A recent study has demonstrated that MYB4 physically interacts with CAM7, and both CAM7 and HY5 bind to the MYB4 promoter to regulate its transcriptional activation (Dutta et al., 2024). To investigate whether MYB4 physically interacts with HY5, we conducted in vitro pull-down assays using poly-His and GST fusion proteins. GST, GST-CAM7, and GST-MYB4 were incubated with Ni-NTA beads to which HY5-6His was bound. The α-GST immunoblot of the pulled down proteins revealed that GST-MYB4 was selectively pulled down by HY5-6His similar to GST-CAM7, used a positive control (Abbas et al., 2014), however not with GST alone ( Figure 1A , Upper Panel). The same membrane was then stripped and re-probed using α-His to examine the equal loading of HY5 ( Figure 1A , Lower Panel). These results suggest that MYB4 and HY5 physically interact with each other.

MYB4 has two distinct domains; the N-terminal domain (9-116 aa) comprises of two MYB domains and the C-terminal domain (117-282 aa) comprises of transcriptional activation domain (Mitra et al., 2019) ( Figure 1B) . To determine the HY5 interacting domain of MYB4, we conducted yeast two-hybrid assays to examine the domain wise protein-protein interaction. As shown in Figure 1C , the C terminal domain of MYB4 interacted with HY5, however not the N-terminal domain. The negative controls involving empty vectors (AD and BD) and their respective combinations with HY5 and MYB4 (full length and truncated) did not exhibit any physical interaction. These results suggest that MYB4 physically interacts with HY5 through the C terminal domain.

To validate this physical interaction in vivo, we performed co-immunoprecipitation assay. The rosette leaves of 30-d old hy5 mutant (Oyama et al., 1997) and HY5 OE line (Lee et al., 2007) were infiltrated with Agrobacterium harbouring either GFP, GFP-CAM7 (positive control, Abbas et al., 2014), or GFP-MYB4 constructs, resulting in transient expression of these proteins. As shown in Figure 1D , HY5 was co-immunoprecipitated with GFP-CAM7 and GFP-MYB4 but not with GFP alone in HY5 OE background. However, in hy5 mutant background, used as negative control, HY5 wasn’t coimmunoprecipitated with any of the transiently expressed proteins. These results further confirm that MYB4 physically interacts with HY5.

MYB4 and HY5 work as positive regulators of photomorphogenesis at various wavelengths of light (Koornneef et al., 1980; Oyama et al., 1997; Ang et al., 1998; Chattopadhyay et al., 1998; Brown and Jenkins, 2008; Dutta et al., 2024). To further investigate the role of MYB4 in photomorphogenesis, we first examined whether MYB4 expression is induced during dark-to-light transition. The semi-quantitative and quantitative PCR analyses revealed that MYB4 expression was induced at various wavelengths of light during dark to light transition. As shown in Figures 2A–D and Supplementary Figure 2 , the level of induction varied with context to time and wavelength of light tested. We then determined the physiological significance of the physical interaction between MYB4 and HY5 through genetic studies. For that, the myb4 was genetically crossed with hy5 mutant, and the homozygous myb4 hy5 double mutant lines were generated Supplementary Figure 3). To determine the genetic relationship between MYB4 and HY5, 6-day-old wild-type (segregated WT of F2 population), myb4, hy5, and myb4 hy5 seedlings were grown in dark and at various wavelengths of light. Dark-grown seedlings exhibited hypocotyl length similar to the WT background ( Figures 3A, B ). The myb4 mutant seedlings displayed significantly increased hypocotyl length as compared to WT under white light (WL), blue light (BL), and far-red light (FRL) ( Figures 3C–H ). No such effect was observed under red light (RL) condition ( Figures 3I, J ). Consistent with the previous findings, hy5 mutant exhibited a significantly elongated hypocotyl compared to WT under all light conditions tested (Ang et al., 1998). The hypocotyl of myb4 hy5 double mutant was found to be significantly more elongated than each of the single mutants in WL, BL, and FRL, however not in RL ( Figures 3C–J ). These results suggest that MYB4 and HY5 work in an additive manner to control the hypocotyl length under WL, BL and FRL.

We then examined whether the genetic interaction between MYB4 and HY5 could influence the expression of light-inducible genes. To determine this, we assessed the transcript levels of RBCS-1A and CAB1 in 6-day-old WT, myb4, hy5, and myb4 hy5 seedlings grown under constant WL (30 μmol m^-2 s^-1). As previously reported, the expression of these genes was significantly decreased in hy5 background as compared to the WT (Ang et al., 1998; Chattopadhyay et al., 1998). However, as shown in Figure 3K , the additional mutation of MYB4 in hy5 mutant, did not alter the expression of RBCS-1A and CAB1, resulting in the expression pattern similar to that of hy5 in myb4 hy5 double mutant. These results collectively suggest that additional mutation in MYB4 in hy5 background does not affect the expression of light inducible genes.The accumulation of chlorophyll and anthocyanin is an important physiological response controlled by light signaling pathways (Lifschitz et al., 1990; Neff and Chory, 1998; Fankhauser and Casal, 2004). It has been reported earlier that hy5 mutant seedlings show a reduced level of chlorophyll and anthocyanin contents (Oyama et al., 1997; Shin et al., 2007). To examine the significance of the genetic interplay between MYB4 and HY5 in the accumulation of chlorophyll and anthocyanin, we assessed the chlorophyll and anthocyanin levels in 6-day-old WT, myb4, hy5, and myb4 hy5 double mutant seedlings grown in WL (30 µmolm^-2s^-1). As shown in Figures 3L, M , while chlorophyll content between myb4 and WT seedlings was similar ( Figure 3L ), an increase in anthocyanin accumulation was observed in myb4 mutants when compared to the WT seedlings ( Figure 3M ). The additional mutation of MYB4 in hy5 mutant seedlings did not alter the level of chlorophyll accumulation ( Figure 3L ), suggesting that HY5 regulates the chlorophyll accumulation independently of MYB4. However, as shown in Figure 3M , the anthocyanin content of myb4 hy5 double mutant was similar to WT background, suggesting that MYB4 and HY5 work antagonistically to regulate the anthocyanin accumulation.

Since HY5 binds to the MYB4 promoter and positively regulates its expression, and given that HY5 and MYB4 physically interact, we aimed to investigate the physiological relevance of this physical interaction in mediating the transcriptional regulation of MYB4. To investigate this, we analysed the DNA-protein interaction between the MYB4 promoter and HY5 in the absence or presence of MYB4. Previously it has been reported that the MYB4 promoter contains three E-Boxes, and HY5 specifically binds to the third E-Box (Dutta et al., 2024). We conducted electrophoretic mobility shift assays (EMSA) using purified GST-HY5 and GST-MYB4 fusion proteins, along with both wild-type and mutated versions of the MYB4 promoter fragments ( Figure 4A ). As shown in Figure 4B , HY5 exhibited strong binding to the wild-type MYB4 promoter fragment, however not to the mutated one. The binding affinity of HY5 to wild-type MYB4 promoter fragment is decreased in the presence of MYB4 ( Figure 4B , lane nos. 6 and 7). The reduction in HY5 binding affinity became more pronounced with increasing concentrations of MYB4. However, the DNA-protein complex of HY5 with the wild-type promoter fragment remained unchanged in the presence of GST alone ( Figure 4B , lane no. 4). Additionally, no DNA-protein complex was formed between MYB4 and the wild-type promoter fragment ( Figure 4B , lane no. 3).

To further gain insight into this regulation, we conducted transient expression assays in Arabidopsis protoplasts using a construct in which the wild-type MYB4 promoter (DNA fragment containing HY5 but not MYB4 binding sites) was fused to the β-glucuronidase reporter gene (Pro AtMYB4-GUS) ( Figure 4C ). As shown in Figure 4D , while HY5 increased the MYB4 promoter activity by ~2.2-fold, MYB4 alone did not alter the promoter activity. However, when both MYB4 and HY5 were introduced into the protoplast, MYB4 could drastically reduce the HY5 mediated increase in MYB4 promoter activity ( Figure 4D ). Since we omitted the MYB4 binding site in the transactivation assay, the negative regulation is probably not due to direct binding to the promoter which supports our EMSA results. These results collectively suggest that the physical interaction between MYB4 and HY5 modulates HY5-mediated transcriptional activation of MYB4.

Previous studies have shown that HY5 protein is stabilized by light (Ang et al., 1998; Wang et al., 2021; Fang et al., 2024) and cytokinin (Vandenbussche et al., 2007). As this study revealed the genetic interplay between HY5 and MYB4 during photomorphogenic growth, we wanted to explore the genetic interaction between MYB4 and HY5 upon cytokinin treatment. To examine this, we first analyzed whether cytokinin alters MYB4 expression. For this, we monitored MYB4 transcript levels in WT seedlings grown on MS plates, with or without 2 µM and 4 µM trans-zeatin. It was observed that at lower light intensity (15 µmol m^-2s^-1) MYB4 expression was significantly reduced at 4 µM trans-zeatin, however not at 2 µM, suggesting that higher concentrations of cytokinin represses MYB4 expression ( Supplementary Figure 4 ). Next, we examined the hypocotyl length of 6-day-old WT, myb4, hy5, myb4 hy5 mutants in both cytokinin-treated and untreated seedlings grown in the darkness. It was observed that the dark-grown seedlings of each genotype studied exhibited similar hypocotyl length in absence or presence of different concentrations of cytokinin. ( Supplementary Figure 5 ). These results suggest that mutation in MYB4 doesn’t alter the cytokinin mediated inhibition of hypocotyl elongation in dark.

To determine the effect of light and cytokinin on hypocotyl growth, we examined hypocotyl length of 6-day-old WT, myb4, hy5, and myb4 hy5 double mutant seedlings grown under lower and higher WL fluences on MS plates, with or without 1 µM, 2 µM, and 4 µM trans-zeatin. As shown in Figures 5A, C , under lower intensity of WL (15 μmol m⁻² s⁻¹), among the various cytokinin concentrations tested, only the highest concentration (4 µM) was able to rescue the etiolated growth of the myb4 mutant. However, at higher WL intensity (60 μmol m⁻² s⁻¹), both 2 µM and 4 µM cytokinin were able to suppress hypocotyl elongation, resulting in hypocotyl length in the myb4 mutant similar to that of the WT ( Figures 5B, D ). Consistent with earlier reports (Vandenbussche et al., 2007), we observed that similar to the control condition, hy5 mutants exhibited elongated hypocotyl as compared to the WT upon cytokinin treatment. Notably, the hypocotyl length of the myb4 hy5 double mutant was significantly higher than that of hy5 mutant under all the conditions tested ( Figures 5A–D ). These results suggest that the additional loss of MYB4 function alters the hy5 phenotype upon cytokinin treatment. At lower intensity of WL and lower cytokinin concentration, MYB4 and HY5 function additively to regulate hypocotyl length which is similar to the effect of WL alone. However, at the same light intensity but higher cytokinin concentration this genetic interaction is altered and MYB4 and HY5 exhibit synergistic mode of interaction. At higher light intensity even lower cytokinin concentration is sufficient for the cytokinin mediated synergistic interaction between MYB4 and HY5. Thus, the genetic interaction between MYB4 and HY5 is modulated by both light intensity and cytokinin levels.

We also measured the primary root length in WT, myb4, hy5, and myb4 hy5 double mutant seedlings grown under cytokinin treatment in darkness and under various intensities of white light. While cytokinin suppresses primary root length in WT in both dark and WL (Liu et al., 2022), our results showed no significant differences in inhibition ratios between WT and myb4 mutant seedlings across all conditions ( Supplementary Figures 6A–C ). Consistent with previous findings by Cluis et al. (2004), the hy5 mutant exhibited resistance to cytokinin treatment under both the intensities of the white light. The myb4 hy5 double mutant displayed primary root inhibition patterns similar to the hy5 mutant in white light, indicating that HY5 works independent of MYB4 in regulating primary root length upon cytokinin treatment ( Supplementary Figures 6B, C ).

Our study shows that while cytokinin rescues the etiolated growth of myb4 seedlings, the primary root length of myb4 remains similar to WT in both presence and absence of cytokinin. The ATP-binding cassette (ABC) transporter subfamily G14 (ABCG14) in Arabidopsis is primarily active in roots and is crucial for transporting cytokinin to the shoots. When ABCG14 expression is disrupted, it leads to significant stunted growth in the shoots, which can be reversed by applying trans-zeatin externally (Ko et al., 2014; Zhang et al., 2014). To explore whether altered ABCG14 levels is responsible for the suppression of the hypocotyl length of myb4, we assessed the transcript levels of ABCG14 in 6-day-old WT, myb4, hy5, and myb4 hy5 seedlings grown under low (15 μmol m^-2 s^-1) and high (60 μmol m^-2 s^-1) white light in MS or in MS with 2 μM and 4 μM trans-zeatin ( Figures 6A, B ). Under control conditions (without cytokinin), at both low (15 μmol m^-2 s^-1) and high intensities (60 μmol m^-2 s^-1) of WL, the expression of ABCG14 in the myb4 mutant is similar to that in the WT, whereas the hy5 mutants exhibit a significant upregulation (~2.2-fold increase) of ABCG14 expression compared to WT ( Figures 6A, B ). An additional mutation in MYB4 could suppress the elevated expression of ABCG14 in the hy5 mutant, resulting in expression levels similar to WT in the myb4 hy5 double mutant. Earlier it was reported that the expression of ABCG14 was induced by cytokinin treatment (Ko et al., 2014). When we assessed the transcript levels of ABCG14 in 6-day-old WT, myb4, hy5, and myb4 hy5 seedlings grown under constant low white light (15 μmol m^-2 s^-1) and under constant high white light (60 μmol m^-2 s^-1) in MS with 2 μM and 4 μM trans-zeatin, we observed that in low light intensity and low cytokinin concentrations, the expression of ABCG14 is significantly reduced in myb4 mutant than that of WT ( Figures 6A, B ). The expression level pattern of ABCG14 remains same like the untreated for both myb4 and hy5 mutant in all other conditions. Interestingly, unlike the untreated conditions, in myb4 hy5 double mutant, the expression of ABCG14 is similar to that of hy5 mutants. This suggests that upon 2 μM cytokinin treatment at low intensity of WL, HY5 works downstream to MYB4 to regulate the expression of ABCG14 ( Figure 6A ).

Sweere et al. (2001) suggested a potential mechanism through which cytokinin and light signaling pathways interact to control hypocotyl elongation through RESPONSE REGULATOR 4 (ARR4). To determine if ARR4 plays a role in the altered response of myb4 mutant seedlings, we measured ARR4 transcript levels in 6-day-old WT, myb4, hy5, and myb4 hy5 seedlings grown under continuous low white light (15 μmol m⁻² s⁻¹) and high white light (60 μmol m⁻² s⁻¹) conditions, both in untreated MS and in MS supplemented with 2 μM and 4 μM trans-zeatin ( Figures 6C, D ). The results show that ARR4 expression was significantly increased in the myb4 mutant but decreased in the hy5 mutant. In the myb4 hy5 double mutant, ARR4 expression was similar to that in the hy5 mutant, indicating that HY5 functions downstream of MYB4 in regulating ARR4 expression. As it was previously reported that the ARR4 gene expression is also induced by cytokinin (Sweere et al., 2001), we checked the transcript levels of ARR4 in 6-day-old WT, myb4, hy5, and myb4 hy5 seedlings grown under constant low white light (15 μmol m^-2 s^-1) and under constant high white light (60 μmol m^-2 s^-1) in MS with 2 μM and 4 μM trans-zeatin. In low light, the pattern of expression of ARR4 is similar to that of the untreated for both myb4 and hy5 mutants. Interestingly, the expression of ARR4 in myb4 hy5 double mutant is similar to that of myb4 mutants, suggesting that under cytokinin treatment in low light, MYB4 works downstream to HY5 to regulate the expression of ARR4. However, in high light, there were no significant differences in the ARR4 transcript in any cytokinin concentrations tested ( Figure 6D ). This suggests that the role of MYB4 and HY5 in the regulation of cytokinin induced ARR4 expression is specific to low intensity of WL.

Seedlings treated with exogenous cytokinins accumulate anthocyanin pigments, with increased transcript levels of CHALCONE SYNTHASE (CHS) gene, a key player in the anthocyanin biosynthesis pathway (Deikman et al., 1995). HY5 enhances anthocyanin biosynthesis by directly binding to CHS promoter resulting in its transcriptional activation (Ang et al., 1998; Chattopadhyay et al., 1998). Conversely, MYB4, a known negative regulator of anthocyanin biosynthesis, indirectly repress CHS expression (Wang et al., 2020). Given these observations, we asked how loss of MYB4 and HY5 functions influence CHS regulation under cytokinin treatment. To answer this, we grew 6-day-old WT, myb4, hy5, and myb4 hy5 seedlings under constant low white light (15 μmol m⁻² s⁻¹) and high white light (60 μmol m⁻² s⁻¹) on MS media without trans-zeatin and with 2 µM and 4 µM trans-zeatin ( Figures 6E, F ). Consistent with previous findings, we observed that in untreated conditions, CHS transcript level was significantly elevated in the myb4 mutant whereas it was significantly reduced in the hy5 mutant. In the myb4 hy5 double mutant, CHS expression was similar to the hy5 mutant, indicating that HY5 acts downstream of MYB4 in regulating CHS expression under control condition ( Figures 6E, F ). However, under both low and high cytokinin treatment across both light intensities, CHS expression was significantly reduced (~2-3 fold) in the myb4 mutant. In contrast, CHS expression patterns were similar in hy5 and myb4 hy5 double mutants across both treated and untreated conditions. These findings suggest that MYB4 plays opposing roles in the regulation of CHS expression depending on the presence or absence of cytokinin and HY5 works downstream to MYB4 in this regulation.

Arabidopsis thaliana plants used in this study were homozygous myb4 mutant (CS26404), a Ds transposon tagged mutant in Ler background (Dutta et al., 2024) and homozygous hy5-ks50 mutant, a T-DNA insertion mutant in Ws background (Oyama et al., 1997). The myb4 hy5 double mutants were created through genetic crosses using standard methods outlined by Yadav et al. (2005) by crossing homozygous myb4 mutant with hy5 homozygous mutant lines. The homozygosity of both the mutation in T2 generation were confirmed by the inability of gene-specific primer sets, which spanned the full length of MYB4 (MYB4FP and MYB4RP primers) and HY5 (HY5FP and HY5RP) to amplify the MYB4 and HY5 gene respectively from genomic DNA isolated from the plants. One homozygous myb4 hy5 double mutant plant, used for different experiments was additionally confirmed by assessing the transcript levels of both MYB4 and HY5 using RT-PCR analyses. To analyse double mutants from different ecotype backgrounds, a segregated WT line of Ws-Ler from the T2 generation was marked as WT and used as a control to compare phenotypic and molecular differences.

Arabidopsis seeds were surface-sterilized and sown on Murashige and Skoog (MS) plates, stratified in the dark at 4°C to break dormancy before being transferred to growth chambers. They were then grown under specific wavelengths of light at designated light intensities at 22 °C. For different phenotypic studies, seedlings were photographed, and their hypocotyl lengths were measured using ImageJ software. For cytokinin responsive experiments, seeds were sown on MS plates with different concentration of trans-Zeatin.

The full-length coding sequences (CDS) of CAM7 and MYB4 were cloned into the pGEX‐4T2 vector to produce fusion proteins with Glutathione S‐transferase (GST). The GST (negative control), GST-CAM7 (positive control) and GST-MYB4 were overexpressed and purified from E. coli BL21 (DE3) using Glutathione Sepharose 4B beads (Amersham Biosciences). The full-length CDS of HY5 was cloned into the pET‐20b (+) vector, adding a 6× Histidine tag to the C-terminus. The HY5-His protein was overexpressed in E. coli BL21 (DE3) and purified with Ni‐NTA Agarose beads (Qiagen). The in vitro binding assay was performed following the protocol from Senapati et al. (2019). For the assay, 2μg of HY5-His was added to microcentrifuge tubes containing Ni-NTA Agarose beads and incubated with in vitro binding buffer (50 mM Tris-Cl pH 7.5, 100 mM NaCl, 0.2% glycerol, 0.1% Triton-X100, 1 mM EDTA, 1 mM PMSF, 0.1% NP-40, and protease inhibitor cocktail; Sigma, Calcutta, India) for 2 hours at 4 °C. After washing off the unbounded proteins, GST, GST-CAM7 or GST-MYB4 was added in equimolar amounts and incubated overnight at 4 °C. Following incubation, the supernatant was collected by centrifugation, and the beads were washed three times with binding buffer. The beads were then resuspended in 5X sodium dodecyl sulfate (SDS) loading buffer and boiled for 10 minutes. Both the supernatant (5%) and pellet fractions were analyzed using 12% SDS–polyacrylamide gel electrophoresis (PAGE) and immunoblotted with anti-GST antibodies to detect protein interactions. The same membrane was then stripped and probed using α-His to examine the equal loading of HY5-His.

For the domain-wise interaction study of MYB4 with HY5, full-length CDS and truncated version of MYB4 was combined with the GAL4 DNA-binding domain (BD-MYB4- bait) and the full-length CDS of HY5 was combined with the GAL4 activation domain (AD-HY5- prey). The fusion constructs were then introduced into yeast AH109 strain using polyethylene glycol/lithium acetate transformation (Clontech), and the growth were assessed of the co-transformed yeast cells on 2D plates (lacking leucine and tryptophan) and 4D plates (lacking leucine, tryptophan, adenine, and histidine). As previously demonstrated by Ang et al., 1998, COP1 and HY5 were used as a positive control for this study.

Co-IP assay was performed as described previously (Senapati et al., 2019) with slight modifications. Full length CDS of MYB4 was cloned into pCAMBIA 1303 to yield a GFP fusion protein. Total protein was extracted from the leaves of 30-d old hy5 and HY5OE plants transiently expressing GFP, CAM7-GFP (Senapati et al., 2019) or MYB4-GFP using Co‐IP buffer (50 mM Tris–HCl pH 7.5, 150 mM NaCl, 5 mM EDTA, 0.1% NP-40, 10% glycerol and protease inhibitor cocktail). 500 μg of total protein for each sample was incubated with 15 μl of anti-GFP monoclonal antibody (Sigma) for 6h at 4 °C. To this 30 μl of pre blocked protein A‐agarose beads (Sigma) was added and incubated for 2 hr at 4 °C. Then the beads were washed three times with Co‐IP buffer, and after the last wash the beads were boiled in 5X protein loading dye for 10 min to elute the protein and along with that 5% input was run in SDS‐PAGE. Immunoblots were developed using anti-GFP antibody (Sigma, dilution- 1:2500) and co-immunoprecipitated protein was detected by stripping and re-probing with anti-HY5 antibody (Srivastava et al., 2015; Sigma, dilution- 1:5000).

Total plant RNA was isolated from the 6-day-old Arabidopsis thaliana seedlings grown under required light conditions using RNeasy plant mini kit (Qiagen). cDNA was synthesized from 1µg of total RNA using RevertAid H Minus First Strand cDNA synthesis Kit (Thermo Scientific). qPCR was then performed using Power SYBR^® Green PCR Master Mix (Applied 509 Biosystems) in StepOnePlusTM Real-Time PCR Systems using respective qPCR gene specific primers. ACTIN2, a housekeeping gene was used as reference gene to normalize the transcript level of the qPCR values. To analyze the relative changes in gene expression, the common 2^-ΔΔCT algorithm was used. First, ΔCt value is calculated by normalizing samples Ct to ACTIN2 Ct values. This value for different samples was then normalized to the Ct values of the experimental control and the ΔΔCt values were obtained. Fold expression was calculated by the formula, 2^-ΔΔCT, which was then plotted on the graph. The primers used for qPCR analysis has been mentioned in Table 1 .

The levels of chlorophyll and anthocyanin were determined following the protocol described by Holm et al. (2002). Specifically, for chlorophyll content estimation, 6-day-old seedlings grown under continuous white light (WL) were weighed and quickly frozen in liquid nitrogen. The frozen tissues were grounded in a 1.5 ml microcentrifuge tube, and chlorophyll was extracted in the dark using 80% acetone until the pellet became colourless. Chlorophyll a and b levels were calculated using MacKinney’s specific absorption coefficients: chlorophyll a = 12.7(A663) - 2.69(A645), and chlorophyll b = 22.9(A645) - 4.48(A663). The total chlorophyll content was expressed as micrograms of chlorophyll a and b per gram of fresh tissue weight.

For anthocyanin content estimation, 6-day-old seedlings grown under continuous WL were weighed, rapidly frozen in liquid nitrogen, and ground. Total plant pigments were extracted overnight in 0.3 mL of 1% HCl in methanol. After adding 0.2 mL of water, chlorophyll was separated from anthocyanin by adding an equal volume of chloroform. The anthocyanin level was assessed by spectrophotometric measurements of the aqueous phase (A530 - A657) and normalized to the total fresh weight of the tissue. Flowering times were determined as described by Yadav et al. (2005).

GST, GST-MYB4, and GST-HY5 proteins were overexpressed in E. coli BL21 (DE3) and affinity purified using Glutathione Sepharose 4B beads (Amersham Biosciences). For the DNA binding assays, wild type MYB4 promoter spanning the 3^rd E box and mutated 3^rd E box generated through primer-based site-directed mutagenesis, were used as probes. All of these fragments were cloned into the pBluescript SK+ vector, followed by PCR amplification and purification to produce the probes. Approximately 100 ng of the DNA fragment was incubated with the purified protein in a reaction mixture containing 1X binding buffer (15 mM HEPES, pH 7.5, 35 mM KCl, 1 mM MgCl2, 1 mM EDTA, pH 8.0, 2% glycerol, and 1 mM DTT) in a total volume of 20 µl. The incubation was carried out at room temperature for 20 minutes. Following incubation, the reactions were resolved on a 7% native polyacrylamide gel, stained with SYBR^® Green EMSA nucleic acid gel stain (Molecular Probe, Invitrogen), and visualized using the iBright 750 imaging system (Invitrogen) for image documentation.

Arabidopsis protoplasts were isolated and transformed following the protocol outlined by Yoo et al. (2007). Wild-type MYB4 promoter fragments containing the three E-boxes were PCR amplified and cloned into the pGAL-UAS-GUS vector. The full-length coding sequences (CDS) of HY5 and MYB4 were individually cloned into the pRTL2 vector to create the effector constructs. Both the reporter and effector constructs were introduced into Col-0 protoplasts. Following transfection, the protoplasts were incubated under 20 µmol/m²/s white light for 10 hours before harvesting. The GUS activity measurement was carried out as mentioned by Yoo et al. (2007) and Chakraborty et al. (2019).