- Research article
- Open Access
GmFLD, a soybean homolog of the autonomous pathway gene FLOWERING LOCUS D, promotes flowering in Arabidopsis thaliana
BMC Plant Biologyvolume 14, Article number: 263 (2014)
Flowering at an appropriate time is crucial for seed maturity and reproductive success in all flowering plants. Soybean (Glycine max) is a typical short day plant, and both photoperiod and autonomous pathway genes exist in soybean genome. However, little is known about the functions of soybean autonomous pathway genes. In this article, we examined the functions of a soybean homolog of the autonomous pathway gene FLOWERING LOCUS D (FLD), GmFLD in the flowering transition of A. thaliana.
In soybean, GmFLD is highly expressed in expanded cotyledons of seedlings, roots, and young pods. However, the expression levels are low in leaves and shoot apexes. Expression of GmFLD in A. thaliana (Col) resulted in early flowering of the transgenic plants, and rescued the late flowering phenotype of the A. thaliana fld mutant. In GmFLD transgenic plants (Col or fld background), the FLC (FLOWERING LOCUS C) transcript levels decreased whereas the floral integrators, FT and SOC1, were up-regulated when compared with the corresponding non-transgenic genotypes. Furthermore, chromatin immuno-precipitation analysis showed that in the transgenic rescued lines (fld background), the levels of both tri-methylation of histone H3 Lys-4 and acetylation of H4 decreased significantly around the transcriptional start site of FLC. This is consistent with the function of GmFLD as a histone demethylase.
Our results suggest that GmFLD is a functional ortholog of the Arabidopsis FLD and may play an important role in the regulation of chromatin state in soybean. The present data provides the first evidence for the evolutionary conservation of the components in the autonomous pathway in soybean.
Flowering at an appropriate time is crucial for seed maturity and reproductive success in all flowering plants. Multiple flowering promotion pathways that respond to both environmental cues and endogenous factors have been evolved in plants to properly regulate flowering time. In the model plant A. thaliana, the photoperiod and vernalization pathways monitor seasonal changes in day length and temperature. These two pathways are responsible for initiating flowering in response to long days or prolonged cold temperatures (vernalization). The autonomous pathway, together with the gibberellin acid (GA) pathway, integrates signals from the developmental state of the plant and promotes flowering constitutively -.
In the photoperiod pathway, CONSTANS (CO) is the key protein and its expression is regulated by GI (GIGANTEA), which is under the control of the circadian clock . During long days, CO protein accumulates to high enough levels to promote floral transition, and as a result, up-regulates the expression of FT (FLOWERING LOCUS T) to initiate flowering ,. Both the vernalization pathway and autonomous pathway promote flowering through repression of the expression of FLC (FLOWERING LOCUS C), a central repressor of flowering in A. thaliana,. FLC is activated by FRI (FRIGIDA), a transcription factor with coiled coil motifs -. Natural A. thaliana winter annuals contain the dominant alleles of FRI and require vernalization, which represses FLC expression, for flowering ,,. Vernalization represses the expression of FLC by regulating the chromatin status of the FLC locus, and the underlying mechanism of this regulation was discussed in several critical reviews ,-. In contrast to winter annuals, natural rapid-cycling accessions do not have the functional FRI. FLC is suppressed constitutively by the so-called autonomous pathway genes, including FCA (FLOWERING LOCUS CA), FY (FLOWERING LOCUS Y), FPA (FLOWERING LOCUS PA), FVE (FLOWERING LOCUS VE), LD (LUMINIDEPENDENS), FLD (FLOWERING LOCUS D), and FLK (FLOWERING LOCUS KH DOMAIN) . Among these autonomous pathway components, FCA, FPA, FY and FLK participate in the RNA regulatory process in controlling flowering ,,, whereas LD, FVE, and FLD are involved in the regulation of the chromatin modification state ,,.
FLD encodes a plant ortholog of the human Lys-Specific Demethylase 1 (LSD1) protein. FLD functions in histone H3K4 demethylation and H3/H4 deacetylation to repress the expression of FLC-. In vivo, FLD is a sumoylation target of SIZ1 [SAP (scaffold attachment factor, acinus, protein inhibitor of activated signal transducer and activator of transcription) and Miz1 (Msx2-interacting zinc finger), SIZ], an E3 ligase in Arabidopsis. SUMO conjugation to FLD inhibits its repression activity for FLC expression and is required for full activation of FLC in a FRI background . Recently, Zhang et al.  reported that FLD expression is regulated by BRZ1 (BRASSINAZOLE-RESISTANT1) in a CYP20-2 dependent manner. Hence FLD may mediate brassinosteroid-controlled flowering regulation in Arabidopsis. FLD physically interacts with HDA6 to act synergistically in controlling the flowering of A. thaliana. In A. thaliana genome, two other FLD homolog, LSD1-LIKE1 (LDL1) and LSD1-LIKE2 (LDL2), act in partial redundancy with FLD to repress FLC expression. However, LDL1 and LDL2 act independently of FLD in the silencing of FWA (FLOWERING WAGENINGEN), a homeodomain-containing transcription factor. The FWA gene is silenced in the sporophyte and only expressed in the female gamete and extra-embryonic endosperm tissue in a maternal-imprinted manner .
Soybean is a typical short-day plant and the photoperiod sensitivity of different soybean cultivars is associated with their distribution range. Hence, soybean is also a short-day model plant for studying photoperiod response, and much progress has been made in identifying functions of the genes in the photoperiod pathway in soybean. To the best of our knowledge, at least ten FT homologs were experimentally identified in the soybean genome . Among these FT homologs, GmFT2a and GmFT5a are thought to be the florigen in soybean, and their expressions are regulated by the PHYA-mediated photoperiodic regulation system  as well as the classical maturity locus E1 encoding a novel plant transcription factor, which plays a pivotal role in controlling soybean flowering . Ectopic expression of GmFT2a and GmFT5a in A. thaliana resulted in premature flowering  and GmFT2a over-expression in soybean resulted in precocious flowering independent of photoperiod . CO has four homologs in soybean, GmCOL1a, GmCOL1b, GmCOL2a and GmCOL2b, and each of them can fully complement the late flowering effect of the co mutant in A. thaliana,. GIGANTEA has three soybean homologs, GmGI1a, GmGI1 and GmGI2, whose responses to circadian clock and photoperiod are different from each other ,,. GmGI1a is the classical maturity locus E2, who has multiple functions involved in the circadian clock and flowering ,. Hence, although soybean is a short-day plant, which is different from A. thaliana, the photoperiod pathway seems to be conserved between these two species .
Based upon the draft sequence of the soybean genome , homologs of autonomous pathway genes were also identified from the genome through bioinformatics analysis ,,. However, study on this group of genes has been limited. In this paper, the functions of the soybean FLD ortholog, GmFLD, were tested experimentally. Heterologous expression of GmFLD in A. thaliana resulted in early flowering of the transgenic plants and could partially complement the late flowering phenotype of fld mutants. In the GmFLD transgenic A. thaliana (Col or fld background), FLC transcript levels decreased and the floral integrator genes FT and SOC1 increased significantly. In the complementing transgenic lines, both histone H3 lysine4 trimethylation (H3K4me3) and H4 acetylation decreased around the transcriptional start site of FLC. Our results suggest that GmFLD is a functional soybean ortholog of FLD and may play an important role in the regulation of the chromatin modification state in soybean.
Soybean has four FLD homologs
In A. thaliana, FLD has other two homologs: LDL1 and LDL2. These homologs act redundantly with FLD to repress FLC transcription . By searching the NCBI soybean genome database using the Arabidopsis FLD protein sequence, four FLD homologs (E value = 0.0) were found: LOC100786453 (Glyma02g18610), LOC100810687 (Glyma09g31770), LOC100783933 (Glyma07g09980), and LOC100809901 (Glyma06g38600) with identity of 73%, 57%, 53% and 52% respectively. The deduced amino acid sequences of these four genes were then blasted against the A. thaliana proteome database (TAIR10), and the results show that LOC100786453 (Glyma02g18610) has the highest homology to the Arabidopsis FLD (73% identity), LOC100783933 (Glyma07g09980) and LOC100810687 (Glyma09g31770) are more similar to LDL1 (65% and 71% identity respectively), and LOC100809901 (Glyma06g38600) is more related to LDL2 (66% identity). Phylogenetic analysis with FLD homologs from different plant species show that plant LSD1 homologs are divided into three subgroups: LOC100786453 (Glyma02g18610) is clustered with the Arabidopsis FLD, LOC100809901 (Glyma06g38600) is in the LDL2 cluster, and LOC100783933 (Glyma07g09980) and LOC100810687 (Glyma09g31770) belong to LDL1 cluster (Figure 1). Hence, LOC100786453 (Glyma02g18610) is designated as GmFLD, LOC100809901 (Glyma06g38600) is designated as GmLDL2, and LOC100783933 (Glyma07g09980) and LOC100810687 (Glyma09g31770) are designated as GmLDL1A and GmLDL1B, respectively.
Both GmFLD and GmLDL2 contain the SWIRM and Amine Oxidase domains (Figure 2) that are characteristic of the LSD1 group of histone demethylases . The SWIRM and amino oxidase domains in GmFLD and GmLDL2 are organized in the same pattern as those in the Arabidopsis FLD and LDL2: SWIRM domain is at the N terminal while the C terminal contains the amine oxidase domain (Figure 2). However, in addition to the SWIRM and the amino oxidase domains, both GmLDL1A and GmLDL1B proteins contain new domains that are not present in the Arabidopsis: LDL1, LDL2 and FLD (Figure 2). GmLDL1A contains a NDA-binding-8 domain between the SWIRM domain and the amino oxidase domain, while GmLDL1B contains TAXi-N and TAXi-C domains at the N-terminal in front of the SWIRM domain. The NAD-binding-8 domain is involved in coenzyme binding , whereas the proteins containing TAXi domains are associated with proteolysis of phytopathogen xylanase secreted by the pathogen to degrade plant cell wall during plant pathogen infection . In addition, both GmLDL1A and GmLDL1B differ from the Arabidopsis LDL1 in that the soybean genes have intron (s) according to the annotations at NCBI and JGI databases (Additional file 1). Taken together, GmLDL1s probably diverged in functions from their counterpart LDL1 during evolution.
GmFLD and GmLDL2 exhibit different expression patterns from their Arabidopsis counterparts: FLD and LDL2
As noted above, among soybean homologs of FLD, GmFLD and GmLDL2 are more conserved in domain type and organization pattern than LDL1A and LDL1B. This suggests that the functions of GmFLD and GmLDL2 may be conserved in soybean. We thus examined whether GmFLD and GmLDL2 expressions in soybean have similar patterns with that of the Arabidopsis FLD and LDL2. Figure 3 shows that the transcripts of both genes could be detected in all tissues tested, including roots, hypocotyls and epicotyls, cotyledons, leaves, young pods, and flowers. This indicates that both genes are widely expressed in soybean. However, the transcript levels vary among different organs. The transcript abundance of both GmFLD and GmLDL2 was high in cotyledons, roots and pods, moderate in seedlings, hypocotyls and epicotyls, and flowers, and very low in true leaves, including unifoliate and trifoliate leaves (Figure 3). Interestingly, levels of both GmFLD and GmLDL2 transcripts were also very low in the shoot apex (Figure 3), which is very different from previous reports showing that the Arabidopsis FLD and LDL2 are preferentially expressed in shoot apex ,.
Both GmFLD and GmLDL2 proteins are localized in nuclei
As putative histone demethylases, GmFLD and GmLDL2 should function in the nucleus. However, a bioinformatics prediction at http://psort.hgc.jp/form.html showed that only GmFLD has putative nuclear localization sites (NLS) and may localize in the nucleus, while GmLDL2 was predicted to localize in the mitochondrial matrix space or cytoplasm. Hence, a transient expression assay was performed to examine the subcellular localization of GmFLD and GmLDL2. The constructs 35S::GmFLD-YFP and 35S::GmLDL2-YFP were used respectively to co-transform rice protoplasts with 35S::Ghd7-CFP, a marker for nuclear localization . Figure 4 shows that yellow fluorescent protein (YFP) signals of both GmFLD and GmLDL2 were clearly overlapped with Cyan Fluorescent Protein (CFP) signals, and no significant fluorescence signals were detected in the cytoplasm. This indicates that both GmFLD and GmLDL2 are localized in the nucleus. This subcellular localization pattern is consistent with the putative functions of GmFLD and GmLDL2 as histone demethylases.
GmFLD but not GmLDL2 promotes flowering in A. thaliana
Since GmFLD and GmLDL2 showed different expression patterns from their Arabidopsis counterparts FLD and LDL2 (Figure 3), we examined whether GmFLD and GmLDL2 could function as a flowering time control, similar to the Arabidopsis FLD. GmFLD and GmLDL2 CDSs driven by the cauliflower mosaic virus (CaMV) 35S promoter were introduced into A. thaliana (Col-0) to assess flowering phenotype of transgenic plants. The transgenic T1 plants expressing GmFLD flowered significantly earlier than wild type plants (Table 1), and the early flowering phenotype was also observed in the progenies of the T1 plants (Figure 5A, 5B, 5C). In contrast to GmFLD, the transgenic plants overexpressing GmLDL2 did not show significant changes in flowering time (Table 1).
GmFLD complements the late flowering phenotype of fld mutant
Since GmFLD could promote flowering in the Arabidopsis wild type background, we further checked whether GmFLD could rescue the late flowering phenotype of the Arabidopsis fld mutant. The 35S::GmFLD construct was introduced into the fld mutant and in the T1 generation, most GmFLD transgenic plants flowered as early as the Col wild type plants (19 out of 27 plants flowered early, Table 1). Homozygous single-copy transgenic lines were screened from these early flowering transgenic plants for further analysis. Flowering phenotype scoring showed that the progeny plants consistently flowered much earlier than the fld plants, but not as early as the Col wild type plants (Figure 5E, 5G). The fld mutants produced approximately 67.6 ± 3.9 leaves (rosette plus cauline leaves) before flowering while transgenic plants produced only 21.9 ± 1.8 leaves, and Col wild type plants produced 14.5 ± 2.3 leaves. The lifecycle of the transgenic plants were also shortened. This was observed by evaluating the time when flower buds became visible or when flowers started to open (Table 1). These results reveal that GmFLD could partially complement the phenotype of the fld mutant. As expected, GmFLD transcript could only be detected in those phenotype-complementary transgenic lines (Figure 5F), whereas in those non-complementary transgenic plants (8 out of 27 T1 plants flowered as late as fld), GmFLD transcripts could not be detected although the transgene does exist in these transgenic T1 plants (data not shown).
GmFLD promotes flowering in A. thaliana through repressing FLC transcription
In A. thaliana, FLD promotes flowering through repressing FLC transcription ,. To assess whether GmFLD promotes flowering through the same mechanism, the transcript levels of FLC and the floral integrator genes, FT and SOC1, acting downstream of FLC in the transgenic A. thaliana (Col-0 or fld background) were analyzed. In the transgenic plants in Col background, FLC transcript level decreased significantly while FT and SOC1 were up-regulated, the SOC1 level were especially increased (Figure 5D). Similar trends were observed in the transgenic plants in the fld background (Figure 5H). Hence, GmFLD promotes flowering in A. thaliana through repressing FLC transcription as its A. thaliana counterpart FLD does. Taken the above results together, our experiments demonstrate that GmFLD is a functional homolog of the Arabidopsis FLD.
GmFLD decreased the levels of histone H3K4me3 and H4 acetylation at the FLC locus
In A. thaliana, FLD represses FLC transcription through affecting the state of H3K4 methylation and H4 acetylation ,,. So we examined the modification state of H3K4 and H4 in FLC chromatin in the transgenic plants that rescued fld mutant phenotype. In the GmFLD complementing plants, the level of H3K4me3 near the transcriptional start site (P3 region) was significantly decreased, and was about half of that in the fld plants. However, we did not find significant changes in the level of H3K4me3 modification in other regions tested (P1, P2, P4) (Figure 6B), which is consistent with previous reports ,. Hence, GmFLD could recover at least partially the H3K4 methylation levels of FLC chromatin in the fld mutant. The acetylation level of H4 in the region around the FLC transcription start site was also decreased significantly, whereas no obvious change was found in the P1, P2 and P4 regions tested (Figure 6C). These results suggest that GmFLD represses FLC transcription possibly through decreasing the modification levels of H3K4me3 and H4 acetylation in FLC chromatin.
Soybean is a typical photoperiod-sensitive crop and photoperiod is an important factor that determines its flowering time. Hence, since the whole genomic sequence of soybean was released , the functions of many photoperiod pathway genes, including GmFTs, GmCOs and GmGIs, have been identified and characterized ,,-,. However, little is known about the functions of autonomous pathway genes in soybean, although most of the A. thaliana autonomous pathway genes have more than one orthologs in soybean as predicted by bioinformatics analysis ,,. In this report, the functions of FLD homologs in soybean were studied by using bioinformatics, genetic and molecular tools. Our results provide solid evidence to support the function and evolution of autonomous pathway genes in plants.
GmFLD is a functional homolog of FLD
In A. thaliana, FLD plays a major role in promoting flowering, while LDL1 and LDL2 play a minor role and act redundantly with FLD,. In palaeopolyploid soybean, both FLD and LDL2 only have a single ortholog, GmFLD and GmLDL2 respectively, and both have functional domains arranged in the same pattern as that in the Arabidopsis FLD and LDL2 (Figure 2). Notably, GmFLD could complement the late flowering phenotype of the A. thaliana fld mutant plants (Figure 5E, 5G). In the transgenic plants, FLC expression is down-regulated and the downstream floral integrator genes SOC1 and FT are up-regulated (Figure 5D, 5H). To our knowledge, no FLD homologs from other plants have been experimentally identified and characterized, although the homologs of several other autonomous pathway genes, including OsFCA and OsFVE of rice, BvFVE and BvFLK of Beta vulgaris, and ZmLD of maize -, were studied. Among them, only BvFLK and OsFVE could complement the late-flowering phenotype of Arabidopsis flk or fve mutant through FLC repression ,. Our ChIP assay further demonstrates that GmFLD is involved in the regulation of the chromatin modification state at FLC locus (Figure 6B, 6C). Hence GmFLD operated in the transgenic A. thaliana in the same manner as the native FLD does ,. Although FLC orthologs were not identified in leguminous plants previously , a recent comparative genomic analysis of soybean flowering genes indicated that soybean has an FLC homolog, GmFLC (Glyma05g28130) . Whether GmFLC is the target of GmFLD and (or) other autonomous pathway genes is unknown at present. However from our results, it is reasonable to propose that GmFLD may repress its target gene expression through regulation of the chromatin modification state to control flowering in soybean. Our results suggest that FLD and GmFLD are functionally related regulators of the chromatin modification state and provide the first experimental evidence for evolutionary conservation of FLD function between Arabidopsis and soybean.
Functional conservation and divergence of FLD homologs in soybean
FLD was identified to physically interact with HDA6, a histone deacetylase involved in gene silencing, to function synergistically in chromatin modification . In the complementing transgenic plants, both the H3K4me3 and H4 acetylation levels decreased as compared to those in the fld mutant (Figure 6B, 6C). This result suggests that, in addition to its histone demethylase function, GmFLD may also interact with histone deacetylase to affect histone acetylation in chromatin. However, in our preliminary study, interaction between GmFLD and the soybean HDA6 homologs was not detected by yeast two-hybrid analysis (data not shown). Furthermore, the expression pattern of GmFLD in soybean is somewhat different from that of FLD in A. thaliana. The Arabidopsis FLD is preferentially expressed in apical meristem regions of roots and shoots, but the transcript level of GmFLD is very low in the shoot apex of soybean (Figure 3A). The key function of FLD in A. thaliana is to promote flowering through repressing the expression of FLC, which is epigenetically silenced by vernalization ,-. Different from A. thaliana, soybean does not require vernalization to induce flowering. Therefore, it is conceivable that GmFLD probably has additional functions other than flowering control in soybean. Further characterization of the soybean GmFLD will be performed in our future work.
In A. thaliana, LDL2 acts redundantly with FLD and LDL1 to repress FLC expression, and LDL2 also has overlapping function with LDL1 to repress sporophytic expression of FWA. However, LDL2 itself plays a minor role in promoting flowering of A. thaliana, and loss-of-function ldl2 mutants do not have significant phenotypic changes . This may explain why heterologous expression of GmLDL2 in A. thaliana did not result in significant flowering phenotype changes. Based on its sequence similarity with LDL2 (Figures 1 and 2), nuclear localization (Figure 4) and expression pattern similar to that of GmFLD (Figure 3B), we propose that GmLDL2 is the functional ortholog of LDL2 and probably acts redundantly with GmFLD to repress the gene expression in soybean. However, its biological roles in soybean still require further investigation. Soybean appears to have two LDL1 orthologs, GmLDL1A and GmLDL1B. Interestingly, GmLDL1A and GmLDL1B gained additional functional domains during evolution. The occurrence of TAXi-N and TAXi-C domains in GmLDL1B also suggests functions in pathogen resistance .
Functional divergence of autonomous pathway genes
Autonomous pathway genes were originally identified from a group of A. thaliana late flowering mutants and their homologs apparently exist widely in plant kingdom ,. As for FLD, two homologs were identified in the genome of Physcomitrella patens (Figure 1), a cryptogam without floral transition. This suggests that some FLD homologs may play pivotal roles in other developmental processes other than flowering. In cells, autonomous pathway components are involved in chromatin modification and RNA processing, which play important roles in multiple physiological processes such as growth and development, response to abiotic stress, etc. ,. Therefore, it is not surprising that some autonomous pathway genes have additional functions in regulating growth and developmental processes other than flowering. For example, double mutant plants, fpa fld, fpa fve, and fpa ld showed pleiotropic effects on growth rate, chlorophyll content, leaf morphology, flower development, and fertility . Furthermore, some experimental evidence show that both FCA and FVE play a role in thermosensory flowering pathway ,, whereas FY is involved in the development of seed dormancy and ABA sensitivity in A. thaliana. On the other hand, some orthologs of the A. thaliana autonomous pathway genes from other species appear to have diversified in function and/or acting mechanism. Rice FCA homolog OsFCA could partially rescue the late flowering phenotype of the Arabidopsis fca mutant, but through the activation of SOC1 rather than FLC down-regulation. The OsFCA also does not have a negative feedback to regulate the OsFCA mRNA level as the Arabidopsis FCA does . In addition, OsFCA has interaction partners in rice, including OsSF1, OsFIK1 and OsMADS8  that were not identified in Arabidopsis. BvFVE1 of sugarbeet showed 72% amino acid identity to FVE, but could not complement the phenotype of the Arabidopsis fve mutant , whereas maize ZmLD not only failed to complement the ld phenotype, but resulted in other developmental defects in A. thaliana. Taken the above together, the biological functions of autonomous pathway genes are complex and it is of great interest to probe the biological functions of autonomous pathway components in other plants in addition to A. thaliana. Our present data provide the first evidence for evolutionary conservation of the components in the autonomous pathway of flowering in soybean.
In soybean, FLD has four homologs, GmFLD, GmLDL2, GmLDL1A, and GmLDL1B. GmFLD is a functional ortholog of the Arabidopsis FLD and may play an important role in regulation of the chromatin modifying state in soybean. GmLDL2 is a functional ortholog of LDL2 and may function redundantly with GmFLD in soybean.
The A. thaliana protein sequences of FLD, LDL1, LDL2, and LDL3 were downloaded from The A. thaliana Information Resource . The FLD protein sequence was used to search NCBI  and JGI phytozome soybean databases  using the blastp algorithm. At this round of search, four FLD homologs (E value = 0.0) were obtained: LOC100786453 (Glyma02g18160), LOC100809901 (Glyma06g38600), LOC100783933 (Glyma07g09980), and LOC100810687 (glyma09g31770). To help infer orthology by bidirectional best hit (BBH) analysis , the soybean protein sequences retrieved through the above analysis were used as queries to blast the TAIR10 proteins dataset . For phylogenetic analysis, putative FLD homologs in other plants were identified and retrieved from NCBI database as described above. The rooted phylogenetic tree was constructed by Mega 5.0  and the conserved protein domains were identified using PFAM 27.0 . The subcellular localization of GmFLD and GmLDL2 was predicted through online analysis .
Plant materials and growth conditions
The soybean cultivar Zhongdou32 (Glycine max L. Merr.) was used in this study. The soybean plants were grown in pots with soil/vermiculite mixture (V/V = 1:1) in a growth chamber in short-day conditions (8 h light and 16 h dark) at 24-26°C. All A. thaliana materials, including Col, fld-1 mutant  and other transgenic lines, were grown in long days conditions (16 h light/8 h dark) at 22°C in soil/vermiculite mixture or - MS agar plates according to experimental requirement. All A. thaliana seeds were stratified for 2 days at 4°C before being moved into the growth chamber.
Expression pattern analysis of soybean FLD homologs
For RNA extraction, plant samples were collected as follows: seedling samples were harvested at the stage when the cotyledons expanded fully. At the unifoliolate stage (when unifoliolates expand fully), hypocotyl, epicotyl, cotyledon and shoot apex (including the apical meristem and leaf primordia) materials were collected. At the flowering stage (when flowers start to open), the flowers, the 1st, 2nd, 3rd and 4th trifoliolates (from bottom to top) were harvested. The root and unifoliolate samples were collected separately at both the unifoliotate and flowering stages. The pods were sampled separately at 7, 14 and 21 days after flowering. All samples were frozen in liquid nitrogen and stored at -80°C until use.
Total RNA was extracted from seedlings, roots, hypocotyls, epicotyls, leaves, flowers, and pods using Trizol reagent (Invitrogen, USA) according to the manufacturer’s instructions. cDNA was synthesized by using Prime Script™ RT regent Kit with gDNA Eraser (Takara, Japan). The real-time quantitative PCR was performed on a C1000 Touch TM Thermal cycler with SYBR Premix Dimer Eraser™ (Takara, Japan). Each assay was quantified in triplicate and normalized using the actin-encoding gene, glyma02g10170, as an internal control. All experiments had three biological replicates. The primers were listed in Additional file 2 (the same for other primers described below).
Subcellular localization assay
Full-length CDSs of GmFLD and GmLDL2 were amplified by RT-PCR from the seedling RNA sample and inserted into the vector pM999 at SacI and NcoI restriction sites to generate the transient expression constructs 35S::GmFLD-YFP and 35S::GmLDL2-YFP. The constructs were sequenced and introduced into rice protoplasts according to the method described by Bart et al.  and Wang et al. . In brief, about 30 μg endotoxin-free construct DNA was used to transform rice protoplasts. The construct 35S::Ghd7-CFP was used as the nuclear localization marker while the 35S::YFP was used as an empty control . The transformed cells were observed and imaged under the confocal laser scanning microscope (Zeiss LSM dater server). For each subcellular location analysis, at least three biological replicates were performed and at least 10 cells were examined in each sample.
Heterologous expression of soybean FLD homologs in A. thaliana
The full length CDSs of GmFLD and GmLDL2 were amplified from the soybean cDNA and cloned into the vector pBI121 at the Xbal and SacI restriction sites, downstream of 35S promoter of cauliflower mosaic virus, to produce the over-expression binary vectors pBI121-GmFLD and pBI121-GmLDL2. After being sequenced, the constructs were introduced into the Agrobacterium tumefaciens strain GV3101 for A. thaliana transformation. The A. thaliana plants (Col wild type or fld mutant) were transformed by the floral dip method . The harvested seeds (T1 generation) were selected on 1/2 MS agar media containing 50 mg/L kanamycin. The positive T1 seedlings were transferred to soil/vermiculite mixture to grow for phenotype assay and collection of T2 seeds. The T2 seeds were sowed on kanamycin plates to examine the copy number of the transgene. Only the single-copy transgenic lines were further propagated for producing T3 transgenic homozygous seeds for further experiments. The flowering time was assessed by numbers of rosette and cauline leaves.
The expression of GmFLD in transgenic A. thaliana was determined by semi quantitative RT-PCR. The expression of FLC, SOC1, and FT in transgenic A. thaliana was examined by using real-time quantitative RT-PCR. The RNA was extracted from the seedlings (ten days old) growing on 1/2 MS agar plates according to the method described above. Each assay was quantified in triplicate and normalized using ACT2 (AT3g18780) as an internal control. All experiments had three biological replicates.
Chromatin immuno-precipitation (ChIP) analysis
ChIP analysis was performed according to the protocols described previously by Jiang et al. . The leaves from four-week-old A. thaliana plants were harvested for experiment. Anti-trimethyl-histone H3K4 and anti-acetyl-histone H4K5K8K12K16 were purchased from Millipore Corporation. The amounts of immuno-precipitated genomic DNA were determined by real-time quantitative PCR. Each assay was quantified in triplicate and normalized using EIF4A1 (AT3g13920) as an internal control. All experiments had three biological replicates.
QH carried out the experiments. YJ performed the statistical analysis. QH and WY drafted the manuscript. HS revised the manuscript. WY conceived of the study and participated in its design and coordination. All authors read and approved the final manuscript.
Andres F, Coupland G: The genetic basis of flowering responses to seasonal cues. Nat Rev Genet. 2012, 13: 627-639. 10.1038/nrg3291.
He Y: Chromatin regulation of flowering. Trends Plant Sci. 2012, 17: 556-562. 10.1016/j.tplants.2012.05.001.
Pose D, Yant L, Schmid M: The end of innocence: flowering networks explode in complexity. Curr Opin Plant Biol. 2012, 15: 45-50. 10.1016/j.pbi.2011.09.002.
Song YH, Ito S, Imaizumi T: Flowering time regulation: photoperiod- and temperature-sensing in leaves. Trends Plant Sci. 2013, 18: 575-583. 10.1016/j.tplants.2013.05.003.
Samach A, Onouchi H, Gold SE, Ditta GS, Schwarz-Sommer Z, Yanofsky MF, Coupland G: Distinct roles of CONSTANS target genes in reproductive development of Arabidopsis. Science. 2000, 288: 1613-1616. 10.1126/science.288.5471.1613.
Michaels SD, Amasino RM: FLOWERING LOCUS C encodes a novel MADS domain protein that acts as a repressor of flowering. Plant Cell. 1999, 11: 949-956. 10.1105/tpc.11.5.949.
Sheldon CC, Rouse DT, Finnegan EJ, Peacock WJ, Dennis ES: The molecular basis of vernalization: the central role of FLOWERING LOCUS C (FLC). Proc Natl Acad Sci U S A. 2000, 97: 3753-3758. 10.1073/pnas.97.7.3753.
Johanson U, West J, Lister C, Michaels S, Amasino R, Dean C: Molecular analysis of FRIGIDA, a major determinant of natural variation in Arabidopsis flowering time. Science. 2000, 290: 344-347. 10.1126/science.290.5490.344.
Gazzani S, Gendall AR, Lister C, Dean C: Analysis of the molecular basis of flowering time variation in Arabidopsis accessions. Plant Physiol. 2003, 132: 1107-1114. 10.1104/pp.103.021212.
Choi K, Kim J, Hwang HJ, Kim S, Park C, Kim SY, Lee I: The FRIGIDA complex activates transcription of FLC, a strong flowering repressor in arabidopsis, by recruiting chromatin modification factors. Plant Cell. 2011, 23: 289-303. 10.1105/tpc.110.075911.
Michaels SD, He YH, Scortecci KC, Amasino RM: Attenuation of FLOWERING LOCUS C activity as a mechanism for the evolution of summer-annual flowering behavior in Arabidopsis. Proc Natl Acad Sci U S A. 2003, 100: 10102-10107. 10.1073/pnas.1531467100.
Kim DH, Doyle MR, Sung S, Amasina RM: Vernalization: winter and the timing of flowering in plants. Annu Rev Cell Dev Biol. 2009, 25: 277-299. 10.1146/annurev.cellbio.042308.113411.
Sheldon CC, Finnegan EJ, Peacock WJ, Dennis ES: Mechanisms of gene repression by vernalization in Arabidopsis. Plant J. 2009, 59: 488-498. 10.1111/j.1365-313X.2009.03883.x.
Amasino R: Seasonal and developmental timing of flowering. Plant J. 2010, 61: 1001-1013. 10.1111/j.1365-313X.2010.04148.x.
Kim DH, Sung S: Genetic and epigenetic mechanisms underlying vernalization. Arabidopsis Book. 2014, 11: e0171-10.1199/tab.0171.
Simpson GG: The autonomous pathway: epigenetic and post-transcriptional gene regulation in the control of Arabidopsis flowering time. Curr Opin Plant Biol. 2004, 7: 570-574. 10.1016/j.pbi.2004.07.002.
Rataj K, Simpson GG: Message ends: RNA 3′ processing and flowering time control. J Exp Bot. 2014, 65: 353-363. 10.1093/jxb/ert439.
He Y, Michaels SD, Amasino RM: Regulation of flowering time by histone acetylation in Arabidopsis. Science. 2003, 302: 1751-1754. 10.1126/science.1091109.
Jiang D, Yang W, He Y, Amasino RM: Arabidopsis relatives of the human lysine-specific demethylase1 repress the expression of FWA and FLOWERING LOCUS C and thus promote the floral transition. Plant Cell. 2007, 19: 2975-2987. 10.1105/tpc.107.052373.
Yu CW, Liu X, Luo M, Chen C, Lin X, Tian G, Lu Q, Cui Y, Wu K: HISTONE DEACETYLASE 6 interacts with FLOWERING LOCUS D and regulates flowering in Arabidopsis. Plant Physiol. 2011, 156: 173-184. 10.1104/pp.111.174417.
Jin JB, Jin YH, Lee J, Miura K, Yoo CY, Kim WY, Oosten MV, Hyun Y, Somers DE, Lee I, Yun DJ, Bressan RA, Hasegawa PM: The SUMO E3 ligase, AtSIZ1, regulates flowering by controlling a salicylic acid-mediated floral promotion pathway and through affects on FLC chromatin structure. Plant J. 2008, 53: 530-540. 10.1111/j.1365-313X.2007.03359.x.
Zhang Y, Li B, Xu Y, Li H, Li S, Zhang D, Mao Z, Guo S, Yang C, Weng Y, Chong K: The cyclophilin CYP20-2 modulates the conformation of BRASSINAZOLE-RESISTANT1, which binds the promoter of Flowering Locus D to regulate flowering in A. thaliana. Plant Cell. 2013, 25: 2504-2521. 10.1105/tpc.113.110296.
Thakare D, Kumudini S, Dinkins RD: The alleles at the E1 locus impact the expression pattern of two soybean FT-like genes shown to induce flowering in Arabidopsis. Planta. 2011, 234: 933-943. 10.1007/s00425-011-1450-8.
Kong F, Liu B, Xia Z, Sato S, Kim B, Watanabe S, Yamada T, Tabata S, Kanazawa A, Harada K, Abe J: Two coordinately regulated homologs of FLOWERING LOCUS T are involved in the control of photoperiodic flowering in soybean. Plant Physiol. 2010, 154: 1220-1231. 10.1104/pp.110.160796.
Xia Z, Watanabe S, Yamada T, Tsubokura Y, Nakashima H, Zhai H, Anai T, Sato S, Yamazaki T, Lü S, Wu H, Tabata S, Harada K: Positional cloning and characterization reveal the molecular basis for soybean maturity locus E1 that regulates photoperiodic flowering. Proc Natl Acad Sci U S A. 2012, 109: E2155-E2164. 10.1073/pnas.1117982109.
Sun H, Jia Z, Cao D, Jiang B, Wu C, Hou W, Liu Y, Fei Z, Zhao D, Han T: GmFT2a, a soybean homolog of FLOWERING LOCUS T, is involved in flowering transition and maintenance. PLoS One. 2011, 6: e29238-10.1371/journal.pone.0029238.
Kim MY, Shin JH, Kang YJ, Shim SR, Lee SH: Divergence of flowering genes in soybean. J Biosci. 2012, 37: 857-870. 10.1007/s12038-012-9252-0.
Wu F, Price BW, Haider W, Seufferheld G, Nelson R, Hanzawa Y: Functional and evolutionary characterization of the CONSTANS gene family in short-day photoperiodic flowering in soybean. PLoS One. 2014, 9: e85754-10.1371/journal.pone.0085754.
Watanabe S, Xia Z, Hideshima R, Tsubokura Y, Sato S, Yamanaka N, Takahashi R, Anai T, Tabata S, Kitamura K, Harada K: A map-based cloning strategy employing a residual heterozygous line reveals that the GIGANTEA gene is involved in soybean maturity and flowering. Genetics. 2011, 188: 395-407. 10.1534/genetics.110.125062.
Li F, Zhang X, Hu R, Wu F, Ma J, Meng Y, Fu Y: Identification and molecular characterization of FKF1 and GI homologous genes in soybean. PLoS One. 2013, 8: e79036-10.1371/journal.pone.0079036.
Fan C, Hu R, Zhang X, Wang X, Zhang W, Zhang Q, Ma J, Fu YF: Conserved CO-FT regulons contribute to the photoperiod flowering control in soybean. BMC Plant Biol. 2014, 14: 9-10.1186/1471-2229-14-9.
Schmutz J, Cannon SB, Schlueter J, Ma J, Mitros T, Nelson W, Hyten DK, Song Q, Thelen JJ, Cheng J, Xu D, Hellsten U, May GD, Yu Y, Sakurai T, Umezawa T, Bhattacharyya MK, Sandhu D, Valliyodan B, Lindquist E, Peto M, Grant D, Shu S, Goodstein D, Barry K, Futrell-Griggs M, Abernathy B, Du J, Tian Z, Zhu L, et al: Genome sequence of the palaeopolyploid soybean. Nature. 2010, 463: 178-183. 10.1038/nature08670.
Jung CH, Wong CE, Singh MB, Bhalla PL: Comparative genomic analysis of soybean flowering genes. PLoS One. 2012, 7: e38250-10.1371/journal.pone.0038250.
Bashton M, Chothia C: The geometry of domain combination in proteins. J Mol Biol. 2002, 315: 927-939. 10.1006/jmbi.2001.5288.
Pollet A, Sansen S, Raedschelders G, Gebruerk K, Rabijns A, Delcour JA, Courtin CM: Identification of structural determinants for inhitition strenth and specificity of wheat xylanase inhibitors TAXI-IA and TAXI-IIA. FEBS J. 2009, 276: 3916-3927. 10.1111/j.1742-4658.2009.07105.x.
Xue W, Xing Y, Weng X, Zhao Y, Tang W, Wang L, Zhou H, Yu S, Xu C, Li X, Zhang Q: Natural variation in Ghd7 is an important regulator of heading date and yield potential in rice. Nat Genet. 2008, 40: 761-767. 10.1038/ng.143.
Jiang D, Gu X, He Y: Establishment of the winter-annual growth habit via FRIGIDA-mediated histone methylation at FLOWERING LOCUS C in Arabidopsis. Plant Cell. 2009, 21: 1733-1746. 10.1105/tpc.109.067967.
Yu X, Michaels SD: The Arabidopsis Paf1c Complex component CDC73 participates in the modification of FLOWERING LOCUS C chromatin. Plant Physiol. 2010, 153: 1074-1084. 10.1104/pp.110.158386.
Wu FQ, Zhang XM, Li DM, Fu YF: Ectopic expression reveals a conserved PHYB homolog in soybean. PLoS One. 2011, 6: e27737-10.1371/journal.pone.0027737.
Lam HM, Xu X, Liu X, Chen W, Yang G, Wong FL, Li MW, He W, Qin N, Wang B, Li J, Jian M, Wang J, Shao G, Wang J, Sun SS, Zhang G: Resequencing of 31 wild and cultivated soybean genomes identifies patterns of genetic diversity and selection. Nat Genet. 2010, 42: 1053-1059. 10.1038/ng.715.
Van Nocker S, Muszynski M, Briggs SK, Amasina RM: Characterizatin of a gene from Zea mays related to the Arabidopsis flowering-time gene LUMINIDEPENDENS. Plant Mol Biol. 2000, 44: 107-122. 10.1023/A:1006472929800.
Lee JH, Cho YS, Yoon HS, Suh MC, Moon J, Lee I, Weigel D, Yun CH, Kim JK: Conservation and divergence of FCA function between Arabidopsis and rice. Plant Mol Biol. 2005, 58: 823-838. 10.1007/s11103-005-8105-8.
Baek IS, Park HY, You MK, Lee JH, Kim JK: Funcitonal conservation and divergence of FVE genes that control flowering time and cold response in rice and Arabidopsis. Mol Cells. 2008, 26: 368-372.
Abou-Elwafa SF, Buttner B, Chia T, Schulze-Buxloh G, Hohmann U, Mutasa-Gottgens E, Jung C, Muller AE: Conservation and divergence of autonomous pathway genes in the flowering regulatory network of Beta vulgaris. J Exp Bot. 2011, 62: 3359-3374. 10.1093/jxb/erq321.
Hecht V, Foucher F, Ferrandiz C, Macknight R, Navarro C, Morin J, Vardy ME, Ellis N, Beltran JP, Rameau C, Weller JL: Conservation of Arabidopsis flowering genes in model legumes. Plant Physiol. 2005, 137: 1420-1434. 10.1104/pp.104.057018.
Veley KM, Michaels SD: Functional redundancy and new roles for genes of the autonomous floral-promotion pathway. Plant Physiol. 2008, 147: 682-695. 10.1104/pp.108.118927.
Jung JH, Seo PJ, Ahn JH, Park CM: Arabidopsis RNA-binding protein FCA regulates MicroRNA172 processing in thermosensory flowering. J Biol Chem. 2012, 287: 16007-16016. 10.1074/jbc.M111.337485.
Kim HJ, Hyun Y, Park JY, Park MJ, Park MK, Kim MD, Lee MH, Moon J, Lee I, Kim J: A genetic link between cold responses and flowering time through FVE in Arabidopsis thaliana. Nat Genet. 2004, 36: 167-171. 10.1038/ng1298.
Jiang S, Kumar S, Eu YJ, Jami SK, Stasolla C, Hill RD: The Arabidopsis mutant, fy-1, has an ABA-insensitive germination phenotype. J Exp Bot. 2012, 63: 2693-2703. 10.1093/jxb/err452.
Jang YH, Park HY, Kim SK, Lee JH, Suh MC, Chung YS, Paek KH, Kim JK: Survey of rice proteins interacting with OsFCA and OsFY proteins which are homologous to the Arabidopsis flowering time proteins, FCA and FY. Plant Cell Physiol. 2009, 50: 1479-1492. 10.1093/pcp/pcp093.
The Arabidopsis Information Resource , [www.arabidopsis.org]
Soybean database at NCBI , [http://www.ncbi.nlm.nih.gov/projects/mapview/map_search.cgi?taxid=3847&build=1.1]
Soybean database at JGI phytozome , [http://phytozome.jgi.doe.gov/pz/portal.html]
Overbeek R, Fonstein M, D’Souza M, Pusch GD, Maltsev N: The use of gene clusters to infer functional coupling. Proc Natl Acad Sci U S A. 1999, 96: 2896-2901. 10.1073/pnas.96.6.2896.
Tamura K, Peterson D, Peterson N, Stecher G, Nei M, Kumar S: MEGA5: molecular evolutionary genetics analysis using maximum likelihood, evolutionary distance, and maximum parsimony methods. Mol Biol Evol. 2011, 28: 2731-2739. 10.1093/molbev/msr121.
Punta M, Coggill PC, Eberhardt RY, Mistry J, Tate J, Boursnell C, Pang N, Forslund K, Ceric G, Clements J, Heger A, Holm L, Sonnhammer ELL, Eddy SR, Bateman A, Finn RD: The Pfam protein families database. Nucleic Acids Res. 2012, 40 (Database Issue): D290-D301. 10.1093/nar/gkr1065.
PSORT Prediction , [http://psort.hgc.jp/form.html]
Bart R, Chern M, Park CJ, Bartley L, Ronald PC: A novel system for gene silencing using siRNAs in rice leaf and stem-derived protoplasts. Plant Methods. 2006, 2: 13-10.1186/1746-4811-2-13.
Wang Y, Peng W, Zhou X, Huang F, Shao L, Luo M: The putative Agrobacterium transcriptional activator-like virulence protein VirD5 may target T-complex to prevent the degradation of coat proteins in the plant cell nucleus. New Phytol 2014. doi:10.1111/nph.12866.,
Clough SJ, Bent AF: Floral dip: a simplified method for Agrobacterium-mediated transformation of Arabidopsis thaliana. Plant J. 1998, 16: 735-743. 10.1046/j.1365-313x.1998.00343.x.
We thank Dr. Yuehui He (National University of Singapore) for providing fld-1 mutant seeds, Dr Jian Xu (National University of Singapore) for providing pM999 plasmid, and Professor Xin’an Zhou (Oilcrops Research Institute, Chinese Academy of Agricultural Sciences) for providing soybean seeds. This work was supported in part by National Natural Science Foundation of China (Grant no. 31270316) to W Yang.
The authors declare that they have no competing interests.