The Maize ZmBES1/BZR1-9 Transcription Factor Accelerates Flowering in Transgenic Arabidopsis and Rice

In model plants, the BRI1-EMS suppressor 1 (BES1)/brassinazole-resistant 1 (BZR1) transcription factors play vital roles in regulating growth, development, and stimuli response. However, the roles of maize ZmBES1/BZR1 members are largely unknown. In this research, the ZmBES1/BZR1-9 gene was ectopically expressed in Arabidopsis and rice for the phenotyping of flowering. We found that the complementation and overexpression of ZmBES1/BZR1-9 in bes1-D mutant and wild type Arabidopsis both resulted in early flowering that was about 10 days shorter than in the untransformed control under long-day conditions. In addition, there was no difference in the rosette leaf number between all transgenic lines and the control. Subsequently, the ZmBES1/BZR1-9 gene was overexpressed in rice. It was found that overexpression lines of rice exhibited early flowering with heading dates that were 8 days shorter compared with untransformed plants. Moreover, the results of RNA-seq and qRT-PCR showed that five flowering-regulated genes, namely At2-MMP, AtPCC1, AtMYB56, AtPELPK1, and AtPRP10, were significantly up-regulated in all complementary and overexpressing lines of Arabidopsis. Meanwhile, the results of RNA-seq showed that 69 and 33 differentially expressed genes (DEGs) were up- and down-regulated in transgenic rice, respectively. Four flowering-related genes, namely OsGA20OX1, OsCCR19, OsBTBN19, and OsRNS4 were significantly up-regulated in transgenic lines. To sum up, our findings demonstrate that ZmBES1/BZR1-9 is involved in controlling flowering and provide insights into further underlying roles of BES1/BZR1s in regulating growth and development in crops.


Introduction
In higher plants, flowering is an important developmental process and marks the transition from vegetative growth to reproductive growth [1]. The photoperiod, vernalization, autonomic, gibberellin, age, and temperature pathways are among the main flowering-regulatory cues in plants. But, as crucial steroidal hormones, brassinosteroids (BRs) have been confirmed to control various biological processes in plants, such as growth and development regulation as well as stress resistance [2,3]. The BR signal is perceived by the receptor BRI1 and the co-receptor BAK1 on the cell membrane and triggers the phosphorylation and dephosphorylation of various downstream proteins, including BSKs (BR SIGNALING KINASES), CDGl (CONSTITUTIVE DIFFERENTIAL GROWTH1F), BSU1 (BRI1-SUPPRESSORF), BIN2 (BR-INSENSITIVE 2) [4][5][6][7]. Importantly, the inactivated BIN2 cannot phosphorylate the BES1 (BRI1-EMS-SUPPRESSOR 1) and BZR1 (BRASSINAZOLE RESISTANT 1) transcription factors, which also can be dephosphorylated by PP2A [6,8,9]. Thereafter, the activated BES1/BZR1s accumulate in the nucleus and regulate the transcription of nuclear genes by directly binding to elements of the E-box (CANNTG) or BRREs (CGTGT/CG) in gene promoters to regulate plant growth and BR synthesis [10][11][12][13].

Expression of ZmBES1/BZR1-9 Accelerates Flowering in Arabidopsis
To explore the function of ZmBES1/BZR1-9 in the floral transition, the days from mination to flowering (DTF) of the complementary lines L9-3 and L9-5 were measu 100 mg/L hygromycin solution, and the insertion of ZmBES1/BZR1-9 in the rice genome was confirmed by PCR and RT-PCR, respectively ( Figure 2). The complementary lines L9-3 and L9-5 of the Arabidopsis mutant were produced in our previous study [21]. Hence, the complementary Arabidopsis lines L9-3 and 9-5, the overexpressing Arabidopsis lines OE9-2 and OE9-3, as well as the overexpressing rice lines R9-1 and R9-5 were used for phenotyping.

Expression of ZmBES1/BZR1-9 Accelerates Flowering in Arabidopsis
To explore the function of ZmBES1/BZR1-9 in the floral transition, the days from germination to flowering (DTF) of the complementary lines L9-3 and L9-5 were measured.

Expression of ZmBES1/BZR1-9 Accelerates Flowering in Arabidopsis
To explore the function of ZmBES1/BZR1-9 in the floral transition, the days from germination to flowering (DTF) of the complementary lines L9-3 and L9-5 were measured. As shown in Figure 3, under long-day (LD) conditions, the L9-3 and L9-5 lines exhibited early flowering and a significantly higher rate of flowering plants in the same growth stage compared with the bes1-D mutant, which showed delayed flowering compared with the WT. However, there was no difference in the total rosette leaf number (RLN) between the bes1-D, L9-3, and L9-5 lines. Meanwhile, the RLN of bes1-D was significantly higher than that of WT. As shown in Figure 3, under long-day (LD) conditions, the L9-3 and L9-5 lines exhibite early flowering and a significantly higher rate of flowering plants in the same growt stage compared with the bes1-D mutant, which showed delayed flowering compared wit the WT. However, there was no difference in the total rosette leaf number (RLN) betwee the bes1-D, L9-3, and L9-5 lines. Meanwhile, the RLN of bes1-D was significantly highe than that of WT. Meanwhile, to validate the early-flowering phenotype regulated by ZmBES1/BZR1 9, we generated the overexpressed lines OE9-2 and OE9-3 and conducted a phenotyp characterization as a complementary assay. Similarly, early flowering was also observe in overexpressing lines under LD conditions. The DTF of the OE9-2 and OE9-3 lines wa about 10 days earlier than that of WT. Compared with WT, the number of flowering plant in OE9-2 and OE9-3 was significantly higher from 55 days after sowing (Figure 4). At th same time, there was also no significant difference in RLN among different lines.
The above results indicate that the heterologous expression of the ZmBES1/BZR1gene accelerates floral transition but does not affect the vegetative growth in Arabidopsis Meanwhile, to validate the early-flowering phenotype regulated by ZmBES1/BZR1-9, we generated the overexpressed lines OE9-2 and OE9-3 and conducted a phenotype characterization as a complementary assay. Similarly, early flowering was also observed in overexpressing lines under LD conditions. The DTF of the OE9-2 and OE9-3 lines was about 10 days earlier than that of WT. Compared with WT, the number of flowering plants in OE9-2 and OE9-3 was significantly higher from 55 days after sowing (Figure 4). At the same time, there was also no significant difference in RLN among different lines.
The above results indicate that the heterologous expression of the ZmBES1/BZR1-9 gene accelerates floral transition but does not affect the vegetative growth in Arabidopsis.

ZmBES1/BZR1-9 Accelerates Flowering in Rice
To further analyze the role of ZmBES1/BZR1-9 in regulating flowering, the transgenic rice lines R9-1, R9-5, and the WT line were grown in Chengdu in summer (LD) and in Sanya in winter (SD). The results of phenotyping showed that there was no difference in heading date between the transgenic lines and the WT under LD conditions in summer. In contrast, compared with the WT, the transgenic lines R9-1 and R9-5 exhibited an earlier heading date and their flowering time was shortened by about 8 days under SD conditions in winter (Figure 7).

ZmBES1/BZR1-9 Accelerates Flowering in Rice
To further analyze the role of ZmBES1/BZR1-9 in regulating flowering, the transgenic rice lines R9-1, R9-5, and the WT line were grown in Chengdu in summer (LD) and in Sanya in winter (SD). The results of phenotyping showed that there was no difference in heading date between the transgenic lines and the WT under LD conditions in summer. In contrast, compared with the WT, the transgenic lines R9-1 and R9-5 exhibited an earlier heading date and their flowering time was shortened by about 8 days under SD conditions in winter (Figure 7).

Discussion
Previous studies show that BR signaling is an important pathway involved in plantflowering regulation. For instance, BR-insensitive and deficient mutants exhibit delayed flowering morphology, confirming that the components of the BR signal control the flowering transition [30,31,[39][40][41]. BES1/BZR1s are key hubs in the BR signal [10,13]. In the present study, we found that the complementation and overexpression of the ZmBES1/BZR1-9 gene in WT and mutant Arabidopsis resulted in early flowering under LD conditions compared with the control (Figures 3 and 4). Nevertheless, there was no significant difference in flowering time between the transgenic Arabidopsis and the control, which both exhibited exuberant vegetative growth of each line under SD conditions (Figure 5). The RLN is a key indicator of Arabidopsis flowering [42], but the transgenic plants showed no difference in RLN under LD and SD conditions ( Figure 5). Meanwhile, transgenic rice exhibited a shorter heading date and about 8 days earlier flowering than WT under SD conditions, but there was no difference under natural LD conditions (Figure 7). This indicates that the ZmBES1/BZR1-9 gene positively regulates flowering but does not affect the vegetative growth process.
It has been confirmed that BES1 acts as a positive regulator of photoperiodic flowering in Arabidopsis [33]. As previously reported, the overexpression of BES1 promotes flowering, and its mutant and RNAi plants delayed flowering and possess significantly higher

Discussion
Previous studies show that BR signaling is an important pathway involved in plantflowering regulation. For instance, BR-insensitive and deficient mutants exhibit delayed flowering morphology, confirming that the components of the BR signal control the flowering transition [30,31,[39][40][41]. BES1/BZR1s are key hubs in the BR signal [10,13]. In the present study, we found that the complementation and overexpression of the ZmBES1/BZR1-9 gene in WT and mutant Arabidopsis resulted in early flowering under LD conditions compared with the control (Figures 3 and 4). Nevertheless, there was no significant difference in flowering time between the transgenic Arabidopsis and the control, which both exhibited exuberant vegetative growth of each line under SD conditions ( Figure 5). The RLN is a key indicator of Arabidopsis flowering [42], but the transgenic plants showed no difference in RLN under LD and SD conditions ( Figure 5). Meanwhile, transgenic rice exhibited a shorter heading date and about 8 days earlier flowering than WT under SD conditions, but there was no difference under natural LD conditions (Figure 7). This indicates that the ZmBES1/BZR1-9 gene positively regulates flowering but does not affect the vegetative growth process.
It has been confirmed that BES1 acts as a positive regulator of photoperiodic flowering in Arabidopsis [33]. As previously reported, the overexpression of BES1 promotes flowering, and its mutant and RNAi plants delayed flowering and possess significantly higher RLN than WT under LD conditions [32,33,43]. Similarly, overexpression of MiRZFP34 resulted in early flowering in transgenic Arabidopsis but no difference in RLN was observed [44]. Likewise, it was previously proven that histone deacetylase HDA703 interacts with OsBZR1 to control rice growth and heading by inhibiting Ghd7 expression [34]. Hence, it is speculated that the different phenotypes of transgenic Arabidopsis and rice expressing ZmBES1/BZR1-9 under LD and SD conditions may be due to the Arabidopsis being an LD plant but the rice being an SD plant [42]. In Arabidopsis, under LD conditions, the CO (CONSTANS) protein accumulated and induced the expression of FT and its homolog TSF (TWIN SISTER OF FT) to promote flowering [45,46]. However, the CO-homologous Hd1 promotes early heading by up-regulating the expression of the FT-homologous Hd3a gene under SD conditions in rice [47,48]. Therefore, we speculate that the ZmBES1/BZR1-9 gene regulates flowering through different photoperiod-mediated pathways in Arabidopsis and rice.
Furthermore, we showed that ZmBES1/BZR1-9 up-regulated the expression of At2-MMP, AtPCC1, AtMYB56, AtGRDP2, and AtPELPK1, as well as of OsGA20OX1, OsCCR19, OsBTBN19, and OsRNS4, in transgenic Arabidopsis and rice, respectively (Figures 6 and 8). This can be explained by the binding of ZmBES1/BZR1-9 to E-box elements in these genes promoters to promote their expression (Table S2) because the ZmBES1/BZR1-9 protein localizes in the nucleus and functions as a transcription factor [21]. It was confirmed that BES1/BZR1s can directly bind to E-box or BRRE elements to regulate the transcription of target genes [10,13]. Previous studies confirmed that at2-mmp mutant, AtPCC1 and AtPELPK1 RNAi plants, and AtGRDP2 knockout lines showed delayed flowering, suggesting their positive roles in regulating flowering [49][50][51][52]. Likewise, PPC1 exhibits a circadian-regulated expression pattern, is involved in light-regulated development via interaction with the COP9 signalosome subunit 5, and regulates the flowering transition in the photoperioddependent pathway [52][53][54]. AtGRDP2 also regulates female gametophyte development via the auxin pathway, and plants overexpressing it show early flowering [50,55]. Meanwhile, BES1 directly represses AtMYB56, which positively regulates the quiescent center and cell division [56,57]. In rice, OsGA20OX1 influences GA levels, is expressed in reproductive meristems, and crosstalks with cytokinin to regulate growth and development [58,59]. GA also regulates flowering [60,61]. OsRNS4 is regulated by phytochrome (pyh) A-, B-, and C-mediated light signals in rice [62]. It has been shown that phyA, B, and C play crucial roles in plant flowering [63][64][65][66]. In addition, OsBTBN19 and its homolog, NPY1, encode BTB domain proteins, which are reported to regulate flowering [67,68], although the OsBTBN19 and OsCCR19 genes were not directly confirmed to regulate flowering.
The study suggests that the ZmBES1/BZR1-9 gene positively regulates flowering in transgenic Arabidopsis and rice and that its overexpression can be used to shorten the flowering period. Meanwhile, the mechanism of ZmBES1/BZR1-9 in regulating flowering in maize is unknown and will be revealed in our future studies.

Plants Materials and Growth Conditions
Arabidopsis thaliana (Col-0) and Oryza sativa (Nipponbare) were used for the overexpression of the ZmBES1/BZR1-9 gene. The complementary lines (L9-3 and L9-5) of the Arabidopsis bes1-D mutant were previously produced in our laboratory [21]. All Arabidopsis plants were grown in the growth chambers under an artificial LD photoperiod (14 h light/10 h dark, LD) or SD photoperiod (10 h light/14 h dark, SD) under 60-70% relative humidity at 22 • C. The rice seedlings were grown in the field in Chengdu in summer (from mid-May to mid-July; LD conditions with 13.4-14 h of light) and in Sanya in winter (from mid-Oct to mid-next March; natural SD conditions with 9.5-12 h light).

Vector Construction and Transformation
The specific primers 1300-F and 1300-R (Table S1) were designed using Primer5.0, synthesized at Sangon Biotech (Shanghai, China), and used to amplify the coding sequence (CDS) of the ZmBES1/BZR1-9 gene from the 35S::ZmBES1/BZR1-9-eGFP plasmid constructed previously [21]. The PCR products and the pCAMBIA1300 plasmid were digested using Hind III and BamH I. After digestion, the PCR products were subsequently cloned into Hind III and BamH I sites of the pCAMBIA1300 plasmid to generate 35S::ZmBES1/BZR1-9 using the ClonExpress II One Step Cloning Kit (Vazyme, Nanjing, China). The ZmBES1/BZR1-9 gene was driven by the 35S promoter and terminated by CaMV Poly(A).
The 35S::ZmBES1/BZR1-9 plasmids were introduced into the Agrobacterium tumefaciens strain GV3101 for the next transformation. To create overexpressed lines of Arabidopsis, the transformation of Arabidopsis and rice was performed by the floral dip method and Agrobacterium-mediated calli transformation, respectively [69,70]. After transformation, the seeds of transgenic Arabidopsis were screened using 50 mg/L kanamycin (Sigma, St. Louis, MI, USA) to screening for transformants. The transgenic rice plants were screened by using 100 ng/mL hygromycin B (Coolaber, Beijing, China). The positive Arabidopsis seedlings with kanamycin resistance and positive rice seedlings with hygromycin resistance were harvested individually. In the T 2 generation, the plants showed a 3:1 segregation for resistance/susceptibility to kanamycin or hygromycin and were self-pollinated to generate T 3 . Then, the seeds of each line were screened using the same methods. The homozygous lines were identified without segregation and used in the next study.

PCR and RT-PCR
The gDNA of the transgenic lines and the WT line was extracted using the CTAB method [71]. To confirm the insertion of ZmBES1/BZR1-9 in the genome of Arabidopsis and rice, a pair of specific primers, 9F and 9R, were designed, synthesized, and used to amplify a 911 bp fragment from ZmBES1/BZR1-9. Moreover, the total RNA of each line was extracted, and the gDNA was removed and reverse-transcribed to cDNA using the RNAiso Plus kit (Takara) and HiScript ® II 1st Strand cDNA Synthesis Kit (+gDNA wiper), respectively. The RT-PCR was performed to detect the transcription of the ZmBES1/BZR1-9 gene in transgenic lines. Meanwhile, the primer pairs qAf/qAr and qGf/qGr were designed, synthesized, and used to amplify AtACTIN2 and OsGAPDH genes, respectively, which were used as the internal controls. The sequences of primers used for RT-PCR are also listed in Table S1.

Phenotyping of Transgenic Lines
To analyze the flowering time of transgenic Arabidopsis, the seeds of the complementary lines, overexpressed lines, WT, and bes1-D mutant were sown in soil and cultured in the growth chambers under the conditions described above. The days from germination to flowering, the total rosette leaf number, and the percentage of flowering plants over the same time period were measured from 25 plants and used to monitor the flowering time, as described by Li et al. [32] and Wang et al. [33]. For the phenotyping of transgenic rice, the days from seed germination to the appearance of the first main panicle were counted and used to detect the heading date, as described by Lu et al. [44]. In each replicate, 20 plants of every line were scored. Three replicates of each experiment were performed in this study. All statistical data were calculated using GraphPad Prism and Microsoft Excel 2017 and were presented as the mean ± SE. Student's t-tests were used to analyze the significance of the data between the transgenic lines and the WT. * and ** represent p < 0.05 and <0.01, respectively.

RNA-Seq Analysis
The RNA-seq analysis was conducted as our previous study [21]. In brief, the total RNA was extracted from two-week-old seedlings of R9-1, R9-2, and WT using the RNAprep Pure Plant Kit. Then, each RNA sample was qualified by testing their quality and integrity and used for sequencing library preparation using the Bioanalyzer 2100 and the VAHTSTM mRNA-seq V2 Library Prep Kit, respectively. The library was sequenced at the Sanshu Biotechnology Company (Shanghai, China) using the Novaseq 6000 system. The sequencing data were analyzed as described by Sun et al. [25]. The raw data were evaluated and filtered by removing sequencing adapters and low-quality reads as well as contaminants to produce clean data using FastQC (version 0.11.2) and Trimmomatic (version 0.36). Then, clean data were aligned with the Arabidopsis genome (TAIR 10) using Hisat2 and used to assemble transcripts of each gene using StringTie. The read counts of each gene assembled by StringTie were used to identify DEGs using DESeq2 with a p-value < 0.05 and |FoldChange| > 2. The GO enrichment analysis was performed using KOBAS accessed on 15 February 2023 (http://kobas.cbi.pku.edu.cn/anno_iden.php).

qRT-PCR
The expression of candidate genes in transgenic lines was analyzed by qRT-PCR using PerfectStart ® Green qPCR SuperMix (TransGen, Beijing, China) in the CFX96 TM Real-Time System (Bio-Rad, Hercules, CA, USA), as described in our previous study [25]. The procedure of qRT-PCR consisted of a two-step temperature cycle with pre-degeneration at 95°C for 30 s, 39 cycles of degeneration at 95 for 5 s, and an extension step at 58 • C for 30 s. The temperature was set to increase to 95°C by 0.5°C/s at the end of each last cycle to differentiate between specific and non-specific amplicons. Likewise, the AtACTIN2 and OsGAPDH genes were amplified using the primer pairs qAf/qAr and qGf/qGr, respectively, and used as the internal controls. The relative expression level was normalized following the 2 −∆∆Ct method [72]. The sequences of these candidates were derived from the National Center for Biotechnology Information (NCBI) or the Arabidopsis Information Resource (TAIR) and used to design primers using primer-BLAST accessed on 5 November 2021 (https://www. ncbi.nlm.nih.gov/tools/primer-blast/index.cgi?LINK_LOC=BlastHome). The primers are listed in Table S1. The candidate genes include AtACTIN2 (AT3G18780), OsGAPDH (Os04g40950), At2-MMP (AT1G70170), AtPCC1 (AT3G22231), AtMYB56 (AT5G17800), AtGRDP2 (AT4G37900), AtPELPK1 (AT5G09530), OsCCR19 (Os09g0419200), OsBNTB19 (Os09g0420900), OsRNS4 (Os09g0537700), and OsGA20OX1 (Os03g0856700).

Conclusions
In conclusion, the objective of the study was to validate the role of the maize ZmBES1/ BZR1-9 transcription factor in regulating flowering time. The ZmBES1/BZR1-9 gene was ectopically expressed in Arabidopsis and rice for phenotyping. Our findings demonstrate that ZmBES1/BZR1-9 is a positive regulator that promotes flowering via multiple photoperiodmediated pathways, but it does not affect vegetative growth. Our results also suggest that the maize ZmBES1/BZR1-9 gene can be used to improve the flowering period via its overexpression and provide a reference for further underlying roles of BES1/BZR1 in regulating growth and development in crops.