To investigate the role of the PEBP/RKIP gene family in regulating the transition from vegetative to reproductive growth in Cymbidium, PEBP orthologs were identified and characterized. A combined RT-PCR and RACE strategy was used to clone FT from C. sinense “Qi Jian Bai Mo”, C. goeringii and C. ensifolium “Jin Si Ma Wei”. CsFT (GenBank accession number HM120862), CgFT (GenBank accession number HM120863) and CeFT (GenBank accession number HM803115) contain 618-bp nucleotides with an open reading frame (ORF) of 531 bp encoding 176 amino acids (AAs), two exons and one intron (161 bp) (Figure 1). The analysis based on AA sequence alignment shows that the three FT AA sequences are identical; they also share a high identity with FT of other plants in GenBank, such as OnFT (94%) from Oncidium Gower Ramsey, Hd3a (79%) from Oryza sativa, and FT (74%) from Arabidopsis thaliana (Figure 2). AA sequence alignment also revealed that CsFT, CgFT and CeFT contain a conserved domain, which is characteristic of the PEBP-RKIP superfamily. The conserved key AA residues Tyr (Y, site 84) and Gln (Q, site 140) in FT homologs were identified in CsFT, CgFT and CeFT protein (Figure 2). The sequence similarity between CsFT, CgFT, CeFT and other FTs indicates that CsFT, CgFT, and CeFT are the putative Cymbidium FT orthologs. The conserved domains were analyzed in FT: LGRQTVYAPGWRQN (14 AAs) and LYN/IYN conserved domain was similar to other FT proteins [14]. The Cymbidium genes characterized here are closely related to the FT gene from monocotyledonous plants (Oncidium and Oryza sativa) [8,15] based on their protein sequences which have a conserved domain LYN (Figure 2). This suggests that CsFT, CgFT, and CeFT are potentially FT orthologs that regulate the transition from vegetative state to flowering and flower initiation in Cymbidium. The AA sequence alignment shown in Figure 2 and the sequences for several other FT orthologs were used to construct a phylogenetic tree for the FT group of genes (Figure 3). CsFT, CgFT and CeFT were grouped within the monocotyledonous FT subgroup and are closely related to Oncidium, followed by Triticum aestivum, Hordeum vulgare and Oryza sativa.

To explore whether the expression of CsFT, CgFT, and CeFT was influenced by daily oscillations in photoperiod, the expression of CsFT, CgFT, and CeFT mRNA was analyzed every 4 h over a 24-h period in long day (LD) and short day (SD) conditions using quantitative real time PCR (Figure 4). CsFT expression (Figure 4A) was different under SD and LD: it was higher under LD than under SD. Expression of CsFT showed an increasing trend under LD, indicating that expression of CsFT was very sensitive to changes in light under LD. CgFT (Figure 4B) was also regulated by light but its expression was significantly higher under SD than under LD, but there was a similar change in pattern under both conditions, with the level highest at the 4th hour of the light period and lowest at the 16th hour. C. ensifolium “Jin Si Ma Wei” CeFT (Figure 4C) showed higher expression under LD than SD with a rhythmic cycle, but peaked at different times (12 h and 36 h of LD, 20 h and 36 h of SD). CeFT mRNA showed highest levels of expression 12 h into the light period and lowest levels at dawn. CgFT expression was thus different from CsFT and CeFT expression and all were regulated by light.

Since the three Cymbidium FTs have the same AA sequences, we then randomly chose only one (CgFT) for functional verification. CgFT driven by the CaMV 35S promoter was transformed into wild-type (WT) A. thaliana plants for functional analysis in order to explore whether Cymbidium FT could regulate the transition to flowering in A. thaliana. 35S::CgFT transgenic A. thaliana T₀ plants were screened on half-strength MS [16] containing 50 μg/mL Kan using a protocol from Cheng et al. [17] (Figure 5A). In total, 45 independent 35S::CgFT transgenic A. thaliana T₁ plants were obtained. All transgenic plants showed identical phenotypes by flowering earlier than WT plants (Figure 5B–D). These 35S::CgFT transgenic plants (Figure 5) flowered at about 15 days after sowing by producing about four rosette leaves (Table 1; Figure 5B). The flowering time of WT A. thaliana plants was more than 30 days and produced nine or ten rosette leaves (Table 1). When transgenic plants were exposed to LD or SD, flowering was obviously promoted under LD (Figure 5E), but was slower under SD (Figure 5F) and ectopic expression of cgFT in A. thaliana was consistent with the AtFT gene expression pattern in A. thaliana [18,19].

To explore whether the early flowering phenotype was correlated with CgFT expression in 35S::CgFT transgenic plants, RT-PCR analysis was performed. As shown in Figure 6, higher CgFT expression was observed in the severe 35S::CgFT transgenic plants more than in the transgenic plants with a less severe or WT phenotype. Further analysis indicated that the promotion of flowering time in severe early flowering 35S::CgFT transgenic plants was also related to significant up-regulation of the flower meristem identity gene AtAP1 in transgenic plants (Figure 6). This result indicates that the function of Cymbidium FT is similar to that of A. thaliana FT in regulating flowering time.

Flowering time is regulated by a special set of flowering genes [20]. A number of studies indicated that in different plant species, FLOWERING LOCUS T (FT) homologous genes are involved in the earliest stages of flower development [21,22]. Our result shows that the structure of Chinese Cymbidium FT homologous genes contains two exons and one intron. This is not similar to other species, such as A. thaliana [10], Oryza sativa [23], Hordeum vulgare [24], Vitis vinifera [22], Populus sp. [25], and Zea mays [26], all of which have four exons and three introns. However, the wheat TaFT and barley HvFT genes have three exons encoding a 177-AA protein [27], in contrast with four exons and three introns in HvFT2, HvFT3, HvFT4 and HvFT5 of barley [24]. There is a loss of an intron in several barley cultivars, wild barley accessions, and wheat, suggesting that this may be a general feature of temperate grasses [24]. However, we deduced that the lack of introns cannot serve as a common feature of winter plants (i.e., plants that require low temperature for vernalization) because C. ensifolium does not need low temperature for vernalization and has a long flowering period (April-October) in China. C. goeringii, however, needs low temperature for vernalization. For C. sinense “Qi Jian Bai Mo”, photoperiod has a major impact on flowering. The origins and importance of spliceosomal introns comprise one of the longest-abiding mysteries of molecular evolution [28]. The rates of loss of existent introns are higher than the rates of gain in eukaryotic evolution, and the rates of loss of introns represent the rate of evolution of a species [29]. It has been suggested that the process of intron birth in early eukaryotes could be fundamentally different from the process in more recent evolution [29]. Interestingly, the Cymbidium FT intron is also different from that of Oncidium, which has four exons and three introns [8]. The loss of an intron in the Cymbidium FT gene may be molecular evidence that Cymbidium is evolutionarily more advanced than Oncidium.

FT gene expression increases dramatically when plants are induced by light [15,30]. Our results showed that expression of CsFT and CeFT were regulated by light and were significantly higher under LD than under SD, but that CgFT expression was significantly higher under SD. These results suggest that C. goeringii may be SD-sensitive plants regulated by light, similarly to FT in rice [15], while C. sinense “Qi Jian Bai Mo” and C. ensifolium “Jin Si Ma Wei” may be an LD-sensitive plant like A. thaliana. The difference in expression pattern of Cymbidium FT orthologs in response to light may be due to the fact that the Cymbidium genus contains species with different photoperiodic requirements, unlike A. thaliana, which is an LD plant, and rice, which is an SD plant. This pattern indicates that the expression of CsFT peaked at the 32nd hour of LD and CeFT expression peaked at the 12th hour of LD. However, this is unlike At FT, whose expression level peaks at the 16th hour of the light period and is lowest at dawn under LD conditions [18,19]. Interestingly, although the AA sequence of CsFT, CgFT and CeFT are identical, they have different expression patterns under LD and SD, indicating that expression of CsFT, CgFT and CeFT may also be affected by other factors, such as the upstream gene CONSTANS (CO)-like gene in the photoperiod pathway and other circadian clock genes [31], leading to diverse and complex FT expression patterns. The expression quantity of CsFT and CeFT are enormously higher than that of CgFT under LD, implying that C. sinense and C. ensifolium may primarily be regulated by photoperiod while C. goeringii may be mainly regulated by other factors, such as low temperature.

FT plays a central role as a florigen in floral induction, and its function is conserved across different plant species. The transgenic plants carrying the 35S::CgFT construct flowered earlier (about 23 days from sowing to flowering) and had fewer rosette leaves (average = 4.2) at flowering than the WT plants (average = 10.4). The early-flowering phenotype observed in 35S::CgFT transgenic plants was similar to that observed in A. thaliana [9,10], Solanum lycopersicum [32], Pharbitis nil [33], Vitis vinifera [22], Cucurbita moschata [34], Populus sp. [25], Oryza sativa [35,36], Oncidium Gower Ramsey [7,8] and Glycine max [37] that ectopically express FT orthologs; in all cases, transgenic plants had a flowering time of less than 25 days and the number of rosette leaves at flowering was fewer than five. In contrast, flowering time in A. thaliana and rice plants could be delayed by using RNAi or miRNA of the FT gene [15,35]. Our results suggest that the function of the FT gene in flowering plants is very conservative and that FT and its homologs are required for flowering plants, regardless of photoperiod or species (dicotyledonous or monocotyledonous). Of importance, the level of the AtAP1 transcripts was related to the expression of CgFT (Figure 6): the expression of AtAP1 was higher when FT expression was higher. This indicates that constitutive expression of CgFT acts similarly to AtFT by regulating the transition from vegetative state to flowering by activating AtAP1.

C. sinense “Qi Jian Bai Mo”, C. goeringii and C. ensifolium “Jin Si Ma Wei” plants used in this study were grown and maintained in pots in a greenhouse of the South China Botanical Garden, Guangzhou, China. The pot substrate consisted of Zhijing stone for orchids (Northridge Enterprise Co., Ltd., Taiwan): sieved peat: shattered fir (1:1:1; v/v/v). Arabidopsis thaliana “Columbia” seeds obtained from our lab were surface sterilized in 70% ethanol for 10 s, then immersed in 0.1% (w/v) HgCl₂ solution for 10 min followed by five rinses with sterile distilled water. A. thaliana seeds were sterilized and placed on agar plates containing half-strength Murashige and Skoog medium [16] at 4 °C for 2 days. The seedlings were then grown in a growth chamber under LD conditions (16-h light/8-h dark) or SD conditions (8-h light/16-h dark) at 22 °C for 10 days, then transplanted to soil under LD or SD conditions. The light intensity in the growth chambers was 150 μmol m⁻² s⁻¹.

Total RNA was extracted from young leaves of all three Cymbidium spp. using the Trizol method. The first cDNAs strands were synthesized by reverse transcription PCR (RT-PCR) using M-MLV reverse transcriptase (TaKaRa Bio. Co., Ltd., Dalian, China). The primers (forward: 5′-TAGGACG AGTGATTGGTGA-3′; reverse: 5′-TCACTTGGACTTGGAGCAT-3′) designed for cDNA fragment amplification in PCR experiments were as described for the OnFT sequence from Genebank (ACC59806.1). The amplified fragments contained a partial FT sequence that showed high sequence identity to the FT homologous gene of other species from NCBI Blast. Gene-specific nested primers for 3′-RACE of FT were 5′-TAGGACGAGTGATTGGTGA-3′ (forward), 5′-TTCAGCAGTAGTGG AGCAG-3′ (reverse) designed using the cDNA fragment of FT from Cymbidium. Gene-specific primers (hiTAIL-PCR) [38] for 5′-RACE of FT were 5′-TGGAGCATCTGGATCTACCATGACCTG-3′, 5′-ACGATGGACTCCAGTCCGGCCGAAAGTCCTGAGGTCATTCCCTCCAC-3′, and 5′-CTCGGC TGCTCCACTACTGCTGAAGGC-3′, designed by using the cDNA fragment and the 3′-RACE of FT from Cymbidium. The cDNAs and genomic DNA sequences for the three FT genes were obtained by PCR amplification using forward (5′-CTGAAGGAAGTGATAGCAA-3′) and reverse (5′-AGCAGT GACGCAAAGGAAA-3′) primers designed by using the full-length sequence of FT from Cymbidium after 5′-RACE. The forward primer for FT contained an XbaI recognition site (5′-TCTAGA-3′) and the reverse primer contained a BamHI recognition site (5′-GGATCC-3′) to facilitate the cloning of the cDNAs.

Quantitative real-time PCR was carried out using SYBR^® Premix Ex Taq™ II (TaKaRa) and ABI 7500 real-time PCR for transcript measurements. Amplification conditions were: 95 °C for 30 s, followed by 40 cycles of amplification (95 °C for 5 s, 60 °C for 34 s, 72 °C for 30 s) and plate reading after each cycle. The primers designed from the full-length sequence of FT from Cymbidium used for quantitative real-time PCR for CsFT, CgFT and CeFT were: forward (5′-AGAGTTGAAG TTGGAGGGAATG-3′) and reverse (5′-GGTCGTTGCTGGGATATCG-3′). The primers for Cymbidium ACTIN (designed using the CsActin sequence, Genebank accession number GU181353) were: forward (5′-AATCCCAAGGCAAACAGA-3′) and reverse (5′-CCATyACCAGAATCCAG-3′). Data were analyzed using ABI 7500 Real-time PCR system Gene Expression software.

A phylogenetic comparison of aAA sequences of different FT-like homologs was performed using GenBank [39] and aligned with Clustal W2 [40]. Phylogenetic trees based on the complete sequences were generated using MEGA4 [41] and constructed by the neighbor joining (NJ) method. Bootstrap values were derived from 1000 replicate runs.

Total RNA was extracted from young leaves using TRIzol Reagent (Invitrogen, Carlsbad, CA, USA) following the manufacturer’s protocol. Each RNA sample was treated with RNase-free DNase (Promega, Madison, WI, USA) following the manufacturer’s protocol in an effort to remove any residual genomic DNA (gDNA). DNase-treated RNA was subjected to reverse transcriptase reactions using oligo-dT primer and PrimeScript™ Reverse Transcriptase (TaKaRa) according to the manufacturer’s protocol. The gene-specific primers (designed by using the full-length sequence of FT from Cymbidium) for FT used in all RT-PCR were 5′-TAGGACGAGTGATTGGTGA-3′ (forward) and 5′-TCACTTGGACTTGGAGCAT-3′ (reverse). The cDNA sequence of the TUBULIN gene in A. thaliana was used as a control and amplified using two primers: 5′-GAGCCTTACAACGCTA CTCTGTCTGTC-3′ (forward) and 5′-ACACCAGACATAGTAGCAGAAATCAA-3′ (reverse) [42]. The primers for A. thaliana AtAP1 used in RT-PCR were 5′-GCACCTGAGTCCGACGTC-3′ (forward) and 5′-GCGGCGAAGCAGCCAAGG-3′ (reverse). PCR was performed with a PCR System LABCYCLER Standard Plus (SENSOQUEST, Hannah, Germany). The following thermocycling conditions were applied: initial denaturation at 94 °C for 1 min; 25 cycles of 94 °C for 30 s, 58 °C for 45 s, and 72 °C for 1 min; final extension at 72 °C for 10 min. The amplified products were separated on a 1.5% agarose gel using 0.1% ethidium bromide (EtBr) and photographed in a Bio Sens SC 710 system.

The full-length cDNA for CgFT was cloned into the binary vector pBI121 (BD Biosciences, Clontech, San Jose, CA, USA) under the control of the CaMV 35S promoter. The orientation of the construct was determined by PCR and used for further plant transformation. Arabidopsis plants were transformed using a floral dip method [43]. Transformants that survived in medium containing kanamycin (50 μg mL⁻¹) were further verified by RT-PCR analyses.