Analysis of the DlGI promoter sequence using the PlantCARE database (Lescot et al., 2002).
Histochemical localization of GUS activity in transgenic Arabidopsis thaliana plants carrying the pDlGI::GUS fusion construct: (A) GUS staining localized in the cotyledons of 7-d-old seedlings; (B) 15-d-old plant, showing strong GUS expression in the cotyledons, hypocotyls, and leaves; (C) GUS staining observed in 21-d-old flower bud; (D) GUS staining localized in flowers (sepals, petals, and stigma) of a 26- to 30-d-old plant; (E) 40-d-old plant siliques showing GUS expression; and (F) no expression in wild type A. thaliana.
GUS expression in transgenic Arabidopsis thaliana seedlings under different light conditions: (A) high light (1100 µmol·m−2·s−1); (B) weak light (400 µmol·m−2·s−1); (C) dark conditions; (D) GUS staining in wild type A. thaliana (negative control); (E) level of GUS transcript in cotyledon, hypocotyl, and root under different intensity of light. Bars indicate the se.
Effect of different photoperiod on the activity of DlGI promoter infiltrated in Nicotiana benthamiana leaves: (A) DlGI promoter activity under long day (16/8 h light/dark) and short day (8/16 h light/dark) photoperiod; (B) DlGI promoter activity measured at different points of the day. The bars indicate the se of three biological replicates. The level of significant differences is indicated with an asterisk (*) and were assessed by t test (*P < 0.05, **P < 0.01, or ***P < 0.001).
Response of the DlGI promoter to different hormones. Leaves of Nicotiana benthamiana were infiltrated with the construct pDlGI::GUS were under the treatments of 8.6 μm auxin, 34.6 μm gibberellin, 100 μm methyl jasmonate, 75.7 μm abscisic acid, and water, as a control (CK). Bars indicate the se of three biological replicates. Letters represents a significant difference at the level of P < 0.05 using least significant difference.
DlELF4 increase the transcriptional activity of DlGI promoter: (A) pDlGI::GUS coinfiltrated with DlELF4-1 and DlELF4-2 separately in Nicotiana benthamiana leaves. (B) The firefly luciferase (LUC) and Renilla luciferase (REN) assay to study the interaction between DlGI promoter and DlELF4 genes. The bars indicate the se of three biological replicates. Letter represents a significant difference at the level of P < 0.05 using least significant difference statistical analysis.
Different factors including gibberellic acid (GA3), auxin (IAA), methyl jasmonate (MEJA), and abscisic acid (ABA) affect the activity of gigentea (GI) promoter.
Click on author name to view affiliation information
In fruit trees, flowering is a key event followed by fruit development and seed production. Gigentea (GI), a clock-associated gene, is known to contribute to photoperiodic flowering and circadian clock control in Arabidopsis thaliana. However, its functions in woody fruit trees remain unclear. In this study, a 2000 bp promoter fragment of the longan (Dimocarpous longan) DlGI gene was isolated from the genomic DNA of longan ‘Honghezi’ by polymerase chain reaction amplification. The DlGI promoter contained two main types of potential cis-acting elements: light-responsive and hormone-responsive elements. The promoter was fused with the β-glucuronidase (GUS) reporter gene of pBI121 to generate the pDlGI:GUS construct. GUS histochemical staining of transgenic A. thaliana revealed that DlGI might play a role in different developmental phases of longan. Exposure of transgenic A. thaliana to varying light intensities showed that the GUS activity increases with increased light intensity. Transient expression of pDlGI::GUS in Nicotiana benthamiana showed that the GUS activity was higher and reached peak a few hours earlier under short-day (SD) than long-day conditions. Exposure to different hormonal treatments revealed that the transcript level of GUS was activated by gibberellin (GA3) and indoleacetic acid (IAA) but suppressed by abscisic acid and methyl jasmonate treatment. In addition, N. benthamiana transient assay and dual-luciferase assay revealed that the presence of early flowering 4 (ELF4) homologs of longan (DlELF4-1 and DlELF4-2) significantly activated the DlGI promoter. The positive response of DlGI promoter to high light-intensity, SD photoperiod, GA3 and IAA signals, and DlELF4 transcription factor suggest that DlGI may function as a circadian clock and play a role in responding to SD conditions and other signals in flower initiation of longan.
Dimocarpus longan (longan) is an evergreen fruit tree that belongs to the Sapindaceae family and grows well in many tropical and subtropical regions of the world. China is top-ranked in terms of the plantation area and fruit production of longan (Matsumoto, 2006; Wu, 2008). Flowering is a crucial developmental phase of plant life, especially in the case of fruit trees, for which the transition of flowering is essential for the fruit set and ripening process (Peng et al., 2008). Generally, longan trees have a single spring flowering period. However, one cultivar of longan, Sijimi, originating in China (Guangxi province), has a perpetual flowering trait and has been successfully used to produce off-season fruit without any additional environmental stimuli (Jia et al., 2014). RNA sequencing results revealed that there were several flowering-time homologs in longan, including gigentea (GI) and early flowering 4 (ELF4), which could be involved in the perpetual flowering traits of ‘Sijimi’ (Jia et al., 2014). Moreover, our previous studies showed that the longan GI gene (DlGI) has a higher expression level during the floral bud physiological differentiation stage (Huang et al., 2017). DlGI is supposed to be involved in the regulation of physiological differentiation of floral buds, inducing floral initiation by affecting endogenous auxin synthesis and transport in longan (Huang et al., 2017).
Multiple flowering time pathways are controlling the vegetative-to-reproductive phase transition, and GI is proposed to have essential functions in promoting the photoperiodic flowering pathways (Mishra and Panigrahi, 2015). In Arabidopsis thaliana, GI plays a vital role in the regulation of flowering time by controlling the daily endogenous rhythms, also known as circadian rhythms (Dodd et al., 2005). GI was found to induce flowering in long-day (LD) photoperiods by regulating miR172 accumulation (Jung et al., 2007). Plants having a defect in maintaining circadian rhythm usually show altered flowering phenotypes (Imaizumi, 2010), and defects in circadian clock components have been shown to affect GI transcription (Mishra and Panigrahi, 2015). The overexpression of GI exhibited early flowering phenotypes (Mishra and Panigrahi, 2015). While mutation in GI delays flowering under LD conditions (Park et al., 1999), short circadian periods (Mizoguchi et al., 2005), long hypocotyls (Araki and Komeda, 1993), and sensitivity to low-temperature exposures (Cao et al., 2005).
Recent studies demonstrated that internal and external signals affect GI expression in A. thaliana (Berns et al., 2014). The transcription of the GI gene is activated by exposure to light and regulated by the circadian clock (Paltiel et al., 2006; Sawa and Kay, 2011). GI participates in flowering induction by extending photoperiods; therefore, GI is a key component in the mechanism leading to day-length discrimination (Sawa et al., 2007). Mutations within the components of the morning loop, such as CCA1 and LHY, result in GI transcription to occur earlier in the circadian cycle, suggesting that these proteins repress or delays the GI transcription when they are expressed in the morning (Mizoguchi et al., 2002). The CCA1 and LHY bind to several motifs that are closely associated in sequence, including “evening element” (EE) (Harmer and Kay, 2005; Mikkelsen and Thomashow, 2009). In addition to light regulation, EE is also found in close proximity of abscisic acid (ABA) ABREL motifs (Mikkelsen and Thomashow, 2009). Berns et al. (2014) suggested a composite effect of ABRELs and EE that contributes to generating the observed pattern of GI transcription in A. thaliana in the evening. ELF4, another essential component of circadian clock input pathways (McWatters et al., 2007). Overexpression of ELF4 shows delayed flowering signs in A. thaliana, while early flowering was observed in ELF4-deficient mutants (McWatters et al., 2007). In A. thaliana, ELF4 has been proposed to interact with GI to control the hypocotyl elongation and flowering time (Kim et al., 2012). ELF4 is epistatic to GI during hypocotyl elongation control, whereas GI is epistatic to ELF4 regarding flowering time regulation (Kim et al., 2012). A study in pea (Pisum sativum) supposed that Late bloomer 1 (LATE1) and Die neutralis (DNE), orthologs of ELF4 and GI, respectively, interact genetically to regulate flowering time (Liew et al., 2009).
Flowering induction in perennial fruit trees is much different from annual plants. Perennial plants sense the seasonal changes and initiate flowering. However, up to now, understanding of the functions of GI homologs in perennial fruit trees have been limited. On the basis of previous research, we speculate that DlGI may play a role in longan physiological differentiation of floral buds by sensing internal and external signals. It would be interesting to investigate the upstream regulatory factors of DlGI. Therefore, in this study, the promoter of DlGI gene was cloned, and bioinformatics analysis, A. thaliana stable transformation, Nicotiana benthamiana transient cotransformation, and dual luciferase assay were conducted. The results show that DlGI responded to photoperiod, light intensity, hormones, and DlELF4 genes. These findings will contribute to a better understanding of DlGI functions and the molecular mechanism of flowering induction in perennial fruit trees.
The seeds of N. benthamiana were grown in pots containing sterilized compost mix and maintained in a growth chamber. Young N. benthamiana leaves were used for transient expression. The A. thaliana ecotype Columbia (Col-0) seeds were directly grown in plug trays containing compost mix and maintained in a growth chamber at 22 to 25 °C, 50% to 70% humidity, and a 16-h photoperiod.
The DlGI promoter sequence was amplified (for primers, see Supplemental Table 1) from longan genomic DNA that had been extracted from leaf samples using the cetyltrimethylammonium bromide method (Turaki et al., 2017). The polymerase chain reaction (PCR) reaction for DlGI promoter was executed as per the following protocol: initial preheating for 30 s at 94 °C; 35 cycles of 94 °C for 30 s, 61 °C for 90 s, and 72 °C for 2 min; and a final elongation step of 72 °C for 5 min. The amplified fragment was cloned into the pMD-18-T vector, and positive colonies were selected at random and sent for sequencing to Biosune, Shanghai, China. PLACE (Higo et al., 1998) and PlantCARE (Lescot et al., 2002) databases were used for predicting the regulatory elements and potential core sequences of DlGI promoter.
The DlGI promoter was cloned into the plant expression vector pBI121, which contained GUS as a reporter gene, using an In-Fusion HD cloning kit (Clontech Laboratories, Beijing, China). The pBI-121 plasmid was digested with restriction endonuclease HindIII and EcoRI. The cut product was recovered by the gel electrophoresis and ligated with the DlGI promoter to obtain a recombinant plasmid pDlGI::GUS. The verified pDlGI::GUS construct was transformed into a Agrobacterium tumefaciens GV3101. The pBI121-DlELF4-1 and pCAMBIA2300-DlELF4-2 plasmids (Fu et al., 2018) were also transformed into A. tumefaciens GV3101.
The pDlGI::GUS transgenic A. thaliana lines were established using the floral dip method on the development of inflorescence stems, described by Zhang et al. (2006). The putative transformants were selected on Murashige and Skoog (MS) medium with (50 mg·L−1) kanamycin. Cetyltriethylammnonium was used to extract genomic DNA from the leaf tissues of transformed plants (Wang et al., 1996). PCR analysis was performed to verify the transgenic plantlets by using DlGI promoter specific-primers (F: 5′ CGCGACTAACATGTATCAATGGTC-3′; R: 5′-CCGCGTATAGGAAAACAAAGAGG-3′). Plants corresponding to the T3 generation were selected for further experiments.
GUS activity was visualized histochemically using the chromogenic substrate 5-bromo-4-chloro-3-indyle-β-D-glucuronide (X-Gluc) for staining, according to the method described by Jefferson et al. (1987). Transgenic A. thaliana plants were incubated in GUS-staining solution (2 mm K3Fe(CN)6, 1 mm X-Gluc, 2 mm K4Fe(CN)6, 100 mg·mL−1 chloramphenicol in 50 mm NaH2PO4 buffer (pH 7.0), 0.1% (v/v) Triton X-100; Sigma-Aldrich, St. Louis, MO), for 2 to 3 d at 37 °C. Plant material was then incubated in 100% ethanol for 4 to 5 h to remove chlorophyll from the plant material.
We selected six independent T2 transgenic lines (genetically unique) for each treatment. The T3 seeds (six lines of each antibiotic-resistant line) obtained were further selected on MS medium supplemented with 50 mg·L−1 kanamycin to obtain plants that were homozygous. Ten-day-old T3 resistant seedlings were transferred into the pots containing the compost mix for further expression analysis at lateral developmental stages and incubated them in the GUS substrate as previously described. GUS activity sites were observed by a stereo fluorescence microscope (SZX16; Olympus, Tokyo, Japan).
The 7-d-old T3 transgenic A. thaliana plants were kept in growth chambers for 3 d at 22 °C under different light intensities, including high light (1100 µmol·m−2·s−1), weak light (400 µmol·m−2·s−1), and dark conditions.
For photoperiod treatment, N. benthamiana plants leaves were infiltrated with A. tumefaciens harboring vector pDlGI::GUS and kept either in LD (16 h/8 h light-dark) conditions with the photoperiod starting at 0800 hr and ending at 0000 hr, or in SD (8 h/16 h light-dark) with the photoperiod starting at 0800 hr and ending at 1600 hr, for 72 h at 22 °C. To observe the effect of different hormones on the transcriptional activity of GUS, N. benthamiana leaves were infiltrated with A. tumefaciens harboring pDlGI::GUS construct and sprayed with hormone treatment [34.6 µm gibberellin (GA3), 8.6 µm IAA, 75.7 µm ABA, 100 µm methyl jasmonate (MeJA)] and sterile water for the control.
To investigate the regulation of DlELF4 by DlGI promoter, A. tumefaciens strains containing either of DlELF4 genes or pDlGI::GUS were separately cultured and diluted to measure the absorbance at a wavelength at A600 of 0.6 to 0.8. A. tumefaciens carrying either DlELF4-1 or DlELF4-2 were mixed separately with a culture carrying pDlGI::GUS at a 1:1 ratio, as described by Shamloul et al. (2014), and infiltrated in N. benthamiana leaves.
The RNA extraction for N. benthamiana and A. thaliana were carried out with the help of plant Total RNA Extraction Kit (Biotech Corp., Beijing, China). An ultraviolet spectrophotometer (ND-1000; Nanodrop Technologies, Wilmington, DE) was used to measure RNA concentrations. The cDNA synthesis was carried out via a SMARTer 50/30 kit (Clontech Laboratories). To analyze the expression level of the GUS gene under a series of different treatments, quantitative PCR (qPCR) was conducted using GUS gene specific–primers, and N. benthamiana actin was used as a reference gene (Supplemental Table 2) (David et al., 2006; Jefferson et al., 1987). The relative gene expression was evaluated by the 2−ΔΔCt method (Livak and Schmittgen, 2001).
The dual-luciferase assay was carried out according to the method described by Hellens et al. (2005) and Espley et al. (2007). Transient assays were performed on N. benthamiana leaves, and plants were kept in a growth chamber for 3 d. Leaf discs of 1 cm diameter were taken as samples for each treatment, and cell lysate was added into the grinded samples, followed by incubation on ice for 5 min. The firefly luciferase (LUC) and Renilla luciferase (REN) assays were accomplished with the help of a dual-luciferase reporter assay kit (Yeasen Biotech Co., Shanghai, China). Luciferase activity was determined with the help of a luciferase assay system, and the setting for fluorescence value measurement reading time was 1000 ms in the luminometer. The LUC to REN ratio was used to quantify promoter activity.
Data are shown as the means with standard errors of three independent biological replicates. The statistical analysis was carried out using a t test and one-way analysis of variance. Least significant difference values were calculated by P = 0.05 to compare the treatment means. In all graphs, the error bars indicate the standard deviation.
A 2000-bp 5′-flanking sequence upstream of the DlGI translation site (ATG) was obtained from longan genomic DNA, covering a previously cloned 1338 bp 5′-flanking sequence (GenBank: KT429718). Thereafter, the PlantCARE and PLACE databases were used to investigate the regulatory elements and potential core sequences of the DlGI promoter (Supplemental Table 1). Many putative cis-acting elements, including 24 CAAT boxes, which are responsible for the tissue-specific activity of the promoter, and 34 TATA boxes, required for critical and precise transcription initiation, were found dispersed over the entire promoter sequence. Two main types of cis-acting elements were found within the promoter sequence: light- and hormone-responsive regulatory elements (Fig. 1). The light-responsive elements include the box II, Box 4, G-box, G-Box, Box I, GAG-motif, GT1-motif, and EE. Whereas the hormone-responsive regulatory elements include the auxin-responsive element (TGA-box), the gibberellin-responsive element (GARE-motif), the MeJA-responsive element (CGTCA-motif) and the abscisic acid-responsive element (ABRE) (Fig. 1). Stress-related elements were also found in the promoter region, for example, LTR, a cis-acting element involved in low-temperature responsiveness.
Citation: Journal of the American Society for Horticultural Science J. Amer. Soc. Hort. Sci. 145, 6; 10.21273/JASHS04946-20
To examine the activity of DlGI promoter in T3 transgenic A. thaliana, the 2000-bp promoter fragment driven by the GUS reporter gene was monitored using histochemical staining during different developmental stages of plant growth. Six-day-old transgenic A. thaliana seedlings grown on MS medium were strongly stained, and the whole plantlet showed the GUS activity except roots (Fig. 2A). Similarly, GUS activity was observed in 14-d-old plants (Fig. 2B). GUS activity was also detected in the flower buds of 21-d-old transgenic A. thaliana (Fig. 2C). Strong GUS staining was observed in the sepals, petals, and stigma of opened flowers of transgenic A. thaliana, but no GUS staining was observed in the anthers (Fig. 2D). GUS staining was also found in the siliques of 40-d-old transgenic A. thaliana (Fig. 2E). As expected, GUS staining was not observed in wild-type A. thaliana plants (Fig. 2F).
Citation: Journal of the American Society for Horticultural Science J. Amer. Soc. Hort. Sci. 145, 6; 10.21273/JASHS04946-20
Bioinformatical analysis of the DlGI promoter sequence revealed that it contains many motifs that are well characterized for their response to light (Table 1), suggesting that different intensities of light might affect the activity of the DlGI promoter. To validate this hypothesis, pDlGI::GUS activity was measured under different light intensities. Strong GUS activity was detected in seedlings grown under high light intensity, except for the root apex, which did not show any activity (Fig. 3A). Seedlings from reduced light intensity also showed GUS activity in cotyledons and the hypocotyl, but the activity was weaker compared with plants grown in high light intensity (Fig. 3B). Reduced GUS activity appeared in the seedlings grown in dark conditions (Fig. 3C). Wild-type A. thaliana was used as a negative control and did not show any GUS activity when grown under a high light intensity (Fig. 3D).
Citation: Journal of the American Society for Horticultural Science J. Amer. Soc. Hort. Sci. 145, 6; 10.21273/JASHS04946-20
GI is considered to be an essential clock gene due to its significance in circadian clock regulation. The transient assays were performed to detect the pDlGI::GUS expression under different photoperiods. The results showed that GUS expression was relatively higher in SD conditions compared with LD photoperiods (Fig. 4A). GUS expression was also measured at different points of the day under both LD and SD photoperiod systems. We found that pDlGI::GUS expression peaks at 1200 hr in SD conditions, whereas in LD, the highest expression was observed ≈1600 hr. Under the LD photoperiod system, GUS expression was delayed by a few hours (i.e., the peak was found at a later time point) compared with the SD photoperiod system. Under both photoperiods, the expression levels were lowest in the morning but increased during the day (before dusk) to a peak value and then decreased to a minimum during the dark period (Fig. 4B). These results depict that variation in daily rhythms and photoperiods has definite effects on the expression of DlGI.
Citation: Journal of the American Society for Horticultural Science J. Amer. Soc. Hort. Sci. 145, 6; 10.21273/JASHS04946-20
Inspection of the DlGI promoter sequence revealed that it contains a few cis-acting elements that are responsive to hormones, such as gibberellin (GARE-motif), auxin (TGA-box), ABA (ABRE), and MeJA (CGTCA-motif). We performed qPCR to detect the pDlGI::GUS expression under different hormonal treatments. GA3 showed the highest GUS expression compared with other hormone treatments, with a 3-fold higher change than the control. GUS expression was also induced by IAA, with an up to 1.5-fold higher change compared with the control (Fig. 5). In contrast, GUS expression was repressed by ABA and MeJA (Fig. 4).
Citation: Journal of the American Society for Horticultural Science J. Amer. Soc. Hort. Sci. 145, 6; 10.21273/JASHS04946-20
The ELF4 gene is known to be involved in many of the same physiological processes as GI. We examined whether there is any direct interaction between DlELF4 and DlGI. Two DlELF4 genes, DlELF4-1 and DlELF4-2, were injected with pDlGI::GUS, respectively, and GUS expression was assayed by qPCR. Results showed that both DlELF4 genes significantly increased the transcriptional activity of the DlGI promoter, especially DlELF4-2 (Fig. 6A). Furthermore, a dual luciferase assay was performed to verify the interaction between DlGI promoter and DlELF4. The results were consistent with the GUS activity assay, showing that both DlELF4 genes increase DlGI promoter activity: DlElF4-2 induced 5-fold higher activity and DlElF4-1 induced 3-fold higher luciferase activity than the control (Fig. 6B).
Citation: Journal of the American Society for Horticultural Science J. Amer. Soc. Hort. Sci. 145, 6; 10.21273/JASHS04946-20
The plant clock uses daily rhythms and recognizes photoperiods in various developmental processes, from seedling growth to flowering (Kim et al., 2012). The expression profile of the DlGI promoter in transgenic A. thaliana suggested that DlGI might play a broader role in different organs and developmental phases of plant growth in longan. A similar kind of GI expression pattern has been observed in A. thaliana (David et al., 2006), Oryza sativa (Hayama et al., 2002), and Glycine max (Li et al., 2013), suggesting DlGI acts as a circadian clock in longan. However, in contrast to GI in A. thaliana, DlGI has a higher expression level in SD conditions, and the expression level reaches its peak earlier in the day under SD conditions compared with LD, suggesting that the expression levels of DlGI are greatly affected by SD-length. Huang et al. (2017) found that DlGI might be involved in the regulation of the physiological differentiation of floral buds. The initiation of flowering in longan occurs during autumn in China, when the photoperiod becomes shorter; therefore, the promoter of DlGI has higher activity under SD conditions, suggesting that DlGI may play a role in responding to SD conditions and inducing flower initiation in longan. Variation in GUS activity under different photoperiods indicates that the conserved motif presented upstream of DlGI may contribute to maximizing the expression at a precise time of the day, thereby exerting control over flowering time. The EE motif present in the promoter represses the transcription of GI during the morning and increases the amplitude of expression in the evening (Berns et al., 2014). The ABRELs are predicted to be recognized by bZIP transcription factors, several of which are implicated in light induction of gene transcription (Chattopadhyay et al., 1998). Thus, the contribution of such elements provides a clue in controlling the distinct subsets of a gene by function, possibly due to differing requirements at different times in the day–night cycle.
Light is a well-known environmental factor that plays a substantial role in controlling flowering time in different plants through applied photoperiod, which depends upon the light quality (wavelength) and intensity (Cerdán and Chory, 2003; Lu et al., 2017). Early flowering was observed in 11 herbaceous plant species when they were subjected for 18 h daily to high-intensity lamps, in contrast to plants grown in an ambient environment (Erwin and Warner, 2000). Light quality has a significant role in flower bud initiation and the transition from vegetative to reproductive state in many fruit crops (Wilkie et al., 2008). Several studies in A. thaliana have revealed that the light quantity and quality affect the GI transcription (Fowler et al., 1999; Mishra and Panigrahi, 2015). Our results showed that pDlGI::GUS activity was predominantly induced by high-intensity light, compared with low light and dark conditions (Fig. 3A–C), suggesting the expression level of DlGI is upregulated by high light intensity. Recent research on fruit crop flowering has indicated that clock genes involved in flower initiation may generally be mediated by both light quality and quantity (Jackson et al., 2011). Flower bud formation in Malus ×domestica is affected by different intensities of light, as fewer flower buds were formed when the light intensity was reduced for 7 weeks (Tromp, 1983). However, the relationship between light intensity and flowering remains unknown in longan and needs further investigation. Light-responsive cis-elements upstream of DlGI may play crucial roles in the response of GUS gene expression in the transgenic seedlings. Studies related to promoters of light-regulated genes have revealed the presence of different LREs, such as GATA, GT1, and G-box, which play vital roles in light-regulated transcription activity (Yadav et al., 2002). The two most common light-responsive cis-elements in the promoter sequence of DlGI were the G-box, also found in the promoter of photosynthesis-related genes involved in light regulation (López-Ochoa et al., 2007), and the GT-1 motif, which is involved in the regulation of light responses and is usually found clustered with I-box (Escobar et al., 2004).
Some reports have suggested that auxin plays an essential role in the circadian rhythm of flowering stem growth (Jouve et al., 1999). This is demonstrated by the fact that auxin response is linked to processes controlled by the circadian clock and that some circadian clock functions can be influenced by hormone signals (Covington and Harmer, 2007). In this study, GA3 and IAA were found to be upregulated and enhanced the transcriptional activity of GUS. The gibberellin class of phytohormones is involved in the control of flowering in many species (Hedden, 2017). GA3 is linked with other floral pathways via GA3-regulated DELLA proteins, creating versatile interacting modules for signaling proteins (Conti, 2017). In longan, the concentration of some hormones, such as IAA, was increased during the floral bud physiological differentiation stage (Yimei, 1997). DlGI expression was also found to be upregulated at the same stage (Huang et al., 2017). The time of change in the concentration of hormones and the change in expression of DlGI coincides with the time when terminal shoots begin flower bud initiation in longan. At the same time, DlGI was predicted to induce floral initiation by affecting endogenous auxin synthesis and transport in longan (Huang et al., 2017); therefore, there may be an IAA-DlGI regulatory loop in longan during flowering induction. However, this hypothesis requires more concrete evidence. Recent studies show that in A. thaliana, exogenous application of ABA has exhibited late-flowering phenotype (Achard et al., 2004), also found to participate in circadian clock function and flowering time control (Hanano et al., 2006). Early-flowering phenotypes have been observed in ABA-deficient mutants, showing that ABA acts as an inhibitor of floral transition (Barrero et al., 2005). MeJA was found to inhibit the flowering of short-day plants such as Chenopodium rubrum (Albrechtova and Ullmann, 1994). Pharbitis nil seedlings placed in a solution of MeJA for a period of 24 h led to a dramatic reduction in the number of formed flower buds (Maciejewska and Kopcewicz, 2002). In this study, abscisic acid and methyl jasmonate were found to reduce the transcriptional activity of the GUS, indicating that they might be inhibitors of flowering in longan.
GI and ELF4 have a similar circadian phase, and they also interact genetically, resulting in identical circadian outcomes in plant growth and flowering time (Kim et al., 2012). Kolmos et al. (2009) found that ELF4 suppresses GI transcription during circadian clock regulation in A. thaliana. Jia et al. (2014) demonstrated that the expression level of DlELF4 (UniGene 4309) was higher in ‘Sijimi’ compared with ‘Lidongben,’ which suggests that ELF4 may be an essential gene involved in perpetual flowering traits in longan. The data presented in this study demonstrate that DlELF4 genes significantly increase the activity of the DlGI promoter (Fig. 6A and B), suggesting that DlELF4 works upstream of DlGI and promotes DlGI expression (Jia et al., 2014). DlELF4 seems to have different functions from ELF4 in A. thaliana, which is known to destabilize GI protein and be a repressor of flowering. It was reported by Fu et al. (2018) that the overexpression of longan DlELF4 genes in A. thaliana exhibited late flowering and adventitious root phenotypes, suggesting that DlELF4 genes, and in particular DlELF-2, may regulate hormone biosynthesis. Therefore, DlELF4 may regulate hormone biosynthesis by upregulating DlGI and could be involved in longan flower initiation. However, this hypothesis needs further study. Taken together, the DlGI–DlELF4 interaction provides an additional novel mechanism that can contribute to the currently known genetic and molecular interactions which control photoperiodic flowering.
In this study, the GI homolog gene was identified as a circadian rhythms gene in the woody fruit tree for the first time. Longan DlGI responds to external changes in photoperiod and light intensity, activated by SD conditions and high light intensity. In addition, DlGI responds to hormones such as GA3, and IAA, and there may be an IAA–DlGI regulatory loop during flowering induction in longan. Moreover, DlELF4 was found to be working upstream of DlGI and activating DlGI expression. The DlELF4–DlGI interaction may provide an additional novel mechanism to the currently known genetic and molecular interactions that control photoperiodic flowering. Our findings not only increase our understanding regarding the upstream regulatory network of DlGI (Supplemental Fig. 1) but also provide more knowledge of molecular mechanisms of flowering induction in woody fruit trees.
Analysis of the DlGI promoter sequence using the PlantCARE database (Lescot et al., 2002).
Histochemical localization of GUS activity in transgenic Arabidopsis thaliana plants carrying the pDlGI::GUS fusion construct: (A) GUS staining localized in the cotyledons of 7-d-old seedlings; (B) 15-d-old plant, showing strong GUS expression in the cotyledons, hypocotyls, and leaves; (C) GUS staining observed in 21-d-old flower bud; (D) GUS staining localized in flowers (sepals, petals, and stigma) of a 26- to 30-d-old plant; (E) 40-d-old plant siliques showing GUS expression; and (F) no expression in wild type A. thaliana.
GUS expression in transgenic Arabidopsis thaliana seedlings under different light conditions: (A) high light (1100 µmol·m−2·s−1); (B) weak light (400 µmol·m−2·s−1); (C) dark conditions; (D) GUS staining in wild type A. thaliana (negative control); (E) level of GUS transcript in cotyledon, hypocotyl, and root under different intensity of light. Bars indicate the se.
Effect of different photoperiod on the activity of DlGI promoter infiltrated in Nicotiana benthamiana leaves: (A) DlGI promoter activity under long day (16/8 h light/dark) and short day (8/16 h light/dark) photoperiod; (B) DlGI promoter activity measured at different points of the day. The bars indicate the se of three biological replicates. The level of significant differences is indicated with an asterisk (*) and were assessed by t test (*P < 0.05, **P < 0.01, or ***P < 0.001).
Response of the DlGI promoter to different hormones. Leaves of Nicotiana benthamiana were infiltrated with the construct pDlGI::GUS were under the treatments of 8.6 μm auxin, 34.6 μm gibberellin, 100 μm methyl jasmonate, 75.7 μm abscisic acid, and water, as a control (CK). Bars indicate the se of three biological replicates. Letters represents a significant difference at the level of P < 0.05 using least significant difference.
DlELF4 increase the transcriptional activity of DlGI promoter: (A) pDlGI::GUS coinfiltrated with DlELF4-1 and DlELF4-2 separately in Nicotiana benthamiana leaves. (B) The firefly luciferase (LUC) and Renilla luciferase (REN) assay to study the interaction between DlGI promoter and DlELF4 genes. The bars indicate the se of three biological replicates. Letter represents a significant difference at the level of P < 0.05 using least significant difference statistical analysis.
Different factors including gibberellic acid (GA3), auxin (IAA), methyl jasmonate (MEJA), and abscisic acid (ABA) affect the activity of gigentea (GI) promoter.
Contributor Notes
This work was supported by The Construction of Plateau Discipline of Fujian Province (102/71201801101), China.
L.Z. is the corresponding author. E-mail: lhzeng@fafu.edu.cn or lhzeng@hotmail.com.
Analysis of the DlGI promoter sequence using the PlantCARE database (Lescot et al., 2002).
Histochemical localization of GUS activity in transgenic Arabidopsis thaliana plants carrying the pDlGI::GUS fusion construct: (A) GUS staining localized in the cotyledons of 7-d-old seedlings; (B) 15-d-old plant, showing strong GUS expression in the cotyledons, hypocotyls, and leaves; (C) GUS staining observed in 21-d-old flower bud; (D) GUS staining localized in flowers (sepals, petals, and stigma) of a 26- to 30-d-old plant; (E) 40-d-old plant siliques showing GUS expression; and (F) no expression in wild type A. thaliana.
GUS expression in transgenic Arabidopsis thaliana seedlings under different light conditions: (A) high light (1100 µmol·m−2·s−1); (B) weak light (400 µmol·m−2·s−1); (C) dark conditions; (D) GUS staining in wild type A. thaliana (negative control); (E) level of GUS transcript in cotyledon, hypocotyl, and root under different intensity of light. Bars indicate the se.
Effect of different photoperiod on the activity of DlGI promoter infiltrated in Nicotiana benthamiana leaves: (A) DlGI promoter activity under long day (16/8 h light/dark) and short day (8/16 h light/dark) photoperiod; (B) DlGI promoter activity measured at different points of the day. The bars indicate the se of three biological replicates. The level of significant differences is indicated with an asterisk (*) and were assessed by t test (*P < 0.05, **P < 0.01, or ***P < 0.001).
Response of the DlGI promoter to different hormones. Leaves of Nicotiana benthamiana were infiltrated with the construct pDlGI::GUS were under the treatments of 8.6 μm auxin, 34.6 μm gibberellin, 100 μm methyl jasmonate, 75.7 μm abscisic acid, and water, as a control (CK). Bars indicate the se of three biological replicates. Letters represents a significant difference at the level of P < 0.05 using least significant difference.
DlELF4 increase the transcriptional activity of DlGI promoter: (A) pDlGI::GUS coinfiltrated with DlELF4-1 and DlELF4-2 separately in Nicotiana benthamiana leaves. (B) The firefly luciferase (LUC) and Renilla luciferase (REN) assay to study the interaction between DlGI promoter and DlELF4 genes. The bars indicate the se of three biological replicates. Letter represents a significant difference at the level of P < 0.05 using least significant difference statistical analysis.
Different factors including gibberellic acid (GA3), auxin (IAA), methyl jasmonate (MEJA), and abscisic acid (ABA) affect the activity of gigentea (GI) promoter.
