The protein sequences of GmMYB133 and other CCA1-like transcription factors in soybean and Arabidopsis were downloaded from Phytozome¹ and TAIR². Multiple sequence alignment was performed by MEGA X and DNAMAN. The protein homology analysis was calculated using the online tool MUSCLE of EMBL-EBI³, and the heat map of the identity was generated using the TBtools software (Chen C. et al., 2020). The phylogenetic tree of GmMYB133 and other CCA1-like transcription factors was constructed by the maximum likelihood method with 1,000 bootstrap replications using MEGA X (Kumar et al., 2018). The number on the line indicates the branch length.

The Arabidopsis Col-0 and soybean (cultivar “Jilin 32”) were used in this study. Generally, Arabidopsis plants were grown in growth chambers under long days (16 h light/8 h dark) at 20°C with 70–80% relative humidity. To generate transgenic lines overexpressing GmMYB133, the full-length cDNA of GmMYB133 was amplified and cloned into the Gateway vector pEarleyGate101 through recombination. Transgenic Arabidopsis was generated using the floral dip method (Tsuda et al., 2012). Positive transgenic lines were screened on murashige and skoog (MS) with 10 μg/ml of glufosinate, followed by real-time PCR (RT-PCR) for confirmation with gene-specific primers. Primer information is listed in Supplementary Table 1.

To generate β-glucuronidase (GUS) reporter lines, the 1,605 bp fragment of ATG upstream of GmMYB133 was cloned and constructed into the Gateway vector pMDC162 to obtain pGmMYB133:GUS, and the floral dip method (Tsuda et al., 2012) was used to generate transgenic Arabidopsis. Positive transgenic lines were screened on MS with 50 μg/ml of hygromycin, followed by RT-PCR for confirmation. Primer information is listed in Supplementary Table 1.

To conduct histochemical GUS assay, the transgenic samples with pGmMYB133:GUS (7-day-old seedling, root, rosette leaf, cauline leaf, inflorescences, pod wall, and seeds) were collected and immersed in the GUS staining solution as previously described (Bian et al., 2017b). After vacuuming for 15 min, the samples were incubated in staining solution at 37°C overnight and then destained with 75% ethanol. To investigate the expression of GmMYB133 under drought and salt, 7-day-old transgenic Arabidopsis seedlings with pGmMYB133:GUS were separately transferred to MS solid medium with 150 mM NaCl or 300 mM mannitol, and samples were collected for histochemical GUS staining and GUS activity assay at 3 days after stress treatment, respectively.

Fluorometric GUS assay was performed according to the method described by Jefferson et al. (1987). In brief, Arabidopsis seedlings were homogenized in 1 ml extraction buffer and centrifuged at 12,000 rpm for 10 min at 4°C. Notably, 100 μl supernatant was used to perform the assay of GUS activity in 400 μl reaction buffer with 1 mM 4-methylumbelliferyl-β-D-glucuronide. Fluorescence was measured using a fluorescence spectrophotometer at an excitation/emission wavelength of 365/455 nm. The protein concentration in the extracts was determined according to the method of bicinchoninic acid (BCA) assay (Smith et al., 1985). GUS activity was counted as picomoles of 4-methylumbelliferone per minute per milligram of protein.

For the analysis of circadian rhythm, vernalized Arabidopsis seeds of untransformed (UT) control and T3 transgenic plants overexpressing GmMYB133 as well as soybean seeds were sown and grown under 12 h light/12 h dark at 22°C for 10 days and then transferred into the condition of constant light. Samples of soybean and Arabidopsis were collected every 4 h from 0 to 72 h, respectively, and then frozen in liquid nitrogen and stored at −80°C for RNA extraction.

Ten-day-old soybean seedlings were subjected to the treatments of cold, drought, and salt stresses as previously described (Li et al., 2015): For cold stress, the ground-above tissues were harvested at 0, 6, 12, 24, 48, and 72 h after seedlings were transferred to 4°C; for drought stress, the ground-above tissues were collected at 0, 2, 4, 6, 8, and 10 days of water stress; for salinity stress, the ground-above tissues were harvested at 0, 6, 12, 24, 48, and 72 h after 200 mM NaCl solution was applied to seedlings. Then, the collected samples were frozen in liquid nitrogen and stored at −80°C for RNA extraction. Each treatment had at least six pots of seedlings, and three biological replicates were performed for each treatment.

Arabidopsis seeds from GmMYB133-overexpressing transgenic lines and UT control were sterilized and planted on MS solid medium, kept at 4°C for 2 days, and grown in a growth chamber under 16 h light/8 h dark at 22°C. At least 30 9-day-old seedlings were used for hypocotyl length assay, and three biological replicates were performed. Statistical significance of the data was analyzed using independent-samples t-test. Error bars indicate SE and p-value < 0.01 (**).

Germination assay of Arabidopsis seeds was conducted as previously described (Li et al., 2018). In brief, Arabidopsis seeds from GmMYB133-overexpressing transgenic lines and UT control were sterilized and sown on MS solid medium with or without 150 mM NaCl, kept at 4°C for 2 days, and then cultured under a photoperiod of 16/8 h (light/dark) at 22°C. The germination rates were monitored for 10 days. Data were analyzed using Graphpad Prism 8 software⁴.

To investigate the response of adult plants to salt stress, 3-week-old transgenic Arabidopsis plants and UT control were supplied with or without 200 mM NaCl, respectively. After salt stress was applied to Arabidopsis plants for 9 days, samples were collected for the extraction and determination of MDA and total chlorophylls. Additionally, 7-day-old transgenic Arabidopsis plants and UT control were subjected to cold stress (−9°C) for 1 h according to the procedure described by Jiang et al. (2020), and phenotypic difference and MDA content were investigated.

Extraction and determination of MDA were conducted according to the method described by Wu et al. (2021). In brief, 1 g of leaves from GmMYB133-overexpressing transgenic lines and UT control were homogenized in 10% trichloroacetic acid (TCA) and centrifuged for supernatant, which subsequently was mixed with 0.6% thiobarbituric acid in 10% TCA. The mixture was boiled for 15 min and placed on ice immediately, followed by a centrifuge at 10,000 g for 20 min. The absorbance of supernatants was spectrophotometrically measured at 450, 532, and 600 nm, and the following formula was used to count the MDA content of each supernatant: [6.459 × (A532 − A600) − 0.569 × A450]/fresh weight. Chlorophylls were extracted from leaves using acetone. Total chlorophyll content was spectrophotometrically measured as previously described (Li X. Y. et al., 2020).

Three biological replicates were performed with three-technique replicates for the extraction and determination of MDA and chlorophylls. Statistical significance of the data was analyzed using independent-samples t-test. Error bars indicate SE and p-value < 0.01 (**).

Total RNAs of Arabidopsis and soybean were extracted using OminiPlant RNA Kit (CWBIO, China) and RNAprep Pure Plant Plus Kit (TIANGEN, China), respectively. The first-strand cDNA was generated using StarScript II First-strand cDNA Synthesis Mix With gDNA Remover Kit (GenStar, China). Quantitative real-time PCR (qRT-PCR) was performed using the Bio-Rad CFX Connect Real-Time PCR Detection System with the reagent of 2 × RealStar Green Fast Mixture (GenStar, China). AtACTIN8 and GmUBIQUITIN-3 (GmSUBI3) were used as the internal reference for Arabidopsis and soybean, respectively. The data were analyzed using Bio-Rad CFX Manager. Three biological replicates with three techniques were conducted for each sample. Primer information is listed in Supplementary Table 1. Statistical significance of the data was analyzed using independent-samples t-test. Error bars indicate SE and p-value < 0.05 (*) or < 0.01 (**).

The evening element was predicted using the online program PlantPAN 3.0 with TFmatrixID_0029 and TFmatrixID_0030.⁵

To perform Yeast One-Hybrid (Y1H) assay, the full-length cDNAs of two genes (GmMYB133 and AtPRR5) and the promoter fragments of eleven genes (AtPRR5, AtPIF4, AtBBX24, AtIAA19, AtIAA29, AtSAUR21, AtSAUR26, AtCAX3, AtEARLI1, AtMPK3, and AtAZI1) were amplified using gene-specific primers (Supplementary Table 1). The full-length cDNAs of GmMYB133 and AtPRR5 were constructed into the vector of pB42AD as baits (pB42AD-GmMYB133 and pB42AD-AtPRR5), while promoter fragments of the above eleven genes were cloned into the vector pLacZi as preys [pLacZi-pAtPRR5-1 and pLacZi-pAtPRR5-2, pLacZi-pAtPIF4(EE), pLacZi-pAtPIF4(G-box), pLacZi-pAtBBX24, pLacZi-pAtIAA19, pLacZi-pAtIAA29, pLacZi-pAtSAUR21, pLacZi-pAtSAUR26, pLacZi-pAtCAX3, pLacZi-pAtEARLI1, pLacZi-pAtMPK3, and pLacZi-pAtAZI1]. Different plasmid combinations were separately co-transfected into yeast competent cell EGY48, including pB42AD-GmMYB133/pLacZi-pAtPRR5-1, pB42AD-GmMYB133/pLacZi-pAtPRR5-2, pB42AD-GmMYB133/pLacZi-pAtPIF4(EE), pB42AD-GmMYB133/pLacZi-pAtPIF4(G-box), pB42AD-GmMYB133/pLacZi-pAtBBX24, pB42AD-GmMYB133/pLacZi-pAtIAA19, pB42AD-GmMYB133/pLacZi-pAtIAA29, pB42AD-GmMYB133/pLacZi-pAtSAUR21, pB42AD-GmMYB133/pLacZi-pAtSAUR26, pB42AD-GmMYB133/pLacZi-pAtCAX3, pB42AD-GmMYB133/pLacZi-pAtEARLI1, pB42AD-GmMYB133/pLacZi-pAtMPK3, pB42AD-GmMYB133/pLacZi-pAtAZI1, sixteen negative controls (pB42AD/pLacZi, pB42AD-GmMYB133/pLacZi, pB42AD-AtPRR5/pLacZi, pB42AD/pLacZi-pAtPRR5-1, pB42AD/pLacZi-pAtPRR5-2, pB42AD/pLacZi-pAtPIF4-EE, pB42AD/pLacZi-pAtPIF4-G-Box, pB42AD/pLacZi-pAtBBX24, pB42AD/pLacZi-pAtIAA19, pB42AD/pLacZi-pAtIAA29, pB42AD/pLacZi-pAtSAUR21, pB42AD/pLacZi-pAtSAUR26, pB42AD/pLacZi-pAtCAX3, pB42AD/pLacZi-pAtEARLI1, pB42AD/pLacZi-pAtMPK3, and pB42AD/pLacZi-pAtAZI1), and one positive control (pB42AD-AtRVE8/placZi-pAtPRR5).

The ChIP-Seq data of AtPRR5 (Nakamichi et al., 2012) were searched and obtained against Arabidopsis thaliana using AtPRR5 as an assayed protein with the Encyclopedia of Plant Genome (ENPG) database.⁶ Enrichment peaks for the binding of AtPRR5 to the genomic regions of AtLHY1, AtCCA1, AtPIF4, AtIAA19, and AtBBX24 were retrieved from the ChIP-Seq data of AtPRR5 and visualized through the program ENPG using the default parameters.

Previously, GmMYB133 was revealed as a CCA1-like player in the regulation of isoflavonoid accumulation (Bian et al., 2018). To provide additional cues of its functional roles, we searched its homologs in soybean and Arabidopsis against the Phytozome and TAIR databases using the amino-acid sequence of GmMYB133 as a query, and sixteen proteins were obtained accordingly, e.g., GmRVE5/6/8/8a and AtRVE3/4/5/6/8. Multiple sequence alignment indicated that GmMYB133 and its homologs harbor a conserved MYB domain and the consensus sequence SHAQ(Y/F)F (Supplementary Figure 1). Phylogenetic analysis showed that all the MYB proteins were clustered into three groups, and GmMYB133 was distributed at the same clade with GmRVE5 (Supplementary Figure 1). In soybean, the protein identities between GmMYB133 and its homologs ranged from 46.44 to 90.94%, and GmMYB133 shared the maximum protein identity with GmRVE5 (Figure 1 and Supplementary Figure 1), which has not been functionally characterized. In Arabidopsis, GmMYB133 was closely related to the RVE8 clade proteins (AtRVE3, AtRVE4, AtRVE5, AtRVE6, and AtRVE8) known as the players to control hypocotyl elongation, flowering, leaf petiole growth, or respond to abiotic stresses (Perez-Garcia et al., 2015; Gray et al., 2017; Li et al., 2019a). The identities between GmMYB133 and the AtRVE8 clade genes were up to 48.44%–56.86% at the protein level with AtRVE5 showing maximum identity (Figure 1).

Previously, the expression pattern of GmMYB133 was investigated in root, stem, leaves, nodules, flowers, and differently developing embryos (Bian et al., 2018). To explore the spatial expression pattern of GmMYB133, its promoter region was cloned and fused with the GUS reporter gene. The transgenic Arabidopsis lines were generated for histochemical staining. As shown in Figures 2A–H, GUS signal was clearly detected in seedling, root, hypocotyl, rosette leaf, cauline leaf, stigma, anther, and pod wall. Further observation indicated that the GUS signal mainly appeared in the vasculature of cotyledons, leaves, and roots. Since GmMYB133 is homologous to the AtRVE8 clade genes, the rhythmic oscillation of GmMYB133 expression was examined under the constant light condition. The gene GmLHY (Glyma.16G017400) was set as a positive control. Consequently, although GmMYB133 showed oscillating expression to some extent, 24 h rhythmicity was not observed for its expression (Figure 2I). As expected, a typical rhythmic expression was observed for GmLHY (Figure 2J).

Several studies showed that CCA1-like MYBs play important roles in response to diverse stresses (Lai et al., 2012; Li et al., 2019a; Chen S. et al., 2020; Yuan et al., 2021). To investigate if GmMYB133 is involved in response to abiotic stresses, its expression pattern was determined under cold, drought, and salt stresses. When soybean seedlings were subjected to 200 mM NaCl stress, the transcript accumulation of GmMYB133 was gradually increased as stress prolongs and reached its maximum level at 48 h after salt stress with 5.69-fold changes as compared with control (Figure 2K). When soybean seedlings were exposed to drought stress, the expression of GmMYB133 remained a stable level until 10 days after stress, and its transcript accumulation was pronounced by 2.25-fold as compared with control (Figure 2L). However, GmMYB133 showed no significant change when cold stress was applied to soybean seedlings (Figure 2M). Furthermore, the above GUS reporter Arabidopsis was used to investigate the expression of GmMYB133 under drought and salt stress. As compared with the normal condition, a stronger GUS signal was observed in young leaves of seedlings under salt stress. No obvious difference was shown between normal condition and drought stress (Supplementary Figure 2A). Meanwhile, GUS activity was measured under drought and salt stress using the whole seedlings. As shown in Supplementary Figure 2B, no significant difference was observed for GUS activity between normal condition and salt stress as well as drought stress. Subsequently, the young leaves of the above seedlings were collected to investigate the expression of the GUS gene using the qRT-PCR approach. As compared with the normal condition, the expression of the GUS gene was significantly enhanced under both salt and drought stress. Especially under salt stress, a 149.5-fold increase was observed (Supplementary Figure 2C).

To investigate its functional roles, GmMYB133-overexpressing Arabidopsis lines were generated. Four transgenic lines (named as OX-16, OX-8, OX-14, and OX-15) were chosen as representatives for further study. As shown in Figure 3A, transgenic plants showed a compact structure with short-petiole leaf, especially OX-8, OX-14, OX-15, and qRT-PCR analysis indicated that GmMYB133 was, in fact, overexpressed in the four transgenic lines (Figure 3B). Consistent with the structural compactness of transgenic lines, a relatively low level of GmMYB133 transcripts was observed in the transgenic line OX16 and high level in OX-8, OX-14, and OX-15 (Figures 3A,B). Further observation indicated that the transgenic lines OX-8, OX-14, and OX-15 showed a clear delay of flowering under the long-day condition (Figure 3C). Noticeably, the hypocotyl lengths of OX-16, OX-8, OX-14, and OX-15 seedlings were decreased to 72.2%, 55.6%, 58.3%, 58.3% of UT control (Figures 3D,E), respectively, suggesting that overexpression of GmMYB133 might inhibit hypocotyl elongation in Arabidopsis.

Numerous studies indicated that endogenous auxin and exogenous light signal antagonistically regulate hypocotyl growth (Xi et al., 2021). To provide some clues about how GmMYB133 can be involved in the regulation of hypocotyl elongation, seven genes were chosen to conduct expression analysis using the qRT-PCR approach, including two light-responsive genes (AtPIF4 and AtBBX24) and five auxin-associated genes such as AtIAA19, AtIAA29, AtSAUR1, AtSAUR21, and AtSAUR26. Molecular and/or genetic study has shown that AtPIF4, AtBBX24, AtIAA19, AtIAA29, and AtSAUR21 are positive regulators for hypocotyl elongation (Jiang et al., 2012; Spartz et al., 2012; Zhou et al., 2013; Shimizu et al., 2016; Li et al., 2021). As shown in Figure 3F, all the seven genes (AtPIF4, AtBBX24, AtIAA19, AtIAA29, AtSAUR1, AtSAUR21, and AtSAUR26) were transcriptionally suppressed by GmMYB133 overexpression. For example, the expressions of AtPIF4, AtIAA19, and AtIAA29 in three GmMYB133-overexpressing lines were decreased to 30.1%–49.1%, 25.0%–29.5%, and 12.1%–18.6% of UT control, respectively (Figure 3F). These observations suggested that GmMYB133 might act as a negative regulator in the regulation of hypocotyl elongation through governing the expressions of light-responsive and auxin-associated genes.

Since GmMYB133 showed salt-stress-inducible expression in soybean, its effect on plant tolerance to salt stress was investigated using the transgenic Arabidopsis with GmMYB133 overexpression. First, seeds of three transgenic lines and UT control were sown on solid MS medium supplemented with or without 150 mM NaCl, and then germination rates were dynamically surveyed from 1 to 10 days after vernalization treatment. As shown in Figures 4A,B, a similar germination tendency was observed between transgenic seeds and UT control under normal conditions, and germination rates reached nearly 100% at 4 days after vernalization treatment. However, in the presence of 150 mM NaCl, seed germination was obviously enhanced by GmMYB133 overexpression. As shown in Figures 4A,C, the germination rates for three transgenic lines were up to 68.0%–80.3%, whereas UT control only showed a 43.0% germination rate. Furthermore, phenotypic and physiological analyses of transgenic plants and UT control were performed in the presence of 200 mM NaCl. Consequently, the GmMYB133-overexpressing lines were insensitive to salt stress as compared with UT control. As indicated in Figures 4D,E, the leaves of three transgenic lines were greener than the UT control, and the content of total chlorophylls was significantly increased by GmMYB133 overexpression with 1.7–2.4-fold changes. Consistently, the MDA contents in the three transgenic lines were decreased to 9.3%–14.6% of UT control (Figure 4F). These results suggested that GmMYB133 might confer plant tolerance to salt stress in Arabidopsis. Additionally, transgenic plants and UT control were subjected to cold stress (−9°C) for 1 h, and phenotypic difference and MDA content were investigated accordingly. As shown in Supplementary Figure 3, no significant difference was observed in the aspects of phenotype and MDA content.

To provide some hints about how GmMYB133 can respond to salt stress, four positive regulators for salt tolerance were chosen to conduct expression analysis using the qRT-PCR approach, including AtCAX3, AtEARLI1, AtMPK3, and AtAZI1. Consequently, the expressions of the four genes were significantly enhanced by GmMYB133 overexpression to a different extent. Especially, the two genes AtCAX3 and AtEARLI1 were transcriptionally elevated by 8.9–66.8- and 6.2–8.2-fold changes, respectively. Previous studies indicated that AtPIF4 is not only involved in the regulation of light-associated development but also acts as a negative regulator in response to salt stress. Furthermore, the expressions of the above four positive regulators and AtPIF4 were investigated in the transgenic lines with GmMYB133 overexpression and UT control under salt stress (150 mM NaCl). Consistently, AtPIF4 was transcriptionally decreased by GmMYB133 overexpression under salt stress, while the expressions of AtEARLI1, AtMPK3, and AtAZI1 were remarkably increased in the transgenic lines. However, no change or a slight decrease was observed for AtCAX3 between the transgenic lines and UT control under salt stress. Further observation indicated that the expression of AtCAX3 was remarkably increased in UT control under salt stress (Figure 4H and Supplementary Figure 4), which is consistent with the previous report that AtCAX3 expression was strongly induced by salt stress (Shigaki and Hirschi, 2000). These results implied that GmMYB133 might confer plant tolerance to salt stress through regulating the expressions of salt tolerance-related genes such as AtEARLI1, AtMPK3, and AtAZI1 in Arabidopsis.

It has been known that hypocotyl elongation and salt tolerance are controlled by the circadian clock (Rawat et al., 2011; Nakamichi et al., 2016; Gray et al., 2017). To explore if GmMYB133 affects the circadian clock, the rhythmic expressions of three core oscillator genes (AtCCA1, AtLHY1, and AtPRR5) were investigated in GmMYB133-overexpressing Arabidopsis lines and UT control under continuous light using the qRT-PCR approach. Consequently, all the examined oscillator genes sustained approximately 24-h circadian rhythm in UT control. However, overexpression of GmMYB133 obviously altered the rhythmic expressions of AtCCA1, AtLHY1, and AtPRR5. As shown in Figures 5A,B, AtCCA1 and AtLHY1 showed a 4-h rhythmic phase shift and reduced oscillation amplitude in transgenic lines as compared with type. Noticeably, the rhythmic expression of AtPRR5 was extremely disrupted by GmMYB133 overexpression. As indicated in Figure 5C, the oscillation amplitude of AtPRR5 was remarkably reduced by GmMYB133 overexpression, and a relatively stable expression level was observed for AtPRR5 in GmMYB133-overexpressing Arabidopsis lines. On the whole, the trough level of AtPRR5 was elevated by GmMYB133 overexpression. These results indicated that GmMYB133 might serve as a regulator of core circadian oscillators, especially AtPRR5, to control diverse biological processes.

Since the expression level and endogenous rhythm of AtPRR5 were significantly altered by GmMYB133 overexpression in Arabidopsis, it is worth exploring if GmMYB133 acts as a transcription factor to directly regulate the expression of AtPRR5. First, two CCA1-like protein binding sites (AAATATCT and AAATATCC) were observed in the promoter region of the gene AtPRR5 (Figure 5D). Thus, Y1H assay was performed using GmMYB133 as bait and two fragments of AtPRR5 promoter (named as pAtPRR5-1 and pAtPRR5-2) as preys. Consequently, like the positive control (pB42AD-AtRVE8/placZi-pAtPRR5), blue colonies were shown when GmMYB133 served as bait and the two fragments of AtPRR5 promoter (pAtPRR5-1 and pAtPRR5-2) as preys. As shown in Figure 5E, the colonies harboring the plasmid combination of GmMYB133 and pAtPRR5-2 showed strong blue color, while a very weak blue color was observed for the colonies with the plasmid combination of GmMYB133 and pAtPRR5-1. No blue color appeared in the four negative controls (pB42AD/pLacZi, pB42AD-GmMYB133/pLacZi, pB42AD/pLacZi-pAtPRR5-1, and pB42AD/pLacZi-pAtPRR5-2) (Figure 5E). These results indicated that physical interaction could occur between GmMYB133 protein and AtPRR5 promoter.

GmMYB133 can affect the expressions of AtPIF4, AtBBX24, AtIAA19, AtIAA29, AtSAUR21, AtSAUR26, AtCAX3, AtEARLI1, AtMPK3, and AtAZI1. To explore whether GmMYB133 acts as a transcription factor to directly target these genes, we first predicted the binding site of CCA1-like MYBs, the evening element, in their promoter regions using the online program PlantPAN 3.0. As shown in Supplementary Figure 5A, all the ten genes harbored 1–4 evening element(s) in their promoter regions. Furthermore, the Y1H assay was performed using GmMYB133 as bait and the promoter fragments with evening elements as prey. Consequently, no blue colony was observed when GmMYB133 served as bait and the promoter fragments of the above ten genes as prey, while blue colonies appeared for the positive control (Supplementary Figure 5B).

To further address GmMYB133-mediated mechanisms that control hypocotyl elongation and response to salt stress in Arabidopsis, we retrieved the ChIP-Seq data of AtPRR5 (Nakamichi et al., 2012) to obtain its target clients through the online program ENPG (see text footnote 6). Consequently, the peaks of the AtPRR5 binding site were observed for two rhythm-associated genes (AtCCA1 and AtLHY1), two light-responsive genes (AtPIF4 and AtBBX24), and AtIAA19 (Figures 6A–E). Further observation indicated that the peaks were mainly distributed in the promoter and/or UTR regions. Furthermore, AtPIF4 was chosen as a representative to confirm the binding of AtPRR5 using Y1H assay. AtPRR5 usually prefers to bind to the element G-box (CACGTG), and it was observed that AtPIF4 harbored one G-box element in its promoter region (Figure 6F). Y1H assay indicated that blue colonies appeared when AtPRR5 served as bait and the fragment of AtPIF4 promoter as prey, while no blue colony was observed for the negative control (Figure 6G). These results suggested that GmMYB133 might affect hypocotyl elongation and salt tolerance directly through regulating the expression of AtPRR5, which subsequently alters the transcriptional accumulations of its direct targets such as AtPIF4, AtBBX24, and AtIAA19.