Purple-fleshed sweet potato cv. A5 and white-fleshed sweet potato cv. Yubeibai were cultivated in the biological garden at South China Normal University, Guangzhou, Guangdong, China. Arabidopsis (Arabidopsis thaliana) used for subcellular localization and for dual luciferase assays was cultivated in a growth chamber with a day/night cycle of 16/8 h at 20 $\pm$ 2 °C.

A small amount of root tissue (0.5 g) was ground to a fine powder in the presence of liquid nitrogen using a mortar and pestle. A plant genomic DNA kit (Cat. No. 4992201, Tiangen, Beijing, China) was used to extract DNA and a HiPure Plant RNA Midi kit (Cat. No. R4152, Magen, Guangzhou, China) was used to extract RNA. In order to eliminate the possibility of DNA contamination, total RNA was treated with DNAse I enzyme from an RNase-free kit (TaKaRa, Kyoto, Japan). DNA concentrations and purity were determined using a BioPhotometer plus (Eppendorf, Hamburg, Germany). cDNA synthesis was performed with oligo (dT) as primer using M-MLV reverse transcriptase according to the manufacturer’s recommendations (cat. no. A5001, Promega, Wisconsin, USA). PCR products were judged on 1.2% agarose gels, ligated into plasmid and sequenced (Sangong, Shanghai, China).

Y1H were performed using the Matchmaker Gold Yeast One-Hybrid Library Screening System according to a previous report [21]. Total RNA was isolated to construct the prey cDNA library (TaKaRa, Japan). A cDNA pool was inserted into pGADT7-Rec. Two amplifications promoter fragments IbWD40-1 (–1136 bp~–585 bp) and IbWD40-2 (–584 bp~–1 bp) were inserted into pAbAi. The pAbAi-bait plasmids were linearized and transformed into the yeast strain Y1HGold. Transformants were selected on plates containing a selective synthetic dextrose medium lacking uracil. The linear pGADT7-Rec vector was co-transformed into the bait yeast strains and selected on synthetic dextrose (SD)/-Leu/AbA plates. The primers used in the Y1H screening are listed in Supplementary Table 1.

Y1H was performed using the Matchmaker Gold Yeast One-Hybrid System as described in a previous study [22]. This identified IbERF1, IbERF10, IbEBF2, IbPDC, IbPGP19, IbUR5GT, IbDRM, IbPPA and IbERF73 as interacting with the IbWD40 promoter. The promoter for IbWD40 was inserted into pAbAi. Complete CDSs for IbERF1, IbERF10, IbEBF2, IbPDC, IbPGP19, IbUR5GT, IbDRM, IbPPA and IbERF73 were separately inserted into the pGADT7, which were then transferred into the bait strain and grown on SD/-Leu/AbA plates. Primers used for Y1H are listed in Supplementary Table 1.

The transcriptional activity of IbERF1, IbERF10, IbEBF2, IbPDC, IbPGP19, IbUR5GT, IbDRM, IbPPA and IbERF73 was investigated using the Matchmaker™ Gold Yeast Two-Hybrid System (Clontech, Shanghai, China). Full-length coding sequences for IbERF1, IbERF10, IbEBF2, IbPDC, IbPGP19, IbUR5GT, IbDRM, IbPPA and IbERF73 were cloned into the pGBKT7 vector to construct bait-BD vectors. The PGBKT7-bait and PGADT7-empty vectors were then co-transferred into the yeast strain Y2HGold and grown on SD/-Trp plates. Positive colonies were transferred and grown on SD/-His-AbA plus plates. Then, positive colonies were transferred and grown on SD/-His-AbA X-a-Gal plus plates to observe the growth and color of the yeast colonies. The interaction between pGADT7-53 and pGBKT7-53 was used as a positive control, and the interaction between pGADT7-53 and pGBKT7 was used as a negative control. The primers used for Y2H are listed in Supplementary Table 1.

Dual-luciferase assays were used to measure the transactivation activities of IbERF1, IbERF10, IbEBF2, IbPDC, IbPGP19, IbUR5GT, IbDRM, IbPPA and IbERF73 on the IbWD40 promoter. In brief, full-length cDNAs for IbERF1, IbERF10, IbEBF2, IbPDC, IbPGP19, IbUR5GT, IbDRM, IbPPA and IbERF73 were inserted into the pGreen II 0029 62-SK vector, with Sac I and Xho I, and the IbWD40 promoter was inserted into the pGreen II 0800-LUC vector with Sac I and Xho I. Both constructs were transformed into Arabidopsis protoplasts as described previously [23]. The ratio of enzyme activity for LUC and REN was measured with an E1910 Dual-Luciferase® Reporter Assay System (Promega, Wisconsin, USA). Three independent experiments were carried out to study each interaction between transcription factors and promoters. For each interaction between transcription factors and promoters, three independent experiments were carried out, with three replicates in each experiment. A luciferase gene from Renilla driven by a 35S promoter in the luciferase vector acted as the positive control. Mixtures containing each transcription factor and empty vector 62-SK were also tested on the promoter as a control. The primers used for the dual-luciferase assay are listed in Supplementary Table 1.

To analyze subcellular localization, the CDSs for upstream transcription factors without the stop codon were amplified and cloned with BamH Ⅰ and Hind Ⅱ into the pCambia1300 vector containing the GFP gene and UBQ promoter. Both constructs were transformed into Arabidopsis protoplasts as described previously [23]. GFP fluorescence was observed using a Zeiss confocal microscope (LSM710). Primers used for the subcellular localization analysis are listed in Supplementary Table 1.

Real-time quantitative PCR was used to evaluate the expression of upstream transcription factors (IbERF1, IbERF10, IbERF73), transcription factors (IbMYB1, IbbHLH2, IbWD40) and structural genes (IbCHI, IbCHS, IbF3H, IbF3’H, IbDFR, IbANS, IbUF3GT) in the fibrous roots, thick roots and storage roots of purple-fleshed sweet potato cv. A5 and white-fleshed sweet potato cv. Yubeibai. RT‒qPCR was then performed using SYBR® Premix Ex Taq™ II (TaKaRa, Kyoto, Japan) in a total 20 µL reaction volume containing 100 ng of template cDNA, 0.5 µM of each primer, and 10 µL of SYBR® Premix Ex Taq™ II. The Bio-Rad CFX96 Real-Time PCR system (BIO-RAD, California, USA) was used according to the manufacturers’ instructions. The amplification program was as follows: one cycle of 95 ${}^{\circ }$C for 10 sec, followed by 40 cycles of 95 ${}^{\circ }$C for 5 sec and 60 ${}^{\circ }$C for 30 sec. IbG14 was used as the internal control and the comparative Ct analysis method was performed. The primers used for RT-qPCR are listed in Supplementary Table 1.

Three biological replicates of each sample were subjected to one-way analysis of variance (ANOVA). Significant differences were calculated using SPSS 21.0 Statistics (SPSS Inc., Chicago, IL, USA) and Tukey’s honest test (p 0.05). Figures were drawn with SigmaPlot 12.3 (Systat Software Inc., San Jose, CA, USA).

Two amplifications of IbWD40-1/2 promoter fragments were ligated into pAbAi vector to generate the pAbAi-bait plasmid. The minimum inhibitory concentration of AbA for the bait strains was determined according to the system user manual. This was found to be 300 ng/mL for the pAbAi-PIbWD40-1 strains. The AbA concentration was 900 ng/mL to inhibited the positive strains (pGADT7-53 + p53-AbAi). Self-activation of PIbWD40-2 could not be suppressed because the pAbAi-PIbWD40-2 strains cannot be inhibited.

A total of 453 colonies of PIbWD40-1 screened positive, and 175 binding proteins for PIbWD40-1 were identified by the Y1H assay. The gene sequences for pIbWD40-1 are shown in Supplementary Table 2. The proteins IbERF1, IbERF10, IbEBF2, IbPDC, IbPGP19, IbUR5GT, IbDRM, IbPPA and IbERF73 were found to interact with the promoter for IbWD40-1.

Y2H assays results showed that pGADT7-53 + pGBKT7-53 and pGBKT7-transcription regulators + pGADT7-empty transformed strains can grow on SD/-Trp, SD/-His-AbA plus, and SD/-His-AbA X-a-Gal plus plates, and the color of emerging yeast colony was blue (Fig. 1). These results indicate that IbERF1, IbERF10, IbEBF2 and IbERF73 had transcriptional activity.

Fig. 1.

Transcriptional activation of transcription regulators in yeast cells. Y2H Gold strains successfully transformed with corresponding vector and grown on the SD/-His/AbA/X-$\alpha$-Gal plates at 30 °C for 3–5 days. The two points of each treatment were grown on a petri dish. Transcription activation was monitored by the growth status of yeast cells using the X-$\alpha$-Gal assay.

Y1H was performed to identify whether the transcription regulators identified above (IbERF1, IbERF10, IbEBF2, IbPDC, IbPGP19, IbUR5GT, IbDRM, IbPPA, IbERF73) interacted with PIbWD40-1. The assay results showed the positive control could grow on SD/-Leu/AbA plates, but not the negative control. The transformed strains grew on SD/-Leu/AbA plates (Fig. 2). These results demonstrate interaction between transcription regulators and PIbWD40-1.

Fig. 2.

Interaction between transcription regulators and PIbWD40-1 studied using yeast one-hybrid assays. Y1H Gold strains successfully transformed with corresponding vector and grown on SD/-Leu/AbA plates at 30 °C for 3–5 days. The two points of each treatment were grown on a petri dish. Interaction was confirmed by the growth status of yeast cells.

Dual-luciferase assays were performed to confirm the interactions between transcription regulators and PIbWD40-1. IbERF1, IbERF10, IbEBF2, IbPDC, IbPGP19, IbDRM, IbPPA and IbERF73 all showed significant activation of PIbWD40-1 (Fig. 3), thus confirming their interaction with PIbWD40-1.

Fig. 3.

The IbERF1, IbERF10, IbEBF2, IbPDC, IbPGP19, IbDRM, IbPPA and IbERF73 proteins activate PIbWD40-1 in dual-luciferase assays. Error bars represent standard deviation (SD). The significance tests are shown as a, b, c, d, e, f, g, h. A different lowercase roman alphabet in the column chart indicates a significant difference (p 0.01).

In the subcellular localization assays, the green fluorescence of the GFP control was observed in the nucleus and cytoplasm. Green fluorescence from IbERF1/IbERF10/IbPDC/IbPGP19/IbPPA/IbERF73-GFP fusions were detected only in the nucleus (Fig. 4), suggesting that IbERF1, IbERF10, IbPDC, IbPGP19, IbPPA and IbERF73 were nuclear proteins. The IbUR5GT, IbDRM and IbEBF2 proteins were expressed in both the nucleus and cytoplasm.

Fig. 4.

Subcellular localization of transcription regulators in Arabidopsis protoplasts. The fusion protein and GFP control were transiently expressed in Arabidopsis protoplasts. Bars represent 20 µm. A: GFP; B: Chloroplast; C: Light field; D: Merged graph.

Real-time PCR was used to analyze the gene expression of transcription regulators, transcription factors and structural genes involved in anthocyanin biosynthesis at different root stages of purple- and white-fleshed sweet potato (Fig. 5). The expression levels of transcription factors and structural genes at different root stages of purple-fleshed sweet potato were found to be higher than in white-fleshed sweet potato. However, the expression of IbERF1 at different root stages was quarter in purple-fleshed sweet potato compare with in white-fleshed sweet potato, indicating negative regulation by IbERF1 of anthocyanin accumulation. The expression of IbERF10 and IbERF73 at different root stages of purple- and white-fleshed sweet potato were not associated with anthocyanin accumulation, indicating the existence of more complex mechanisms in the regulation of anthocyanin biosynthesis.

Fig. 5.

Relative expression levels of transcription regulators, IbWD40 and structural genes involved in anthocyanin biosynthesis at different root stages of purple-fleshed and white-fleshed sweet potato. Fibrous roots (diameter 2 mm), Thick roots (2 mm diameter $>$5 mm), Storage roots (diameter $>$5 mm).