FLOWERING LOCUS T (FT) is the key flowering integrator in Arabidopsis (Arabidopsis thaliana), and its homologs encode florigens in many plant species regardless of their photoperiodic response. Two FT homologs, GmFT2a and GmFT5a, are involved in photoperiod-regulated flowering and coordinately control flowering in soybean. However, the molecular and genetic understanding of the roles played by GmFT2a and GmFT5a in photoperiod-regulated flowering in soybean is very limited. In this study, we demonstrated that GmFT2a and GmFT5a were able to promote early flowering in soybean by overexpressing these two genes in the soybean cultivar Williams 82 under noninductive long-day (LD) conditions. The soybean homologs of several floral identity genes, such as GmAP1, GmSOC1 and GmLFY, were significantly upregulated by GmFT2a and GmFT5a in a redundant and differential pattern. A bZIP transcription factor, GmFDL19, was identified as interacting with both GmFT2a and GmFT5a, and this interaction was confirmed by yeast two-hybridization and bimolecular fluorescence complementation (BiFC). The overexpression of GmFDL19 in soybean caused early flowering, and the transcription levels of the flowering identity genes were also upregulated by GmFDL19, as was consistent with the upregulation of GmFT2a and GmFT5a. The transcription of GmFDL19 was also induced by GmFT2a. The results of the electrophoretic mobility shift assay (EMSA) indicated that GmFDL19 was able to bind with the cis-elements in the promoter of GmAP1a. Taken together, our results suggest that GmFT2a and GmFT5a redundantly and differentially control photoperiod-regulated flowering in soybean through both physical interaction with and transcriptional upregulation of the bZIP transcription factor GmFDL19, thereby inducing the expression of floral identity genes.
Citation: Nan H, Cao D, Zhang D, Li Y, Lu S, Tang L, et al. (2014) GmFT2a and GmFT5a Redundantly and Differentially Regulate Flowering through Interaction with and Upregulation of the bZIP Transcription Factor GmFDL19 in Soybean. PLoS ONE 9(5): e97669. https://doi.org/10.1371/journal.pone.0097669
Editor: Erik Souer, Vrije Universiteit Amsterdam, Netherlands
Received: March 20, 2014; Accepted: April 14, 2014; Published: May 20, 2014
Copyright: © 2014 Nan et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: The authors confirm that all data underlying the findings are fully available without restriction. All data are included within the manuscript.
Funding: National Natural Science Foundation of China (31071445, 31171579, 31201222 and 31371643); the Open Foundation of the Key Laboratory of Soybean Molecular Design Breeding, Chinese Academy of Sciences; the “Hundred Talents” Program of the Chinese Academy of Sciences; the Strategic Action Plan for Science and Technology Innovation of the Chinese Academy of Sciences (XDA08030108); and the Heilongjiang Natural Science Foundation of China (ZD201001, JC201313). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Plants integrate various environmental signals, such as photoperiod and temperature, to ensure flowering under those conditions that optimize seed production. In Arabidopsis thaliana (Arabidopsis), multiple pathways converge on a small number of floral integrator genes, which include the floral promoters FLOWERING LOCUS T (FT) and TWIN SISTER OF FT (TSF), to integrate photoperiod, temperature, vernalization, and light quality signaling . FT and TSF are members of a family of proteins similar to the mammalian phosphatidylethanolamine-binding domain protein (PEBP) , . In addition to the FT-like proteins, the plant PEBP family consists of two other phylogenetically distinct groups of proteins, the TERMINAL FLOWER 1 (TFL1)-like proteins and the MOTHER OF FT AND TFL (MFT)-like proteins –. FT and TSF act redundantly to promote flowering under long-day (LD) photoperiods, , . Arabidopsis FT and TSF proteins produced in the phloem ,  and are transported to the shoot apex, where they dimerize with the bZIP transcription factor FD to activate the expression of SUPPRESSOR OF OVEREXPRESSION OF CONSTANS 1 (SOC1) ,  and the floral meristem identity genes APETALA1 (AP1) and LEAFY (LFY) , . FT-like proteins from various species function in a manner similar to that of FT regarding the induction of flowering, transport in the phloem, and interaction with FD-like proteins –, suggesting that this general mechanism is likely widely conserved across flowering plants. However, the rice FT ortholog Hd3a interacts with OsFD1 indirectly through a 14-3-3 protein to form a ternary trimer known as the florigen activation complex in the nuclei of the shoot apex, where it activates the expression of OsMADS15, an AP1 homolog that regulates flowering , .
Soybean, Glycine max (L.) Merr., is basically a short-day (SD) plant: its flowering is induced when the day length becomes shorter than a critical length. Soybean is grown at a wide range of latitudes, from at least North 50° to South 35°, although the cultivation area of each soybean cultivar is restricted to a very narrow range of latitudes. This wide adaptability has most likely been generated by genetic diversity at a large number of the major genes and quantitative trait loci that control flowering behavior. Nine major genes, E1 to E8 and J, that control flowering time and maturity have been previously identified in soybean –. Among these genes, E1 has been cloned using a map-based approach and is assumed to be a legume-specific transcription factor containing a putative nuclear localization signal and a B3 distantly related domain ; E2 has been identified as an ortholog of the Arabidopsis GIGANTEA gene ; and E3 and E4 have been confirmed as PHYA homologs using map-based cloning  and a candidate gene approach , respectively. Many allelic variations occur at the E1, E3 and E4 loci, and their allelic combinations condition soybean flowering time, regulate preflowering and postflowering photoperiod responses, and contribute greatly to the wide adaptability of soybean , . Two FT homologs, GmFT2a and GmFT5a, are involved in the transition to flowering and these two genes coordinately control flowering in soybean . The maturity genes E1, E2, E3 and E4 downregulate GmFT2a and GmFT5a expression to delay flowering and maturation under LD conditions, suggesting that GmFT2a and GmFT5a are the soybean flowering integrators and major targets in the control of flowering , , , . In addition, two SOC1 homologs, GmSOC1 and GmSOC1-like, have been molecular characterized in soybean: the overexpression of GmSOC1 partially rescued the late-flowering phenotype of the soc1-1 Arabidopsis mutant under LD conditions , while the overexpression of GmSOC1-like promoted flowering in Lotus corniculatus . These results suggest that the two soybean SOC1 homologs may function as floral activators in soybean. A soybean AP1 homolog, GmAP1, has also been isolated in soybean and is specifically expressed in the flower, especially in the sepals and petals. This gene caused early flowering and the alteration of floral organ patterns when ectopically expressed in tobacco . Despite the economic importance of soybean, knowledge regarding its molecular mechanisms of flowering remains limited. Here, we report that the overexpression of GmFT2a and GmFT5a in soybean can promote early flowering by activating the expression of floral identity gene homologs such as GmAP1, GmLFY and GmSOC1. The GmFT2a and GmFT5a proteins interact with the bZIP transcription factor GmFDL19. The overexpression of GmFDL19 in soybean can also induce the expression of floral identity genes. Additionally, we show that the bZIP transcription factor GmFDL19 is able to bind with the ACGT cis-element of the GmAP1 promoter. Our results suggest that the putative flowering model FT/FD-AP1 is well conserved in the legume soybean and that GmFDL19 may act as the key component in the photoperiod-regulated flowering pathway controlled by GmFT2a and GmFT5a.
Materials and Methods
Plant materials and growth conditions
The soybean cultivars Harosoy, Williams 82 and Dongnong 50 were used in this study. All plants were grown in a growth chamber (Conviron ADAPTIS-A1000, Canada) at a consistent temperature of 25°C and an average photon flux of 300 µmol m−2s−1, supplied by T5 fluorescent lamps. Day length regimes were 12L/12D for SD and 16L/8D or 18L/6D for LD. Tissue-specific expression was analyzed using the cultivar Harosoy grown under SD. Total RNA was isolated from trifoliate leaves, shoot apices, roots, flowers, flower buds, and roots. For the temporal expression analysis, pieces of young, fully developed trifoliate leaves and shoot apices were bulk sampled at 4 hours after dawn from 4 individual plants grown under SD every five days from 10 DAE until 25 DAE. The trifoliate leaves and shoot apices from 4 plants of both transgenic and untransformed lines were bulk sampled at 4 hours after dawn at 20 DAE under the LD condition and stored until total RNA extraction.
Soybean genetic transformation.
The cDNA sequences of Harosoy GmFT2a/5a and GmFDL19 were first cloned into the pEASY-T1 vector (Transgene, Beijing, China). XbaI/SacΙ-digested fragments were then inserted at multiple cloning sites in the pTF101.1 vector, and the transgenes were driven by the cauliflower mosaic virus 35S promoter . The GmFT2a/5a-pTF101 and GmFDL19-pTF101 constructs were used to transform the cultivars Williams 82 and Dongnong 50, respectively, following the cotyledonary node method . T0, T1 and T2 transformants were screened by daubing 160 mg/L glufosinate into the preliminary leaves of the seedlings. Herbicide-resistant T2 plants were subjected to molecular and phenotypic analysis.
RT-PCR and quantitative RT-PCR analyses
Total RNA was isolated and cDNA was synthesized as described in Kong et al. . RT-PCR of GmFDL19, GmAP1 (a, b, c, d), GmSOC1a, GmSOC1b, GmLFY1, GmLFY2 and Tubulin (as an internal control) was conducted using cDNAs synthesized from total RNA. PCR conditions were as follow: one cycle of 5 min at 94°C; 30 cycles of 30 sec at 94°C, 30 sec at 55°C to 60°C (depending on the gene), and 30 sec at 72°C; and a final extension of 10 min at 72°C. RT-PCR was performed using homolog-specific primers to easily separate the RT-PCR products (approximately 500 bp) from the fragments amplified from genomic DNAs (>1 kb). The RT-PCR products were separated by electrophoresis in a 1% agarose gel and visualized with EtBr under UV light. Quantitative RT-PCR was performed as described in . The quantitative RT-PCR mixture was prepared by mixing a 1 µl aliquot of the reaction mixture from the cDNA synthesis, 5 µl of 1.2 µM primer premix, 10 µl SYBR Premix ExTaq Perfect Real Time (TaKaRa Bio), and water to a final volume of 20 µl. The analysis was conducted using the DNA Engine Opticon 2 System (Bio-Rad). The PCR cycling conditions were as follow: 95°C for 10 sec, 55°C to 60°C (depending on the gene) for 20 sec, 72°C for 20 sec, and 78°C for 2 sec. This cycle was repeated 40 times. Fluorescence quantification was conducted before and after the incubation at 78°C to monitor the formation of primer dimers. The mRNA level of the Tubulin gene was used as a control for the analysis. A reaction mixture without reverse transcriptase was also used as a control to confirm that no amplification occurred from genomic DNA contaminants in the RNA sample. In all of the PCR experiments, the amplification of a single DNA species was confirmed using both melting curve analysis of the quantitative PCR and gel electrophoresis of the PCR products. The primers used for qRT-PCR and RT-PCR are listed in Table S1 and Table S2, respectively.
Identification of soybean FD, AP1, LFY and SOC1 homologs.
The database used for these searches is available at Phytozome (http://www.phytozome.net/soybean). Starting with the Arabidopsis FD, AP1, LFY and SOC1 protein sequences, TBLASTN searches were conducted against the soybean (Glycine max) gene index (release 1.0). The top 18 FD-like gene sequences producing high-scoring segment pairs were chosen and investigated further. Primers were designed to amplify cDNAs for each of the top 18 FD-like genes (Table S2). Seven pairs of full-length CDS PCR (Table S3) primers were designed to amplify the seven expressed FD-Like genes, and restriction sites (underlined) were included in the oligos to facilitate the cloning of the PCR products into the yeast vector pGADT7. A multiple sequence alignment and a neighbor-joining phylogenetic tree were constructed using DDBJ (http://clustalw.ddbj.nig.ac.jp/) online ClustalW software and Treeview 2.0. The tree was based on the full-length amino acid region including the bZIP domain and the SAP motif (Figure S1). The bootstrap percentage supports are indicated at the branches of the tree.
Yeast two-hybridization assays
The yeast cloning vectors pGBKT7 and pGADT7, the control vectors pGADT7-T and pGBKT7-53, and the yeast strain Y2H used in the yeast-two hybridization assays were obtained from the Clontech company (http://www.clontech.com/). The yeast two-hybridization assays were performed according to the manufacturer's instructions. Soybean full-length CDS of GmFT2a and GmFT5a were inserted into pGBKT7 vectors to generate fused GAL4 DNA binding domains as the soybean baits. Full-length CDS of the three soybean FD-like genes containing the SAP motif (GmFDL19, GmFDL08 and GmFDL15) were cloned into pGADT7 to generate fused GAL4 DNA activation domains as the soybean preys, and the full-length CDS of Arabidopsis FD was also cloned into pGADT7 to generate a positive control. Table S4 lists the primers and restriction sites used to generate the yeast bait and prey constructs. The bait and prey plasmids were cotransformed into the yeast strain Y2H using the lithium acetate method and selected on SD medium lacking leucine (Leu) and tryptophan (Trp). After 4 days of incubation at 30°C, the yeast cells were replated on selection plates with SD medium lacking Leu, Trp, histidine (His) and adenine (Ade) but including the X-α-gal substrate for the interaction test.
Bimolecular fluorescence complementation
The full-length CDS of GmFT2a and GmFT5a were amplified using the primer pairs GmFT2a-NE-F/R and GmFT5a-NE-F/R, respectively (Table S5), and were then inserted into the pUC_SPYNE  vector, which contains the DNA encoding the N-terminus of YFP. The full-length cDNAs of GmFDL08, GmFDL15, GmFDL19 and FD were amplified using the primer pairs listed in Table S5 and then inserted into the pUC_SPYCE  vector, which contains the DNA encoding the C-terminus of YFP. The recombined pUC_SPYNE/CE plasmids were cotransformed into Arabidopsis protoplasts using polyethylene glycol–mediated transfection, as described previously . YFP-dependent fluorescence was detected 24 h after transfection using a confocal laser-scanning microscope (Zeiss LSM 510 Meta).
Electrophoretic mobility shift assay
The full-length coding region of GmFDL19 was amplified by PCR using the primer pair GmFDL19-29b-F/R (Table S6). The PCR product and the pET29b plasmid (Novagene, WI, USA) were digested with NdeI and SalI. After ligation, the construct was transformed into the E. coli competent cell line BL21 (DE3) (Transgene, Beijing, China) according to the manufacturer's instructions. The recombinant GmFDL19 protein was purified using the His tag purification nickel ion system (Kangweishiji, Beijing, China). EMSA was conducted using the recombinant GmFDL19 protein and the DNA products of the GmAP1a promoter obtained by hybridizing the forward and reverse complementary oligos containing the ACGT core sequence (Table S6). The EMSA assay was conducted using the EMSA kit (Invitrogen, www.Invitrogen.com, Cat #E33075). The DNA-protein complex samples were loaded into a TBE gradient 6% polyacrylamide native gel (Bio-Rad Laboratories, www.bio-rad.com) at 200 V for 45 minutes. The DNA in the gel was stained using SYBR Green, provided in the same kit, and visualized using the GE Typhoon LFA 9500 Imaging System (GE Healthcare Life Science, USA).
Overexpression of GmFT2a and GmFT5a causes precocious flowering in soybean
Two FT homologs, GmFT2a and GmFT5a, are involved in photoperiod-regulated flowering, and these two genes coordinately control flowering in soybean . To determine how these two FT homologs regulate soybean flowering, GmFT2a and GmFT5a were genetically transformed into the soybean cultivar Williams 82 under the control of the cauliflower mosaic virus (CaMV) 35S promoter. The overexpression of GmFT2a and GmFT5a caused the early flowering of Williams 82 even under noninductive LD (16L/8D) conditions (Figure 1A, B, C, D and Figure 1E, F, G, H). The transgenic GmFT2a T2 overexpression line #2-1-1 flowered at approximately 33 days after emergence (DAE) and the transgenic GmFT5a T2 overexpression line #5-1 flowered at approximately 35 DAE; however, the untransformed Williams 82 flowered at approximately 57 DAE (Figure 1I). These data suggested that both GmFT2a and GmFT5a are able to induce early flowering in soybean under noninductive LD conditions.
(A) The close shot of the transgenic plant in (B) shows the precocious flowers at the axils of the trifoliate leaves. (B) A transgenic GmFT2a plant showing precocious flowering at the axils of the trifoliate leaves. (C) A wild-type Williams 82 plant. (D) The close shot of the wild type Williams 82 plant in (C) does not show flowers at the axils of the trifoliate leaves. (E) The close shot of the transgenic plant in (F) shows the precocious flowers at the axils of the trifoliate leaves (F) A transgenic GmFT5a plant showing precocious flowering at the axils of the trifoliate leaves. (G) A wild-type Williams 82 plant. (H) The close shot of the wild-type Williams 82 plant in (G) does not show flowers at the axils of the trifoliate leaves. (I) Days to flowering from the emergence of the transgenic plants and wild-type plants. Averages and standard errors are calculated from four T2 plants for each construct and 5 Williams 82 plants. Double asterisks indicate significant differences from the corresponding wild-type Williams 82 at P<0.01.
GmFT2a and GmFT5a upregulate floral meristem identity genes
All flowering pathways converge onto floral integrators, including FT and SOC1, and induce the expression of floral meristem identity genes, including AP1 and LFY , . Several genes involved in the determination of flowering time have recently been isolated and characterized in soybean, including GmAP1, GmSOC1, GmSOC1-like and GmLFY , , . By searching the soybean reference genome using Phytozome (http://www.phytozome.net/soybean), four AP1 homologs were identified and designated as GmAP1b (Glyma01g08150), GmAP1c (Glyma08g36380), GmAP1d (Glyma02g13420) and GmAP1a (Glyma16g13070, GmAP1), the last of which has been characterized previously . Spatial RT-PCR analyses suggested that GmAP1a, GmAP1b and GmAP1c were mainly transcribed in reproductive organs such as shoot apices, flower buds and flowers under SD (12L/12D) conditions in the cultivar Harosoy, with the transcription of GmAP1a being the most prominent; however, the transcription of GmAP1d was not detected in any tissues (Figure 2A). Two LFY homologs, designated as GmLFY1 (Glyma04g37900)  and GmLFY2 (Glyma06g17170), were also identified in the soybean genome. GmLFY1 was transcribed mainly in developing pods and seeds and was not detected in leaves or the SAM (shoot apex meristem) (Figure 2A), suggesting that the gene might contribute to seed development in soybean instead of flowering, as was previously reported . Similarly to the AP1 homologs, GmLFY2 was also transcribed in shoot apices, flower buds and flowers (Figure 2A). Two soybean SOC1 homologs, GmSOC1a (Glyma18g45780)  and GmSOC1b (Glyma09g40230) , could be identified from the soybean genome. Both of these genes were highly expressed in shoot apices, leaves, flower buds and roots but were weakly expressed in flowers and pods (Figure 2A), in agreement with the expression patterns observed for SOC1 in multiple organs of Arabidopsis . In total, six floral meristem identity genes, GmAP1a, GmAP1b, GmAP1c, GmSOC1a, GmSOC1b and GmLFY2, were constantly expressed in the shoot apices of the cultivar Harosoy from 10 DAE to 25 DAE under SD (12L/12D) conditions before the floral bud formation stage (floral buds formed at 25 DAE), and GmSOC1a and GmSOC1b were also constantly expressed in the leaves of the cultivar (Figure 2B). The constantly high expression levels of these genes in shoot apices before the floral bud formation stage most likely indicate their involvement in the flowering transition of soybean.
(A) Transcript levels of eight soybean flowering-related genes (GmAP1a, GmAP1b, GmAP1c, GmAP1d, GmSOC1a, GmSOC1b, GmLFY1, GmLFY2) in leaves and shoot apices of the soybean cultivar Harosoy under SD (12L/12D) conditions; Tubulin is included as an endogenous control. Samples were collected from 10 DAE to 25 DAE. DAE: days after emergence. L: leaves; S: shoot apex. (B) Tissue-specific expression analyses of eight flowering-related genes by RT-PCR under SD (12L/12D) conditions. L: leaves, S: shoot apices, F: flowers, FB: flower buds, P: pods, R: roots.
In Arabidopsis, FT controls photoperiod-regulated flowering by activating the downstream flower meristem identity genes AP1, LFY and SOC1 –, –. To determine whether GmFT2a and GmFT5a induce early flowering in soybean by regulating the orthologs of AP1, LFY and SOC1, the expression levels of GmAP1 (a, b, c), GmSOC1a, GmSOC1b and GmLFY2 were determined using quantitative PCR in the leaves and shoot apices of the transgenic GmFT2a and GmFT5a overexpression soybean lines. In the transgenic GmFT2a soybean line #2-1-1, the expression levels of GmFT2a, GmAP1 (a, b, c), GmSOC1a, GmSOC1b and GmLFY2 were significantly higher in the shoot apices than were levels in untransformed Williams 82 (Figure 3A), while the expression of GmFT5a remained unchanged. However, in the transgenic GmFT5a soybean line #5-1, the expression levels of GmFT5a, GmAP1 (a, b, c), and GmSOC1b were significantly higher in the shoot apices than were levels in untransformed Williams 82 (Figure 3B), while the expression levels of GmFT2a, GmSOC1a and GmLFY2 were unchanged (Figure 3B). These results suggest that the flowering genes GmAP1s, GmSOC1a, GmSOC1b and GmLFY2 are downstream of GmFT2a and GmFT5a and that these genes are most likely differentially involved in the GmFT2a and GmFT5a–induced early flowering of soybean. In addition, the expression of GmFT2a in the GmFT5a transgenic line and the expression of GmFT5a in the GmFT2a transgenic line were both unchanged in the leaves and shot apices, suggesting that GmFT2a and GmFT5a promote soybean flowering in a redundant manner (Figure 3).
(A) Expression analyses of GmFT2a, GmFT5a and flowering-related genes in transgenic GmFT2a plants (#2-1-1) and wild-type Williams 82 plants (WT). (B) Expression analyses of GmFT5a, GmFT2a and flowering-related genes in transgenic GmFT5a plants (#5-1) and wild-type Williams 82 plants (WT). The white and black columns represent relative expression in leaves and shoot apices, respectively. Asterisks and double asterisks indicate significant differences between transgenic and WT plants at 0.01<P<0.05 and P<0.01, respectively.
GmFT2a and GmFT5a interact with GmFDL19
FT and its homologs are widely understood to move from leaves to shoot apices, where they interact with FDs to form FT/FD complexes that bind to the promoter of AP1 , , . In this study, we found that GmFT2a and GmFT5a promote flowering by inducing the expression of GmAP1s, GmSCO1s and GmLFY2, and it is easily assumed that GmFTs also require a partner such as GmFD to regulate the downstream flowering genes in soybean. Taking the amino acid sequence of FD from Arabidopsis as the query, we searched for orthologs of FD in the soybean genome using Phytozome and identified 18 high-scoring candidate GmFDLs (GmFD-Like) genes (Figure S1). Expression analyses were conducted using RT-PCR for all eighteen selected GmFDLs, of which seven GmFDLs were transcribed both in leaves and shoot apices (Figure 4A). A multiple sequence alignment of the seven soybean FD-like proteins with the FDs from other species shows a conserved bZIP domain of 42 amino acids (N-X7-R-X9-L-X6-L-X6-L) (Figure S2) and an SAP (RXXS/TAP) motif (Figure 4B), SAP motif has been reported as a putative binding sequence for FT . Phylogenetic analysis was conducted based on the amino acid sequences of the 18 candidate GmFDLs, and the proteins were grouped into three clades (Figure S1, Table S7). GmFDL02, GmFDL04 and GmFDL0602 were divided into the FD clade, but these three GmFDLs did not transcribe or share an SAP motif. GmFDL08, GmFDL13, GmFDL15, GmFDL19 and GmFDL20 were divided into the wheat TaFDL2 clade, with GmFDL08, GmFDL15 and GmFDL19 sharing the classic SAP motif; these three GmFDLs were therefore tested for interactions with soybean GmFT2a and GmFT5a. GmFDL06 and GmFDL12 were divided into the AREB and ABI5 cluster; these two proteins may represent the stress-related bZIP transcription factors in soybean.
(A) Transcript levels of seven GmFDLs in leaves and shoot apices of the soybean cultivar Harosoy under SD (12L/12D) conditions; Tubulin is included as an endogenous control. Samples were collected from 10 DAE to 25 DAE. L: leaves; S: shoot apices. (B) Multiple alignment of the amino acid sequences in the SAP motif region of the FDs from soybean and other species. The SAP motif is a putative sequence for FT binding.
GmFT2a and GmFT5a promoted early flowering in Arabidopsis , , so we first examined the interactions of FD with GmFT2a and GmFT5a using the yeast two-hybridization assay. The results indicated that both GmFT2a and GmFT5a were able to interact with FD (Figure 5A). Autoactivation tests of GmFT2a and GmFT5a in yeast confirmed that both proteins were unable to activate the reporter genes when used alone as bait (data not presented). The yeast two-hybridization assays of the three SAP motif proteins among the GmFDLs, GmFDL08, GmFDL15 and GmFDL19, with GmFT2a and GmFT5a revealed that only GmFDL19 was able to interact with GmFT2a and GmFT5a (Figure 5A). To validate the results of the yeast two-hybridization tests, in vivo BiFC analyses were conducted. These results confirmed that both FD and GmFDL19 could interact with GmFT2a and GmFT5a in the nuclei of Arabidopsis protoplasts (Figure 5B). Our results suggest that GmFT2a and GmFT5a promote early flowering in soybean, most likely through interacting with GmFDL19 and upregulating downstream floral identity genes in a manner similar to that of FT in Arabidopsis. GmFDL19 was constantly transcribed both in leaves and shoot apices, with increasing expression levels in the shoot apices following the growth stage from 10 DAE to 25 DAE before floral bud formation (25 DAE) (Figure 4A). In addition, the expression of GmFDL19 was upregulated by GmFT2a in the transgenic overexpression line #2-1-1, while the transcription of GmFDL19 was unchanged in the transgenic GmFT5a overexpression line #5-1 (Figure 3A, B). The expression patterns and protein interactions of GmFDL19 strongly support this protein as the candidate soybean FD ortholog, which may participate differentially in the early flowering of soybean promoted by GmFT2a and GmFT5a. That is, GmFT2a promotes soybean early flowering through both transcriptional upregulation of and physical interaction with GmFDL19, but GmFT5a only promotes flowering through physical interaction with the protein.
(A) Yeast two-hybridization assays; FD was also included because both GmFT2a and GmFT5a promoted early flowering in Arabidopsis. After cotransformation of the baits and preys, an equal amount of yeast clones were plated on SD-Leu-Trp and SD-Leu-Trp-His-Ade+X-α-gal selective plates, and the plates were incubated at 30°C until the emergence of the yeast clones. (B) BiFC (bimolecular fluorescence complementation) assays to confirm the results of the yeast two-hybridization assays. Arabidopsis protoplasts cotransformed with constructs of FTs or FDs fused to the N-terminal (YN) and C-terminal (YC) halves of YFP, respectively (as indicated), were imaged using a confocal microscope after incubation at room temperature (20°C to 25°C) over 18 hours. Images are shown as YFP, merged YFP and bright field. Scale bars indicate 20 µm.
Overexpression of GmFDL19 causes early flowering in soybean
The expression patterns and protein interactions of GmFDL19 suggest that the protein might be involved in the GmFT2a- and GmFT5a-regulated flowering pathway in soybean. To determine the functions of GmFDL19 in soybean flowering, overexpression of GmFDL19 driven by the 35S promoter was genetically transformed into the cultivar Dongnong 50 (Figure 6A, B). Under LD (16L/8D) conditions, this cultivar flowered early, at approximately 30 DAE. To observe clearer flowering differences between the transgenic T2 line #12-1 and untransformed Dongnong 50, we evaluated the flowering times of both lines under longer photoperiod LD (18L/6D) conditions. Under these conditions, the transgenic line #12-1 flowered, on average, at approximately 43 DAE, while Dongnong 50 flowered at 55 DAE, indicating that GmFDL19 is able to promote flowering in soybean (Figure 6A, B, C, D, E).
(A) The close shot of the transgenic plant in (B) shows the precocious flowers at the axils of the trifoliate leaves (B) A transgenic GmFDL19 plant showing precocious flowering at the axils of the trifoliate leaves. (C) A wild-type Dongnong 50 plant. (D) The close shot of the wild-type Dongnong 50 plant in (C) does not show flowers at the axils of the trifoliate leaves. (E) Number of days to flowering in transgenic and wild-type plants. Averages and standard errors are calculated from five independent T2 plants and five Dongnong 50 plants. (F) Expression analyses of GmFDL19 and flowering-related genes in transgenic GmFDL19 plants (#12-1) and wild-type Dongnong 50 plants (WT); because these flowering related genes are transcribed mostly in shoot apices, shoot apex samples were collected from transgenic and wild-type Dongnong 50 plants at 40 DAE under LD (18L/6D) conditions. Asterisks and double asterisks indicate significant differences from the corresponding wild-type Dongnong 50 at 0.01<P<0.05 and P<0.01, respectively.
We have demonstrated that GmFT2a and GmFT5a promote flowering by inducing the expression of flowering-related genes and that GmFDL19 transgenic plants flower earlier than untransformed Dongnong 50 plants under an LD photoperiod. We next wished to determine whether these flowering-related genes were also upregulated in the shoot apices of GmFDL19 transgenic plants. As expected, the transcription levels of GmFDL19, as well as those of the flowering related genes GmAP1s, GmSOC1s and GmLFY2, were significantly higher in the shoot apices of the transgenic GmFDL19 line #12-1 than in those of untransformed Dongnong 50 (Figure 6F). However, the expression levels of GmFT2a and GmFT5a in the transgenic GmFDL19 line #12-1 were unchanged and were only faintly detected in both #12-1 and Dongnong 50 under the LD (18L/6D) conditions (Figure 6F). Considering the upregulation of GmFDL19 by GmFT2a, as well as the fact that GmFDL19 has the same effect on the upregulation of GmAP1s, GmSOC1s and GmLFY2 as do GmFT2a, the results suggest that GmFDL19 may act downstream of GmFT2a in the regulation of the flowering transition in soybean. In addition to the transcriptional upregulation of GmFDL19 by GmFT2a, GmFDL19 interacts with both GmFT2a and GmFT5a, suggesting that GmFDL19 may be required for GmFT2a- and GmFT5a-regulated flowering in soybean.
GmFDL19 binds to the GmAP1a promoter in vitro
FT-like proteins interact with FD-like proteins to form FT/FD complexes, which bind to the core ACGT cis-elements located at the promoters of AP1-like genes in Arabidopsis, rice and wheat , , . To determine whether this mechanism is conserved in soybean, we conducted an electrophoretic mobility shift assay (EMSA) to test the binding of GmFDL19 with the promoter of GmAP1a. GmAP1a was selected for the binding assay because it showed a higher expression level than did the other two homologs, GmAP1b and GmAP1c (Figure 2, Figure 3 and Figure 6F). This gene contains seven ACGT core elements and one CArG box, representing the putative binding site for MADS domain transcription factors in its promoter (Figure 7). The EMSA results demonstrated that GmFD19 binds to the ACGT core sequences in vitro. As the negative control, the CArG box could not be bound by GmFDL19 (Figure 7). These results suggest that the FT/FD-AP1 pathway is well conserved in soybean and that GmFDL19 serves as an important component of GmFT2a- and GmFT5a-regulated flowering in the legume.
Potential bZIP binding sites presented in the GmAP1a promoter were used as probes in binding reactions with the purified recombinant GmFDL19 protein. The eight probes included seven potential bZIP binding sites and a CArG-box as negative control. (1) T-box (AACGTT), (2) A/C hybrid of A and C-box (TACGTC), (3) T/C hybrid of T and C-box (AACGTC), (4) CArG-box (CCNNNNNNNNGG), (5) G/A hybrid of G and A-box (CACGTA), (6) T/A hybrid of T and A-box (AACGTA), (7) G/A hybrid of G and A-box (CACGTA), (8) T-box (AACGTT). The scheme below indicates the positions of the various bZIP binding sites.
Photoperiod-regulated flowering is integrated through GmFT2a and GmFT5a in soybean
Soybean is adapted to wide range of latitudes, from at least North 50° to South 35°, and its wide adaptability can most likely be attributed to the genetic diversity at many of the major genes (E1 to E8 and J) and unclassified quantitative trait loci controlling its flowering and maturity. Of the major maturity genes, E1 to E4 delay flowering and maturity under LD but have no effects on flowering and maturity under SD . The molecular identities of E1 to E4 have recently been characterized . The E1 gene has the largest effect of the maturity genes on delaying flowering and maturity under LD conditions , , ,  and has been cloned using a map-based approach, revealing it to be a legume-specific transcription repressor with a putative nuclear localization signal (NLS) and a B3 distantly related domain . The repression of flowering by E1 is most likely due to its suppression of the transcription of GmFT2a and GmFT5a under LD conditions , . E1 is transcribed mainly in vegetative organs such as cotyledons and leaves , which is consistent with the transcription sites of GmFT2a and GmFT5a . The diurnal circadian rhythm of E1 transcription contains two peaks in the leaves within 24 hours , of which the first transcription peak overlaps with the transcription peaks of GmFT2a and GmFT5a at 4 hours after dawn . These results suggest that GmFT2a and GmFT5a are most likely the direct targets of E1 regulation, but this hypothesis requires further evidence for verification. E3 and E4 encode the light receptor phytochrome A (PHYA) proteins GmPHYA3 and GmPHYA2, respectively , . The expression levels of GmFT2a and GmFT5a were additively suppressed by E3 and E4 , perhaps indirectly via E1 function, as shown in genetic studies revealing that E3 and E4 have epistatic effects on E1 under LD conditions . E2 was identified molecularly as an ortholog of the Arabidopsis GIGANTEA gene . The E2 gene mainly controls flowering time through the regulation of GmFT2a, not GmFT5a . Taken together, these results indicate that the photoperiod-regulated flowering pathway in soybean converges at GmFT2a and GmFT5a.
GmFT2a and GmFT5a redundantly and differentially regulate flowering in soybean
An SD-to-LD transfer experiment demonstrated the differences in photoperiod response between GmFT2a and GmFT5a. The expression of GmFT2a was strictly regulated by photoperiodic changes from SD to LD, whereas the response of GmFT5a to photoperiodic changes was gradual, and its expression was maintained at low levels even after the plants were transferred to LD . These findings suggest that, in addition to the phyA-mediated photoperiod response, a second regulatory mechanism may be involved in the differences of expression pattern between GmFT2a and GmFT5a. In addition to E3 and E4, E2 influences the mRNA abundance of FT homologs. Watanabe et al. (2011) found a clear association between flowering time and GmFT2a expression in two sets of near isogenic lines (NILs) for the E2 locus ; dysfunctional e2 alleles promoted GmFT2a expression and conditioned earlier flowering. However, these authors did not observe significant differences in GmFT5a expression between the NILs. These results suggest that the E2 gene (GmGIa) mainly controls flowering time through the regulation of GmFT2a . The different responses to photoperiodic changes observed between GmFT2a and GmFT5a  may thus be caused by the involvement of the GI (E2)-regulated pathway in GmFT2a expression, but not in GmFT5a expression. Under the phyA (E3 and E4)-mediated photoperiodic regulation system, GmFT2a and GmFT5a may redundantly and strongly induce flowering under shorter day lengths, but GmFT5a alone may promote flowering in a photoperiod-independent manner under longer day lengths.
The expression analyses of GmFT2a and GmFT5a in their respective transgenic overexpression lines further demonstrate that GmFT2a and GmFT5a are not regulated by each other, suggesting that GmFT2a and GmFT5a induce soybean flowering redundantly (Figure 3A, B). GmFT2a significantly upregulates downstream floral identity genes such as GmAP1 (a, b, c), GmSOC1a, GmSOC1b and GmLFY2. However, GmFT5a only upregulates GmAP1 (a, b, c) and GmSOC1b and has no effect on GmSOC1a and GmLFY2 (Figure 3A, B). In addition, the expression of GmFDL19 was upregulated by GmFT2a in the transgenic overexpression line #2-1-1 while the transcription of the gene was unchanged in the transgenic GmFT5a overexpression line #5-1 (Figure 3A, B). A hypothesis for the system was developed: GmFT2a, in combination with GmFDL19, triggers the upregulation of GmLFY2, and GmLFY2 then feeds back directly to further upregulate GmFDL19. However, this hypothesis requires further confirmation. These results suggest that GmFT2a and GmFT5a induce soybean flowering differentially and redundantly. The GmFT2a-regulated flowering pathway and the GmFT5a-regulated flowering pathway may be integrated in the SAM and are redundantly balanced in a very complex manner to ensure precise flowering in paleopolyploid soybean. These two FT homologs may therefore coordinately and redundantly control flowering in soybean.
GmFDL19 may be involved in GmFT2a- and GmFT5a-regulated flowering in soybean
FT and its homologs are widely known to move from leaves to shoot apices, where they interact with FDs to form FT/FD complexes that bind to the promoter of AP1 and induce flowering in many plant species , , . In this study, we report that the bZIP transcription factor GmFDL19 is able to physically interact with two soybean FT homologs, GmFT2a and GmFT5a, as confirmed by both yeast two-hybridization in vitro and BiFC in vivo. The binding of GmFDL19 with the cis-elements in the promoter of the AP1 soybean ortholog GmAP1a was further confirmed by EMSA in vitro. Our results further extend the regulatory module of FT/FD-AP1 in the legume species soybean. In addition to the interaction of GmFDL19 with GmFT2a and GmFT5a, GmFT2a is able to upregulate the transcription of GmFDL19 in shoot apices. The transcription levels of GmFT2a and GmFT5a are not regulated by GmFDL19, suggesting that GmFDL19 functions downstream of GmFT2a and GmFT5a. The floral identity genes GmAP1 (a, b, c), GmSOC1a, GmSOC1b and GmLFY2 are upregulated by GmFDL19, GmFT2a and GmFT5a in their respective transgenic overexpression lines. Taken together, these results suggest that GmFDL19 may be involved in GmFT2a- and GmFT5a-regulated flowering in soybean.
In summary, we propose a molecular network of photoperiod-regulated flowering in soybean (Figure 8). Under SD conditions, the E3, E4 and E1 genes do not express in leaves where GmFT2a and GmFT5a are able to transcribe at high levels, and the GmFT2a and GmFT5a proteins are then transported from the leaves to the shoot apices, where they bind with GmFDL19 to induce the expression of flowering-related genes (GmAP1, GmSOC1, GmLFY), thus leading to early flowering. Under LD conditions, E3 and E4 genes are highly transcribed in leaves, where they epistatically induce the high expression of the E1 gene, thereby suppressing the expression levels of GmFT2a and GmFT5a. The expression levels of flowering related genes are not upregulated due to lack of FT proteins, leading to a late-flowering phenotype.
(A) Model of GmFT2a and GmFT5a regulating the expression of GmAP1a. The horizontal dotted line represents the GmAP1a promoter, and the black vertical bars indicate the eight exons of GmAP1a. The green oval represents the GmFDL19 protein, and this protein can bind to the T-box, G-box or hybrid box (white rectangle) in the GmAP1a promoter. The orange oval represents the interactions of GmFT2a and GmFT5a with GmFDL19. (B) A proposed molecular network for photoperiod-regulated flowering in soybean.
Phylogenetic relationship of soybean FD-like proteins and FDs from other species constructed using the neighbor-joining method with the program CLUSTAL W. Bootstrap percentage supports are indicated at the branches of the tree. The seven red filled rectangles indicate the bZIP domain of seven expressed FD-like genes in soybean, and the red rectangles indicated the SAP motif contained in soybean FD-like proteins and FDs from other species. The locus IDs or accession numbers of these FDs are presented in Table S7.
Conserved bZIP domain of the seven soybean FD-like proteins and FDs from other species.
Primers for isolation of seven GmFDLs.
Primers for yeast two-hybridization assays.
We thank Dr. Kan Wang, Iowa State University of Science and Technology for providing soybean transformation vector pTF101.1 and Agrobacterium strain EHA101; Miss Dandan Zhang, Miss Jiajia Deng, Mr. Hang Ren and Miss Lingli Kong for generating transgenic soybean lines.
Conceived and designed the experiments: HYN BHL FJK. Performed the experiments: HYN DC DYZ YL SJL LLT. Analyzed the data: HYN XHY BHL FJK. Contributed reagents/materials/analysis tools: XHY BHL FJK. Contributed to the writing of the manuscript: HYN FJK.
- 1. Amasino R (2010) Seasonal and developmental timing of flowering. Plant J 61: 1001–1013.
- 2. Kardailsky I, Shukla VK, Ahn JH, Dagenais N, Christensen SK, et al. (1999) Activation tagging of the floral inducer FT. Science 286: 1962–1965.
- 3. Kobayashi Y, Kaya H, Goto K, Iwabuchi M, Araki T (1999) A pair of related genes with antagonistic roles in mediating flowering signals. Science 286: 1960–1962.
- 4. Bradley D, Ratcliffe O, Vincent C, Carpenter R, Coen E (1997) Inflorescence commitment and architecture in Arabidopsis. Science 275: 80–83.
- 5. Mimida N, Goto K, Kobayashi Y, Araki T, Ahn JH, et al. (2001) Functional divergence of the TFL1-like gene family in Arabidopsis revealed by characterization of a novel homologue. Genes Cells 6: 327–336.
- 6. Yoo SY, Kardailsky I, Lee JS, Weigel D, Ahn JH (2004) Acceleration of flowering by overexpression of MFT (MOTHER OF FT AND TFL1). Mol Cells 17.
- 7. Yamaguchi A, Kobayashi Y, Goto K, Abe M, Araki T (2005) TWIN SISTER OF FT (TSF) acts as a floral pathway integrator redundantly with FT. Plant Cell Physiol 46: 1175–1189.
- 8. Yoo SJ, Chung KS, Jung SH, Yoo SY, Lee JS, et al. (2010) BROTHER OF FT AND TFL1 (BFT) has TFL1-like activity and functions redundantly with TFL1 in inflorescence meristem development in Arabidopsis. Plant J 63: 241–253.
- 9. Michaels SD, Himelblau E, Kim SY, Schomburg FM, Amasino RM (2005) Integration of flowering signals in winter-annual Arabidopsis. Plant Physiol 137: 149–156.
- 10. Jang S, Torti S, Coupland G (2009) Genetic and spatial interactions between FT, TSF and SVP during the early stages of floral induction in Arabidopsis. Plant J 60: 614–625.
- 11. Takada S, Goto K (2003) TERMINAL FLOWER2, an Arabidopsis homolog of HETEROCHROMATIN PROTEIN1, counteracts the activation of FLOWERING LOCUS T by CONSTANS in the vascular tissues of leaves to regulate flowering time. Plant Cell 15: 2856–2865.
- 12. Yoo SK, Chung KS, Kim J, Lee JH, Hong SM, et al. (2005) Constans activates suppressor of overexpression of constans 1 through Flowering Locus T to promote flowering in Arabidopsis. Plant Physiol 139: 770–778.
- 13. Abe M, Kobayashi Y, Yamamoto S, Daimon Y, Yamaguchi A, et al. (2005) FD, a bZIP protein mediating signals from the floral pathway integrator FT at the shoot apex. Science 309: 1052–1056.
- 14. Wigge PA, Kim MC, Jaeger KE, Busch W, Schmid M, et al. (2005) Integration of spatial and temporal information during floral induction in Arabidopsis. Science 309: 1056–1059.
- 15. Lifschitz E, Eviatar T, Rozman A, Shalit A, Goldshmidt A, et al. (2006) The tomato FT ortholog triggers systemic signals that regulate growth and flowering and substitute for diverse environmental stimuli. Proc Natl Acad Sci USA 103: 6398–6403.
- 16. Lin MK, Belanger H, Lee YJ, Varkonyi-Gasic E, Taoka KI, et al. (2007) FLOWERING LOCUS T protein may act as the long-distance florigenic signal in the cucurbits. Plant Cell 19: 1488–1506.
- 17. Tamaki S, Matsuo S, Wong HL, Yokoi S, Shimamoto K (2007) Hd3a protein is a mobile flowering signal in rice. Science 316: 1033–1036.
- 18. Li C, Dubcovsky J (2008) Wheat FT protein regulates VRN1 transcription through interactions with FDL2. Plant J 55: 543–554.
- 19. Taoka K, Ohki I, Tsuji H, Furuita K, Hayashi K, et al. (2011) 14-3-3 proteins act as intracellular receptors for rice Hd3a florigen. Nature 476: 332–335.
- 20. Kobayashi K, Yasuno N, Sato Y, Yoda M, Yamazaki R, et al. (2012) Inflorescence meristem identity in rice is specified by overlapping functions of three AP1/FUL-like MADS box genes and PAP2, a SEPALLATA MADS box gene. Plant Cell 24: 1848–1859.
- 21. Bernard R (1971) Two major genes for time of flowering and maturity in soybeans. Crop sci 11: 242–244.
- 22. Buzzell R (1971) Inheritance of a soybean flowering response to fluorescent-daylength conditions. Can J Genet Cytol 13: 703–707.
- 23. Buzzell R, Voldeng H (1980) Inheritance of insensitivity to long daylength. Soybean Genet Newsl 7: 26–29.
- 24. McBlain B, Bernard R (1987) A new gene affecting the time of flowering and maturity in soybeans. J Hered 78: 160–162.
- 25. Bonato ER, Vello NA (1999) E6, a dominant gene conditioning early flowering and maturity in soybeans. Genet Mol Biol 22: 229–232.
- 26. Cober ER, Voldeng HD (2001) A new soybean maturity and photoperiod-sensitivity locus linked to E1 and T. Crop Sci 41: 698–701.
- 27. Cober ER, Molnar SJ, Charette M, Voldeng HD (2010) A new locus for early maturity in soybean. Crop sci 50: 524–527.
- 28. Ray JD, Hinson K, Mankono J, Malo MF (1995) Genetic control of a long-juvenile trait in soybean. Crop sci 35: 1001–1006.
- 29. Xia Z, Watanabe S, Yamada T, Tsubokura Y, Nakashima H, et al. (2012) Positional cloning and characterization reveal the molecular basis for soybean maturity locus E1 that regulates photoperiodic flowering. Proc Natl Acad Sci USA 109: E2155–E2164.
- 30. Watanabe S, Xia Z, Hideshima R, Tsubokura Y, Sato S, et al. (2011) A map-based cloning strategy employing a residual heterozygous line reveals that the GIGANTEA gene is involved in soybean maturity and flowering. Genetics 188: 395–407.
- 31. Watanabe S, Hideshima R, Xia Z, Tsubokura Y, Sato S, et al. (2009) Map-based cloning of the gene associated with the soybean maturity locus E3. Genetics 182: 1251–1262.
- 32. Liu B, Kanazawa A, Matsumura H, Takahashi R, Harada K, et al. (2008) Genetic redundancy in soybean photoresponses associated with duplication of the phytochrome A gene. Genetics 180: 995–1007.
- 33. Tsubokura Y, Matsumura H, Xu M, Liu B, Nakashima H, et al. (2013) Genetic variation in soybean at the maturity locus E4 is involved in adaptation to long days at high latitudes. Agronomy 3: 117–134.
- 34. Xu M, Xu Z, Liu B, Kong F, Tsubokura Y, et al. (2013) Genetic variation in four maturity genes affects photoperiod insensitivity and PHYA-regulated post-flowering responses of soybean. BMC Plant Biol 13: 91.
- 35. Kong F, Liu B, Xia Z, Sato S, Kim BM, et al. (2010) Two coordinately regulated homologs of FLOWERING LOCUS T are involved in the control of photoperiodic flowering in soybean. Plant physiol 154: 1220–1231.
- 36. Thakare D, Kumudini S, Dinkins RD (2011) The alleles at the E1 locus impact the expression pattern of two soybean FT-like genes shown to induce flowering in Arabidopsis. Planta 234: 933–943.
- 37. Zhong X, Dai X, Xv J, Wu H, Liu B, et al. (2012) Cloning and expression analysis of GmGAL1, SOC1 homolog gene in soybean. Mol Biol Rep 39: 6967–6974.
- 38. Na X, Jian B, Yao W, Wu C, Hou W, et al. (2013) Cloning and functional analysis of the flowering gene GmSOC1-like, a putative SUPPRESSOR OF OVEREXPRESSION CO1/AGAMOUS-LIKE 20 (SOC1/AGL20) ortholog in soybean. Plant Cell Rep 32: 1219–1229.
- 39. Chi Y, Huang F, Liu H, Yang S, Yu D (2011) An APETALA1-like gene of soybean regulates flowering time and specifies floral organs. J Plant Physiol 168: 2251–2259.
- 40. Paz MM, Shou H, Guo Z, Zhang Z, Banerjee AK, et al. (2004) Assessment of conditions affecting agrobacterium-mediated soybean transformation using the cotyledonary node explant. Euphytica 136: 167–179.
- 41. Flores T, Karpova O, Su X, Zeng P, Bilyeu K, et al. (2008) Silencing of GmFAD3 gene by siRNA leads to low α-linolenic acids (18: 3) of fad3-mutant phenotype in soybean [Glycine max (Merr.)]. Transgenic Res 17: 839–850.
- 42. Walter M, Chaban C, Schütze K, Batistic O, Weckermann K, et al. (2004) Visualization of protein interactions in living plant cells using bimolecular fluorescence complementation. Plant J 40: 428–438.
- 43. Sheen J (2002) Phosphorelay and transcription control in cytokinin signal transduction. Science 296: 1650–1652.
- 44. Blázquez MA (2000) Flower development pathways. J Cell Sci 113: 3547–3548.
- 45. Lee JH, Yoo SJ, Park SH, Hwang I, Lee JS, et al. (2007) Role of SVP in the control of flowering time by ambient temperature in Arabidopsis. Genes Dev 21: 397–402.
- 46. Meng Q, Zhang C, Huang F, Gai J, Yu D (2007) Molecular cloning and characterization of a LEAFY-like gene highly expressed in developing soybean seeds. Seed Sci Res 17: 297–302.
- 47. Lee J, Lee I (2010) Regulation and function of SOC1, a flowering pathway integrator. J Exp Bot 61: 2247–2254.
- 48. Ruiz-García L, Madueño F, Wilkinson M, Haughn G, Salinas J, et al. (1997) Different roles of flowering-time genes in the activation of floral initiation genes in Arabidopsis. Plant Cell 9: 1921–1934.
- 49. Moon J, Lee H, Kim M, Lee I (2005) Analysis of flowering pathway integrators in Arabidopsis. Plant Cell Physiol 46: 292–299.
- 50. Searle I, He Y, Turck F, Vincent C, Fornara F, et al. (2006) The transcription factor FLC confers a flowering response to vernalization by repressing meristem competence and systemic signaling in Arabidopsis. Genes Dev 20: 898–912.
- 51. Cober E, Tanner J, Voldeng H (1996) Genetic control of photoperiod response in early-maturing, near-isogenic soybean lines. Crop sci 36: 601–605.
- 52. Xia Z, Zhai H, Liu B, Kong F, Yuan X, et al. (2012) Molecular identification of genes controlling flowering time, maturity, and photoperiod response in soybean. Plant Syst Evol 298: 1217–1227.
- 53. Liu B, Fujita T, Yan Z-H, Sakamoto S, Xu D, et al. (2007) QTL mapping of domestication-related traits in soybean (Glycine max). Ann Bot 100: 1027–1038.