Cloning and sequence alignment of the full-length cDNA of P sDFR in the red and pink petals
Based on the unigene sequence of PsDFR obtained from transcriptome sequencing of P. suffruticosa ‘Shima Nishiki’, the full-length cDNA of its red and pink petals was cloned and sequenced, respectively. The results of electrophoresis showed that the target fragments with single band and the same size were obtained from the petals of these two colors. Furthermore, it was found that there is no difference in the full-length cDNA sequence of this gene between the red and pink petals, and both of their lengths are 1095 bp (Fig. 2).
Cloning and sequence alignment of the genomic DNA of PsDFR in the red and pink petals
In view of the above obtained full-length cDNA sequence of PsDFR, the genomic DNA of the red and pink petals in P. suffruticosa ‘Shima Nishiki’ was cloned and sequenced. Considering the possible complexity of introns in genomic DNA, it is relatively difficult to clone the full-length genomic DNA at one time. Here, the amplification of this gene was performed by four times. The results of electrophoresis showed that the target fragments with a single band and same size were obtained from the petals of these two colors (Fig. 3a). Sequencing results showed that the genomic DNA sequences of the red and pink petals were identical. According to statistics, the full-length genomic DNA of PsDFR was 2896 bp, including 6 exons and 5 introns. Among them, the third intron sequence was relatively complex, including two TATA repeats (Fig. 3b).
Cloning and sequence alignment of the promoter region of PsDFR in the red and pink petals
Based on these cDNA and genomic DNA sequences of PsDFR, firstly, the promoter region of PsDFR in the red petals was cloned by chromosome walking method. In this study, the total fragment length of promoter region of PsDFR obtained was 4748 bp, which was amplified by five times. The length of these five fragments was 1698 bp, 743bp, 1434 bp, 951 bp and 421 bp, respectively (Fig. 4a).
By referring to the promoter sequence of PsDFR obtained from the red petals in the previous step, the promoter region of PsDFR in the pink petals was also cloned and verified by the method of segmented amplification. The length of the fragments amplified was 875 bp, 1493 bp, 1489 bp and 1336 bp, respectively (Fig. 4b). Sequencing results showed that the promoter region sequences of PsDFR in the red and pink petals were all the same (Fig. 4c).
Analysis of cis-acting elements in promoter region of PsDFR
In view of the identified promoter sequence of PsDFR, the cis-acting elements in the promoter sequence with 2000 bp of the start codon (ATG) upstream of PsDFR were analyzed by using plantCARE online database. The results showed that many elements were predicted in promoter region of PsDFR, including light response element, abscisic acid response element, gibberellin response element, etc (Table 1). Among them, there were three MYB-binding sites at these positions of -405 bp, -503 bp and − 1758 bp of the start codon upstream, respectively (Table 1 and Fig. 5).
Table 1
Cis-acting elements in promoter region of PsDFR.
Element name | Position from ATG | Sequence (5’-3’) | Function |
A-BOX | -553 | CCGTCC | cis-acting regulatory element |
ABRE | -151, -180, -481, etc | ACGTG | cis-acting element involved in the abscisic acid responsiveness |
ARE | -1749/complementary strand | AAACCA | cis-acting regulatory element essential for the anaerobic induction |
BOX-4 | -254/complementary strand | ATTAAT | part of a conserved DNA module involved in light responsiveness |
CAAT-box | -119/complementary strand,-653, etc | CCAAT | common cis-acting element in promoter and enhancer regions |
CAT-box | -1969 | GCCACT | cis-acting regulatory element related to meristem expression |
CGTCA-motif | -562/complementary strand,-978 | CGTCA | cis-acting regulatory element involved in the MeJA-responsiveness |
G-Box | -124/complementary strand, etc | CACGTG | cis-acting regulatory element involved in light responsiveness |
GT1-motif | -19,-361, etc | GGTTAA | light responsive element |
MBS | -1758 | CAACTG | MYB binding site |
MYB-like | -405,-503, etc | TAACCA | MYB binding site |
TATA-box | -80/complementary strand,-255, etc | ATTATA | core promoter element |
TATC-box | -1803/complementary strand | TATCCCA | cis-acting element involved in gibberellin-responsiveness |
TCCC-motif | -341 | TCTCCCT | part of a light responsive element |
Determination of methylation level of CpG island and promoter region of PsDFR in the red and pink petals
Based on the cloning and sequence alignment of the PsDFR gene in the red and pink petals, it was preliminarily concluded that the double-color formation of P. suffruticosa ‘Shima Nishiki’ should not be caused by transposon insertion into PsDFR or base deletion of PsDFR. In addition, previous studies also have shown that the methylation level of key transcriptional regulatory regions of structural genes related to anthocyanin biosynthesis may affect the formation and difference of flower, leaf and fruit color, especially in CpG island (the GC content of this region is more than 50%, which is usually distributed in the promoter and the first exon region) [31–33]. If the methylation level of some key regions of promoter or CpG island is very high, the corresponding gene expression can be greatly reduced. Sometimes, it may be completely inhibited.
In this study, we first predicted the location of CpG island in the promoter and first exon region of PsDFR by using MethPrimer online software. The results showed that there is a 105 bp CpG island (− 9 bp to 96 bp) near the start codon (ATG) (Fig. 6a).
Firstly, in order to verify whether the color difference of double-color flowers is determined by the differential methylation level of CpG island of PsDFR, the methylation level of this position of PsDFR in the red and pink petals was detected. The results showed that C-base of all samples sent for sequencing in these petals of two colors changed to T-base after sulfite treatment. Therefore, it is concluded that the methylation of CpG island of PsDFR did not occur in the red or pink petals (Fig. 6b).
In addition, based on the prediction analysis of cis-acting elements in the promoter region of PsDFR, the methylation level of the key transcription regulatory region and the region containing MYB-binding sites of the start codon (ATG) upstream (-10 bp to -822 bp and − 1732 bp to -2062 bp) was further determined. The results showed that methylation of many sites (CG/CHG/CHH) occurred in these regions of the red and pink petals, and there were a certain degree of differences in some sites. However, on the whole, there was no big difference and obvious regularity for their methylation levels of the red and pink petals (Fig. 6c).
Expression analysis of PsMYB114L and PsMYB12L in different tissues
In order to understand the tissue specificity of the PsMYB114L and PsMYB12L genes, the expression level of these two genes in seven different tissues (root, stem, leaf, flower, sepal, stamen and pistil) in P. suffruticosa ‘Shima Nishiki’ was quantitatively analyzed. The qRT-PCR results showed that the expression level of PsMYB114L was the highest in leaves, followed by roots, flowers, sepals and pistils, and the lowest in stems and stamens. The expression level of PsMYB12L was relatively high in sepals, leaves and flowers, followed by roots and stems, and relatively lower in pistils and stamens (Fig. 7).
Y1H validation of PsMYB114L and PsMYB12L binding to PsDFR promoter
Firstly, the pGADT7 empty vector, the full-length coding sequence (CDS) of PsMYB114L/PsMYB12L (restriction sites: Sac I and Mlu I), the pHIS2 empty vector and the promoter fragment of PsDFR (restriction sites: Nde I and Xho I) were successfully amplified using the corresponding primers (Supplemental Table S1) and double-digested (Fig. 8a), and then recombined, respectively. Subsequently, these three recombinant vectors of pGADT7-PsMYB114L, pGADT7-PsMYB12L and pHIS2-PsDFR (Fig. 8b) were verified successfully by PCR and sequencing.
In addition, the results of co-transformation showed all Y187 strains of the treatment group I (pGADT7-PsMYB114L, pHIS2-PsDFR), treatment group II (pGADT7-PsMYB12L, pHIS2-PsDFR), control group III (pGADT7, pHIS2-PsDFR), negative control group IV (pGADT7, pHIS2) and positive control group V (pGADT7-Rec2-53, pHIS2-p53) could grow normally on the medium without Leu-Trp (SD/-Leu/-Trp), respectively (Fig. 8c).
Finally, the results on the medium without His-Leu-Trp (SD/-His /-Leu /-Trp) showed that under the concentration of 0 mM and 50 mM 3-AT, all the other groups grew well except for the negative control group IV. Furthermore, under the concentration of 100 mM and 200 mM 3-AT, the results suggested that in addition to the normal growth of the positive control group V, the treatment group II (pGADT7-PsMYB12L and pHIS2-PsDFR) also showed good growth status, while the treatment group I (pGADT7-PsMYB114L and pHIS2-PsDFR) and the other control groups did not grow (Fig. 8c). As for PsMYB114L and PsMYB12L, it was preliminarily concluded that only PsMYB12L can be combined with the promoter of PsDFR.
Double luciferase report validation of PsMYB114L and PsMYB12L on PsDFR promoter activation
Based on the method of seamless cloning, the full length coding sequence (CDS) of PsMYB114L and PsMYB12L (BamH I and EcoR I) and the promoter fragment of PsDFR (Sal I and Hind III) with restriction sites were obtained with the special primers (Additional file 1: Table S1) using the plasmids of recombinant vector constructed by Y1H assay as templates. At the same time, the pGreenII62-SK empty vector and pGreenII0800-LUC empty vector were double-digested with special enzymes (Fig. 9a), and then recombined with the corresponding candidate genes, respectively. Subsequently, these three recombinant vectors of pGreenII62-SK-PsMYB114L, pGreenII62-SK-PsMYB12L and pGreenII0800-LUC-PsDFR (Fig. 9b) were also verified successfully by PCR and sequencing.
Tobacco leaves were respectively injected with the combination of Agrobacterium tumefaciens solution of the recombinant vectors, including 4 control and 2 treatment groups, and then the relative activity of luciferase was detected. As for PsMYB114L and PsDFR, it was found that their relative activities of luciferase were lower in the control group containing two empty vectors (pGreenII62-SK, pGreenII0800-LUC) and the single empty vector (pGreenII62-SK-PsMYB114L, pGreenII0800-LUC), and their relative activities of luciferase were relatively higher but the differences were very small (Fig. 9c) in the control group containing single empty vector (pGreenII62-SK, pGreenII0800-LUC-PsDFR) and the treatment group 1 (pGreenII62-SK-PsMYB114L, pGreenII0800-LUC-PsDFR).
As for PsMYB12L and PsDFR, their results of luciferase activity showed a similar trend compared with the above results of PsMYB114L and PsDFR. However, the relative activity of luciferase in the treatment group 2 (pGreenII62-SK-PsMYB12L, pGreenII0800-LUC-PsDFR) was about 2.3 times higher than that in the control group containing single empty vector (pGreenII62-SK, pGreenII0800-LUC-PsDFR) (Fig. 9d). In summary, these results indicated that only PsMYB12L could activate the expression of PsDFR.