APETALA2-like Floral Homeotic Protein Up-Regulating FaesAP1_2 Gene Involved in Floral Development in Long-Homostyle Common Buckwheat

Abstract

In the rosid species Arabidopsis thaliana, the AP2-type AP2 transcription factor (TF) is required for specifying the sepals and petals identities and confers a major A-function to antagonize the C-function in the outer floral whorls. In the asterid species Petunia, the AP2-type ROB TFs are required for perianth and pistil development, as well as repressing the B-function together with TOE-type TF BEN. In Long-homostyle (LH) Fagopyrum esculentum, VIGS-silencing showed that FaesAP2 is mainly involved in controlling filament and style length, but FaesTOE is mainly involved in regulating filament length and pollen grain development. Both FaesAP2 (AP2-type) and FaesTOE (TOE-type) are redundantly involved in style and/or filament length determination instead of perianth development. However, neither FaesAP2 nor FaesTOE could directly repress the B and/or C class genes in common buckwheat. Moreover, the FaesAP1_2 silenced flower showed tepal numbers, and filament length decreased obviously. Interestingly, yeast one-hybrid (Y1H) and dual-luciferase reporter (DR) further suggested that FaesTOE directly up-regulates FaesAP1_2 to be involved in filament length determination in LH common buckwheat. Moreover, the knockdown of FaesTOE expression could result in expression down-regulation of the directly target FaesAP1_2 in the FaesTOE-silenced LH plants. Our findings uncover a stamen development pathway in common buckwheat and offer deeper insight into the functional evolution of AP2 orthologs in the early-diverging core eudicots.

1. Introduction

The ABC model of flower development explains how three classes of homeotic genes confer identity to the four types of floral organs in Arabidopsis thaliana [1,2]. According to this model, APETALA2 (AP2) and AGAMOUS (AG) represent A- and C-class genes that act in an antagonistic fashion to specify the perianth (sepal and petal) and reproductive organs (stamen and pistil), respectively. AP2 binds the AT-rich target sequence lying at the AG second intron and directly restricts AG expression in the sepals and petals [3]. Moreover, AP2 also antagonizes AG activity in the center of the flower to regulate floral stem cells and maintains shoot apical meristem (SAM) activity in part by keeping WUSCHEL (WUS) expression active [4,5]. As a flowering repressor, AP2 is negatively regulated by miR172 and MADS-box transcription factor FRUITFULL (FUL), as well as directly repressing the expression of flowering-promoting transcript AP1 to control the timing of the floral transition [6,7,8]. In addition, the AP2 orthologs (BrAP2a and BnAP2) from Brassica Crops showed high conservation in specifying sepal development [9]. Losses in the function of BrAP2a or BnAP2 will result in a sepal–carpel modification phenotype. The AP2 family can be divided into the euANT, basalANT, and euAP2 lineages [10]. The euAP2 lineage can be further divided into two different classes called the TOE and AP2 types [11]. Within the euAP2 clade, AP2 controls flower development and floral organ identity, while TOE1 and TOE3 (TARGET OF EAT1/3) regulate flowering [12,13]. However, in the asterid species Petunia (Petunia hybrida), the TOE-type gene BLIND ENHANCER (BEN) acts as a floral repressor and represses the C-function together with BLIND (BL), while the AP2-type REPRESSOR OF B-FUNCTION (ROB) genes are required for perianth and pistil development, as well as repress the B-function (but not the C-function) together with BEN in the first floral whorl [11]. The BEN and ROB act as major regulators of nectary size [14]. In rice, OsMADS1 and miR172s/AP2s formed a regulatory network involved in regulating the elongation of the lemma and the palea (homologous to the eudicot sepal) [15]. In orchids, the Cymbidium ensifolium AP2-like gene CeAP2 shows an antagonistic expression pattern with AG ortholog and is negatively regulated by miR172 [16], which suggests a conservation regulatory network of AG ortholog and miR172s/CeAP2. Previous studies suggested that euAP2 genes were recruited to ovule, fruit, and floral organ development during angiosperm evolution, and changes in the role of euAP2/TOE3 homologs during development are most likely due to changes in regulatory regions [17].

Common buckwheat (Fagopyrum esculentum) (Polygonaceae) belongs to the order Caryophyllales (an early-diverging core eudicots clade) and has distylous flowers with single-whorl showy undifferentiated perianth (tepals), representing an obvious difference with most core eudicots flowers [18,19]. Moreover, heteromorphic self-incompatibility (HSI) due to its distylous flowers (pin and thrum) results in successful mating only between flowers with different morphs, which makes it a major obstacle for establishing pure lines, fixation of agronomically useful genes, and genomics-assisted breeding of common buckwheat [20,21]. Here, we have developed a self-compatible (SC) common buckwheat line with uniform long homostyle (LH) flowers (Figure 1) that could freely outcross with pin, thrum, and LH plants [22]. Hence, the LH common buckwheat can easily be the fixation of useful agronomical traits, which also makes it an excellent candidate line to explore the molecular mechanism of floral development and evolution in the buckwheat genus. Here, we separately cloned a TOE-type gene FaesTOE, an AP2-type gene FaesAP2, and an AP1 orthologous gene FaesAP1_2 from LH common buckwheat. Yeast one-hybrid (Y1H) assays were performed to screen whether FaesTOE or FaesAP2 could directly regulate ABC MADS-box genes of common buckwheat to control floral development and floral organ identity. Seven common buckwheat ABC MADS-box gene promotors and a FaesELF3 gene promotor [pFaesAP1_1 (MK327554.1); pFaesAP1_2; pFaesAP3_1 (MK956946.1); pFaesAP3_2 (MN016952.1); pFaesPI_1 (OM032614.1); pFaesPI_2 (OM032615.1); pFaesAG and pFaesELF3 (OP572282.1)] containing AT-rich target sequences that we isolated before were screened [19,22,23,24,25]. Only FaesTOE can directly bind pFaesAP1_2. The dual-luciferase reporter (DR) and VIGS assays further suggested that FaesTOE directly activates FaesAP1_2 (but not the C-functional gene FaesAG) to regulate filament length. Moreover, VIGS-silencing LH common buckwheat showed that FaesAP2 is mainly involved in controlling filament and style length, but FaesTOE is mainly involved in regulating filament length and pollen grain development. Both FaesTOE and FaesAP2 genes share redundant functions in regulating filament length determination in LH common buckwheat. Our results also extend the genetic framework for studying floral diversity in the early-diverging core eudicots and offer deeper insight into the evolutionary history of the AP2 family.

2. Results

2.1. Isolation and Sequence Analysis of APETALA2 Homologous Genes, FaesAP1_2, and Its Promoter from Long-Homostyle Common Buckwheat

Two AP2 homologous genes, FaesAP2 and FaesTOE, were isolated from LH common buckwheat. The 1668 bp FaesAP2 cDNA contains a 1374 bp ORF (Open Reading Frame, ORF) encoding 457 amino acids (aa) (KM386628.1). The 1286 bp FaesTOE cDNA contains a 1047 bp ORF encoding 348aa (PP357007). Phylogenetic tree analysis grouped the FaesAP2 and FaesTOE into the euAP2 lineage (Figure 2A). Moreover, the FaesAP2 gene was an ortholog of Arabidopsis AP2 and was designated as FaesAP2 (Fagopyrum esculentum AP2), while the FaesTOE gene was an ortholog of Arabidopsis TARGET OF EAT1 (TOE1) and was named FaesTOE (Fagopyrum esculentum TOE1).

An AP1 homologous gene, FaesAP1_2, and its promoter (pFaesAP1_2) were also isolated from LH common buckwheat. The 1026 bp FaesAP1_2 cDNA contains a 750 bp ORF encoding 249aa (PP357009). Phylogenetic tree analysis grouped the FaesAP1_2 into the euAP1 lineage (Figure 2B) and revealed that the FaesAP1_2 shared the same evolution branch with another common buckwheat AP1-like protein, FaesAP1_1 (AKI81897.1). The gene was an ortholog of Arabidopsis AP1 and was designated as FaesAP1_2 (Fagopyrum esculentum AP1_2).

The 662 bp FaesAP1_2 promoter (pFaesAP1_2) fragment (−593/+69) was cloned from LH common buckwheat, and the putative transcription start site and cis-acting regulatory elements of the pFaesAP1_2 were shown in Figure S1. The pFaesAP1_2 contains a CCAATBOX1 for CONSTANS protein binding to regulate flowering [26]. The pFaesAP1_2 also contains three GTGANTG10-boxes and four POLLEN1LELAT52-boxes, which are usually found in the promoter region of stamen-development genes [27,28]. Moreover, the pFaesAP1_2 promoter contains a CArG-box for recognition and binding by floral homeotic MADS-box transcription factors [29]. In addition, pFaesAP1_2 has two key AACAAA-/TTTGTT-motifs (−557/−552 and −128/−123) for floral homeotic AP2-like protein recognition and action [3]. Moreover, a MYBPLANT motif for flower-specific MYB protein recognition and binding was also found in the pFaesAP1_2 region [30]. These cis-acting elements indicated that FaesAP1_2 may be activated by the upstream transcription factors and be involved in the regulation of flowering and/or floral organ development.

In addition, there is a PYRIMIDINEBOXOSRAMY1A-box [31] and a DRECRTCOREAT motif [32,33] lying at the pFaesAP1_2 region, which suggests that the FaesAP1_2 expression may also be induced by stress (such as drought, high salt, cold, etc.) and Gibberellin (GA) levels.

2.2. FaesTOE Up-Regulate FaesAP1_2 Involved in Floral Organ Development

Previous studies showed that the floral homeotic protein AP2 transcription factor could bind the AACAAA-/TTTGTT-motifs and participate in the regulation of flower development [3]. In order to assay whether FaesAP2 and/or FaesTOE directly regulate floral homeotic MADS-box gene or not, seven MADS-box gene promotor regions containing AACAAA-/TTTGTT-motifs, such as pFaesAP1_1 (−1004/−999, and −921/−916), pFaesAP1_2 (−128/−123), pFaesAP3_1 (−620/−615), pFaesAP3_2 (+117/+122), pFaesPI_1 (−884/−879), pFaesPI_2 (−329/−324, and +56/+61), and pFaesAG (−879/−874, and −718/−713) were screened. In addition, an EARLY FLOWRING3 homologue promoter pFaesELF3 containing AACAAA-/TTTGTT-motifs (+93/+98, +194/+199, +320/+325, +529/+534) was also detected. Yeast one-hybrid (Y1H) assays showed that only FaesTOE could directly bind the target fragment (AACAAA-/TTTGTT-motif) of pFaesAP1_2 (Figure 3A). In addition, the dual-luciferase reporter assays (DR) further showed that FaesTOE up-regulates FaesAP1_2 in plants (Figure 3C). When the reporter construct carried the pFaesAP1_2 (−450/+35) promoter region, the ratio of LUC/REN expression was significantly increased (p < 0.05) (Figure 3B,C).

2.3. Expression Analysis of FaesAP2 and FaesTOE in LH Flower F. esculentum

FaesAP2 was mainly expressed in all flower organs (tepal, stamen, pistil) and young fruit of the LH F. esculentum (Figure 4A), while FaesTOE was mainly expressed in root, stamen, pistil, and young fruit. In addition, the expression level of FaesAP2 in the tepal or pistil was significantly higher than that of the stamen (p < 0.05, LSD), while the expression level of FaesTOE in the stamen was significantly higher than that of the other floral organ (tepal or pistil) (p < 0.05, LSD) (Figure 4A). However, the expression level of FaesAP2 in pistil was significantly higher than the expression level of FaesTOE in pistil (p < 0.05, LSD), while the expression level of FaesTOE in stamen was significantly higher than the expression level of FaesAP2 in stamen (p < 0.05, LSD) (Figure 4A). Obvious expressions of FaesAP2 and FaesTOE were detected when pistils and stamens emerged in LH floral buds (Figure 4B,C). In addition, FaesAP2 expression increased constantly during the filament’s rapid elongating stage and achieved its peak until anthers at the mononuclear microspore stage (Figure 4B,C—L2, L3). Then, the FaesAP2 expression began to drop sharply at mature pollen appearance (p < 0.05, LSD) (Figure 4B,C—L4). However, FaesTOE expression is maintained at a high level until anther at the mononuclear microspore stage (Figure 4B,C—L1, L2, L3) and begins to drop as the mature pollen appearance (Figure 4B,C—L4).

2.4. Characterization of FaesAP2, FaesTOE and FaesAP1_2-Silenced Plants

In order to explore the functions of FaesAP2, FaesTOE, and FaesAP1_2, a virus-induced gene silencing (VIGS) technique via the Tobacco Rattle Virus (TRV) system was performed to knock down their expression in red and white flower LH F. esculentum, respectively. Eight TRV2-FaesAP2-treated, ten TRV2-FaesTOE-treated, and six TRV2-FaesAP1_2-treated LH plants producing flowers with the biggest phenotypic changes were obtained, respectively. Both strongly affected TRV2-FaesAP2-treated and TRV2-FaesTOE-treated red flower LH plants produced flowers with style- and filament-length decreased obviously (Figure 5(Aa,Ad)). Moreover, some VIGS-FaesAP2 treatment LH flowers have obvious short styles and part of short stamens with anthers at around the same height as the stigmas (Figure 5(Ab,Ac)). However, some VIGS-FaesTOE treatment LH flowers have normal long styles but abnormal short stamens consisting of short filaments and empty anthers (male sterile anther) (Figure 5(Ae)), or flowers have normal long styles but part abnormal tepals and stamens, which displayed part stamens with anther homeotically converted into tepaloid structure or sterile anther on the top (Figure 5(Af)). In addition, strongly affected TRV2-FaesAP1_2-treated white LH flowers have normal styles but part abnormal short stamens with an obvious reduction in filament length (Figure 5(Ag)). Moreover, some TRV2-FaesAP1_2 treatment red LH flowers showed tepal numbers and filament length decreased, and long style numbers changed (Figure 5(Ah,Ai)).

In addition, the expression levels of FaesAP2, FaesTOE, and FaesAP1_2 were separately detected by qRT-PCR in the gene-silenced plants. The results showed that FaesTOE expression was significantly decreased in the flowers of FaesTOE-silenced LH plants (Figure 5C, orange column). However, the FaesAP1_2 expression was significantly decreased in both the flowers of FaesAP1_2-silenced (Figure 5C green column) and FaesTOE-silenced LH plants (Figure 5C orange column) (p < 0.01, LSD). All these suggested that the knockdown of FaesTOE expression could result in expression down-regulation of the directly target FaesAP1_2 in the FaesTOE-silenced LH plants (Figure 5C).

3. Discussion

In the rosid species A. thaliana, AP2 confers A functions and represses expression of C-class gene AG in the out two floral whorls to specify perianth (sepal and petal) identity, as well as directly represses the expression of flowering promoting transcripts AP1 to control the timing of the floral transition [3,6,7,8], while TOE1 and TOE3 (TARGET OF EAT1/3) regulate flowering [12,13]. Both AP2 and TOE are negatively regulated by miR172 [7]. In Brassica Crops (relative species of A. thaliana), the AP2-like genes (BrAP2a and BnAP2) play a key role in sepal modification [9]. Losses in the function of BrAP2a or BnAP2 could result in a sepal–carpel modification phenotype. In Prunus mume (Rosaceae), a 49 bp deletion-mediated miR172 target site loss in AP2-like PmAP2 (APETALA2-like gene) resulted in the transformation of stamens to petals [34]. However, mutation disrupting the miR172 target site of the orthologs of the TOE-type gene in peach and rose also resulted in the conversion of stamens to petals and even loss of floral determinacy (the double-flower trait) [35,36]. In the asterid species Petunia, the TOE-type gene BEN acts as a floral repressor and represses the C-function, while the AP2-type ROB genes are required for perianth and pistil development, as well as repressing the B-function (but not the C-function) together with BEN in the first floral whorl [11]. In the asterid species tomato (Solanum lycopersicum), the TOE-type gene SlTOE1 regulates inflorescence development by repressing STM3 and the tomato SOC1 homolog TM3 [37]. In carnation, a disruption of the miR172 target sequence of the TOE-type ortholog causes a dominant double-flower phenotype [38]. All these data indicate that the floral organ identity genes AP2 or TOE orthologs show obvious functional diversity among core eudicots.

In the rosid species A. thaliana, AP1 activates floral meristem identity (FMI) and specifies sepal and petal identities [39]. AP1 down-regulates ZP1 and ZFP8 in floral meristems, lifting the repression on the class B (AP3 and PI) and C (AG) genes, which leads to specific stamens and carpel identities [40]. In the asterid species Petunia, AP1 orthologs are required for inflorescence meristem identity and act as B-function repressors in the first floral whorl, together with the AP2-type ROB genes [11,41]. All these suggest that AP1 orthologs work together with AP2-type ROB genes involved in a perfect sepal development pathway in Petunia.

In common buckwheat, a small-scale gene duplication event occurs in AP1 orthologs, resulting in FaesAP1_1 and FaesAP1_2 in F. esculentum. FaesAP1_1 activates flowering and regulates tepal (perianth) development [24], while FaesAP1_2 is mainly involved in filament length determination. In common buckwheat, both FaesAP2 (AP2-type) and FaesTOE (TOE-type) are redundantly involved in style and/or filament length determination. Moreover, FaesTOE also seems to play a role in other developments. However, neither FaesAP2 nor FaesTOE could directly repress the B and/or C class genes in common buckwheat. Interestingly, FaesTOE directly up-regulates FaesAP1_2 to be involved in filament length determination in common buckwheat. Moreover, the knockdown of FaesTOE expression could result in expression down-regulation of the directly target FaesAP1_2 in the FaesTOE-silenced LH plants. Our previous study also found another filament pathway in common buckwheat through FaesAP3_1 (B-class gene) and FaesELF3 [22]. However, how these genes work together to specify a perfect stamen still remains unclear. Our findings uncover a pathway controlling common buckwheat filament length and offer deeper insight into the functional evolution of AP2 orthologs in the early-diverging core eudicots.

4. Materials and Methods

4.1. Plant Material

Long-homostyle (LH) common buckwheat lines with red/white flowers were grown under natural conditions in Jingzhou, Hubei Province, China. The roots, stems, juvenile leaves, tepals, stamens, pistils, and 6-day-old fruits (achenes) of LH plants were separately dissected, immediately frozen in liquid nitrogen, and stored at −80 °C until used. Moreover, the LH floral buds at sequential developmental stages were sampled. Each sample was divided in half; one was immediately frozen in liquid nitrogen and stored at −80 °C until used, and the other was fixed in FAA [38% formaldehyde: acetic acid: 70% ethanol = 1:1:18 (v/v)] for cytomorphological examination. Nicotiana benthamiana for the dual-luciferase reporter assay was planted in the growth chamber at 22 °C under long-day conditions (16 h light, 8 h dark).

4.2. Isolation and Characterization of APETALA2 Floral Homeotic Genes Long-Homostyle Common Buckwheat

Total RNA was extracted from LH floral buds using an EASYspin Plant RNA Kit (Aidlab, Hong Kong, China) according to the manufacturer’s protocol. The first-strand cDNA of 3′ RACE was prepared, and then the 3′ end cDNA sequences of FaesAP2 and FaesTOE were separately amplified with gene-specific forward primers 3RGSPAP2 and 3RGSPTOE (Supplementary Table S1) using the 3-full RACE Core Set Ver. 2.0 kit (TaKaRa, Kusatsu, Japan) according to the manufacturer’s protocol, but with 90 s annealing at 58 °C. The forward primer design referenced the AP2 homologous sequences (F01.PB24091 and F01.PB47570) identified before (BioProject ID: PRJNA517031). The phylogenetic tree was constructed using MEGA version 5.05 with the neighbor-joining (NJ) method. The NJ tree was constructed with 1000 bootstrap replications. All the APETALA2 homologous transcription factors (TF) with complete sequences were obtained from the NCBI Genbank (Supplementary Table S2).

4.3. Isolation and Sequence Analysis of FaesAP1_2 and FaesAP1_2 Promoter (pFaesAP1_2) from Long-Homostyle Common Buckwheat

Total RNA was extracted from LH floral buds, and the first-strand cDNA of 3′ RACE was prepared according to the protocol described above. The 3′ end cDNA sequence of FaesAP1_2 was isolated with gene-specific primer 3RGSPAP1_2 (Supplementary Table S1) but with 60s annealing at 58 °C. The forward primer design referenced the AP1 homologous sequence (F01.PB26508) identified before (BioProject ID: PRJNA517031). The phylogenetic analysis of FaesAP1_2 was referenced to the method described before. The putative FaesAP1_2 protein sequence and other AP1/FUL-like proteins were selected for constructing phylogenetic trees obtained from the NCBI Genbank (Supplementary Table S3). In addition, the AGL6-like and SEP-like proteins were also included as outgroups because previous studies grouped them into the AP1/SEP/AGL6 superclade [41,42].

Buckwheat genomic DNA was extracted from LH plant juvenile leaves using the CTAB Plant Genomic DNA Rapid Extraction Kit (Aidlab, Beijing, China) according to the manufacturer’s protocol. The FaesAP1_2 5′ flanking region was isolated from LH buckwheat genomic DNA by using the Genome Walking Kit (TaKaRa, Japan) following the manufacturer’s protocol and with gene-specific primers DpAP1_2SP1, DpAP1_2SP2, and DpAP1_2SP3 (Supplementary Table S1) for the walking sequence. The putative transcription start site of FaesAP1_2 was searched according to the methods described by Solovyev et al. [43], and the cis-acting elements located at the FaesAP1_2 promoter region were searched in the PLACE database [44].

4.4. Yeast One-Hybrid Assay

A yeast one-hybrid (Y1H) assay was performed using the Yeast One-Hybrid Media Kit (Coolaber, Shanghai, China). The pFaesAP1_1 (MK327554.1); pFaesAP1_2; pFaesAP3_1 (MK956946.1); pFaesAP3_2 (MN016952.1); pFaesPI_1 (OM032614.1); pFaesPI_2 (OM032615.1); pFaesAG and pFaesELF3 (OP572282.1) region containing TTTGTT and/or AACAAA motifs were separately cloned into pAbAi plasmid to construct pBait-AbAi vectors with the gene-specific primer pairs Y1HpFaesAP1_1F and Y1HpFaesAP1_1R for pFaesAP1_1 (−1171/−657), Y1HpFaesAP1_2F and Y1HpFaesAP1_2R for pFaesAP1_2 (−450/+35), Y1HpFaesAP3_1F and Y1HpFaesAP3_1R for pFaesAP3_1 (−622/+114), Y1HpFaesAP3_2F and Y1HpFaesAP3_2R for pFaesAP3_2 (−272/+125), Y1HpFaesPI_1F and Y1HpFaesPI_1R for pFaesPI_1 (−1289/−687), Y1HpFaesPI_2F and Y1HpFaesPI_2R for pFaesPI_2 (−402/+31), Y1HpFaesAGF and Y1HpFaesAGR for pFaesAG (−1102/−692), as well as Y1HpFaesELF3F and Y1HpFaesELF3R for pFaesELF3 (−24/+554) (Supplementary Table S1) [3]. Two common buckwheat AP2-like genes, FaesAP2 and FaesTOE, cDNAs containing full-length ORFs (Open Reading Frame, ORF), were separately cloned into the pGADT7 plasmid to construct prey vectors with the gene-specific primer pairs Y1HFaesAP2F and Y1HFaesAP2R for FaesAP2, as well as Y1HFaesTOEF and Y1HFaesTOER for FaesTOE (Supplementary Table S1). The linearized plasmids pAbAi-pFaesAP1_1, pAbAi-pFaesAP1_2, pAbAi-pFaesAP3_1, pAbAi-pFaesAP3_2, pAbAi-pFaesPI_1, pAbAi-pFaesPI_2, pAbAi-pFaesAG, and pAbAi-pFaesELF3 were separately transformed into the Y1H Gold-competent cells to generate bait-reporter strains. The transformants were selected on SD/-Ura media and screened for the minimal inhibitory concentration of Aureobasidin A (AbA) for the bait-reporter strains. Two prey plasmids, pGADT7-FaesAP2 and pGADT7-FaesTOE, were separately transformed into the bait yeast strains for Y1H according to the manufacturer’s protocol. Colonies were incubated and selected on the medium (SD/-Leu/AbA) with the bait of minimum AbA resistance for 3 days at 30 °C.

4.5. Dual-Luciferase Reporter Assay

The pFaesAP1_2 region was cloned into the pGrenen0800-LUC vector to generate the reporter plasmid pGreen0800-pFaesAP1_2 with the gene-specific primer pairs Dual-pFaesAP1_2F and Dual-pFaesAP1_2R (Supplementary Table S1). The full-length ORF of FaesTOE was cloned into the pGreenII 62-SK vector to generate the effector vector with the gene-specific primer pairs Dual-FaesTOEF and Dual-FaesTOER (Supplementary Table S1). The reporter vector and effector vector were separately introduced into Agrobacterium strain GV3101 (pSoup). The effector and the reporter Agrobacterium cultures were mixed together at a ratio of 2:1 and infiltrated into N. benthamiana leaves. Empty pGreenII 62-SK was cotransformed with the pGreen0800-pFaesAP1_2 reporter as a negative control. The treated tobacco plants were cultured in the dark for 24 h before being transferred to the growth chamber at 22 °C under long-day conditions (16 h light, 8 h dark) for 3 days. Firefly luciferase and renilla luciferase were assayed using the Duo-LiteTM Luciferase Assay System (Vazyme, Nanjing, China) following the manufacturer’s protocol. Each combination had three biological replicates.

4.6. Expression Analysis of FaesAP2 and FaesTOE

The total RNA of each sample was extracted according to the above method, but the first-strand cDNA was synthesized for quantitative real-time PCR (qRT-PCR) by using the HiScript^® II Q RT SuperMix for qPCR kit (Vazyme, Nanjing, China) following the manufacturer’s protocol. The relative expressions of FaesAP2 and FaesTOE were separately detected in root, stem, juvenile leaf, tepal, stamen, pistil, and 6-day-old fruit of LH lines by qRT-PCR referring to Ma et al. [22], but with the gene-specific forward primer qFaesAP2F and reverse primer qFaesAP2R for FaesAP2, and with the gene-specific forward primer qFaesTOEF and reverse primer qFaesTOER for FaesTOE, respectively (Supplementary Table S1). In addition, FaesAP2 and FaesTOE expressions were also detected in LH floral buds at sequential developmental stages with the qRT-PCR suggested above. The developmental stages of LH floral buds were confirmed in the paraffin section suggested by Ma et al. [22]. The amplicons of the F. esculentum actin gene (Genbank accession number: HQ398855.1) were amplified as the internal control with primer pairs qFaesactinF and qFaesactinR. The experiments were repeated three times for each sample.

4.7. VIGS Assay in Long-Homostyle Buckwheat

A 487 bp fragment of the FaesTOE cDNA, a 484 bp fragment of the FaesAP2 cDNA, and a 381 bp fragment of the FaesAP1_2 cDNA were separately cloned into the TRV2-mediated VIGS vectors with XbaI and SacI restriction enzymes, but with the gene-specific primer pairs TRV2-FaesTOEF and TRV2-FaesTOER for FaesTOE, TRV2-FaesAP2F and TRV2-FaesAP2R for FaesAP2, and TRV2-FaesAP1_2F and TRV2-FaesAP1_2R for FaesAP1_2, respectively.

Then, the TRV2-FaesTOE, TRV2-FaesAP2, TRV2-FaesAP1_2, TRV2, and TRV1 empty vector plasmids were separately introduced into Agrobacterium strain GV3101. The TRV-mediated VIGS vectors were infiltrated into the leaves of the two-true-leaves stage seedlings of LH lines mediated by Agrobacterium tumefaciens strain GV3101 with the method suggested by Liu et al. [45]. All the infected seedlings were cultivated in darkness for 24 h and then moved to the greenhouse at a temperature of 22 °C under short-day conditions. The expression levels of FaesTOE, FaesAP2, and FaesAP1_2 were separately detected by qRT-PCR in the flowers of VIGS-silencing plants. Plants were treated for each construct to make sure that flowers with the strongest phenotypic changes were seen in at least three individuals.

5. Conclusions

Common buckwheat belongs to the order Caryophyllales (an early-diverging core eudicots clade) and produces distylous flowers with single-whorl showy undifferentiated perianth (tepals), representing an obvious difference with most core eudicots flowers. Here, we have developed a self-compatible (SC) common buckwheat line with uniform long-homostyle (LH) flowers that can easily fix useful agronomical traits, which makes it an excellent candidate line to explore the molecular mechanisms of floral development and evolution in the Fagopyrum genus. In this study, two euAP2 orthologs, FaesAP2 and FaesTOE, and an AP1 orthologous gene, FaesAP1_2, were isolated and identified from LH common buckwheat. Moreover, Y1H was performed to screen whether FaesTOE or FaesAP2 could directly regulate ABC MADS-box genes identified by us before. However, neither FaesAP2 nor FaesTOE could directly repress the B and/or C class genes in common buckwheat. The Dual-Luciferase Reporter (DR) and VIGS Assay further suggested that FaesTOE directly activates FaesAP1_2 (but not the C-functional gene FaesAG) to regulate filament length. Moreover, VIGS-silencing LH common buckwheat showed that FaesAP2 is mainly involved in controlling filament and style length, but FaesTOE is mainly involved in regulating filament length and pollen grain development. Both the FaesTOE and FaesAP2 genes redundantly function in regulating filament length determination in LH common buckwheat. In addition, the FaesAP1_2 silenced flower showed tepal numbers, and filament length decreased obviously. Moreover, the knockdown of FaesTOE expression could result in expression down-regulation of the direct target FaesAP1_2 in the FaesTOE-silenced LH plants. Our findings uncover a stamen development pathway in common buckwheat and offer deeper insight into the evolutionary history of the AP2 family.