Plant materials and growth conditions
The seeds of the Antirrhinum majus cultivar ‘Maryland True Pink’ ((Pan American Seed Co., West Chicago, IL, United States) were planted in a growth chamber at the Beijing University of Agriculture under the following conditions: 22±2℃ and 12 h/12 h light/dark photoperiod at 200±50 μmol·m-2·s-1 with a relative humidity of 70%. Nicotiana benthamiana plants in this study were grown under the following conditions: 24ºC with a photoperiod of 16 h/8 h (day/night).
The different developmental stages of flowers included S1 (full coloration stage, flower buds), S2 (initial blooming stage, from 2 to 4 days before flower opening), S3 (half blooming stage, consisting of flowers from 1 day before anthesis), and S4 (full blooming stage, consisting of 6 to 8-day-old flowers) (Figure S.1), and the experiments were conducted when the snapdragon plants reached S4. Light treatments were conducted with (LED light 200±50 μmol·m-2·s-1 ) blue light (BL) and polychromatic light without blue light (EB), and flowers under constant darkness (DK) were used as a control. The flowers were sampled after 1 h and 6 h of light exposure. Three to five snapdragon plants with the same growth status were sampled in each treatment; samples comprised all petals from each plant in each treatment. Measurements of each sample were taken three times.
Virus-induced gene silencing (VIGS) and transient overexpression
To suppress AmPIF4, a 300-bp fragment of the AmPIF4 gene was inserted into the pNC-TRV2 vector at the SifI restriction site to make the pTRV2-AmPIF4 constructs (Tab S.1). The final expression vectors, pNC-TRV1 and pNC-TRV2, were transferred into Agrobacterium tumefaciens strain GV3101 using the freezing/heat shock method [72, 73]. We collected overnight Agrobacterium cultures by centrifugation at 6,000 g for 15 min and resuspended pellets in infiltration buffer to a final O.D. 600 = 1.0; cell suspensions containing pTRV2-AmPIF4 or pNC-TRV2 and pNC-TRV1 were infiltrated into the petals of D3. After infiltration, the flowers were kept under dark conditions for one day and then maintained in a 12 h/12 h light/dark cycle at 16℃ for 4 to 5 days. Total floral volatile compounds were collected and analyzed at the full-bloom stage using gas chromatography–mass spectrometry (GC–MS). A total of three to five biological replicates were performed.
To generate the AmPIF4 overexpression vector, the full-length coding sequence of AmPIF4 was amplified by PCR from the genomic cDNA, inserted into the pNC-Cam1304-35S cloning vector, and then transformed into GV3101 [74]. The empty vector was used as a control.
Yeast one-hybrid assay
A yeast one-hybrid assay was performed [75] to examine the binding of AmPIF4 protein to the promoters of TPS and downstream structural genes of the MEP pathway (AmOCS, AmMYS, AmDXS, and AmDXR). The full-length AmPIF4 was fused to the pGAL4 activation vector (pGADT7-AD) to generate AD-PIF4 as prey. Empty pGADT7 vector was used as a negative control. A 2,000-bp genomic fragment upstream of the OCS, MYS, DXS, and DXR start codons was PCR-amplified from genomic DNA and then cloned into the 007VS vector. The fragments were then recombined into pAbai to generate the OCSpro-pAbai, MYSpro-pAbai, DXSpro-pAbai, and DXRpro-pAbai expression vectors as baits. The pAbAi vectors harboring the constructs were linearized by digestion with BstbI and integrated into the genome of the Y1H Gold strain. AD-PIF4 or empty vector was subsequently transformed into the yeast strain containing the pAbAi constructs. Yeast cells were grown for 48–96 h on SD/-Leu/-Ura selection plates and then transferred to the selection plates containing aureobasidin A (AbA) at indicated concentrations.
Dual-luciferase (Dual-LUC) assay
We performed transient transcription dual-LUC assays using N. benthamiana plants following previously described methods [76]. This assay was performed to further test the activation of AmOCS, AmMYS, AmDXS, and AmDXR. We cloned the full-length coding sequence of AmPIF4 into pNC-Cam1304-35S, which generated the effector plasmid 35Spro-PIF4, and free 35S vector was used as the control. The promoter sequences were inserted into the pNC-Green-Luc vector (OCSpro-LUC, MYSpro-LUC, DXSpro-LUC, and DXRpro-LUC) as reporter constructs to drive the firefly LUC reporter gene (Fig. 3B); they were then co-transformed with 35Spro:PIF4 or free 35S vector into N. benthamiana leaves using an Agrobacterium-mediated method. Nicotiana benthamiana, which was used for the dual-luciferase reporter assay, was grown in a greenhouse for 3 days after infiltration and exposed to dark conditions for 24 h. The Renilla luciferase gene, driven by the cauliflower mosaic virus 35S promoter, was used as an internal control. The ratio of LUC/REN was calculated as an indicator of the final transcriptional activity. Three biological replicates were conducted for each combination.
Yeast two-hybrid assay
For the yeast two-hybrid assay, the cDNA of AmCRY1, AmCRY2, and AmCOP1 was cloned into pGADT7, and the cDNA of AmPIF4 was cloned into pGBKT7 [77]. The primers used are shown in Tab S.1. Y2H Gold cells containing both BD-AmPIF4 and AD-AmCRY1/AD-AmCRY2/AD-AmCOP1 were placed on SD/-Leu/-Trp medium for 3 days at 28℃. The transformed yeast cells were incubated on SD/-Ade/-His/-Leu/-Trp medium for another 3 days to test for protein interactions. The empty vectors pGADT7 and pGBKT7 were used as empty controls. The activation activity was confirmed via yeast growth on the aforementioned plates. Each yeast colony expressing MEL1 turned blue with the addition of X-α-Gal substrate (Clontech, TaKaRa).
Bimolecular fluorescence complementation (BiFC)
The full-length coding sequence of AmPIF4 was amplified and cloned into pNC-BiFC-Enn. The full-length coding sequences of AmCRY1, AmCRY2, and AmCOP1 were amplified and cloned into pNC-BiFC-Enc. The recombinant constructs were transformed into GV3101. We collected overnight Agrobacterium cultures by centrifugation at 6,000 g for 5 min and resuspended the cells in infiltration buffer (10 mM MgCl2, 10 mM MES, 200 mM AS, pH 5.8) to a final O.D. 600 = 0.6-0.8. The suspensions were mixed and co-infiltrated into the leaves of N. benthamiana. Yellow fluorescent protein signals were imaged after 3 to 5 days of cultivation by confocal microscopy (LSM 880 Zeiss, Germany). The empty vector pNC-BiFC-Enn was used as the negative control [78].
Luciferase complementation imaging (LCI)
The Firefly LCI assay was performed following the methods of [79]. We fused AmPIF4 or AmCRY1/CRY2/COP1 into the vector pCAMBIA1300-nLUC or pCAMBIA1300-cLUC. All vectors were transformed into Agrobacterium strain GV3101 and subsequently transiently cotransformed into N. benthamiana leaves. After 3 days, the abaxial side of the infiltrated leaves was sprayed with 1 mM luciferin and then kept in the dark for 10 min. The LUC activity was monitored with a Tanon-5200 system. The empty cLUC and nLUC vectors were used as the controls. The primers used in this experiment are shown in the Supplementary Table.
GC–MS analysis of floral volatiles
The petal volatile components of snapdragon and their content were assayed qualitatively and quantitatively by headspace solid-phase microextraction and GC–MS. Floral scent compounds were trapped and desorbed followed [40]. The volatiles were examined qualitatively by comparing the mass spectra and retention time with the NIST 2008 mass spectra library and standard samples. Total ion chromatograms were analyzed to detect flower volatiles. The volatiles were quantitatively analyzed using the internal standard and peak area normalization methods. When collecting the volatile components of samples, 10 ng·mL−1 3-octanol was added as the internal standard. The relative content of the target volatiles was calculated by comparing the peak area of volatile terpenoids with those of internal standards. Three to five replications were performed in each group.
RNA isolation, cDNA synthesis, and qRT-PCR
After collecting plant tissues, total RNA was extracted using a TransZol Up Plus RNA kit (TransGen Biotech ER501-01). First-strand cDNA was synthesized from 1 μg of total RNA using the TransScript® Uni All-In-One First-Strand cDNA Synthesis SuperMix for qPCR (TransGen Biotech AU341-02). The PerfectStart® Uni RT&qPCR Kit (TransGen Biotech AUQ-01) was used for qRT-PCR, which was performed on a CFX384 Real-Time System (BIO-RAD). The expression of the target genes was normalized to the expression of the AmUBIgene [80]. Relative gene expression was calculated using the 2 -△△CT formula [81]. Primers used for analysis are shown in Tab S.1. Three biological replicates were performed, and error bars indicate the SD of the three biological replicates.
Statistical analysis
Statistical analyses were performed in Microsoft Excel, SPSS 19.0, and GraphPad Prism version 6.0. Different letters (a, b, c) in the figures indicate significant differences determined by a test for multiple comparisons following ANOVA (P < 0.05). T-tests were conducted using GraphPad Prism; stars indicate the level of significance (*0.01< p< 0.05, and **p< 0.01). All experiments were conducted in triplicate.
Accession numbers
Sequence data in this article can be found in the snapdragon reference genome http://bioinfo.sibs.ac.cn/Am/ under the following accession numbers: AmPIF4 (Am03g17770), AmCRY1 (Am01g07180), AmCRY2 (Am01g60020), and AmCOP1 (Am01g50610).