The transcription factor complex CmAP3-CmPI-CmUIF1 modulates carotenoid metabolism by directly regulating carotenogenic gene CmCCD4a-2 in chrysanthemum

Abstract

Carotenoids are one of the most important pigments for the coloring in many plants, fruits and flowers. Recently, significant progress has been made in carotenoid metabolism. However, the specific understanding on transcriptional regulation controlling the expression of carotenoid metabolic genes remains extremely limited. Anemone-type chrysanthemum, as a special group of chrysanthemum cultivars, contain elongated disc florets in capitulum, which usually appear in different colors compared with the ray florets since accumulating distinct content of carotenoids. In this study, the carotenoid composition and content of the ray and disc florets of an anemone-type chrysanthemum cultivar 'Dong Li Fen Gui' were analyzed by high-performance liquid chromatography-tandem mass spectrometry (HPLC-MS/MS) and the key structural gene CmCCD4a-2, of which differential expression resulted in the distinct content of carotenoids accumulated in these two types of florets, was identified. Then the promoter sequence of CmCCD4a-2 was used as bait to screen a chrysanthemum flower cDNA library and two transcription factors, CmAP3 and CmUIF1 were identified. Y2H, BiFC and Y3H experiments demonstrated that these two TFs were connected by CmPI to form CmAP3-CmPI-CmUIF1 TF complex. This TF complex regulated carotenoid metabolism through activating the expression of CmCCD4a-2 directly. Furthermore, a large number of target genes regulated directly by the CmAP3-CmPI-CmUIF1 TF complex, including carotenoid biosynthetic genes, flavonoid biosynthetic genes and flower development-related genes, were identified by DNA-affinity purification sequencing (DAP-seq), which indicated that the CmAP3-CmPI-CmUIF1 TF complex might participate in multiple processes. These findings expand our knowledge for the transcriptional regulation of carotenoid metabolism in plants and will be helpful to manipulating carotenoid accumulation in chrysanthemum.

Keywords: CmCCD4a-2; CmUIF1-CmPI-CmAP3 transcription factor complex; carotenoid; chrysanthemum; transcriptional regulation.

Figure 1

Characteristics of capitulum developmental stages of “Dong Li Fen Gui” and analysis of carotenoid components and contents. a. The development of the capitulum of “Dong Li Fen Gui” chrysanthemum was divided into five stages (S1–S5). R1–R5 and D1–D5 represent ray florets and disc florets of “Dong Li Fen Gui” at different developmental stages. Bar, 1 cm. b. Carotenoid components and contents of ray florets (R1–R5) and disc florets (D1–D5) of “Dong Li Fen Gui” at different developmental stages. The data are presented as the mean ± SD from at least three biological replicates (^*P < 0.05; ^**P < 0.01; Student’s t-test).

Figure 2

Key carotenoid metabolic genes identified based on comparative transcriptomics analysis. a. The ray and disc florets of “Dong Li Fen Gui” at the S3 stage were used for RNA sequencing. b. Heatmap analysis of all obtained genes based on the transcriptome data. C. Transcripts of all carotenoid metabolic genes in ray and disc florets of “Dong Li Fen Gui” at the S3 stage based on the transcriptome data.

Figure 3

Gene expression analysis of CmCCD4a-2 in the ray and disc florets of “Dong Li Fen Gui” at different developmental stages. a. Identification of key carotenoid metabolic genes that caused differences in the carotenoid contents of ray and disc florets of “Dong Li Fen Gui” by heatmap analysis using semi-quantitative RT-PCR data. b. The expression profile of CmCCD4a-2 was analyzed by qRT-PCR. All data are presented as the mean ± SD from at least three biological replicates (^*P < 0.05; ^**P < 0.01; Student’s t-test).

Figure 4

Identification of the candidate upstream transcription factors CmAP3 and CmUIF1. a. Subcellular localization of CmAP3 and CmUIF1 in tobacco leaves. The green GFP fluorescence signal was imaged by a laser scanning confocal microscope 48 hours after injection. b. The expression profiles of CmAP3 and CmUIF1 were analyzed by qRT-PCR. c. Y1H showed the binding of CmAP3 and CmUIF1 to the CmCCD4a-2 promoter. d. Effect of CmAP3 and CmUIF1 on the activity of the CmCCD4a-2 promoter by dual-luciferase assay. All data are presented as the mean ± SD from at least three biological replicates (^*P < 0.05; ^**P < 0.01; Student’s t-test).

Figure 5

Identification of the intermediate “bridge” protein CmPI. a. The expression profile of CmPI was analyzed by qRT-PCR. The data are presented as the mean ± SD from at least three biological replicates (^*P < 0.05; ^**P < 0.01; Student’s t-test). b. Y1H assay showing that CmPI was not able to interact with the CmCCD4a-2 promoter. c. Protein interaction analysis between CmPI and CmAP3/CmUIF1 by Y2H assay. d. A BiFC assay confirmed the interaction between CmPI and CmAP3/CmUIF1 in Nicotiana benthamiana leaves. The YFP fluorescence signal was imaged by a laser scanning confocal microscope 48 hours after injection.

Figure 6

Effect of the CmAP3-CmPI-CmUIF1 TF complex on the activity of the CmCCD4a-2 promoter. a. Y3H was carried out to detect the interaction among CmAP3, CmPI, and CmUIF1. b. A dual-luciferase assay showed that the CmAP3-CmPI-CmUIF1 heterotrimer effectively activated the promoter of CmCCD4a-2. All data are presented as the mean ± SD from at least three biological replicates (^*P < 0.05; ^**P < 0.01; Student’s t-test).

Figure 7

Downstream target genes regulated by the CmAP3-CmPI-CmUIF1 TF complex were identified by DAP-seq. a. Genome-wide analysis of the CmAP3/CmPI/CmUIF1-binding peaks. Conserved motifs of CmAP3 (b), CmPI (c), and CmUIF1 (d) identified from the DAP-seq data. e. Venn diagram showing peak overlap between the putative target genes related to CmAP3, CmPI, and CmUIF1-binding peaks. f. KEGG enrichment analysis of all putative target genes regulated by the CmAP3-CmPI-CmUIF1 TF complex.

Figure 8

A regulatory model of differential carotenoid accumulation in ray and disc florets of chrysanthemum. The CmAP3-CmPI-CmUIF1 TF complex is abundant in the ray florets and can bind directly to the CmCCD4a-2 promoter and effectively activate its transcription, resulting in the accumulation of only trace or low amounts of carotenoids. Low transcript abundance of CmAP3 and CmPI, important components of the CmAP3-CmPI-CmUIF1 TF complex, was detected in the disc florets. There was not enough TF complex to effectively activate the expression of CmCCD4a-2, and this led to the accumulation of large amounts of carotenoids. The color saturation of the circles (deep/light) represents the expression levels of CmAP3, CmPI, and CmUIF1 (high/low).