-PIF7, GFP-PIF7(2A), GFP-PIF7(5A), GFP-PIF7(2D), GFP-PIF7(5D), or GFP in tobacco leaves (Author response image 1C), 14-3-3λ was unable to change the localization of any version of PIF7. Moreover, 14-3-3 λκ mutation and treatment of R18 didn’t change the localization of GFP-PIF7 under white light in Arabidopsis (Figure 4). Taken together, tobacco system was used to determine the interaction of 14-3-3s and PIF7 preferentially in cytoplasm and localization of PIF7 derivatives, whereas Arabidopsis system is used to address the questions of shade and phosphorylation dependency.

Author response image 1
(a) Quantification of ratios between nuclear and cytoplasmic intensities of YFP/GFP in tobacco leaf cells (n>10) which co-expressed 14-3-3λ-cYFP and nYFP-PIF7 or only expressed GFP-PIF7.

(b) Shade light doesn’t affect the localization of GFP-PIF7 expressed in tobacco leaf cells. (c) Co-expressed 14-3-3λ doesn’t affect the localization of PIF7 derivatives in tobacco leaf cells. Subcellular localization of GFP, GFP-PIF7, GFP-PIF7 (2A), GFP-PIF7 (5A), GFP-PIF7 (2D) and GFP-PIF7 (5D) co-expressed with 14-3-3λ or empty vector in tobacco leaf cells. The expression levels of GFP-PIF7 derivatives and 14-3-3λ-FLAG was detected using anti-GFP antibody and anti-FLAG antibody, respectively.

https://doi.org/10.7554/eLife.31636.024

4) The R18 peptide treatment and 14-3-3 mutant experiments did not support your model. Unlike mutations of 14-3-3 binding site in PIF7 (PIF7(2A) and PIF7(5A)), both R18 treatment and 14-3-3 mutations had no effects on the PIF7 localization or target gene expression under white light. They only slightly increased the N/C ration after shade treatment. This needs to be resolved. Perhaps a positive control (another protein whose subcellular localization affected by 14-3-3 binding) is required to make sure the R18 treatment is effective, and higher order 14-3-3 mutants might show more dramatic effects. Similarly, the effects of R18 and 14-3-3 mutations on PIF7 phosphorylation status are weak and variable (Figure 4C and 4F), whereas PIF7(2A) had a strong effect on dephosphorylation under white light conditions (Figure 3C). The quantitative change of phosphorylation of PIF7(2A) (Figure 3C) suggests an indirect effect of the mutations on phosphorylation of additional sites, which may contribute to the change of nuclear localization and phenotypes.

We thank the reviewers for pointing out these issues. Actually, our model proposed that 14-3-3s retain the phosphorylated PIF7 in cytoplasm under shade, possibly through protection from phosphatases. Our results (Figure 5) support 14-3-3s mainly work during the transition from phosphorylated to dephosphorylated PIF7, not under white light. If 14-3-3s’ function is to escort phosphorylated PIF7 to cytoplasm, we should have observed longer hypocotyl in white light in which phosphorylated PIF7 is dominant, and more cytoplasmic localization of PIF7 in 14-3-3s mutant background than that in Col-0 under white light. But in fact, neither of them happens.

The effect of R18 is significant on shade-induced nuclear-translocation and dephosphorylation of PIF7 (Figure 5). Moreover, the effect of R18 treatment on shade-induced gene expression and hypocotyl elongation reproduces that of 14-3-3 mutations (Figure 6). So we believe that R18 treatment is effective. We also followed the reviewer’s suggestion and tested the effect of R18 on localization of BES1-GFP, as shown in Author response image 2. The subcellular localization of BES1 is affected by 14-3-3 binding (Dev Cell. 2007 13(2):177-89; Mol. Cells. 2010 29:283-290) and also affected by R18 treatment.

Six 14-3-3 proteins interacted with PIF7. There are thirteen 14-3-3 proteins in Arabidopsis, which lead to functional abundance and difficulty to study loss of function of 14-3-3s. Although we agree with the reviewer that higher order 14-3-3-mutants might show more dramatic effects, it’s technically very challenging to construct a higher order mutant with a lot of isoforms. On the other hand, the double mutant of 14-3-3λκ significantly increased the N/C ratio after shade treatment, supporting our hypothesis of 14-3-3s’ effect on PIF7.

We agree with the reviewers that there are additional sites beside S139 and S141 that may contribute to the change of nuclear localization and phenotypes.

Author response image 2
The effect of R18 on the localization of BES1-GFP.

BES1-GFP transgenic plants (a gift from Dr. Xuelu Wang’s lab) grown under white light were treated with R18 or R18(Lys) for 3 hr after 10 min of vacuum. ImageJ was used to quantify the fluorescence intensities. Ratios between the nuclear and cytoplasmic signal intensities were calculated from at least 10 cells for each treatment.

https://doi.org/10.7554/eLife.31636.025

5) The 14-3-3 binding to PIF7(2D) and PIF7(5D) needs to be tested quantitatively in vitro and in vivo, and the phenotypes of plants expressing PIF7(2D) and PIF7(5D) should be analyzed. It has been reported that 14-3-3 binding requires phosphorylation and cannot be mimicked by S-to-D substitution. If PIF7(5D) cannot bind to 14-3-3s, it would suggest that cytoplasmic localization is independent of 14-3-3, though dependent on phosphorylation (negative charge). In this case, it would be important to test whether PIF7(2D) and PIF7(5D) rescue the pif7 mutant and change localization in response to shade, and you may want to test whether 14-3-3 inhibits PIF7 through other mechanisms such as DNA-binding or transcription activity.

We have now included the data about PIF7(2D)-Flash transgenic line in pif7-1 background. As shown in Figure 4C, E, F and G of our revised manuscript, PIF7(2D)-Flash mainly localizes in cytoplasm and can’t complement the shade defective gene expression and phenotype of pif7-1, indicating that S-to-D substitution could mimic phosphorylation of PIF7. However, in our BiFC and GST pull-down assay, PIF7(2D) can’t interact with 14-3-3s (Figure 3—figure supplement 3), probably because the interaction of 14-3-3s and PIF7 can’t be mimicked by S-to-D substitution as the reviewer mentioned. So we would like to conclude that the cytoplasmic localization of PIF7 is dependent on the phosphorylation state, not 14-3-3s under white light. But 14-3-3s delay the shade-induced nuclear- translocation and dephosphorylation of PIF7 (Figure 5), possibly through protection from phosphatases during the transition from white light to shade. We did test the effect of 14-3-3s on PIF7’s transcriptional activity in tobacco leaves as the reviewer suggested. As we expected, co-expressed 14-3-3λ or 14-3-3κ didn’t change the activation of PIF7 on YUCCA8 promoter, indicating the inhibitions of 14-3-3λ and 14-3-3κ on PIF7 are not through the regulation of the transcriptional activity (Author response image 3).

Author response image 3
The effect of 14-3-3λ and 14-3-3κ on PIF7-mediated activation of YUCCA8 promoter.

The effector constructs contain the CaMV 35S promoter fused to the transcription factor PIF7 or 14-3-3s. The reporter construct contains the 2.2 Kb upstream of the translation initiation site of YUCCA8 fused to the LUC reporter gene. Both effector and reporter were co-expressed in tobacco leaf cells. The ratio of Luc to Ren from leaves transfected reporter and empty effector was normalized as 1.

https://doi.org/10.7554/eLife.31636.026

6) Overexpression from the 35S promoter was used for both subcellular localization and in vivo interaction experiments. Reviewers are concerned that overexpression may contribute to artifacts on both cytoplasmic localization and interaction with 14-3-3 proteins. It is important to repeat these experiments using PIF7 expressed from its native promoter at its normal level.

We have been working on generating native promoter driven PIF7 expressed transgenic lines for a while. We amplified 1019 bp before ATG of PIF7 as native promoter according to the published paper (Kidokoro et al., Plant Physiol., 2009) and constructed pPIF7::PIF7-Flash. But unfortunately, we didn’t get complementary line during the screen of pPIF7::PIF7-Flash transgenic lines in pif7-1 background so far. The length of PIF7 promoter we used might be not long enough. Although we agree with the reviewer better to repeat the localization experiment using PIF7 expressed from its native promoter, we need more studies on the length of native promoter before we use it for further study.

However, overexpressed PIF7(2A) and PIF7(2D) displayed the different localization compared to the overexpressed PIF7, indicating the effect of localization is not an artifact of overexpression. Interactions between PIF7 and 14-3-3s have been verified in Y2H, BiFC, GST-pulldown assays and CoIP. We also noticed that overexpression lines have been used for several 14-3-3s’ clients related studies [Dev Cell. 2007 13(2):177; Dev Cell. 2011 21(5):825; Plant Cell. 2014, 26(3):1166]. So we feel confident about our conclusion based on meaningful data.

7) The interaction between PIF7 and 14-3-3 should be analyzed by co-immunoprecipitation, or quantitative IP-MS (Figure 2—figure supplement 1 was not quantitative), in Arabidopsis under light and shade conditions.

We have followed the reviewer’s suggestion and did Co-IP experiment. The data confirmed the binding of 14-3-3s to the phosphorylated PIF7 (revised Figure 2D).

[Editors' note: further revisions were requested prior to acceptance, as described below.]

The main issues are raised by your new data showing that the PIF7(2D) mutation abolished the binding to 14-3-3 λ and 14-3-3 κ, but unexpectedly lead to cytoplasmic retention of PIF7.

Thanks for raising this issue. We hope our newly included data will convince the reviewers of our conclusions.

1) The Discussion of the role of 14-3-3s must be consistent with the results. There are overstatements about this in the current manuscript.

We have modified the Discussion (third and fourth paragraphs).

2) Stronger evidence is needed to clarify the role of 14-3-3s in regulating PIF7. As the effects of R18 treatment and 14-3-3 mutations are subtle, the results need to be quantified with at least three biological replicates. Further, treatment of the 14-3-3 mutants with R18 and R18(Lys) should be tested for possibly stronger effects on the dephosphorylation and nuclear localization of PIF7.

Effects of R18 treatment and 14-3-3 mutations are significant on regulating PIF7’s localization and de-phosphorylation. Quantification of the shade-induced nuclear accumulation of GFP-PIF7 was calculated from at least 10 cells that came from different seedlings (Figure 5B and E). Biological replicates have been newly added in Figure 4C and F to further confirm the effects on de-phosphorylation of PIF7. We also quantitated the ratio of phosphorylated PIF7 to total PIF7 and presented in Figure 5—figure supplement 2. The shade-induced de-phosphorylation of PIF7 is significantly differentiated between R18 and R18 (Lys) treatment, and the effect is even stronger between 14-3-3 mutant and Col-0.

Thanks for the suggestion to treat 14-3-3 mutants with R18. Shade-induced nuclear translocation and de-phosphorylation of PIF7 is a fast process, which is correlated with the quick changes of gene expression and the measurable growth after just 30-40 min shade exposure. The effects of R18 and 14-3-3 mutations accelerate this fast process. We have shown the significant effects by individual treatment of R18 or in 14-3-3 mutants. But to capture the probable stronger effects by both factors have been very challenging due to the handling time and resolution of detective methods.

3) Is it possible that PIF7(2D) binds to some other 14-3-3 family members in vivo? One test of this possibility is to immunoprecipitate PIF7 and PIF7(2D) followed by immunoblotting using the anti-14-3-3 antibodies, which may detect other isoforms. Another test is IP-MS as explained below.

Thanks for this suggestion. We tried Co-IP in vivo and Y2H experiments. The new data in Figure 3A and C shows that 14-3-3s bind to the PIF7 (2D), not PIF7 (2A) in in vivo and in the yeast system. The antibody we used is not isoform specific, so we cannot distinguish 14-3-3 family members yet.

4) The IP-MS experiment (Figure 2—figure supplement 1) is invalid due to the lack of a negative control. The IP-MS experiment needs to be re-performed with a negative control using quantitative mass spectrometry to show specificity. If PIF7(2D) is included and compared to wild type PIF7 under the shade condition or compared to PIF7(2A), the protein that mediates cytoplasmic retention of PIF7(2D) can potentially be identified.

We totally agree with the reviewer. We did IP-MS, using the different light treated transgenic seedlings, when we started the project, to qualitatively identify PIF7’s binder. But soon a substantial amount of evidence revealed the interaction of PIF7 and 14-3-3s. Therefore, for the sake of clarity and concision, we would rather remove the 14-3-3’s MS data and keep the identification of phosphorylated sites from IP-MS in our revised manuscript.

It is a fascinating idea to compare the partners in PIF7(2A) or PIF7(2D) complex using IP-MS. Our new Co-IP experiment has shown the different levels of 14-3-3s bound with PIF7, PIF(2A) and PIF7(2D). We believe that there are more components besides 14-3-3s involved in the translocation of PIF7, such as kinase and phosphatase. The on-going ID of these components would likely greatly forward our understanding of PIF7-related shade-signaling pathway, but to characterize these components requires substantial work and deserves a separate study.

Reviewer #1:

The revision has improved the manuscript. Particularly, the co-IP experiment (Figure 2D) provides convincing evidence for in vivo interaction that is light/shade-dependent.

However, the new data showing the lack of 14-3-3 binding to PIF7(2D) suggests that binding by 14-3-3 is not essential for cytoplasmic retention and inactivation of PIF7, and additional 14-3-3-independent mechanisms may exist. This should be clearly discussed. The statement "Here we demonstrate… an essential role of the 14-3-3 proteins in cytoplasmic-retention of PIF7 in Arabidopsis" is no longer supported by the new data.

The new data in Figure 3A and C shows that 14-3-3s bind to the PIF7 (2D), not PIF7 (2A), in in vivo and in the yeast system. The related sentence has been modified in the last paragraph of the Introduction.

The mass spectrometry data (Figure 2—figure supplement 1) lacks a proper negative control and thus is invalid. Such IP-MS often detects many non-specific background proteins, and a negative control is essential to get meaningful results. I suggest this data be removed. As the interaction has been confirmed by Y2H, BiFC, co-IP, and site-mutagenesis, removing the MS data does not affect the conclusion while avoiding below-standard results.

We totally agree with the reviewer. We did IP-MS, using the different light treated transgenic seedlings, when we started the project, to qualitatively identify PIF7’s binder. But soon a substantial amount of evidence revealed the interaction of PIF7 and 14-3-3s. Therefore, for the sake of clarity and concision, we would rather remove the 14-3-3’s MS data and keep the identification of phosphorylated sites from IP-MS in our revised manuscript.

Reviewer #3:

The revised manuscript by Huang et al. addressed some issues raised by reviewers, but the functional roles of 14-3-3 proteins in the regulation of PIF7 phosphorylation and localization are still not clear and not supported by experimental evidence. One of major molecular functions of 14-3-3 proteins is to retain their target proteins in the cytoplasm, thereby preventing them from translocating to the nucleus. However, several experimental evidences (see below) suggest that the 14-3-3 proteins are not directly involved in the PIF7 localization although they seem to directly bind to the phosphorylated PIF7.

1) The 14-3-3 proteins interact with PIF7 specifically under white light (not shade).

2) Neither R18 treatment nor 14-3-3 mutations affected the PIF7 localization under white light.

We understand the paradoxical discrepancy of interaction and function in white light and have been working on this question from the beginning.

Under our white light conditions, there is no significant effect of R18 and 14-3-3 mutations on PIF7 localization and de-phosphorylation in Figure 5. Together with the hypocotyl length in our white light (Figure 6), we do not claim the role of 14-3-3s under white light. We think the cytoplasmic localization of PIF7 is dependent on the phosphorylation, possibly not dependent on 14-3-3s. However, these results could be influenced by 1) strong white light radiance; 2) the low potency of R18 as an inhibitor of 14-3-3s and 3) the redundancy of thirteen 14-3-3s in Arabidopsis. To clearly demonstrate the role of 14-3-3s under white light, to our knowledge so far, it is necessary to obtain, at least, high order 14-3-3 mutants, stronger inhibitor without side-effect and change the light condition. In contrast, 14-3-3s significantly delay the shade-induced translocation and de-phosphorylation of PIF7 (Figure 5), and consequently enhance shade-induced hypocotyl elongation. The weak shade phenotype might be caused by functional redundancy of 14-3-3 proteins. It’s also possible that compensatory increase of other isoforms occurs in 14-3-3 λκ. 14-3-3 proteins sequester phosphorylated PIF7 in cytoplasm possibly through protection from phosphatases. All these experimental results support the role of 14-3-3s through the interaction with PIF7 during the transition from white light to shade. We have added the related discussion (Discussion, fourth paragraph) to the revised version.

3) PIF7(2D) does not interact with the 14-3-3 proteins, but it is mainly localized in the cytoplasm.

We conducted Co-IP and Y2H experiments and showed that 14-3-3s bind to the PIF7 (2D). Actually, more 14-3-3 proteins were co-immunoprecipitated with PIF7 (2D) than that with PIF7 (2A) in Arabidopsis seedlings, which is consistent with the cytoplasmic retention of PIF7 (2D).

Since the cytoplasmic localization of PIF7 is not dependent on the 14-3-3 proteins (as the authors also suggested in "Responses to reviewers' comments"), the manuscript (especially Discussion part) should be re-written to clearly define the role of 14-3-3 proteins. The title should be changed; the 14-3-3 proteins only slightly affect the PIF7 localization in the very limited condition.

The major and novel findings in our work include at least 4 points: 1) shade rapidly induce nuclear localization of PIF7; 2) Phosphorylation sites of PIF7 are important for its localization and function; 3) 14-3-3s interact with PIF7; 4) 14-3-3s delay the nuclear translocation and de-phosphorylation of PIF7 and negatively regulate SAS. These findings are supported by experimental evidences and appropriately reflected in title: “Shade-induced nuclear localization of PIF7 is regulated by phosphorylation and 14-3-3 proteins in Arabidopsis”.

In addition, although authors claim the 14-3-3 proteins somehow affect the shade-induced PIF7 localization and dephosphorylation, the effects of R18 and 14-3-3 mutations on the PIF7 localization are very subtle (Figure 4A, D) and their effects on the dephosphorylation of PIF7 are not obvious (Figure 4C, F). Since these are the only effects of 14-3-3 proteins on PIF7 that the authors showed in this manuscript, they should provide more convincing data.

Shade-induced nuclear translocation and de-phosphorylation of PIF7 is a fast process, which is correlated with the quick changes of gene expression and measurable growth after just 30-40 min shade exposure. The effects of R18 and 14-3-3 mutations accelerate this fast process. Based on Figures 5B, 5E and newly added Figure 5—figure supplement 2, significant effects of R18 and 14-3-3 mutations have been quantified with more experimental repeats.

https://doi.org/10.7554/eLife.31636.028

Article and author information

Author details

  1. Xu Huang

    1. State Key Laboratory of Genetic Engineering, School of Life Sciences, Fudan University, Shanghai, China
    2. Institute of Plant Biology, School of Life Sciences, Fudan University, Shanghai, China
    Contribution
    Conceptualization, Data curation, Formal analysis, Validation, Investigation, Visualization, Methodology
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0002-2711-9920
  2. Qian Zhang

    1. State Key Laboratory of Genetic Engineering, School of Life Sciences, Fudan University, Shanghai, China
    2. Institute of Plant Biology, School of Life Sciences, Fudan University, Shanghai, China
    Contribution
    Conceptualization, Formal analysis, Validation, Investigation, Methodology
    Competing interests
    No competing interests declared
  3. Yupei Jiang

    1. State Key Laboratory of Genetic Engineering, School of Life Sciences, Fudan University, Shanghai, China
    2. Institute of Plant Biology, School of Life Sciences, Fudan University, Shanghai, China
    Contribution
    Formal analysis, Validation, Investigation, Methodology
    Competing interests
    No competing interests declared
  4. Chuanwei Yang

    1. State Key Laboratory of Genetic Engineering, School of Life Sciences, Fudan University, Shanghai, China
    2. Institute of Plant Biology, School of Life Sciences, Fudan University, Shanghai, China
    Contribution
    Formal analysis, Validation, Investigation, Methodology
    Competing interests
    No competing interests declared
  5. Qianyue Wang

    1. State Key Laboratory of Genetic Engineering, School of Life Sciences, Fudan University, Shanghai, China
    2. Institute of Plant Biology, School of Life Sciences, Fudan University, Shanghai, China
    Contribution
    Validation
    Competing interests
    No competing interests declared
  6. Lin Li

    1. State Key Laboratory of Genetic Engineering, School of Life Sciences, Fudan University, Shanghai, China
    2. Institute of Plant Biology, School of Life Sciences, Fudan University, Shanghai, China
    Contribution
    Conceptualization, Resources, Data curation, Formal analysis, Supervision, Funding acquisition, Investigation, Methodology, Writing—original draft, Project administration, Writing—review and editing
    For correspondence
    linli@fudan.edu.cn
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0003-4840-5245

Funding

National Natural Science Foundation of China (31470374)

  • Lin Li

National Natural Science Foundation of China (31500973)

  • Lin Li

National Key Research and Development Program of China (2017YFA0503800)

  • Lin Li

The funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.

Acknowledgements

We thank Dr. Hongquan Yang (Fudan University) for sharing the cDNA library for yeast two hybrid screens, Dr. Yan Guo (China Agricultural University) and Dr. Honghong Hu (Huazhong Agricultural University) for sharing seeds of 14-3-3 mutants and 14-3-3 λ transgenic lines, Dr. Xuelu Wang (Huazhong Agricultural University) for sharing seeds of BES1-GFP and Dr. Liang Cai (Fudan University) for technical assistance relating to spanning-disk confocal microscopy. Dr. Yanhong Li and Miss Lin Huang at Proteomics Platform (Fudan University) are acknowledged for their help in MS characterization of phosphor-sites of PIF7. This work was supported by National Key R and D Program of China (grant 2017YFA0503800) and by the National Natural Science Foundation of China (grants 31470374 and 31500973).

Reviewing Editor

  1. Zhi-Yong Wang, Carnegie Institution for Science, United States

Publication history

  1. Received: August 30, 2017
  2. Accepted: April 11, 2018
  3. Accepted Manuscript published: June 21, 2018 (version 1)
  4. Version of Record published: July 9, 2018 (version 2)
  5. Version of Record updated: July 31, 2018 (version 3)

Copyright

© 2018, Huang et al.

This article is distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use and redistribution provided that the original author and source are credited.

Metrics

  • 2,378
    Page views
  • 612
    Downloads
  • 1
    Citations

Article citation count generated by polling the highest count across the following sources: Crossref, Scopus, PubMed Central.

Download links

A two-part list of links to download the article, or parts of the article, in various formats.

Downloads (link to download the article as PDF)

Download citations (links to download the citations from this article in formats compatible with various reference manager tools)

Open citations (links to open the citations from this article in various online reference manager services)

Further reading

    1. Cell Biology
    2. Plant Biology
    Andreas S Richter et al.
    Research Article
    1. Cell Biology
    2. Plant Biology
    Maria A Prusicki et al.
    Tools and Resources Updated