Antagonistic regulation of PIF4 and HY5 activity by BBX28 regulates thermomorphogenesis

Abstract

ELONGATED HYPOCOTYL 5 (HY5) and PHYTOCHROME INTERACTING FACTOR 4 (PIF4) compete for binding to common promoter elements of growth-related genes. While HY5 suppresses these genes to inhibit elongation, under high ambient temperature, PIF4 promotes their expression. The molecular regulation of this differential binding is critical for adaptive growth. Under elevated temperature (28°C), BBX28 expression and accumulation are enhanced, and bbx28 seedlings exhibit hyposensitivity to temperature-mediated hypocotyl elongation. Our genetic analyses indicate that BBX28 and PIF4 act in the same pathway and form a positive feed-forward loop. PIF4 promotes BBX28 expression under elevated temperature. BBX28 in turn interacts with PIF4 and promotes PIF4-mediated activation of auxin signaling genes by binding to their promoters. Under elevated temperature, BBX28 interacts with HY5 and inhibits its binding to the same sites on the promoters of IAA19 and SAUR19. BBX28 therefore regulates the HY5-PIF4–mediated antagonistic regulation of growth-related genes as per environmental conditions. This potential of BBX28 to fine-tune adaptive morphological plasticity may contribute to enhancing climate resilience.

INTRODUCTION

With the prevalence of global warming, temperature stress has become one of the most crucial environmental factors affecting plant growth and development. Temperature stress can be categorized into high temperature stress or heat stress and high ambient temperature stress (1, 2). Heat stress is a condition where plants get exposed to lethal levels of temperature, leading to protein denaturation, excessive reactive oxygen species accumulation, disruption of photosynthesis, and under severe conditions, bleaching and plant death (3, 4). However, high ambient temperature typically refers to 4° to 5°C above optimal temperature, which can modify plant morphology by a process known as thermomorphogenesis. These morphological changes include hypocotyl elongation, petiole elongation, hyponastic movement of leaves, root elongation, early flowering, and early leaf senescence (1, 5). Hypocotyl elongation is believed to keep the leaves and apical meristem above the hot soil surface, and hyponasty provides a cooling effect to plants by reducing direct radiation from the Sun to the leaves (6, 7). Hence, these phenotypic changes help in making plants more adaptive toward a slight increase in temperature of the surrounding. Understanding the molecular basis of thermomorphogenesis is therefore critical for developing climate-resilient crops under global warming scenarios.

Morphological changes induced by high ambient temperature are governed by a coordinated signaling cascade involving temperature sensors, downstream transcription factors, hormonal regulators, and cell elongation machinery (1, 5, 7). Many components of thermomorphogenesis overlap with light signaling pathways, reflecting the tight integration of light and temperature cues in plant development (8, 9). Evolutionarily, temperature signaling is thought to have emerged from preexisting light signaling pathways, as early land plants encountered simultaneous changes in light intensity and temperature. Rather than evolving entirely new mechanisms, plants adapted light-responsive components that are inherently sensitive to temperature because of their structural properties (10, 11).

Photoreceptors are the most common example of integrators of light and temperature signaling. Phytochromes, especially PHYTOCHROME B (phyB), which are red light photoreceptors, also act as thermosensors (12, 13). phyB exists in two morphological forms, Pr and Pfr (14–16). Temperature can accelerate the active Pfr–to–inactive Pr conversion. This temperature-based conversion is known as thermal reversion (13, 17). Other photoreceptors, including CRYPTOCHROMES (CRYs), ZEITLUPE (ZTL), UV RESISTANCE LOCUS 8 (UVR8), and phyA, further contribute to sensing the high ambient temperature (12, 18–20). The bZIP transcription factor ELONGATED HYPOCOTYL 5 (HY5), which is a key positive regulator of photomorphogenesis, also regulates thermomorphogenesis but in a negative manner. In the light signaling pathway, HY5 is induced in a phyB-dependent manner, while its stability is negatively regulated by the COP/DET/SPA proteins (21–23). hy5 mutants exhibit hypersensitivity to high ambient temperature, and HY5 suppresses the expression of growth-related genes under high ambient temperature (24–26).

At the center of this integrated network is PHYTOCHROME INTERACTING FACTOR 4 (PIF4), a master transcription factor that consolidates inputs from photoreceptors, circadian components, and hormonal cues (27). In Arabidopsis, under ambient temperature conditions (20° to 23°C), phyB in its active Pfr form interacts with PIF4 and restricts its activity, partly through sequestration in photobodies (28). Elevated temperature (27° to 29°C) accelerates photobody disassembly and the release of active PIF4, thereby promoting elongation growth (17, 29). In addition to direct regulation by photoreceptors, PIF4 also gets regulated by circadian clock components (1, 30). EARLY FLOWERING 3 (ELF3), a core component of the Evening Complex (ELF3-ELF4-LUX), represses PIF4 transcription and can directly inhibit its activity (31). At elevated temperature (27° to 29°C), ELF3 undergoes phase separation into inactive condensates, relieving repression on PIF4 and enabling growth responses (32). Additional clock components, including TIMING OF CAB EXPRESSION 1 (TOC1), PSEUDO-RESPONSE REGULATOR 5 (PRR5), and GIGANTEA (GI), also regulate PIF4 at transcriptional and posttranslational levels, highlighting its circadian control (30, 33–35). Components of the COP/DET/SPA E3 ubiquitin ligase complex that play a key role in negatively regulating photomorphogenesis promote PIF4 stability in the dark to promote thermomorphogenesis (25, 36, 37). The CUL3-E3 ubiquitin ligase LIGHT-RESPONSE BRIC-A-BRACK/TRAMTRACK/BROAD (LRB) degrades HY5 and promotes PIF4-mediated thermomorphogenesis (38). There is some evidence to suggest that the B-box–containing BBX proteins, important regulators of the light signaling pathway, can modulate thermomorphogenesis (39–41). BBX18 and BBX23 are known to regulate the ELF3 protein level by promoting its ubiquitination and degradation to promote thermomorphogenesis (42). Furthermore, BBX24 and BBX25 bind and repress ELF3 to promote PIF4 accumulation and activate thermomorphogenesis (43). In contrast, BBX21 suppresses thermomorphogenesis by enhancing phyB- and ELF3-mediated inactivation of PIF4 (44).

While a lot is known about the regulation of PIF4 stability under high ambient temperature, modulation of its activity is less characterized. PIF4 acts as a central transcriptional regulator coordinating multiple physiological responses under high ambient temperature (27). It promotes growth by activating auxin biosynthesis YUCCA (YUC) genes, including YUC8 and YUC9, and Tryptophan Aminotransferase of Arabidopsis (TAA) family members like TAA1. It also activates auxin signaling genes such as Indole-3 Acetic Acid Inducible (IAA) genes like IAA19 and Small Auxin Up-regulated RNA (SAUR) genes like SAUR19, driving hypocotyl elongation and hyponastic growth (45, 46). Several of these auxin biosynthesis and signaling genes contain the cis-regulatory G-box (CACGTG) element in their promoters. PIF4 can directly bind to these G-boxes and regulate the expression of these genes (27, 47). Direct regulation of IAA and SAUR genes by HY5 is also well established in light-mediated development (48). Several studies suggest that PIF4 and HY5 regulate light- and temperature-dependent growth in opposite ways (25, 26). PIF4 and HY5 antagonistically regulate a common set of target genes by competing for binding to the same cis-regulatory elements within their promoters (26, 47, 48). However, what regulates the differential binding of PIF4 or HY5 on the common promoter elements under different temperature conditions is not known.

BBX28 is a member of the group V B-box (BBX) family of zinc-finger transcription factors in Arabidopsis thaliana. Previous studies have shown that BBX28 regulates multiple developmental processes including early plant development through the COP1-HY5 module as well as later plant development like flowering (24, 49, 50). Here, we report that under high ambient temperature, BBX28 expression and accumulation are enhanced. Our study reveals that BBX28, which is known to inhibit photomorphogenesis (50), positively regulates thermomorphogenesis. bbx28 seedlings are hyposensitive to temperature-mediated hypocotyl elongation, whereas the overexpressors show enhanced hypocotyl elongation under elevated temperature (28°C). BBX28 promotes thermosensory growth in a PIF4-dependent manner. PIF4 promotes BBX28 expression, and BBX28 in turn interacts with PIF4 to promote PIF4-mediated activation of auxin signaling. BBX28 interacts with HY5 and inhibits its binding to the promoters of IAA19 and SAUR19 under elevated temperature (28°C). BBX28 thus forms a key regulator of the competitive binding and antagonistic regulation of growth-inducing genes by HY5 or PIF4, thereby offering adaptive plasticity under changing environmental conditions.

RESULTS

High ambient temperature induces BBX28 to promote thermomorphogenesis

BBX28 exhibits a gradual increase in expression levels under high ambient temperature with approximately eightfold up-regulation after 24 hours under 28°C (Fig. 1A). pBBX28:GUS expressing seedlings exposed to 28°C show enhanced GUS expression after 24 hours of treatment (Fig. 1B and fig. S1B). In Col-0 plants, BBX28 protein levels (detected using an anti-BBX28 antibody raised specifically to detect endogenous BBX28) also show a gradual increase upon exposure to 28°C with high accumulation seen after 24 hours of treatment (Fig. 1C and fig. S1C). Next, we investigated the physiological relevance of the increased expression and accumulation of BBX28 at higher temperature. We compared temperature-regulated morphological changes such as hypocotyl elongation, petiolar elongation, and hyponasty in loss- and gain-of-function mutants of BBX28 and the wild type (WT). Col-0, bbx28-1, and 35S:BBX28 seedlings were kept under 22°C for 2 days and then shifted to 28°C for 5 days under long-day condition before imaging. bbx28-1 is hyposensitive, while 35S:BBX28 is hypersensitive to temperature-mediated hypocotyl elongation (Fig. 1, D to F). Similar hyposensitivity is seen in another mutant allele bbx28 in the Col-3 background (fig. S1, D and E), and hypersensitivity was observed in another overexpressor line 35S:GFP-BBX28 (fig. S1, F and G). We also checked hypocotyl elongation in short day conditions and found that bbx28-1 is hyposensitive to the increase in temperature, while 35S:BBX28 exhibits longer hypocotyl length compared to Col-0 (fig. S1, H and I). Furthermore, bbx28-1 is insensitive to the temperature-mediated increase in hyponastic angle, while the overexpressor, which already exhibits a high hyponastic angle of around 20° at 22°C, further enhances the angle to ∼60° at 28°C (Fig. 1, G and H). The hyponastic angle in the WT is ∼25° at 28°C (Fig. 1, G and H). Col-0 and bbx28-1 exhibit similar petiole length; however, 35S:BBX28 exhibits substantially longer petiole length compared to Col-0 (Fig. 1I). The petiole phenotypes of bbx28-1 and 35S:BBX28 suggest that BBX28 might not be necessary because of its redundancy with other BBXs but sufficient to modulate petiole elongation under high temperature. On the other hand, we believe that overexpression can regulate petiole elongation indirectly via some other factor that might be regulated by additional modulators besides BBX28. Propidium iodide staining indicates that at 28°C, the increase in cell length in the hypocotyl of 7-day-old 35S:BBX28 seedlings is significantly greater compared to Col-0, while it is less in bbx28-1 (Fig. 1, J and K). Together, these results suggest that BBX28 promotes thermosensory growth.

Fig. 1. BBX28 is induced under high ambient temperature and promotes thermomorphogenesis.

(A) RT-qPCR showing the relative expression of BBX28 in 5-day-old Col-0 seedlings transferred to 28°C for 6, 12, and 24 hours. The graph represents n = 4 with two technical replicates each. Error bars represent SEM. A one-way analysis of variance (ANOVA) was performed to check the significance level. Each alphabet represents a different significance group. (B) GUS expression in pBBX28:GUS seedlings grown for 5 days in a long day, treated at 22° and 28°C for 24 hours before taking images, and stained with 0.5 mM X-gluc. (C) Western blot showing protein levels of BBX28 in Col-0 using an anti-BBX28 antibody after 6, 12, and 24 hours (h) of treatment with 22° or 28°C. 0 hours indicates the BBX28 protein expression level before the plants were transferred to 28°C or maintained at 22°C. Actin is used as a loading control. Ladder size is provided to the left of the blots. Numbers below the blot represent the relative intensity of BBX28 versus actin. (D to F) Col-0, bbx28-1, and 35S:BBX28 seedlings grown for 5 days under 22° and 28°C and their hypocotyl lengths. (G) Pictorial representation of hyponasty and petiole length of Col-0, bbx28-1, and 35S:BBX28. The illustration beside (F) represents the hyponastic angle. Hyponastic angle (H) and petiole length (I) of the genotypes after 5 days of treatment. n = 3. (J and K) Propidium iodide–stained hypocotyls of Col-0, bbx28-1, and 35S:BBX28 seedlings grown for 5 days under 22° and 28°C. The error bar represents SEM for all graphs, except for (H) where it represents SD. n = 3. In (E), (H), (I), and (K), alphabets above graphs represent statistical groups as determined by a two-way ANOVA using Tukey’s post hoc test. In (F), Student’s t test was performed to check the significance level. **P < 0.005 and ****P < 0.0001.

PIF4 promotes BBX28 expression by directly binding to its promoter under high ambient temperature

PIF4 is a central regulator of thermomorphogenesis. We analyzed publicly available RNA sequencing datasets and found 13 BBX genes, including BBX28, to be down-regulated in pif4-2 (51). Furthermore, chromatin immunoprecipitation (ChIP) sequencing data indicate that BBX28 is a direct target of PIF4 (52). To validate these data, we performed reverse transcription quantitative polymerase chain reaction (RT-qPCR) to check BBX28 expression in Col-0 and pif4-2 grown under 22° and 28°C, respectively. While Col-0 showed three- to fourfold up-regulation of BBX28 under high ambient temperature, no induction was observed in pif4-2 (Fig. 2A). Similarly, luciferase assays using pBBX28:LUC as a reporter and 35S:PIF4 as an effector showed that PIF4 enhances the promoter activity of BBX28 at 28°C, while at 22°C, there is no significant change (Fig. 2B). This further confirms that PIF4 positively regulates the expression of BBX28 at 28°C. To examine whether PIF4 regulates the protein level of BBX28, we performed Western blotting with an anti-BBX28 antibody. Our data suggest that compared to Col-0, the BBX28 protein level is reduced in pif4-2 at 28°C (Fig. 2, C and D). As PIF4 is a transcription factor with a DNA binding ability, we hypothesized that PIF4 might directly bind to the promoter of BBX28 to regulate its transcription. The promoter of BBX28 contains a G-box (CACGTG) at −824 to −830 position, which is a potential binding site for PIF4. We performed in vitro electrophoretic mobility shift assay (EMSA) and found that PIF4 specifically binds to the G-box in the promoter of BBX28 and the binding is abolished when the G-box is mutated (Fig. 2E). To verify this binding in vivo, we performed ChIP using pPIF4:PIF4-GFP seedlings exposed to 22° or 28°C and checked the enrichment of PIF4 on the BBX28 promoter using a green fluorescent protein (GFP) antibody (Fig. 2F). We found that PIF4 is ∼25-fold enriched in one replicate and ∼10-fold enriched in the second replicate on the G-box–containing promoter fragment of BBX28 at the higher ambient temperature (Fig. 2F and fig. S1J). The TUBULIN promoter was used as the control for PIF4 binding (fig. S1K). All these data suggest that PIF4 promotes BBX28 expression by directly binding to the G-box in its promoter.

Fig. 2. PIF4 promotes BBX28 expression by directly binding to the G-box element in its promoter.

(A) BBX28 expression levels in Col-0 and pif4-2 at 22° and 28°C after 24 hours of exposure to 28°C. The graph represents data from n = 4 with two technical replicates each. Error bars represent SEM. Asterisks denote the significance level calculated using a t test with P < 0.05. ns, not significant. (B) Promoter activity of BBX28 using pBBX28:LUC as a reporter and 35S:PIF4 as an effector after treating the Col-0 protoplast at 22° and 28°C posttransfection. n = 3. The asterisk denotes the significance level calculated using a t test with P < 0.05. (C) Western blot showing protein expression levels of BBX28 in 5-day-old Col-0 and pif4-2 seedlings treated for 24 hours at 22° and 28°C. (D) Quantification of BBX28 protein expression levels in Col-0 and pif4-2 at 22° and 28°C with respect to actin. n = 3. A two-way ANOVA was performed to check the significance. Alphabets above the graphs represent statistical groups. (E) EMSA using a 50-bp fragment of the promoter of BBX28 containing the G-box as a probe. The mutated G-box was used as a negative control. Reaction components and the promoter probes containing the G-box element and mutated G-box are indicated above the blot. The asterisk represents the PIF4-probe complex. (F) ChIP assay showing the fold enrichment of PIF4-GFP on the BBX28 promoter using a GFP antibody. The position of the G-box relative to the start site of BBX28 CDS is mentioned above the graph. pPIF4:PIF4-GFP seedlings were grown for 5 days under long-day conditions and exposed to 22° or 28°C for 24 hours before fixing the samples for chromatin isolation. Four technical replicates from one biological replicate were used to get error bars representing SD. The second replicate is given in fig. S1J. Alphabets above the bar represent statistical groups derived using a two-way ANOVA with Tukey’s post hoc test.

BBX28 regulates thermomorphogenesis primarily in a PIF4-dependent manner

bbx28 is hyposensitive to a high ambient temperature–mediated elongation response similar to pif4, although less severe. Furthermore, PIF4 directly binds and regulates the expression of BBX28. These data suggest a potential genetic interaction between BBX28 and PIF4. To investigate this, we crossed bbx28-1 and 35S:BBX28 with pif4-2 to generate bbx28-1pif4-2 and 35S:BBX28/pif4-2, respectively, and compared their phenotype to Col-0, bbx28-1, and pif4-2 under 22° and 28°C. A high ambient temperature–mediated hypocotyl elongation response is severely defective in pif4-2 (Fig. 3, A and B). Under 28°C, the hypocotyl length of bbx28-1pif4-2 was similar to that of pif4-2, suggesting that BBX28 regulates hypocotyl elongation under high ambient temperature in a PIF4-dependent manner (Fig. 3, A to C). In 35S:BBX28/pif4-2 grown under 28°C, the hypocotyl length was similar to that of Col-0 (Fig. 3, A to C). Given that overexpression of BBX28 in the pif4-2 background rescues the insensitivity of pif4-2 to high ambient temperature, BBX28 might act downstream of PIF4 in the same pathway. Similarly, the hyponastic angle in bbx28-1pif4-2 did not show significant variation from bbx28-1 and pif4-2 (Fig. 3, D and E). In 35S:BBX28/pif4-2, overexpression of BBX28 rescues the hyponastic angle in pif4-2 to Col-0 levels (Fig. 3, D and E). We also observed that the increased hypocotyl cell length in 35S:BBX28 grown at 28°C is partially rescued in 35S:BBX28/pif4-2 seedlings (Fig. 3, F and G). This suggests that the BBX28-mediated hypocotyl elongation response under high ambient temperature is partially PIF4-dependent. Next, we checked the expression of several auxin biosynthesis and signaling genes known to have a role in cell elongation and thermomorphogenesis in Col-0, bbx28-1, and 35S:BBX28 (Fig. 3, H and I, and fig. S2). Of all the genes tested, we found that the expression of IAA19 and SAUR19 is significantly down-regulated in the mutant and up-regulated in 35S:BBX28 compared to Col-0 at 28°C (Fig. 3, H and I, and fig. S2). The expression of these genes is also down-regulated at 28°C in pif4-2 and the double mutant bbx28-1pif4-2 (Fig. 3, H and I). The up-regulation in expression in 35S:BBX28 at 28°C is reduced to Col-0 levels in 35S:BBX28/pif4-2, suggesting that BBX28 regulates IAA19 and SAUR19 in a PIF4 dependent manner (Fig. 3, H and I). Together, these results suggest that BBX28 regulates thermomorphogenesis primarily in a PIF4-dependent manner.

Fig. 3. BBX28 and PIF4 act in the same pathway to regulate thermomorphogenesis.

(A to C) Representative images and hypocotyl length of 2-day-old, long-day-grown seedlings of Col-0, pif4-2, bbx28-1, bbx28-1pif4-2, 35S:BBX28, and 35S:BBX28-1pif4-2 grown under 28°C for 5 days or kept at 22°C. (D and E) Representative images and quantification of hyponastic angle in the indicated genotypes after 5 days of exposure to 22° or 28°C. n = 3. Twenty-five to 30 seedlings were imaged for each biological replicate. (F and G) Propidium iodide–stained hypocotyl and hypocotyl cell lengths of Col-0, bbx28-1, pif4-2, 35S:BBX28, and 35S:BBX28pif4-2 seedlings under 22° or 28°C. n = 3. A minimum of 10 seedlings were imaged for each biological replicate. The error bar represents SEM, and alphabets above graphs represent statistical groups obtained using a two-way ANOVA with Tukey’s post hoc test. (H and I) Transcript levels of IAA19 and SAUR19 in indicated genotypes after treating 5-day-old, long-day-grown seedlings at 22° or 28°C for 24 hours. The graph represents data from three biological replicates with two technical replicates each. Error bars represent SEM. Alphabets above bars represent statistical groups derived using a two-way ANOVA with Tukey’s test.

BBX28 interacts with PIF4 and promotes PIF4-mediated activation of IAA19 and SAUR19

We further checked whether BBX28 can regulate the transcription or protein accumulation of PIF4 but did not find any significant regulation (fig. S3, A to C). To investigate how BBX28 and PIF4 together regulate thermomorphogenesis, we next checked whether BBX28 physically interacts with PIF4. Our yeast-two hybrid assay and bimolecular fluorescence complementation (BiFC) assay suggested that the two proteins bind to each other (Fig. 4, A and B). We further validated this interaction by in vitro and semi–in vivo pull-down assays (Fig. 4, C and D). For the in vitro pull-down assay, we used maltose-binding protein (MBP)–PIF4 recombinant protein as bait and HIS-TF-BBX28 recombinant protein as prey. For the semi–in vivo pull-down assay, HIS-TF-BBX28 recombinant protein was used as bait, and crude extract of PIF4-GFP was used as prey. Both the assays indicated a physical interaction between BBX28 and PIF4 (Fig. 4, C and D). We also tested the interaction between BBX28 and other known regulators of thermomorphogenesis like PIF7 but could not detect any interaction (fig. S4C). We further found that BBX28 is unable to directly bind to the promoters of IAA19 and SAUR19 (fig. S4, A and B). Our previous results indicated that BBX28 regulates the expression of IAA19 and SAUR19 in a PIF4-dependent manner. We hypothesized that the physical interaction between BBX28 and PIF4 might modulate PIF4 activity. To verify this, we checked the expression of IAA19 and SAUR19 in pPIF4:PIF4-GFP and pPIF4:PIF4-GFP/bbx28-1. We observed that the expression of both genes at 28°C is significantly reduced upon mutating BBX28 in the pPIF4:PIF4-GFP background (Fig. 4, E and F). To further confirm whether BBX28 regulates the direct binding of PIF4 to the promoters of downstream genes, we performed ChIP. We checked whether PIF4 is enriched on the promoters of IAA19 and SAUR19 at 28°C and whether mutation in BBX28 can modulate this binding. We found that PIF4 is enriched on promoter fragments of IAA19 and SAUR19 containing the G-box under high ambient temperature (Fig. 4, G and H). However, in pPIF4:PIF4-GFP/bbx28-1, PIF4 enrichment at 28°C is significantly reduced (Fig. 4, G and H). We did not find any enrichment on coding sequence (CDS) probes used as the control (fig. S3, D and E). This suggests that BBX28 promotes binding of PIF4 on the promoters of IAA19 and SAUR19 to regulate their expression. PIF4 and BBX28 therefore form a feed forward loop to regulate thermomorphogenesis.

Fig. 4. BBX28 interacts physically with PIF4 and promotes its activity.

(A) Yeast two-hybrid assay showing the interaction of BBX28 with PIF4. PIF4ΔN54-BD represents PIF4 protein from which 54 amino acids were removed from the N terminus to remove autoactivation. (B) BiFC assay to check the interaction between BBX28 and PIF4 using BBX28 fused with nYFP (pCL112) and PIF4 fused with cYFP (pCL113). Both constructs were infiltrated together in N. benthamiana leaf, with empty vectors in different combinations as mentioned on the left side of the panels and imaged using a confocal microscope. Blue dots represent DAPI (4′,6-diamidino-2-phenylindole) signals, indicating the nucleus, and the green dot represents the YFP signal, indicating an interaction. (C) In vitro pull-down assay using MBP-PIF4 and HIS-TF-BBX28 recombinant proteins as bait and prey, respectively. Empty MBP was used as a negative control. Input and pulled proteins of HIS-TF-BBX28 are detected using anti-His. Asterisks represent the band of interest. (D) Semi–in vivo pull-down assay using HIS-TF-BBX28 as bait and PIF4-GFP as prey. Pulled PIF4 was detected using a GFP antibody. Equal loading of plant protein (prey) was checked using an actin antibody, and the HIS-TF-BBX28 and HIS-TF (baits) empty control was detected using a His antibody. (E and F) qPCR to check transcript levels of IAA19 and SAUR19 in indicated genotypes after exposing the seedlings to 28°C for 24 hours. The graph represents n = 3 with two technical replicates each. Error bars represent SEM. (G and H) ChIP assay using pPIF4:PIF4-GFP and pPIF4:PIF4-GFP/bbx28-1 seedlings treated at 28°C for 24 hours before fixing the samples for chromatin isolation. n = 2 each with two technical replicates were used for the ChIP assays. The error bar represents SEM. Alphabets above graphs represent statistical groups as determined by a two-way ANOVA using Tukey’s post hoc test.

BBX28 interacts with HY5 and inhibits its transcriptional activity

BBX28 and HY5 physically interact as found in our yeast-two hybrid assay and also reported earlier (fig. S4C) (50). We also verified the interaction between BBX28 and HY5 via coimmunoprecipitation assay. In 35S:BBX28-Myc seedlings treated at 28°C for 24 hours, we found that BBX28 interacts with HY5, which was absent in bbx28-1 (Fig. 5E). Lin et al. (50) also reported that BBX28 negatively regulates the photomorphogenesis-promoting activity of HY5. Previous studies have shown that hy5 is hypersensitive to high ambient temperature and HY5 negatively regulates the expression of growth-related genes under elevated temperature (24–26). All these prompted us to ask whether BBX28 can modulate HY5 activity to regulate thermomorphogenesis. To investigate this, we first made a cross between bbx28-1 and hy5-215. We found that bbx28-1hy5-215 has similar hypocotyl length to hy5-215 (Fig. 5, A and B). Our RT-qPCR data indicated that the expression of IAA19 and SAUR19 is equally up-regulated in hy5-215 and bbx28-1hy5-215 at 28°C (Fig. 5, C and D). This suggests that in the absence of HY5, the growth-related genes are activated at 28°C, irrespective of the presence or absence of BBX28. We next wanted to check what happens when HY5 is overexpressed and whether BBX28 can modulate HY5 activity under high ambient temperature conditions. We performed a luciferase assay using pIAA19:LUC and pSAUR19:LUC constructs as reporters and 35S:HY5 and 35S:BBX28 as effectors under 28°C. We observed that when HY5 is overexpressed, the growth genes are not activated (Fig. 5, F and G). However, when the protoplasts were transfected with 35S:BBX28, induction of pIAA19:LUC and pSAUR19:LUC was seen. Overexpression of BBX28 could induce promoter activation even when HY5 was overexpressed, suggesting that BBX28 can inhibit HY5 transcriptional activity (Fig. 5, F and G). We also performed luciferase assay using protoplasts from the pif4-2 mutant to investigate whether the up-regulation of IAA19 and SAUR19 by BBX28 is PIF4-dependent. We found that when PIF4 is mutated, BBX28 is unable to induce IAA19 and SAUR19 expression (fig. S5, D and E). We further performed EMSA to check whether BBX28 can affect the binding ability of HY5 on the promoters of IAA19 and SAUR19 and found that HY5 binding to both IAA19 and SAUR19 promoters in the G-box is reduced upon increasing the concentration of BBX28 (Fig. 5, H and I). This suggests that the physical binding of BBX28 with HY5 might inhibit its binding to the IAA19 and SAUR19 promoters. HY5 and PIF4 bind to the same G-box element in the promoters of IAA19 and SAUR19. We therefore hypothesized that the absence of PIF4 might enhance the binding of HY5 to the promoter elements. Our ChIP data indicated that HY5 enrichment on the promoters of IAA19 and SAUR19 genes is enriched in pif4-2 compared to Col-0 (Fig. 5, J and K). We further confirmed that this enrichment is not due to the change in HY5 levels in Col-0 and pif4-2 (fig. S5A). Given that BBX28 can interact with both HY5 and PIF4, we hypothesized that BBX28 might modulate the differential binding of PIF4 and HY5 to the promoter G-box elements. Our ChIP data indicate that in the absence of BBX28, HY5 enrichment is enhanced on the growth gene promoters under high ambient temperature (Fig. 5, L and M). We could not detect any further enrichment in HY5 binding when PIF4 is also absent in addition to BBX28 (Fig. 5, L and M). The CDS of IAA19 and SAUR19 was used as a negative control for HY5 enrichment (fig. S5, B and C). This suggests that BBX28 might be the key regulator to outcompete HY5 in favor of PIF4 in binding and activating growth genes under high ambient temperature (Fig. 6).

Fig. 5. BBX28 acts in the same pathway as HY5 and inhibits its binding on IAA19 and SAUR19 promoters.

(A) Representative images of seedlings and (B) hypocotyl length and RT-qPCR analysis showing transcript levels of IAA19 (C) and SAUR19 (D) in Col-0, bbx28-1, hy5-215, and bbx28-1hy5-215 at 22° and 28°C. The graph represents data from three biological replicates with two technical replicates each. Error bars represent SEM. Alphabets above graphs represent statistical groups as determined by a two-way ANOVA using Fisher’s least significant difference test. (E) In vivo coimmunoprecipitation assay showing the interaction between 35S:BBX28-MYC and HY5. MYC-tagged BBX28 and the coimmunoprecipitated HY5 were detected using anti-MYC and anti-HY5 antibodies, respectively. bbx28-1 was used as a negative control, and actin served as an internal control. (F and G) Luciferase assay using pIAA19:LUC and pSAUR19:LUC as reporters and 35S:HY5 and 35S:BBX28 as effectors in different combinations, as mentioned below the graphs. Asterisks show significance levels with P < 0.05, as calculated using an unpaired t test. n = 2. (H and I) EMSA blots showing binding of HY5-His and effect of MBP-BBX28 on its binding on the probes of IAA19 and SAUR19, as indicated above the blot. The Sequences of probes and their mutated variants are mentioned further above. The asterisk indicates the HY5-probe complex. (J to M) ChIP showing the enrichment of HY5 on the G-box element in the promoters of IAA19 and SAUR19 at 22° and 28°C in Col-0 and pif4-2 (J and K) and Col-0, bbx28-1, and bbx28-1pif4-2, respectively (L and M). Five-day-old seedlings of respective genotypes are treated at 28°C and under control temperature conditions before fixing the samples. Two biological replicates, each with two or three technical replicates, were used for the ChIP assays. The error bar represents SEM. Alphabets above bars show statistical groups as determined by a two-way ANOVA using Tukey’s post hoc test.

Fig. 6. Mechanism of BBX28-mediated regulation of thermosensory growth.

High ambient temperature promotes PIF4 accumulation, which binds to the G-box element of BBX28 and promotes its expression. High ambient temperature promotes PIF4 accumulation, which binds to the G-box element of BBX28 and promotes its expression. BBX28, upon getting accumulated, promotes PIF4 binding on the promoters of IAA19 and SAUR19 and inhibits HY5 binding on these promoters under high ambient temperature to regulate thermomorphogenesis. Created in BioRender. H. Kumar (2026), https://biorender.com/x35sa3n.

DISCUSSION

Here, we identified BBX28 as a positive regulator of thermomorphogenesis. We observed that high ambient temperature induces BBX28 expression and gradual protein accumulation with a maximum effect after 24 hours (Fig. 1, A to C). In our phenotypic studies using the mutant and overexpressor of BBX28, we found that overexpression of BBX28 enhances hypocotyl elongation and hyponasty, while its mutation reduces these thermomorphogenic traits (Fig. 1, D to H). We found that increased hypocotyl elongation in 35S:BBX28 is due to cell elongation in hypocotyl cells (Fig. 1, J and K). During hyponasty, cells of the abaxial side of the petiole elongate faster than those of the adaxial side, leading to the upward movement of the leaves. Although we did not check the cell length in the petiolar region of leaves, BBX28, which is also expressed in the petiole, might regulate this process (fig. S1B). Our results further indicate that BBX28 can regulate thermomorphogenesis under both long days and short days (Fig. 1, D to F, and fig. S1, H and I). In our study, we found that BBX28 promotes thermomorphogenesis by increasing transcription of few selected auxin signaling genes such as IAA19 and SAUR19 (Fig. 3, H and I, and fig. S2) (46, 53). We further found that PIF4 promotes BBX28 expression by directly binding to the G-box in its promoter. Furthermore, BBX28 physically interacts with PIF4 to promote its binding to the promoters of IAA19 and SAUR19 and induce their expression in a PIF4-dependent manner (Fig. 4). Hence BBX28 and PIF4 activate each other through a positive feed forward loop to regulate thermomorphogenesis. This multilevel regulation is emerging to be a common theme in the working of several BBX proteins (54–56).

Photomorphogenesis and thermomorphogenesis are distinct developmental programs that regulate plant growth in response to light and temperature, respectively (9–11). Despite being triggered by different environmental cues, their downstream signaling pathways exhibit strong similarity and share several common response genes. PIF4 serves as a central regulator of thermomorphogenesis, which integrates the temperature cues with inputs from other pathways to modulate growth responses (27, 53). Similarly, photomorphogenesis is governed by a complex regulatory cascade in which HY5 functions as a central hub (57). Several studies have indicated that both PIF4 and HY5 participate in light-dependent and temperature-dependent growth regulation, although in opposite ways (25, 26). While HY5 inhibits the growth-related genes under light to promote photomorphogenesis, PIF4 activates them to promote skotomorphogenic growth under darkness. HY5 is abundant under light conditions, whereas PIF4 levels are elevated in darkness (23, 58). The combined action of several key regulators helps maintain optimum levels of these proteins and modulate their activity as per the available light and temperature conditions. Some of the key regulators include the E3 ubiquitin ligase COP1 and the photoreceptor phyB. COP1 accumulates in the nucleus in the dark and degrades HY5 to promote skotomorphogenesis (23). On the other hand, PIF4 is degraded in the light in a phyB-dependent manner (17, 29). In thermomorphogenesis, PIF4 acts as the key regulator, and it is directly and indirectly regulated by several factors at the transcriptional and translational levels. CCA1, LHY, BZR1, and TCP positively regulate the transcription of PIF4, whereas ELF3, TOC1, and BBX21 are negative regulators of PIF4 expression (31–33, 44, 59–61). The stability of the PIF4 protein is promoted by the COP-DET-SPA complex (23), while phyB and ELF3 are the key negative regulators of PIF4 stability (32, 37). Many BBX proteins, like BBX18, BBX21, BBX23, BBX24, and BBX25, promote PIF4-mediated thermomorphogenesis by indirectly inhibiting ELF3 (42–44). While a lot is known about the regulation of PIF4 stability under high ambient temperature, modulation of its activity is less characterized. A few examples of regulation of PIF4 activity include those by ELF3, CRY1 under blue light, and HFR1 (31, 59, 62). hy5 mutants are hypersensitive to high ambient temperature, and HY5 inhibits the expression of growth-related genes under high ambient temperature, suggesting that HY5 is a negative regulator of thermomorphogenesis (24–26). HY5 is induced in a phyB-dependent manner, while its stability is negatively regulated by the COP/DET/SPA proteins (21–23).

The antagonistic roles of PIF4 and HY5 in regulating thermomorphogenesis seem to be regulated at multiple levels. While there is some evidence of HY5 negatively regulating PIF4 transcription under high ambient temperature, the lack of such regulation has also been reported (26). Some previous reports suggest that HY5 protein abundance decreases at 28°C, leading to an increase in PIF4 transcription and protein accumulation, thereby regulating thermomorphogenesis (25, 38). We checked PIF4 and HY5 expression levels as well as protein levels at different time points (6, 12, and 24 hours). The 24-hour time point indicates the time just after the transition to light following the dark period. Seedlings harvested at this time point are exposed to very early light just before and during the harvesting process. We found significant up-regulation of PIF4 expression under 28°C compared to 22°C at 12 and 24 hours but not at 6 hours (fig. S5I). This probably represents the gradual temperature-mediated induction of PIF4, as reported earlier as well (26). HY5, on the other hand, is induced only at 24 and 6 hours (fig. S5J). The induction of HY5 at 24 and 6 hours might represent an early light- and temperature-responsive event, while the absence of induction at 12 hours might reflect regulation by BBX28 under elevated temperature to inhibit HY5 autoregulation. In addition, at the protein level, we found an increase in the levels of PIF4 and a decrease in HY5 levels at 28°C compared to 22°C at the 6- and 12-hour time points (fig. S5K). At the 24-hour time point, we found decreased PIF4 accumulation, which is probably because of the negative regulation of PIF4 protein by components of evening complex during the night. The accumulation of HY5 at the 24-hour time point might reflect early light- and temperature-mediated regulation (fig. S5L). These results suggest that under high ambient temperature, PIF4 and HY5 are regulated at multiple levels. Furthermore, PIF4 and HY5 antagonistically regulate a common set of target genes by competing for binding to the same cis-regulatory elements within their promoters (26, 47, 48). Previous studies have shown that PIF4 and HY5 compete for binding to the same G-box elements in the promoters of auxin signaling genes (7, 26, 47, 48). PIF4 promotes their expression, while HY5 represses them. Thus, the differential occupancy of the same promoter elements by PIF4 or HY5 determines the transcriptional output and hence suggests direct competition between the two proteins. However, what regulates the differential binding of PIF4 or HY5 on the common promoter elements under different temperature conditions remains elusive. Here, we show that BBX28 is a key determining factor that tilts the balance in favor of PIF4 under high ambient temperature to promote thermomorphogenesis. The BBX28-HY5 interaction leading to negative regulation of HY5 activity seems to be a common theme in photomorphogenesis and thermomorphogenesis [(50) and this study].

Our study provides information regarding the molecular mechanisms underlying thermomorphogenesis, highlighting the role of BBX28 as a key regulator in this pathway. BBX28 acts at the level of transcriptional complex formation and promoter occupancy and thus modulates the activity of PIF4 or HY5 in regulating growth-inducing genes. The discovery of BBX28 as a modulator of PIF4-HY5 competition adds an additional layer to the understanding of how plants balance light and temperature signaling to optimize growth and development. Under high ambient temperature, BBX28 tilts the balance in favor of PIF4, promoting PIF4 activity and suppressing HY5 activity to regulate thermomorphogenesis. BBX28 therefore regulates the HY5-versus-PIF4 competitive promoter binding and antagonistic regulation of growth-related genes as per environmental conditions. This potential of BBX28 to fine-tune adaptive morphological plasticity can have significant implications for understanding plant adaptation to environmental stress and developing strategies for sustainable agriculture in the face of climate change.

MATERIALS AND METHODS

Plant material and growth conditions

For this study, we have used the Col-0 ecotype as WT control plants, unless otherwise stated. bbx28-1 (SAIL_828_G11/CS877171) of the Col-0 background, bbx28 (SAIL_412_A09) of the Col-3 background, pif4-2 (SAIL_1288_E07), and hy5-215 were ordered from the ABRC seed stock center and used as knockdown mutant lines of respective genes. 35S:BBX28 and 35S:GFP-BBX28 are overexpressor lines of BBX28 that are generated in the lab. The 35S:BBX28-Myc line generated in the lab was used to check the protein level. pPIF4:PIF4-GFP and pPIF4:GUS lines were provided by A. Ranjan (63). All the double mutants, pPIF4:PIF4-GFP/bbx28, pPIF4:GUS/bbx28-1, bbx28-1/hy5-215, bbx28-1pif4-2, and 35S:BBX28/pif4-2 lines, are generated in the lab by crossing.

For thermomorphogenesis assays, seedlings of different genotypes were first germinated for 2 days under ambient long-day conditions and then transferred to 28°C for 5 days under the same photoperiod to examine morphological changes in aerial tissues, including hypocotyl and cotyledons. For gene expression, ChIP, and protein abundance analyses, seedlings were grown under standard long-day conditions for 5 days and subsequently shifted to 28°C for 24 hours before sample collection. Any deviations from these standard growth conditions are described in the corresponding figure legends and Results.

Construction of plasmid and transgenic lines

All the transgenic lines generated and used here including 35S:BBX28, 35S:GFP-BBX28, and 35S:BBX28-Myc are in the Col-0 background. To produce the transgenic lines 35S:BBX28 and 35S:GFP-BBX28, the CDS of BBX28 was amplified with primers containing B1 B2 extensions (table S1) and cloned into pDONR207 entry and, ultimately, to pCAMBIA1300 and pGWB6 destination vectors using gateway cloning. BBX28-pCAMBIA1300 and BBX28-pGWB6 were then transformed into Agrobacterium. Col-0 plants were transformed by floral dipping into a solution of transformed Agrobacterium, as explained in (64). A similar method was used to generate the 35S:BBX28-Myc line. The stop codon was removed while amplifying the CDS of BBX28. The destination vector pGWB17 was used to generate the BBX28-pGWB17 construct, resulting in fusion of a 4×Myc epitope tag at the C terminus of BBX28.

To generate the MBP-PIF4 construct, the full-length CDS of PIF4 was amplified and cloned into the pMAL-c2X vector using the conventional restriction digestion method with enzymes Eco R1 and Xba 1. To generate the HIS-TF-BBX28 construct, the full-length CDS of BBX28 was amplified using cDNA and cloned into the pCOLD-TF vector using Bam H1-Eco R1 sites. To generate YFP^N-BBX28, YFP^C-PIF4, and YFP^C-CO constructs, full-length CDSs of BBX28, PIF4, and CO are amplified and cloned in the pDONR207 vector using BP clonase and the respective pCL112 and pCL113 vectors using LR clonase. The same BBX28-pDONR207 construct was used to generate the AD vector BBX28-pGADT7, N-terminal GFP tag vector BBX28-pGWB6, and 35S constitutive promoter vector BBX28-pCAMBIA1300 with respective vectors using LR clonase in separate reactions. To generate pBBX28:GUS, gateway cloning was used. The promoter of BBX28 is amplified and cloned in the pDONR207 entry vector and lastly to the pGWB3 destination vector. The promoters of IAA19 and BBX28 are amplified and cloned in Kpn 1-Pst 1 and SAUR19 in the Kpn 1-Bam H1 site of pGREEN II 0800-LUC to prepare pIAA19:LUC, pBBX28:LUC, and pSAUR19:LUC constructs, respectively, for our luciferase assay. We made PIF4-pGADT7 earlier in our lab using the full-length PIF4 CDS, and it was giving autoactivation in yeast, so we removed 54 amino acids from the N terminus of PIF4, as suggested in a previous report (65). The fragment was cloned in the entry vector pDONR207 and lastly in the destination vector pGBKT7 using gateway cloning.

Hypocotyl length measurement

Arabidopsis seeds of Col-0, bbx28-1 mutant, and 35S:BBX28 overexpressor line as well as bbx28-1pif4-2 and 35S:BBX28/pif4-2 are inoculated in round plates having ¹/₂ MS medium, 1% sucrose, and 0.8% agar. After inoculation, plates were kept at 4°C for 3 days of stratification and then kept in a growth chamber having 22°C and ∼90 μmol s⁻¹ for 2 days for proper germination of all seeds. After germination, of two plates of each genotype, one is left in a 22°C chamber, and other one was transferred to the chamber with the same fluence of light but an increased temperature of 28°C for giving plants high ambient temperature for 5 days. After 5 days, plates were taken out, and imaging of 20 to 30 seedlings of each genotype was carried out by putting seedlings on a slide and observed under the bright-field microscope Leica M165FC. After imaging, hypocotyl length measurements were performed using ImageJ software.

Confocal microscopy

Seedlings of the above treatment were stained for 1 min with 10 μM propidium iodide and washed three or four times before imaging with the confocal microscope Olympus FV-3000. A 561-nm laser was used to excite propidium iodide with an emission wavelength of 617 nm. Images are then analyzed using ImageJ software, and the hypocotyl cell length is measured using a manual scale bar.

The relative protein level of PIF4 was analyzed by confocal microscopy in pPIF4:PIF4-GFP and pPIF4:PIF4-GFP/bbx28-1 lines under 22° and 28°C. To enhance PIF4 protein accumulation, seedlings were grown under simulated sunlight at 22°C for 1 day and then shifted to simulated sunlight at 28°C for 3 days, as described in (66). The GFP signal was excited by a 488-nm laser, and emission was observed between 510 and 525 nm.

Measurement of hyponasty

Twenty to 30 seedlings of Col-0, bbx28-1, 35S:BBX28, pif4-2, bbx28-1pif4-2, and 35S:BBX28/pif4-2 were inoculated in square plates containing ¹/₂ MS + 1% sucrose media in each biological replicate and kept in an upright way to let seedlings grow at 22° and 28°C under long-day condition. Images of all the seedlings were taken using a bright-field microscope (Leica M165FC), and the petiole length was quantified from the base of leaf lamina to the start of the attachment point of the leaf with the hypocotyl using ImageJ.

Same seedlings were used to measure the hyponastic angle by measuring the angle at which cotyledons are raised from the horizontal plane of the seedling. The measurement style is shown beside Fig. 1F and modified from (53, 67). The angle is measured using the angle plug-in of ImageJ software from the tip of cotyledons to the attachment point of the leaf and hypocotyl relative to the horizontal plane perpendicular to the hypocotyl.

RNA isolation and RT-qPCR

For checking the induction of BBX28 at high ambient temperature, Col-0 seeds were inoculated after stratification and kept at 22°C in a plant growth chamber for 5 days. After 5 days of optimum growth, one plate is kept at 28°C in a separate chamber, and the other plate is kept at the same chamber at 22°C. Samples were collected at different time points for 24 hours. For checking the expression of different genes, seeds of Col-0, bbx28-1, pif4-2, bbx28-1pif4-2, hy5-215, bbx28-1hy5-215, 35S:BBX28, and 35S:BBX28/pif4-2 are sown in ¹/₂ MS + 1% sucrose plates, and the same procedure is used. Samples collected after different time points were ground in liquid N₂, and total RNA was isolated using TRIzol. RNA was quantified, cDNA was prepared from ∼1000 ng of RNA using the TAKARA primescript 1st strand cDNA synthesis kit, and 1 μl of cDNA was further used to perform RT-qPCR using TAKARA SYBR Green. For all the RT-qPCR experiments, the number of biological replicates is three or four. Two technical replicates were performed for each biological replicate. Error bars indicate the standard error of mean (SEM), unless otherwise stated in the figure legend. Primers used in qPCR are listed in table S1.

Expression analysis by the GUS (β-glucuronidase) assay

The pBBX28-pGWB3 line was prepared using the floral dipping method of Col-0 plants and selected until T3 generation. Seeds of pBBX28:GUS were inoculated in two ¹/₂ MS + 1% sucrose plates and were let to grow in a growth chamber in optimum condition for 5 days. pPIF4:GUS and pPIF4:GUS/bbx28-1 seedlings were also grown in the similar manner for 5 days. A five-day-old seedling of one plate was shifted to 28°C, and other was kept in an optimum temperature of 22°C. Five to 10 seedlings were harvested at each time point from both plates and kept for bleaching in 90% acetone. After bleaching, seedlings were washed with a GUS staining solution without a substrate (X-gluc) for 15 min and then treated in the dark on ice with a staining solution with a X-gluc substrate for 30 min under vacuum. Samples were kept at 37°C overnight in the same solution for enzymatic reaction and color development. The next day, samples were dehydrated in 20, 35, and 50% ethanol and fixed in formalin-aceto-alcohol fixative for 1 hour before visualizing under a microscope.

Antibody generation

A BBX28 antibody is generated using a 55-amino-acid-long unique fragment of BBX28, which lacks sequence homology with any other BBX protein. The amplified fragment is encoded by bases from 232 to 397 of the BBX28 CDS. The amplified product was cloned into pET28(a)+ using Bam H1 and Hind III sites. The construct was transformed into BL21 cells and induced at 37°C for 3.5 hours. After induction, protein was purified using Ni-NTA beads and concentrated using a 3-kDa Merck concentrator. One hundred micrograms of protein at a concentration of 1 mg/ml was given to the animal facility three times to inject into mice and generate the antibody.

Total plant protein isolation and protein blot quantification

Seedlings were grown for 5 days under ambient temperature before treating at 28°C for 24 hours under a long-day photoperiod of 16-hour light/8-hour darkness. Any change in the basic condition was mentioned in the figure legend of that result. The tissue was ground into a fine powder in liquid N₂, and total protein was extracted using buffer [50 mM tris-HCl (pH 6.8), 0.5% (w/v) SDS (sodium dodecyl sulfate), 10% glycerol, 150 mM NaCl, and 2 M dithiothreitol (DTT)] and 1× protease inhibitor cocktail (Sigma-Aldrich, St Louis, MO). The crude solution obtained was centrifuged at 12,000g for 20 min, and equal amounts of proteins were separated in 10% SDS–polyacrylamide gel electrophoresis. Anti-GFP (Invitrogen; 1:1000) and anti-ACTIN (Sigma-Aldrich; 1:1000) were used to perform Western blotting. Similar buffer was used to extract protein of Col-0 at different time points as mentioned in Fig. 1 and 5-day-old Col-0 and pif4-2 after 24 hours at 22° or 28°C (long day), and Western blot analysis was performed for respective figures using anti-BBX28 (1:500), anti-HY5 (Agrisera; 1:1000), anti-PIF4 (Agrisera-Goat; 1:1000), and anti-ACTIN (Sigma-Aldrich; 1:1000) antibodies.

Quantification of Western blot signals was carried out using ImageJ. Mean gray values of individual bands were measured and corrected by subtracting the background signal. The corrected intensity of each target protein band was normalized to the corresponding ACTIN signal, and the resulting values were used to calculate relative protein abundance.

Yeast-two hybrid assay

A protein-protein interaction study using yeast model was performed as described earlier in (68). BBX28-pGADT7 and PIF4ΔN54-pGBKT7 were transformed in yeast strain AH109 with empty vectors as well as in combination. The transformants were selected using auxotrophic media like medium lacking Leu and Trp, i.e., DDO (-Trp-Leu), and interaction was confirmed using more stringent selection medium lacking Trp, Leu, and Ade, i.e., TDO.

BiFC assay

BiFC was performed as descried earlier (69). The pCL112 and pCL113 plasmids with YFP^N and YFP^C fused to the CDS of BBX28, PIF4, and CO were transformed into GV3101 Agrobacterium cells; cells were harvested and resuspended in activation media; and 0.25OD cells were mixed as the indicated transformant pairs and were infiltrated into Nicotiana benthamiana leaves. A confocal laser scanning microscope (Olympus FV3000) was used to detect the yellow fluorescent protein (YFP) fluorescence signals after 2 days of darkness. YFP fluorescence was excited by a 488-nm laser, and emission was observed between 510 and 525 nm.

Pull-down assay

In vitro pull-down assay was performed as demonstrated earlier (56). Briefly, HIS-TF-BBX28 and MBP-PIF4 proteins were purified using Ni-NTA beads and MBP beads, respectively. MBP and MBP-PIF4 were bound to the MBP beads and used as bait, while prey proteins HIS-TF-BBX28 were allowed to bind to the bead-bound bait proteins. The protein bound to the beads was separated using 8% SDS gel, followed by blotting using anti-His (MA1-21315; Invitrogen) and anti-MBP antibodies (MA5-15922, Invitrogen).

For semi–in vivo pull-down assay, we grew pif4-2 mutant and pPIF4:PIF4-GFP plants in ¹/₂ MS + 1% sucrose in long-day condition at 22°C for 5 days and then shifted to 28°C for 24 hours before collecting the samples to enhance the PIF4-GFP levels. Samples were crushed in GTEN [1% glycerol, 20 mM tris (pH 7.5), 1 mM EDTA, and 150 mM NaCl] buffer with a protease inhibitor, and the supernatant was collected. Concentrations of plant proteins were calculated using the Bradford assay (Bio-Rad). Ni-NTA beads were used to bind lysate-containing bait proteins HIS-TF-BBX28 and HIS-TF as negative control in separate tubes, and 500 μg of plant proteins was incubated with the bound beads for 4 to 5 hours. Beads were then washed with wash buffer [20 mM tris (pH 7.5) 150 mM NaCl, 50 mM imidazole, 0.1% Tween 20, and 1 mM DTT] for five times, and then the complex was eluted using elution buffer (20 mM tris, 150 mM NaCl, and 250 mM imidazole). 4× SDS loading was carried out, and samples were heated at 98°C for 2 to 3 min and loaded on an SDS gel. The membrane after transfer was blotted with anti-GFP (Invitrogen), anti-actin (Sigma-Aldrich), and anti-HIS (MA1-21315; Invitrogen) antibodies.

Coimmunoprecipitation assay

For coimmunoprecipitation assays, total proteins were extracted from 5-day-old, long-day-grown bbx28-1 and 35S:BBX28-myc seedlings treated at 28°C for 24 hours using extraction buffer containing 50 mM tris-HCl (pH 7.5), 1 mM DTT, 150 mM NaCl, 0.1% Tween 20, 2 mM NaF, 2 mM MG132, and protease inhibitor cocktail. The extracts were clarified by centrifugation at 12,000g for 20 min at 4°C. Equal amounts of total protein were incubated with 3 μg of Myc antibody at 4°C for 12 to 15 hours with gentle rotation, followed by incubation with Protein A magnetic beads for an additional 2 hours at 4°C. The beads were washed five times with extraction buffer +0.5% Triton-X, and the bound proteins were eluted by boiling in 2× SDS sample buffer at 98°C. The supernatant was collected after binding as input. Immunoprecipitated proteins and input samples were separated by SDS–polyacrylamide gel electrophoresis and analyzed by immunoblotting using the indicated antibodies.

Luciferase assay

The protoplast was isolated using the method described in previous study (70). pIAA19:LUC, pBBX28:LUC, and pSAUR19:LUC constructs are used as reporters, and 35S:BBX28, 35S:HY5, and 35S:PIF4 constructs are used as effectors for respective experiments. Vectors of the reporter and effector were transfected in the isolated protoplast by polyethylene glycol–mediated transfection and incubated for 16 hours in the dark under 28°C. The internal control is Renilla luciferase. The Promega kit (E1910) was used to measure luciferase activity.

Electrophoretic mobility shift assay (EMSA)

MBP-BBX28 protein was induced with 0.2 mM isopropyl-β-d-thiogalactopyranoside at 28°C for 8 hours, and the induced protein was purified using affinity purification with MBP beads. His-HY5 was also purified using Ni-NTA beads. Purified proteins were used to perform EMSA on the biotinylated probes of IAA19 and SAUR19 containing light-responsive G-box elements. MBP-PIF4 was induced with 0.2 mM isopropyl-β-d-thiogalactopyranoside at 28°C for 8 hours, and the induced protein was purified using affinity purification with MBP beads. MBP without any fused protein was also purified and used as the negative control for MBP-BBX28 and MBP-PIF4. The purified protein was used to perform EMSA on the biotinylated probes of the BBX28 G-box element. The EMSA reaction was performed as described earlier (71) using LightShift Chemiluminescent EMSA reaction (Thermo Fisher Scientific).

ChIP qPCR

Seeds of mentioned genotypes in the figures were sterilized and grown on a ¹/₂ MS + 1% sucrose plate for 5 days under ambient long-day condition and then treated at 28°C for 24 hours before collecting samples. The harvested sample was cross-linked by fixing in 1% formaldehyde, and chromatin was isolated using a series of extraction buffers as mentioned in (72, 73), followed by sonication using five cycles of 30 s ON/30 s OFF at 4°C. Fragments of chromatin were then bound to the antibody and bead to pull the chromatin bound to protein. The DNA-protein-antibody complex was reverse cross-linked followed by protease treatment and DNA purification. The enrichment was checked through qPCR primers specific to the binding region of protein. Data analysis was done using fold enrichment calculation relative to input.

Statistical analysis

GraphPad Prism 8.0.2 is used to make graphs and perform statistical analyses. The description of statistical analysis is briefly explained in respective figure legends.

Supplementary Materials

This PDF file includes:

Figs. S1 to S5

Table S1