Characterization of a CONSTANS-like Gene from Pigeon Orchid (Dendrobium crumenatum Swartz) and its Expression under Different Photoperiod Conditions

Abstract

Orchids are economically valuable as cut flowers and in pot plant markets. However, a juvenility phase that is too long is the main disadvantage for commercial orchids. To understand the gene involving floral transition controls in orchids, a CONSTANS-like (COL) gene in the photoperiodic flowering pathway was isolated from Dendrobium crumenatum (pigeon orchid). The cDNA isolated has an open reading frame (ORF) of 969 bp, encoding 322 amino acids. Sequence alignment based on amino acid sequences revealed that the Dendrobium crumenatum COL (DcCOL) shared a high identity with COL isolated from other plant species including Phalaenopsis COL (85%), Oryza sativa Hd1 (70%), Erycina pusilla COL5 (EpCOL5) (66%), and Arabidopsis thaliana CO (39%). DcCOL has three conserved domains (CCT, B-box I, and B-box II domains) and is classified in group I CO/COL by phylogenetic analysis in the Arabidopsis B-box zinc finger protein family. Quantitative real-time RT-PCR demonstrated DcCOL was expressed in all stages of development and all tissue types with the highest expression in floral buds and opened flowers of mature orchids. The expression pattern under photoperiod pathway demonstrated a diurnal expression. The DcCOL was accumulated in the dark in all photoperiodic conditions, i.e., long, neutral, and short days suggesting that the regulation of DcCOL was controlled in a circadian rhythm-dependent manner. The results suggested that photoperiodism is not the main factor in D. crumenatum floral induction control. This DcCOL expression pattern coincided with the D. crumenatum flowering behavior in which the flowering occurs before dawn and lasts for only 24 h implying the function of DcCOL is related to flowering.

Introduction

Orchidaceae is the largest family of angiosperms with the number of species exceeding 25000 (Atwood, 1986). The members of this plant family have high diversity in both vegetative and floral morphologies (da Silva et al., 2014; Xu et al., 2006). Many orchid genera are economically valuable, namely Phalaenopsis (Fu et al., 2011; Hsiao et al., 2011; Su et al., 2011), Cymbidium (Wu et al., 2009), Oncidium (Chang et al., 2011), and Dendrobium (da Silva et al., 2014). However, these commercial orchids have characteristics that are disadvantageous to genomic studies, including long juvenility, polyploidy, a large genome size, and a high chromosome number (Pellegrino et al., 2009; Russell et al., 2010). One of the main challenges in orchid research and orchid breeding programs is their prolonged vegetative phase (Yu and Goh, 2001). A better understanding of floral transition controls in orchids would enable shorter juvenility periods and facilitate enhanced orchid breeding programs.

Dendrobium crumenatum Swartz (pigeon orchid) is a member of the Dendrobium genus that can flower more than once a year (Holttum, 1949; Meesawat and Kanchanapoom, 2007). This pigeon orchid has unique features such as white flowers and fragrant and gregarious flowering (Holttum, 1949; Meesawat and Kanchanapoom, 2007). Its flowering cycles can be triggered throughout the year, but its flower is short-lived (Beaman et al., 2001; Yap et al., 2008). In natural conditions, the flowering of this orchid is induced by a prolonged cool period, i.e., heavy rainfall (Arditti, 1979; Holttum, 1949). Full flower opening is observed before dawn approximately nine days after cold induction, and the flowers last for only 24 h (Yap et al., 2008). The Dendrobium orchid species is well studied in floral developmental pathways. However, the molecular mechanism underlying the floral transition process in this orchid is still unknown (Xu et al., 2006; Yap et al., 2008).

Floral transition, described as the transition switch from the vegetative to the reproductive phase is important for flowering plants (Boss et al., 2004). This change from vegetative to reproductive growth is a critical developmental process in a plant’s life cycle. The transition is controlled by both endogenous signals and environmental stimuli. In Arabidopsis thaliana, four major pathways of floral transition, including photoperiod, autonomous, hormonal, and vernalization pathways were identified (Borner et al., 2000; Mouradov et al., 2002; Xiang et al., 2012). These processes are integrated by the function of two central flowering regulators, CONSTANS (CO) and FLOWERING LOCUS C (FLC). In the photoperiod pathway, light signals are perceived by photoreceptors, which process the physical signals. These photoreceptors signal GIGANTEA (GI) gene and then trigger the CO gene (Fowler et al., 1999; Hicks et al., 2001; Simpson and Dean, 2002; Suarez-Lopez et al., 2001; Zhang et al., 2011). CO is a transcription factor that promotes flowering by directly inducing expression of FLOWERING LOCUS T (FT) and indirectly induces expression of SUPPRESSOR OF OVEREXPRESSION OF CO1 (SOC1) (Komeda, 2004; Samach et al., 2000; Tiwari et al., 2010; Zhang et al., 2011). This CO transcription factor contains three conserve domains: two B-box domains and one CCT (CO, CO-like, TOC1) domain (Robson et al., 2001). The two B-box domains are a class of zinc-finger transcription factor and are located near the amino terminus (N-terminus). These B-box domains have a function in protein-protein interaction and transcription regulation (Chou et al., 2013; Gangappa and Botto, 2014; Suarez-Lopez et al., 2001; Thomas, 2006). The CCT domain locates near carboxyl terminus (C-terminus) which has a function in protein-protein interactions, transcriptional regulation, and protein transport (Chou et al., 2013; Gangappa and Botto, 2014; Putterill et al., 1995; Robson et al., 2001; Strayer et al., 2000). The CCT domain of Arabidopsis CO was demonstrated to bind directly to the FT promoter (Gangappa and Botto, 2014; Tiwari et al., 2010). In Arabidopsis, seventeen members of the CO gene family have been identified including one CO and sixteen CO-like (COL) genes (Robson et al., 2001; Zhang et al., 2011). The Arabidopsis B-box zinc finger protein family (BBX) was classified into five groups based on phylogenetic analysis and structure classification: Group I contains two B-box domains (B-box I and B-box II) and a CCT domain, Group II contains B-box I, high variation B-box II, and a CCT domain; Group III contains only one B-box domain (B-box I) and a CCT domain, Group IV contains B-box I and B-box II but no CCT domain, Group V contains only a single B-box I domain (Gangappa and Botto, 2014; Khanna et al., 2009). These BBX transcription factors have various functions involving developmental and growth regulation including seedling photomorphogenesis, photoperiodic regulation of flowering, shade avoidance, and responses to biotic and abiotic stresses (Gangappa and Botto, 2014). CO and COL genes have been studied in various plant species, i.e., A. thaliana (Robson et al., 2001), barley (Griffiths et al., 2003), wheat (Nemoto et al., 2003), and sugar beet (Chia et al., 2008).

To date, 854 COL genes from 81 species have been deposited in the PlantTFDB database (Liu et al., 2015). COL genes have been identified in three orchid species, Erycina pusilla (Chou et al., 2013), Phalaenopsis hybrida (Zhang et al., 2011), and Oncidium ‘Gower Ramsey’ (Chang et al., 2011). In Oncidium ‘Gower Ramsey’, seven putative COLs were identified through 454 GS-FLX Pyrosequencing (Chang et al., 2011). In Erycina pusilla, 12 EpCOL were identified through Illumina sequencing. These EpCOL genes were divided into four groups (I–IV) according to their structural domains (Chou et al., 2013). Another COL from an orchid that has been identified was Phalaenopsis hybrida COL (PhalCOL). This PhalCOL was classified in the group I COL which contains two conserved B-box domain and one CCT domain. Overexpression of the PhalCOL in tobacco demonstrated the early flowering phenotype (Zhang et al., 2011). It is clear that a number of COLs are present in plant genomes and each of them have different functions, although next generation sequencing technologies are used for identification of all the COLs in plants, traditional methods of cDNA amplification and cloning are required for characterization of these genes.

One of the main floral induction factors in various Dendrobium orchids, including D. crumenatum, is known to be cold treatment. However, another factor that is required for floral transition in many flowering plants is photoperiod. To examine whether photoperiodism plays a role in floral induction in this tropical orchid, a CO/COL gene known to have function in floral transition via the photoperiodic pathway was isolated and characterized. The expression of this D. crumenatum CO/COL gene was investigated under different day length conditions. The results provided here will improve our understanding of the regulation of orchid flowering through photoperiods and contribute knowledge for further analysis of flowering control in Dendrobium orchids.

Materials and Methods

Plant materials

Dendrobium crumenatum plants were maintained under natural light in a greenhouse at Kasetsart University. The temperature of the greenhouse was approximately 30–35°C during the day time and 28–32°C during the night time throughout the year. The reproductive stage of the plant was determined by the presence of a leafless inflorescence stem produced at the distal part of the pseudobulb (Fig. 4C). In the photoperiod experiment, D. crumenatum specimens that contained inflorescence stems were transferred into a growth chamber at a photosynthetic photon flux density (PPFD) of 26 μmol·m⁻²·s⁻¹. All plants were treated with a neutral day condition (12 h light, 32°C and 12 h dark, 30°C) for three days. The plants were then separated into three sets to be treated with long day, neutral day, and short day conditions for two days before the leaves were collected for quantitative RT-PCR analysis. In the long day condition, plants were exposed to 16 h light, 32°C and 8 h dark, 30°C. In the neutral condition, plants were exposed to 12 h light, 32°C and 12 h dark, 30°C. In short day condition, plants were exposed to 8 h light, 32°C and 16 h dark, 30°C. In all conditions, one leaf (1 × 2 cm) per plant was harvested every four hours for two consecutive days. The harvested leaves were immediately submerged in liquid nitrogen and stored at −80°C.

CONSTANS-like cloning from D. crumenatum

Total RNA samples were extracted from leaves of reproductive stage plants using TRIzol (Invitrogen, USA) according to the manufacturer’s manual. D. crumenatum CO-Like (DcCOL) cDNA was synthesized using Viva 2-steps RT-PCR Kit (Vivantis, Malaysia). The PCR reaction included 1 μg of total RNA, 40 mM Oligo d(T)18 primer, 10 mM dNTPs. Water was added to 10 μL and incubated at 65°C for 5 min. The reaction was placed on ice for 2 min. Later, 1x buffer M-MuLV and M-MuLV reverse transcriptase 100 U were added to the reaction and the volume was adjusted to 20 μL with water and incubated at 42°C for 90 min and 70°C 10 min. PCR was then performed in the reaction containing 1 μL of cDNA, 5 μM of each specific primers, W2-CO-F1 (5'-GTCTGGCTCTGCGAGGTGTG-3') and W2-CO-R1 (5'-ATGAAAGCAGCGGTGGCG-3'), 0.1 mM dNTPs, 1x Vi buffer, and 2.5 U Taq polymerase under the following conditions: initial denaturation for 5 min at 94°C, 40 amplification cycles of denaturation for 1 min at 94°C, annealing for 30 s at 60°C, extension for 1 min at 72°C, and final extension for 5 min at 72°C. The W2-CO-F1 and W2-CO-R1 primers were designed from the conserved regions of Dendrobium loddigesii (GU301089.1) and Phalaenopsis hybrid (FJ469986.1) COL genes retrieved from Genbank database. The amplified cDNA fragment was cut and purified with a Gel/PCR DNA Fragments Extraction Kit (FAVOGEN, Taiwan). A purified fragment 75 ng was cloned into pGEM-T Easy vector 25 ng (Promega, USA) as described in the manufacturer’s protocol and transformed into DH5α Escherichia coli by a heat shock technique. Positive clones were selected on LB agar media containing ampicillin, isopropyl β-D-1-thiogalactopyranoside (IPTG), and X-gal and further verified using colony PCR with GoTaq^® Green (Promega). The positive clones were then sent for sequencing (Macrogen, South Korea). Full-length DcCOL was amplified using a SMARTer RACE cDNA Amplification kit (Clontech, USA). Both 3' and 5' RACE were performed in 5 μL reactions with total RNA of 0.5 μg, 1.2 mM 3' RACE CDS primer A for 3' RACE and 1.2 μM 5'-CDS Primer A for 5' RACE. Nuclease-free water was added to 2.38 μL for 3' RACE and 1.38 μL for 5' RACE. The reactions were incubated at 72°C for 3 min and chilled at 42°C for 2 min. DTT 4 mM was then added to the reaction followed by 2 mM dNTP mix, 2x first strand buffer, RNase inhibitor 5.2 U, and SMARTScribe Reverse Transcriptase 50 U. SMARTer II A oligo 1.2 μM was only added to the 5' RACE reaction. Both reactions were then incubated at 42°C for 90 min and 70°C for 10 min. PCR was then performed in a 25 μL reaction comprised of 1.25 μL of either 3' or 5' RACE cDNA, 1x Advantage 2 PCR Buffer, 2 mM dNTP Mix, 25x Advantage 2 Polymerase Mix, 0.2 μM 10X Universal Primer A Mix (UPM) Long, and 5 mM primers. For 3' RACE, a W2-CO-F1 (5'-GTCTGGCTCTGCGAGGTGTG-3') primer was used under the following thermal cycling conditions: initial denaturation for 5 min at 94°C, 40 amplification cycles of denaturation for 1 min at 94°C, annealing for 30 s at 60°C, extension for 1 min at 72°C, and final extension for 5 min at 72°C. For 5' RACE-PCR, W2-CO-R1 (5'-ATGAAAGCAGCGGTGGCG-3') primer was used under the following thermal cycling condition: five amplification cycles of 30 s at 94°C and 72°C 3 min, five amplification cycles of denaturation for 30 s at 94°C, annealing for 30 s at 68°C and extension for 3 min at 72°C and 25 amplification cycles of denaturation for 30 s at 94°C, annealing for 30 s at 68°C and extension for 3 min at 72°C. PCR products from 5' and 3' RACE-PCR were checked on 1% agarose gel. The amplified sequences were cloned and sequenced as mentioned above. DNA sequences from 3' and 5' RACE were blasted against the GenBank database using BLAST tools (Johnson et al., 2008). The 3' and 5' DNA sequences were then assembled using CAP3 (Huang and Madan, 1999).

Protein sequence analysis

The nucleotide sequence of COL from D. crumenatum was translated by the ExPASy translation tool (Gasteiger et al., 2003). The translated protein was blasted against Genbank database using blastp (Johnson et al., 2008). Conserved domains of DcCOL were identified using multiple sequence alignment. Amino acid sequences of DcCOL and COL from other plant species i.e. Dendrobium loddigisii COL (ADU05552.1), Phalaenopsis hybrida COL (ACS94258.1), Cymbidium sinense COL (ADN97077.1), C. goeringii COL (ADQ27228.1), C. ensifolium COL (ADQ27229.1), Erycina pusilla COL5 (AGI62029.1), Boehmeria nivea COL2 (AHI45076.1), Solanum lycopersicum COL (AAS67376.1), Arabidopsis thaliana CO/COL1-5 (NP_197088.1, NP_186887.1, NP180052.1, NP197875.2, NP_568863.1, NP_197088.1), Oryza sativa Hd1 (ABB17664.1), Zea mays Hd1 (ABW82153.1), and Triticum aestivum Hd1 (BAC92733.1) were aligned using MAFFT (Katoh and Standley, 2013). Protein motifs of DcCOL and other COL were predicted using MEME (Bailey et al., 2015).

Phylogenetic analysis

MAFFT-Aligned amino acid sequences of DcCOL and CO/COL from other plant species were used to find the best model for constructing a phylogenetic tree using the model test tool in MEGA6 (Tamura et al., 2013). A phylogenetic tree of these CO/COL proteins was then constructed using Neighbor-Joining (NJ) method with a JTT + G model using MEGA6 (Tamura et al., 2013). Bootstrap values were derived from 1000 replicates.

Gene expression analysis

Total RNA was extracted from roots, pseudobulbs, and leaves of juvenile and reproductive plants of D. crumenatum. In addition, total RNA was also extracted from inflorescence, floral buds, and flowers of D. crumenatum in the reproductive stage. All D. crumenatum plants used in this experiment were cultured under the greenhouse conditions mentioned previously. Three plants were used for each stage of development. The RNA extraction was performed as mentioned previously. Total RNA was reverse transcribed into cDNA using a RevertAid First Strand cDNA Synthesis kit (Thermo Scientific, USA). The first strand synthesis reaction included 1 μg of total RNA, 5x reaction buffer, 10 mM dNTP Mix, RiboLock RNase Inhibitor 20 U, and RevertAid M-MuLV RT 200 U. The reaction was incubated for 60 min at 42°C and terminated at 70°C for 5 min. Quantitative real-time RT-PCR was conducted on an Eppendorf Mastercycler^® ep realplex4 S real-time RT-PCR instrument (Eppendorf, USA) with Luminaris Color HiGreen qPCR Master Mix (Thermo scientific). The quantitative real-time RT-PCR reaction included cDNA template 3 μL, 0.3 μM of forward/reverse primers, Maxima SYBR green qPCR Master Mix(2x) no ROX 12.5 μL, and water was added up to 25 μL. Ubiquitin was amplified as an internal control using DOUbi-F (5'-AGGCTAAGAGGTGGAACAATGATC-3') and DOUbi-R (5'-ATCAGCAAGCTGCTTGCCTGCAT-3') primers. DcCOL gene product was detected using W3-CO-F (5'-CGAACGACTGCTTCTCTGAT-3') and W3-CO-R (5'-GCCTGTTCTTCCTCTTCTCC-3') primers. Data analysis was performed according to RealPlex software (Eppendorf). The PCR reaction was pre-treated with 50°C 2 min followed by initial denaturation for 10 min at 95°C, denaturation for 15 s at 95°C, annealing for 30 s at 58°C, and a last extension for 30 s at 72°C. Relative expression was calculated by the ΔΔCT method.

For photoperiod response analysis, Total RNA was extracted from the leaves of plants treated with long day, neutral day, and short day conditions at 4 h intervals across 48 h, starting from 0 h. The RNA extraction and quantitative real-time RT-PCR were performed as described above. Four replications were performed for each time point.

Results and Discussion

DcCOL protein sequence analysis

First strand DNA derived from Dendrobium curmentaum (KT894671) was 969 bp, and encoded 322 amino acids with a molecular weight of 34.25 kDa. The deduced DcCOL amino acid sequence shared a high identity with COL from Dendrobium (92%), Cymbidium (86%), Phalaenopsis COL (85%), Oryza sativa Hd1 (70%), Erycina COL5 (66%), Boehmeria nivea COL2 (44%), Solanum lycopersicum COL (41%), and Arabidopsis thaliana CO (39%). This result demonstrated that DcCOL has the highest homology with the closely related species, D. lodigisii, and more similar with monocot than dicot confirming that DcCOL was one of the COL members in the Dendrobium orchid. In addition, according to amino acid sequence alignment, three conserved regions were detected in the DcCOL sequence. The first region, B-box1, comprised approximately 40 amino acids, starting from C20 to H59. The second region, B-box2 comprised 38 amino acids starting from C65 to H102. The last region, CCT, which was the highest conserved region comprised 43 amino acids starting from R246 to R288 (Fig. 1). MEME suite was used to perform motif analysis to delineate the functional motifs in DcCOL protein. The predicted motifs of DcCOL included the known CCT (motif 1) and the B-box domains (motif 2 and motif 3; Fig. 2).

Fig. 1

Multiple sequence alignment of DcCOL. Multiple sequence alignment of DcCOL and COL from other plant species was performed with MAFFT (Katoh and Standley, 2013). Conserved B-boxes, VP motif, and CCT domain are shown in highlighted areas. The nuclear localization signal in the CCT domain is shown in a box.

Fig. 2

Putative motifs in various COL proteins as identified by MEME. The boxes represent different putative motifs. The expected values for all motifs were calculated by MEME (Bailey et al., 2015). Black, white, and gray boxes represent CCT, B-box I, and B-box II motifs, respectively. Group I CO/COL contains all three motifs, CCT, B-box I, and B-box II motifs. Group II contains a CCT domain, one conserved B-box I and one highly variable B-box II. Group III contains CCT and B-box I motifs.

According to the presence and arrangement of putative motifs, DcCOL was classified into group I B-box zinc finger protein. The consensus sequence detected in DcCOL B-box1 was C-X₂-C-X₁₀-C-X₇-C-X₂-C-X₄-H-X₈-H and B-box2 was C-X₂-C-X₈-C-X₇-C-X₂-C-X₄-H-X₈-H (Fig. 1). There were seven conserved residues, including cysteine and histidine that acted as one zinc atom binding per one B-box in DcCOL protein structure. The B-box domains have been demonstrated to be involved in the protein-protein interaction (Torok and Etkin, 2001). The consensus sequence of B-box1 and B-box2 in Arabidopsis and rice and EpCOL5 are as follows: Arabidopsis; B-box1: C-X₂-C-X_{7 - 8}-C-X₂-D-X-A-X-L-C-X₂-C-D-X₃-H, B-box2: C-X₂-C-X₃-P-X₄-C-X₂-D-X₃-L-C-X₂-C-D-X₃-H, Rice B-box1: C-X₂-C-X₈-C-X₇-C-X₂-C-X₄-H-X₈-H, B-box2: C-X₂-C-X₈-C-X₇-C-X₂-C-X₄-H-X₈-H, EpCOL5; B-box1: C-X₂-C-X₁₃-C-X₇-C-X₂-C-X₄-H-X₈-H, B-box2: C-X₂-C-X₈-C-X₇-C-X₂-C-X₄-H-X₈-H. Interestingly, the DcCOL B-box consensuses were identical to other Dendrobium and Cymbidium COLs but were slightly different from EpCOL5. The B-box1 of EpCOL5 contained five additional amino acids in addition to DcCOL (Fig. 1). Most of the CO/COL that contained both B-box1 and B-box2 domains were reported to be involved in photoperiodic control of flowering e.g. Arabidopsis CO (Valverde, 2011), Oryza sativa Hd1 (Yano et al., 2000), Lolium perenne LpCO (Martin et al., 2004), and Hordeum vulgare HvCO1 (Campoli et al., 2012). However, the CO/COL proteins that contain only B-box1 were demonstrated to be involved in other pathways other than the flowering control (Gangappa and Botto, 2014). Therefore, the functional B-box2 domain has been suggested to be crucial for flower induction (Campoli et al., 2012; Griffiths et al., 2003; Liu et al., 2015; Martin et al., 2004). Although both the AtCO and OsHD1 contain both B-box1 and B-box2 domains, the composition of amino acids in these two domains are slightly different. Both of these proteins have a function in induction of flowering, but their regulation pathways are different. Hd1 was demonstrated to promote flowering in short day (SD) (Hayama et al., 2003; Kim et al., 2008). While AtCO promotes flowering in long day (LD) (Putterill et al., 1995; Samach et al., 2000; Zhang et al., 2011). The different regulation patterns suggested that other parts of this protein are responsible for the regulation of flowering. Another important domain in the DcCOL was the CCT domain. Most of the CCT domain-containing proteins have functions associated with environmental responses to light, temperature, or the circadian clock (Wenkel et al., 2006). The CCT domain was demonstrated to have two crucial functions in the CO/COL proteins; a nuclear localization and a protein-protein interaction (Robson et al., 2001). The nuclear localization signals (NLSs) can be classified into two groups, monopartite and bipartite. The monopartite NLSs contain K-(K/R)-X-(K/R) as a putative consensus sequence (Crocco and Botto, 2013; Lange et al., 2007). The bipartite NLSs contain a putative consensus sequence of (K/R)-(K/R)-X10-12-(K/R) (Crocco and Botto, 2013). The NLSs identified in the DcCOL, and other members of CO/COL analyzed was bipartite (Fig. 1). The previous study showed that bipartite NLSs has a nuclear uptake function in CO and COL3 (Crocco and Botto, 2013; Datta et al., 2006; Robson et al., 2001). This segment of the CCT domain was demonstrated to localize GFP to the nucleus. This CCT domain was demonstrated to have functions in both transcriptional regulation and nuclear localization. A mutation analysis of the CCT domain delays flowering without affecting the nuclear localization function (Kim et al., 2013; Robson et al., 2001). Apart from these three domains in the CO/COL, other domains were also identified in the CO/COL proteins. The VP Motif (Fig. 1) located between the B-box2 and CCT was identified in the structural group I (Crocco and Botto, 2013; Gangappa and Botto, 2014). This motif has an upstream stretch of 4–5 negatively charged residues and the core sequence V-P-E/D-φ-G (φ = hydrophobic residue) (Holm et al., 2001). It was demonstrated to interact with CONSTITUTIVE PHOTOMORPHOGENIC1 (COP1), an E3 ubiquitin ligase that represses photomorphogenesis in the dark (Datta et al., 2006). Even though all group I CO/COL members contain these four conserved motifs, their functions are diverse across all members, suggesting that the position responsible for the diversified function of CO/COL is unlikely to be found in these conserved regions.

Phylogenetic analysis of CO/COL

To examine the relationship among CO/COL genes, amino acid sequences of 37 CO/COL from 13 plant species were used to construct a phylogenetic tree. The phylogenetic tree of these CO/COL sequences can be grouped into three clades based on their structural domains (Fig. 3). Clade I (73 BP) comprised of CO/COL was categorized in group I B-box as most of the proteins in this clade contain two highly conserved B-boxes, except for EpCOL11, that contained only one B-box. This clade was comprised of DcCOL, AtCO, OsHd1, ZmHd1 and TsHd1, AtCOL1, 2, 3, 4, 5, BnCOL2, BvCOL2, PhalCOL1, 2, EpCOL5, 11, and all other orchid COLs. Clade II (100 BP) comprised of group III CO/COL containing one B-box and a CCT domain. This group III CO/COL was comprised of AtCOL6, 7, 8, EpCOL7 and 12. Clade III (99 BP) comprised of group II CO/COL containing two B-boxes, one highly conserved and one highly variable B-box. This group II CO/COL was comprised of AtCOL9, 10, 11, 12, 13, 14, 15, EpCOL3, 4, 6, 8, and 10. The topology and classification of these CO/COL proteins were similar to the trees demonstrated in previous studies (Chou et al., 2013; Zhang et al., 2011) supporting the classification of DcCOL in the group I COL. However, even though all group I CO/COL proteins contain similar B-boxes and CCT domain structures, they were demonstrated to have various functions (Gangappa and Botto, 2014; Liu et al., 2015). AtCO and AtCOL5 were demonstrated to induce flowering in Arabidopsis (Hassidim et al., 2009; Suarez-Lopez et al., 2001); however, the other Arabidopsis group I CO/COL, AtCOL1, and AtCOL2 were demonstrated to have little effect on flowering (Ledger et al., 2001). Nonetheless, the circadian clock was shown to have an effect on the expression of these two genes (Ledger et al., 2001). Although not all group I CO/COLs have a direct effect on flowering, all were regulated by light (Gangappa and Botto, 2014).

Fig. 3

Phylogenetic tree of CO/COL proteins from different plant species. The phylogenetic tree was generated by Neighbor-Joining (NJ) using MEGA6.0 (Tamura et al., 2013) with a bootstrap of 1000. The CO/COL were separated into three groups according to their structural domains. Group I contains CCT, B-box I, and B-box II domains. Group II contains a CCT domain, B-box I, and a diverged B-box II. Group III contains CCT and a B-box I domain. DcCOL was classified into the first group. Numbers above branches indicate bootstrap values.

DcCOL expression analysis

DcCOL expression was examined in various tissue types and under different light conditions to assess the function of DcCOL in flowering. Quantitative real-time RT-PCR demonstrated that the DcCOL was expressed in all organs in both juvenile and reproductive stages (Fig. 4). In the juvenile stage, expression of DcCOL was detected at the highest level in pseudobulbs and the lowest in the leaves. In the reproductive phase, the highest expression was detected in the open flowers and the lowest in the pseudobulbs. The expression of DcCOL was approximately two fold higher in the reproductive stage than in the juvenile stage, and the highest level was found in open flowers. This vegetative stage expression of COL was also demonstrated in E. pusilla (Chou et al., 2013) and P. hybrida (Zhang et al., 2011). The expression of EpCOL and PhalCOL genes was detected in different tissues including floral buds, opened flowers, peduncles, capsules, leaves, and roots (Chou et al., 2013; Zhang et al., 2011). The CO/COL genes were reported to be expressed throughout the life cycle of the plant from embryonic development through to the reproductive phase in almost all tissue types (Tiwari et al., 2010). CO/Hd1 proteins in Arabidopsis and rice were demonstrated to be expressed in the leaves which is a key location of these proteins to modulate the transcription of FT/Hd3a (Liu et al., 2015; Turck et al., 2008). However, DcCOL expression was detected at a higher level in roots and pseudobulbs than in the leaves of the juvenile stage (Fig. 4) which is not surprising considering the morphology of this epiphytic orchid, D. crumenatum. This orchid has large pseudobulbs that produce a long stem containing 2–10 small leaves. The pseudobulbs and roots of most epiphytic orchids, including Dendrobium, are the main locations of photosynthesis. The expression of PhalCOL corresponded to its morphology as it processes big green leaves that grow on the stem and does not contain pseudobulbs. The PhalCOL expression was detected at the highest level in the leaves than in the roots and stems during the entire vegetative development (Zhang et al., 2011). This evidence suggested that COL can be expressed differently among species depending on the habit of the plants. The expression of DcCOL in the juvenile phase leaves was lower than in the reproductive stages, suggesting that DcCOL is required in the reproductive stage possibly for floral induction through the photoperiod pathway that occurred in the leaves.

Fig. 4

Expression of DcCOL in different organs at vegetative and reproductive stages. Quantitative real-time RT-PCR (A) of DcCOL was performed using Ubiquitin (DcUbi) as an internal control gene. Juvenile (B) and reproductive (C) stages of D. crumenatum can be distinguished by the presence of an inflorescence stem (arrow) that appeared only in the adult plants. Error bars represent standard error (n = 3). Jr; Juvenile root, Jp; Juvenile pseudobulb, Jl; Juvenile leaf, Rr; Reproductive root, Rp; Reproductive pseudobulb, Rl; Reproductive leaf, Inf; Inflorescence, Fb; Flower bud, Flo; Open flower.

To assess whether DcCOL expression is controlled by the photoperiod pathway, the expression patterns of DcCOL under different light conditions were investigated. Quantitative real-time RT-PCR of DcCOL was carried out in the mature D. crumenatum specimens under LD, neutral day (ND), and SD conditions. The expression levels of DcCOL under LD and ND were higher than under SD. The DcCOL mRNA expression was observed at four hour intervals across 48 h (Fig. 5). The expression levels of DcCOL under all conditions were decreased at the light phase and increased gradually to the peak at 16 Zeitgeber time (ZT). This could partly be a result of the pretreatment with the ND condition as the expression in all treatments demonstrated the exact same pattern. However, on the second day of the light treatment (24–48 ZT), the expression levels of the DcCOL more closely coincided with the day-night cycle. In all treatments, the DcCOL expressions were decreased to the basal level at the light phase and spiked up to the highest level at exactly 4 h after darkness. The expression patterns of DcCOL in all conditions were relatively similar, suggesting that day length does not affect flowering in D. crumenatum. This result coincides with the flowering behavior of this orchid. D. crumenatum is a free flowering species which can flower throughout the year (Holttum, 1949). The diurnal expression of DcCOL also coincides with the flowering habit of D. crumenatum as its flower opens before dawn and lasts for only 24 h (Yap et al., 2008). A similar diurnal expression pattern of DcCOL was also observed in OsHd1, and BnCOL2 in which both have a function in flowering control (Izawa et al., 2002; Liu et al., 2015). Taken together, the results from protein structure analysis, phylogenetic and expression patterns suggested that DcCOL can be classified in group I CO similar to Arabidopsis CO, rice Hd1, and BnCOL2 that are all suggested to have a function involved in flowering regulation. Moreover, expression analysis of DcCOL demonstrated a diurnal expression pattern and accumulation of transcript during the dark similar to the expression pattern in OsHd1 and BnCOL2, suggesting that DcCOL may also have a function in flowering control in the orchid D. crumenatum.

Fig. 5

Expression of the DcCOL gene under different day length conditions. Expression of the DcCOL gene under long day, neutral day, and short day conditions were compared. The expression level of DcCOL in D. crumenatum was determined by quantitative real-time RT-PCR using specific primers. Day and night periods are labeled at the bottom of each chart using white and black bars, respectively. Transcripts of DcCOL were determined using four replicates and normalized with Ubiquitin. Experimental time refers to Zeitgeber time (ZT). Error bars represent standard error (n = 4).

The flowering control in D. crumenatum is a complex process involving the following steps: 1) An induction of inflorescence development on the terminal portion of stems. 2) Development of flower buds on the stem. 3) The flower buds develop until all flower parts are formed and the anther is almost fully developed and then undergo dormancy. 4) Upon cold induction, these dormant flower buds open (Arditti, 1979). Floral transition in most flowering plants involves two floral enabling (endogenous and hormonal) and two floral promoting pathways (vernalization and photoperiod). The first three steps of D. crumenatum flowering control are likely to be controlled by the floral enabling pathway to develop the inflorescence stem. The last step of D. crumenatum flowering control is controlled by the vernalization pathway (Arditti, 1979; Holttum, 1949; Meesawat and Kanchanapoom, 2007). Taken together, the floral induction behavior in the natural habitat of D. crumenatum and the expression patterns of the light-responsive gene DcCOL demonstrated in this study, the photoperiod pathway does not seem to have a crucial role in controlling the expression of DcCOL and the floral induction in D. crumenatum. Instead, the expression of DcCOL was controlled by the dark and light period during the day, corresponding to the flowering time which only occurs just before the sun rises and lasts for only one day. This similar flowering behavior was also observed in morning glory, Ipomoea nil (Yamada et al., 2007). A differential subtraction screening of cDNA of I. nil flower from t = 0 (8:00 a.m.) to 4 h (11:00 a.m.) showed the expression of the CO-Like gene was dramatically decreased suggesting involvement of COL in the flowering process in this species (Yamada et al., 2007). Not only floral induction, but also flower opening require environmental factors such as temperature and circadian rhythm (van Doorn and van Meeteren, 2003; van Doorn and Kamdee, 2014). Therefore, COL, which is already known for its function in floral transition, may also have an indirect function in flower opening as demonstrated in this study. Another factor that controlled flowering in D. crumenatum was demonstrated to be induction by vernalization. Cold induction was detected in various tropical epiphytic orchid species e.g. Phalaenopsis aphrodite (Jang et al., 2015), Dendrobium Second Love (Campos and Kerbauy, 2004), and D. nobile (Yen et al., 2008). In Phalaenopsis aphrodite, diurnal fluctuation of high day and low night temperatures or a cool temperature at night promoted flowering of this orchid. A low temperature, but not photoperiod, was demonstrated to induce flowering by directing the expression of the two main floral integrator genes, FT and FD (Jang et al., 2015).

Conclusions

This study demonstrated that the expression of DcCOL, the main photoperiodic responsive gene in D. crumenatum, was not affected by the photoperiod, but its expression responded to light and dark periods in a daily cycle. Although the direct effect of DcCOL expression and floral induction was not demonstrated in this study, the expression patterns reported here coincide with the flowering behavior of this orchid in the natural environment, suggesting the function of DcCOL in flowering control.

Acknowledgements

We would like to thank Graduate School, Kasetsart University, Bangkok and the Kasetsart University Research and Development Institute (KURDI), for financial support for this project.

Literature Cited

• Arditti, J. 1979. Aspects of the physiology of orchids. Academic Press, London.
•  Atwood,  J. T. 1986. The size of the Orchidaceae and the systematic distribution of epiphytic orchids. Selbyana 9: 171–186.
•  Bailey,  T. L.,  J.  Johnson,  C. E.  Grant and  W. S.  Noble. 2015. The MEME Suite. Nucl. Acids Res. 43: W39–49.
• Beaman, T. E., J. J. Wood and R. S. Beaman. 2001. Orchids of Sarawak. Natural History Publications (Borneo) and Kew Publishing, Malaysia.
•  Borner,  R.,  G.  Kampmann,  J.  Chandler,  R.  Gleissner,  E.  Wisman,  K.  Apel and  S.  Melzer. 2000. A MADS domain gene involved in the transition to flowering in Arabidopsis. Plant J. 24: 591–599.
•  Boss,  P. K.,  R. M.  Bastow,  J. S.  Mylne and  C.  Dean. 2004. Multiple pathways in the decision to flower: Enabling, promoting, and resetting. Plant Cell 16: S18–S31.
•  Campoli,  C.,  B.  Drosse,  I.  Searle,  G.  Coupland and  M.  von Korff. 2012. Functional characterisation of HvCO1, the barley (Hordeum vulgare) flowering time ortholog of CONSTANS. Plant J. 69: 868–880.
•  Campos,  K. O. and  G. B.  Kerbauy. 2004. Thermoperiodic effect on flowering and endogenous hormonal status in Dendrobium (Orchidaceae). J. Plant Physiol. 161: 1385–1387.
•  Chang,  Y. Y.,  Y. W.  Chu,  C. W.  Chen,  W. M.  Leu,  H. F.  Hsu and  C. H.  Yang. 2011. Characterization of Oncidium ‘Gower Ramsey’ transcriptomes using 454 GS-FLX pyrosequencing and their application to the identification of genes associated with flowering time. Plant Cell Physiol. 52: 1532–1545.
•  Chia,  T. Y.,  A.  Muller,  C.  Jung and  E. S.  Mutasa-Gottgens. 2008. Sugar beet contains a large CONSTANS-LIKE gene family including a CO homologue that is independent of the early-bolting (B) gene locus. J. Exp. Bot. 59: 2735–2748.
•  Chou,  M. L.,  M. C.  Shih,  M. T.  Chan,  S. Y.  Liao,  C. T.  Hsu,  Y. T.  Haung,  J. J.  Chen,  D. C.  Liao,  F. H.  Wu and  C. S.  Lin. 2013. Global transcriptome analysis and identification of a CONSTANS-like gene family in the orchid Erycina pusilla. Planta 237: 1425–1441.
•  Crocco,  C. D. and  J. F.  Botto. 2013. BBX proteins in green plants: insights into their evolution, structure, feature and functional diversification. Gene 531: 44–52.
•  da Silva,  J. A. T.,  S.  Aceto,  W.  Liu,  H.  Yu and  A.  Kanno. 2014. Genetic control of flower development, color and senescence of Dendrobium orchids. Sci. Hortic. 175: 74–86.
•  Datta,  S.,  G.  Hettiarachchi,  X.-W.  Deng and  M.  Holm. 2006. Arabidopsis CONSTANS-LIKE3 is a positive regulator of red light signaling and root growth. Plant Cell 18: 70–84.
•  Fowler,  S.,  K.  Lee,  H.  Onouchi,  A.  Samach,  K.  Richardson,  B.  Morris,  G.  Coupland and  J.  Putterill. 1999. GIGANTEA: a circadian clock‐controlled gene that regulates photoperiodic flowering in Arabidopsis and encodes a protein with several possible membrane‐spanning domains. EMBO J. 18: 4679–4688.
•  Fu,  C.-H.,  Y.-W.  Chen,  Y.-Y.  Hsiao,  Z.-J.  Pan,  Z.-J.  Liu,  Y.-M.  Huang,  W.-C.  Tsai and  H.-H.  Chen. 2011. OrchidBase: a collection of sequences of the transcriptome derived from orchids. Plant Cell Physiol. 52: 238–243.
•  Gangappa,  S. N. and  J. F.  Botto. 2014. The BBX family of plant transcription factors. Trends Plant Sci. 19: 460–470.
•  Gasteiger,  E.,  A.  Gattiker,  C.  Hoogland,  I.  Ivanyi,  R. D.  Appel and  A.  Bairoch. 2003. ExPASy: the proteomics server for in-depth protein knowledge and analysis. Nucl. Acids Res. 31: 3784–3788.
•  Griffiths,  S.,  R. P.  Dunford,  G.  Coupland and  D. A.  Laurie. 2003. The evolution of CONSTANS-like gene families in barley, rice, and Arabidopsis. Plant Physiol. 131: 1855–1867.
•  Hassidim,  M.,  Y.  Harir,  E.  Yakir,  I.  Kron and  R.  Green. 2009. Over-expression of CONSTANS-LIKE 5 can induce flowering in short-day grown Arabidopsis. Planta 230: 481–491.
•  Hayama,  R.,  S.  Yokoi,  S.  Tamaki,  M.  Yano and  K.  Shimamoto. 2003. Adaptation of photoperiodic control pathways produces short-day flowering in rice. Nature 422: 719–722.
•  Hicks,  K. A.,  T. M.  Albertson and  D. R.  Wagner. 2001. EARLY FLOWERING3 encodes a novel protein that regulates circadian clock function and flowering in Arabidopsis. Plant Cell 13: 1281–1292.
•  Holm,  M.,  C. S.  Hardtke,  R.  Gaudet and  X. W.  Deng. 2001. Identification of a structural motif that confers specific interaction with the WD40 repeat domain of Arabidopsis COP1. EMBO J. 20: 118–127.
•  Holttum,  R. E. 1949. Gregarious flowering of the terrestrial orchid Bromheadia finlaysoniana. Gard. Bull. (Singapore) 12: 295–302.
•  Hsiao,  Y.-Y.,  Z.-J.  Pan,  C.-C.  Hsu,  Y.-P.  Yang,  Y.-C.  Hsu,  Y.-C.  Chuang,  H.-H.  Shih,  W.-H.  Chen,  W.-C.  Tsai and  H.-H.  Chen. 2011. Research on orchid biology and biotechnology. Plant Cell Physiol. 52: 1467–1486.
•  Huang,  X. and  A.  Madan. 1999. CAP3: A DNA sequence assembly program. Genome Res. 9: 868–877.
•  Izawa,  T.,  T.  Oikawa,  N.  Sugiyama,  T.  Tanisaka,  M.  Yano and  K.  Shimamoto. 2002. Phytochrome mediates the external light signal to repress FT orthologs in photoperiodic flowering of rice. Genes & development 16: 2006–2020.
•  Jang,  S.,  S.-C.  Choi,  H.-Y.  Li,  G.  An and  E.  Schmelzer. 2015. Functional characterization of Phalaenopsis aphrodite flowering genes PaFT1 and PaFD. PloS ONE 10: e0134987.
•  Johnson,  M.,  I.  Zaretskaya,  Y.  Raytselis,  Y.  Merezhuk,  S.  McGinnis and  T. L.  Madden. 2008. NCBI BLAST: a better web interface. Nucl. Acids Res. 36: W5–W9.
•  Katoh,  K. and  D. M.  Standley. 2013. MAFFT multiple sequence alignment software version 7: improvements in performance and usability. Mol. Biol. Evol. 30: 772–780.
•  Khanna,  R.,  B.  Kronmiller,  D. R.  Maszle,  G.  Coupland,  M.  Holm,  T.  Mizuno and  S.-H.  Wu. 2009. The Arabidopsis B-box zinc finger family. Plant Cell 21: 3416–3420.
• Kim, S. K., H. Y. Park, Y. H. Jang, J. H. Lee and J. K. Kim. 2013. The sequence variation responsible for the functional difference between the CONSTANS protein, and the CONSTANS-like (COL) 1 and COL2 proteins, resides mostly in the region encoded by their first exons. Plant Sci. 199–200: 71–78.
•  Kim,  S. K.,  C. H.  Yun,  J. H.  Lee,  Y. H.  Jang,  H. Y.  Park and  J. K.  Kim. 2008. OsCO3, a CONSTANS-LIKE gene, controls flowering by negatively regulating the expression of FT-like genes under SD conditions in rice. Planta 228: 355–365.
•  Komeda,  Y. 2004. Genetic regulation of time to flower in Arabidopsis thaliana. Annu. Rev. Plant Biol. 55: 521–535.
•  Lange,  A.,  R. E.  Mills,  C. J.  Lange,  M.  Stewart,  S. E.  Devine and  A. H.  Corbett. 2007. Classical nuclear localization signals: definition, function, and interaction with importin α. J. Biol. Chem. 282: 5101–5105.
•  Ledger,  S.,  C.  Strayer,  F.  Ashton,  S. A.  Kay and  J.  Putterill. 2001. Analysis of the function of two circadian-regulated CONSTANS-LIKE genes. Plant J. 26: 15–22.
•  Liu,  T.,  S.  Zhu,  Q.  Tang and  S.  Tang. 2015. Identification of a CONSTANS homologous gene with distinct diurnal expression patterns in varied photoperiods in ramie (Boehmeria nivea L. Gaud). Gene 560: 63–70.
•  Martin,  J.,  M.  Storgaard,  C. H.  Andersen and  K. K.  Nielsen. 2004. Photoperiodic regulation of flowering in perennial ryegrass involving a CONSTANS-like homolog. Plant Mol. Biol. 56: 159–169.
•  Meesawat,  U. and  K.  Kanchanapoom. 2007. Understanding the flowering behavior of pigeon orchid (Dendrobium crumenatum Sw.). Orchid Sci. Biotech. 1: 6–14.
•  Mouradov,  A.,  F.  Cremer and  G.  Coupland. 2002. Control of flowering time: Interacting pathways as a basis for diversity. Plant Cell 14: S111–S130.
•  Nemoto,  Y.,  M.  Kisaka,  T.  Fuse,  M.  Yano and  Y.  Ogihara. 2003. Characterization and functional analysis of three wheat genes with homology to the CONSTANS flowering time gene in transgenic rice. Plant J. 36: 82–93.
•  Pellegrino,  G.,  F.  Bellusci and  A.  Musacchio. 2009. Genetic integrity of sympatric hybridising plant species: the case of Orchis italica and O. anthropophora. Plant Biol. (Stuttg) 11: 434–441.
•  Putterill,  J.,  F.  Robson,  K.  Lee,  R.  Simon and  G.  Coupland. 1995. The CONSTANS gene of Arabidopsis promotes flowering and encodes a protein showing similarities to zinc finger transcription factors. Cell 80: 847–857.
•  Robson,  F.,  M. M.  Costa,  S. R.  Hepworth,  I.  Vizir,  M.  Pineiro,  P. H.  Reeves,  J.  Putterill and  G.  Coupland. 2001. Functional importance of conserved domains in the flowering-time gene CONSTANS demonstrated by analysis of mutant alleles and transgenic plants. Plant J. 28: 619–631.
•  Russell,  A.,  R.  Samuel,  V.  Klejna,  M. H. J.  Barfuss,  B.  Rupp and  M. W.  Chase. 2010. Reticulate evolution in diploid and tetraploid species of Polystachya (Orchidaceae) as shown by plastid DNA sequences and low-copy nuclear genes. Ann. Bot. 106: 37–56.
•  Samach,  A.,  H.  Onouchi,  S. E.  Gold,  G. S.  Ditta,  Z.  Schwarz-Sommer,  M. F.  Yanofsky and  G.  Coupland. 2000. Distinct roles of CONSTANS target genes in reproductive development of Arabidopsis. Science 288: 1613–1616.
•  Simpson,  G. G. and  C.  Dean. 2002. Arabidopsis, the Rosetta stone of flowering time? Science 296: 285–289.
•  Strayer,  C.,  T.  Oyama,  T. F.  Schultz,  R.  Raman,  D. E.  Somers,  P.  Más,  S.  Panda,  J. A.  Kreps and  S. A.  Kay. 2000. Cloning of the Arabidopsis clock gene TOC1, an autoregulatory response regulator homolog. Science 289: 768–771.
•  Su,  C.-L.,  Y.-T.  Chao,  Y.-C. A.  Chang,  W.-C.  Chen,  C.-Y.  Chen,  A.-Y.  Lee,  K. T.  Hwa and  M.-C.  Shih. 2011. De novo assembly of expressed transcripts and global analysis of the Phalaenopsis aphrodite transcriptome. Plant Cell Physiol. 52: 1501–1514.
•  Suarez-Lopez,  P.,  K.  Wheatley,  F.  Robson,  H.  Onouchi,  F.  Valverde and  G.  Coupland. 2001. CONSTANS mediates between the circadian clock and the control of flowering in Arabidopsis. Nature 410: 1116–1120.
•  Tamura,  K.,  G.  Stecher,  D.  Peterson,  A.  Filipski and  S.  Kumar. 2013. MEGA6: Molecular Evolutionary Genetics Analysis Version 6.0. Mol. Biol. Evol. 30: 2725–2729.
•  Thomas,  B. 2006. Light signals and flowering. J. Exp. Bot. 57: 3387–3393.
•  Tiwari,  S. B.,  Y.  Shen,  H. C.  Chang,  Y.  Hou,  A.  Harris,  S. F.  Ma,  M.  McPartland,  G. J.  Hymus,  L.  Adam and  C.  Marion. 2010. The flowering time regulator CONSTANS is recruited to the FLOWERING LOCUS T promoter via a unique cis‐element. New Phytol. 187: 57–66.
•  Torok,  M. and  L. D.  Etkin. 2001. Two B or not two B? Overview of the rapidly expanding B-box family of proteins. Differentiation 67: 63–71.
•  Turck,  F.,  F.  Fornara and  G.  Coupland. 2008. Regulation and identity of florigen: FLOWERING LOCUS T moves center stage. Annu. Rev. Plant Biol. 59: 573–594.
•  Valverde,  F. 2011. CONSTANS and the evolutionary origin of photoperiodic timing of flowering. J. Exp. Bot. 62: 2453–2463.
•  van Doorn,  W. G. and  C.  Kamdee. 2014. Flower opening and closure: an update. J. Exp. Bot. 65: 5749–5757.
•  van Doorn,  W. G. and  U.  van Meeteren. 2003. Flower opening and closure: a review. J. Exp. Bot. 54: 1801–1812.
•  Wenkel,  S.,  F.  Turck,  K.  Singer,  L.  Gissot,  J.  Le Gourrierec,  A.  Samach and  G.  Coupland. 2006. CONSTANS and the CCAAT box binding complex share a functionally important domain and interact to regulate flowering of Arabidopsis. Plant Cell 18: 2971–2984.
• Wu, Z., P. Raven and D. Hong. 2009. Flora of China. Vol. 25 (Orchidaceae). Science Press, Beijing, and Missouri Botanical Garden Press, St. Louis.
•  Xiang,  L.,  X.  Li,  D.  Qin,  F.  Guo,  C.  Wu,  L.  Miao and  C.  Sun. 2012. Functional analysis of FLOWERING LOCUS T orthologs from spring orchid (Cymbidium goeringii Rchb. f.) that regulates the vegetative to reproductive transition. Plant Physiol. Biochem. 58: 98–105.
•  Xu,  Y.,  L. L.  Teo,  J.  Zhou,  P. P.  Kumar and  H.  Yu. 2006. Floral organ identity genes in the orchid Dendrobium crumenatum. Plant J. 46: 54–68.
•  Yamada,  T.,  K.  Ichimura,  M.  Kanekatsu and  W. G.  van Doorn. 2007. Gene expression in opening and senescing petals of morning glory (Ipomoea nil) flowers. Plant Cell Rep. 26: 823–835.
•  Yano,  M.,  Y.  Katayose,  M.  Ashikari,  U.  Yamanouchi,  L.  Monna,  T.  Fuse,  T.  Baba,  K.  Yamamoto,  Y.  Umehara and  Y.  Nagamura. 2000. Hd1, a major photoperiod sensitivity quantitative trait locus in rice, is closely related to the Arabidopsis flowering time gene CONSTANS. Plant Cell 12: 2473–2483.
•  Yap,  Y.-M.,  C.-S.  Loh and  B.-L.  Ong. 2008. Regulation of flower development in Dendrobium crumenatum by changes in carbohydrate contents, water status and cell wall metabolism. Sci. Hortic. 119: 59–66.
•  Yen,  C. Y.-T.,  T. W.  Starman,  Y.-T.  Wang and  G.  Niu. 2008. Effects of cooling temperature and duration on flowering of the nobile Dendrobium orchid. HortScience 43: 1765–1769.
•  Yu,  H. and  C. J.  Goh. 2001. Molecular genetics of reproductive biology in orchids. Plant Physiol. 127: 1390–1393.
•  Zhang,  J.-X.,  K.-L.  Wu,  L.-N.  Tian,  S.-J.  Zeng and  J.  Duan. 2011. Cloning and characterization of a novel CONSTANS-like gene from Phalaenopsis hybrida. Acta Physiol. Plant 33: 409–417.