Flavonoid biosynthetic genes contain YABBY-binding motifs
The cloned promoter sequences of flavonoid biosynthetic genes used in the present study, AaPAL, AaCHS, AaCHI, AaFLS, AaFSII, and genes regulating anthocyanins AaDFR, AaLDOX, and AaUFGT, were analyzed for putative YABBY-binding sequences37 using PlantPAN 3.0 (PlantPAN; http://PlantPAN.itps.ncku.edu.tw). Recent work, including ChIP and RNA-seq studies on YABBY-binding motifs present in soya bean and protein-binding microarrays in A. thaliana, has shown that these sites vary greatly among different species. YABBY-binding motifs are represented by AT-rich sites in Arabidopsis with consensus binding sequences defined as AATNATAA and AATNATTA. The homologous YABBY-binding motifs found in the promoter sequences are shown in Fig. 2a, with the positions marked by numbers. Except for DFR, YABBY-binding motifs were identified in all promoters.
a Putative YABBY-binding sites predicted by PLANTPAN3.0 are shown. Positions on plus and minus strands are represented by numbers above and below the promoter sequence, respectively. YABBY-binding sites were found in all promoters except the DFR promoter sequence, for which no predicted YABBY-binding site was predicted. b Heatmap showing the expression profile of YABBY family genes, as well as flavonoid biosynthetic genes in six tissues. The color scale at the top represents the RPKM (reads per kilobase per million mapped reads) values. AaYABBY5 (marked with a black dot) was selected as a potential transcription factor that might regulate flavonoid biosynthesis because of its similar expression pattern to the important flavonoid biosynthetic genes
Global expression profile of YABBY family genes and selection of AaYABBY5 as a potential transcription factor regulating flavonoid biosynthesis
The transcriptome data of six different tissues of A. annua were previously generated by our lab46. Plant secondary metabolites are usually synthesized in a species- or tissue-specific manner, and secondary metabolites, including flavonoids, are synthesized in trichomes47. To identify potential YABBY family genes that might regulate flavonoid biosynthesis, a heatmap was constructed to compare the expression of YABBY genes and flavonoid biosynthetic genes across tissues; trichomes, buds, stems, roots, leaves, and seeds (Fig. 2b).
We found two YABBY genes clustered with flavonoid pathway genes, showing higher expression in trichomes and buds. Among the YABBY family genes, AaYABBY5 (marked with a black dot) was found to be a candidate transcription factor that showed a transcription profile parallel to that of PAL, CHI, DFR, and FLS and showed higher expression in trichome, bud, stem, and leaf tissues. In our previous findings, we found that AaYABBY5 regulates DBR2 and CYP71AV1 that are involved in the artemisinin biosynthetic pathway48. Here, we found that AaYABBY5, DBR2, and CYP71AV1 showed similar expression patterns, i.e., higher expression in trichomes and buds, with a progressive decline in their expression in the stem tissues.
As this study focused on regulating flavonoid regulation, we hypothesized that AaYABBY5 might regulate PAL, CHS, CHI, and FLS, which showed similar expression patterns in trichomes and/or buds. Interestingly, PAL and DBR2 (the YABBY5 target gene) showed similar expression in trichomes, buds, stems, roots, and leaves. DFR and FSII expression was found more in buds than in trichomes. F3’H expression was very different from the other genes present in the flavonoid pathway. LDOX and UFGT are enzymes that are involved in anthocyanin biosynthesis. In A. annua, anthocyanins are limited to stem tissues28, and the expression of LDOX and UFGT was found to be higher in stem tissues than in other tissues. In a previous study, real-time PCR analysis of different tissues revealed that AaYABBY5 transcripts are also found in the stem tissues of A. annua,48; therefore, we speculated that it might also regulate the LDOX and/or UFGT genes. Overall, it was speculated that YABBY5 might regulate PAL, CHI, CHS, and FLS, which are involved in early flavonoid biosynthesis, and LDOX and UFGT, which are present in the late flavonoid (anthocyanin) pathway. Further experiments were carried out to test this hypothesis.
AaYABBY5 significantly activates the promoters of AaPAL, AaCHI, AaCHS, and AaUFGT in transiently transformed N. benthamiana
Knowing putative YABBY-binding sites in the AaPAL, AaCHI, AaCHS, AaDFR, AaFLS, AaFSII, AaLDOX, and AaUFGT promoter sequences, a dual-luciferase assay was performed, where AaYABBY5 inserted in pEarleyGate 104-YFP was used as an effector and promoter sequences inserted into pGreenII 0800-LUC were used as reporters (Fig. 3a). Equal-sized infiltrated leaf discs for each combination of the reporter with the effector AaYABBY5 were analyzed by commercially available dual-LUC reagents (Promega, USA). Values greater than a twofold increase were taken into consideration, and lower values were negated. A significant increase in relative LUC/REN values was found for the AaPAL, AaCHS, AaCHI, and AaUFGT promoters. AaYABBY5 exhibited a 7.4-fold increase in the activity of the PAL promoter, a 3.2-fold increase in the activity of the CHS promoter, a 3.4-fold increase in the activity of the CHI promoter and a three fold increase in the activity of the UFGT promoter (Fig. 3b–d, i). The fold change was calculated from comparative values of each effector/reporter, and an empty vector was used as a negative control/reporter. The increase in the LUC/REN values corresponds to the intensity of LUC signals driven by the respective promoters in the presence of the AaYABBY5 protein. Based on these results, it was hypothesized that AaYABBY5 could activate these promoter sequences in A. annua either directly, by binding to putative YABBY-binding motifs, or indirectly, through some protein–protein interactions.
a Schematic representation of constructs used to prepare effector and reporter strains. The AaYABBY5 open-reading frame fused to the yellow fluorescent protein (YFP-N) in pEG104 was used as the effector. YFP-N was used as a negative control. Promoters of PAL, CHS, CHI, DFR, FLS, FSII, LDOX, and UFGT were fused with the LUC gene at its N-terminus as reporter constructs. From b–i showing relative LUC activities obtained for each combination of the reporter with effector. Significant increases were found with the PAL, CHI, CHS, and UFGT promoters. Data show the mean values ± SD of four independent infiltrations. Error bars show the standard deviation for n = 4. **P < 0.01. *P < 0.05. The gene sequence of AaYABBY5 can be found in NCBI GenBank under accession number MK675289. The promoter sequences of DFR, PAL, CHS, CHI, FLS, FSII, LDOX, and UFGT have been submitted to the National Center for Biotechnology Information (NCBI), and accession numbers have been assigned as MW558943, MW558944, MW915581, MW558945, MW464242, MW464241, MW464239, and MW464240, respectively
AaYABBY5 directly binds to promoter regions of AaPAL, AaCHS, AaCHI, and AaUFGT in the EGY48 yeast strain
Transactivation assays using N. benthamiana revealed that AaYABBY5 mediated a significant increase in the activities of PAL, CHS, CHI, and UFGT promoters in vivo. Therefore, to determine the molecular basis of this regulation and whether AaYABBY5 directly activates them, a Y1H assay was performed. The experiment demonstrated the binding of the pB42AD-AaYABBY5 fusion protein (blue color appearance), but not pB42AD alone (no color), to the PAL, CHS, CHI, and UFGT promoter sequences, indicated by the activation of the lacZ reporter gene, which produces ß-galactosidase and cleaves the X-gal present in growth medium to a compound with a blue-colored phenotype. The experiment was repeated three times to validate the results. No colored phenotype was found for the DFR, FLS, FSII, and LDOX promoters (Fig. 4b).
a Sketch map of prey and bait constructs used to perform the Y1H assay. The coding sequence of the prey protein AaYABBY5 was cloned into the pB42AD vector under the GAL1 inducible promoter sequence as a fusion to the NLS; nuclear localization sequence, AD; activation domain, and HA (hemagglutinin) epitope tag, whereas promoters of PAL, CHS, CHI, DFR, FLS, FSII, LDOX, and UFGT were cloned into the placZ2µ vector as fusions with the lacZ reporter gene to form the bait strains. b AaYABBY5 directly bound to full-length PAL, CHS, CHI, and UFGT promoters in yeast cells cotransformed with these bait strains along with pB42AD-AaYABBY5, as shown by the blue-colored phenotype of yeast clones, but not with empty pB42AD. FLS, FSII, and LDOX did not show positive results. c–j show the relative expression of the PAL, CHS, CHI, DFR, FLS, FSII, LDOX, and UFGT genes in selected AaYABBY5 OE, AaYABBY5 AnT, 35S, and wild-type (W) A. annua plants. Gene expression was found to increase in AaYABBY5-OE plants, whereas in AaYABBY5 AnT. plants, a decrease in the expression of PAL, CHS, CHI, DFR, FLS, FSII, LDOX, and UFGT was found. β-Actin was used as an internal control. The graph shows the mean values ± SD of three experimental replicates. Error bars show the standard deviation for the sample, n = 3. Statistical significance was determined using the Student’s t test. **P < 0.01, *P < 0.05
The results of the Y1H assay were consistent with the findings of the dual-luciferase reporter assay. In other plants, including A. thaliana, flavonoid genes, and their transcriptional regulation are well-studied. The A. thaliana FIL gene has been previously reported to be a positive regulator of anthocyanins by activating the MYB75 gene42. To our knowledge, for the first time, we found the molecular basis of the transcriptional regulation of flavonoid biosynthetic pathway genes, PAL, CHS, CHI, and UFGT in A. annua. These results indicated that AaYABBY5 might have the potential to regulate both flavonoid and anthocyanin biosynthesis in A. annua. Therefore, these contents were measured and compared in AaYABBY5 OE and AaYABBY5 AnT. A. annua plants.
AaYABBY5-overexpressing A. annua plants showed a consistent significant increase in the expression of genes from AaPAL to AaUFGT
After analyzing the binding of AaYABBY5 to the promoters of early (flavanone) and late (anthocyanin) flavonoid biosynthetic genes, it was important to study its functions in A. annua. As expected, the gene expression analysis revealed an increase in the expression of all genes under study. The comparative expression of these genes indicated that AaCHS showed higher expression than all other genes under study. Similarly, the expression of DFR, FSII, and UFGT was lower than that of PAL, CHS, CHI, FLS, and LDOX (Fig. 4c–j).
Although DFR, FLS, FSII, and LDOX were not activated by AaYABBY5, a higher expression of these genes in AaYABBY5-OE plants was found. It is proposed that the increased flux provided by PAL, CHS, and CHI activates downstream pathway enzymes by increasing the concentration of substrates for the enzymes acting downstream: DFR, FLS, FSII, and LDOX. AaYABBY5 overexpression not only increased the expression of its direct target genes but also the flavonoid pathway under study.
AaYABBY5 antisense A. annua plants showed a significant decrease in the expression of flavonoid biosynthetic genes
The AaYABBY5 protein activates flavonoid biosynthetic genes, and the increased expression of AaYABBY5 led to a dramatic increase in the expression of flavonoid-regulating genes. Transcript levels of PAL, CHI, CHS, DFR, FLS, FSII, LDOX, and UFGT were analyzed in AaYABBY5 antisense RNA-containing plants. As expected, a significant decrease in the expression of the genes under study was found in AaYABBY5 AnT. plants (Fig. 4c–j). Overall, from these findings, a clear understanding of AaYABBY5-regulated flavonoid biosynthesis was obtained. To validate the above findings, flavonoid and anthocyanin concentrations were measured and compared among AaYABBY5 OE, AaYABBY5 AnT, wild-type/control plants, and vector-containing plants.
AaYABBY5 positively regulates flavonoid biosynthesis
Flavonoids are polyphenolic plant secondary metabolites that are classified into different types. In this study, we found that AaYABBY5 positively regulates the PAL, CHI, CHS, and UFGT genes. The results of real-time PCR also verified the increased expression of these genes in AaYABBY5 overexpression plants. These findings revealed a positive behavior of AaYABBY5 toward flavonoid biosynthesis in A. annua. To justify this, flavonoid contents from AaYABBY5 overexpression plants, AaYABBY5 antisense plants, and control plants were measured using the aluminum chloride (AlCl3) colorimetric method with quercetin as a standard. As expected, the results revealed an increased concentration of flavonoids in AaYABBY5 overexpression plants compared to AaYABBY5 antisense or control plants (Fig. 5a), proving the function of AaYABBY5 as a positive regulator of flavonoid biosynthesis.
a Comparative analysis of total flavonoid content in AaYABBY5-OE plants, AaYABBY5 AnT. plants, vector-containing (35S), and wild-type plants, showing a significant increase in the total flavonoid content of AaYABBY5 OE plants compared to AaYABBY5 AnT. plants and control ones. Similarly, AnT. plants showed a decrease in the concentration of flavonoids. b–d Phenotypic analysis of transgenic plants showing deep purple pigmentation in the stems of AaYABBY5-OE plants, whereas no purple-color phenotype was found in wild-type and/or antisense AaYABBY5 plants. Scale bars = 8 cm in (b) and (c) and 4 cm in (d). e Comparative analysis of anthocyanin concentrations in AaYABBY5-OE plants, AaYABBY5 AnT. plants, and vector-containing (35S) and wild-type plants, showing a significant increase in the total anthocyanin content of AaYABBY5-OE plants compared to AaYABBY5 AnT. plants and control ones. In contrast, AaYABBY5 AnT. plants showed a decrease in the concentration of anthocyanins. f Coloric representation of anthocyanin enrichment in the anthocyanin extract of AaYABBY5-OE plants showing the red color of the anthocyanin extract, an indication of an increased amount of anthocyanins. Scale bars = 4 cm. g, h Phenotypic analysis showing broader lamina in AaYABBY5-OE plants (ii) compared to AaYABBY5 AnT. plants (i). Scale bars = 3.5 cm and 10 cm in g and h, respectively. The graph shows the mean values ± SD of three experimental replicates. Error bars show standard deviation for sample, n = 3. Statistical significance was determined using Student’s t test. **P < 0.01, *P < 0.05
AaYABBY5 promotes anthocyanin biosynthesis, displaying a purple phenotype in A. annua stems
Anthocyanins are secondary metabolites widely present in plant species and are responsible for the purple, bluish, and pinkish pigmentation of different plant parts. In this study, we observed deep purple pigmentation in the stems of AaYABBY5 overexpression plants, whereas AaYABBY5 antisense plants and control plants showed no purple phenotype (Fig. 5b, c).
To test whether AaYABBY5 is involved in the regulation of anthocyanin biosynthesis in A. annua, we measured and compared the total anthocyanin content of stem extracts of control plants and pHB/wild-type, AaYABBY5-overexpressing, and AaYABBY5 antisense plants. It is known that anthocyanins produce a red color when treated with acids. Consistent with this, the anthocyanin extracts from purple-colored AaYABBY5-OE plants were red-colored in acidic medium (Fig. 5f). As expected, a significant increase in the concentration of anthocyanins was found in the stems of plants overexpressing AaYABBY5 compared to the control and AaYABBY5 antisense plants (Fig. 5e). Previous investigations have demonstrated that the YABBY family TF AtFIL is a positive regulator of anthocyanin biosynthesis in the model plant A. thaliana through activating the MYB75 promoter; however, no direct link to the genes regulating anthocyanins or flavonoids was reported in A. annua42. MYC2 has been reported to be responsible for an increase in the anthocyanin content, giving a purple phenotype to AaMYC2-overexpressing stems28; however, the mechanism underlying this regulation was not studied. To our knowledge, we provide the first molecular basis of flavonoids, including anthocyanin regulation by the YABBY family transcription factor AaYABBY5. It was supposed that AaYABBY5 activates anthocyanin biosynthesis through the direct activation of UFGT and upstream pathway genes.
AaYABBY5 overexpression results in broader leaf lamina and increased trichome density
The primary function of YABBY family transcription factors found in seed plants is the control of leaf development, increasing the size of the leaf lamina, and maintaining organ polarity50; therefore, it was important to determine whether AaYABBY5 overexpression and/or its downregulation in A. annua affected leaf morphology. Phenotypic analysis of AaYABBY5-OE plants, AaYABBY5 AnT. plants, and control plants showed that the leaves of AaYABBY5-OE plants have broader leaf lamina, whereas in AaYABBY5 AnT. plants, the leaf lamia was reduced. Leaves of AaYABBY5 AnT. plants were radialized compared to that of OE plants and/control plants (Fig. 5g, h). Trichomes are the sites of secondary metabolite synthesis, and studies have reported that flavonoids are synthesized in trichomes47.
Trichome development is a part of leaf development. Knowing that AaYABBY5 regulates leaf lamina, the trichome densities on the leaf surfaces were compared among transgenic plants. The trichome densities on leaf surfaces from AaYABBY5 overexpression plants, AaYABBY5 antisense plants, and control plants with an empty vector were calculated. A significant increase in the trichome densities of AaYABBY5 overexpression plants was found compared to control plants, whereas for AaYABBY5 antisense plants, trichome density was found to significantly decrease. These results show that AaYABBY5 is also a positive regulator of trichome development (Fig. 6a, b).
a Trichome densities on the leaf surface of transformed and control plants were calculated, and images were captured using fluorescence microscopy (Olympus, Japan). The trichome densities of AaYABBY5-OE plants were higher than those of the control and AaYABBY5 AnT plants. Scale bars represent 200 µm. b Graphical representation of trichome densities of control, AaYABBY5 AnT., and AaYABBY5-OE plants calculated as numbers per millimeter square (no./mm2). Data = mean values ± standard deviation for n = 4. Error bars represent standard deviation for n = 4. Student’s t test with paired and two-tailed distribution methods. * represents P < 0.05. c The AaYABBY5 protein directly targets the promoters of PAL, lying upstream of the flavonoid biosynthetic pathway. The PAL gene, once activated, provides increased metabolic flux toward the first committed step of flavonoid biosynthesis. AaYABBY5 also directly binds to and activates the CHS and CHI promoters, providing increasing concentrations of flavonones (total flavonoid content). Going through multistep reactions, in the final step, anthocyanidins, which are the precursors of anthocyanins, are converted into anthocyanins by AaYABBY5 through the direct activation of the AaUFGT promoter. AaYABBY5 also regulates trichome number and leaf lamina growth through an unknown mechanism
AaHD1 (homeodomain protein 1) is involved in the initiation of both glandular and nonglandular trichomes in A. annua.49; however, no protein interactions between AaYABBY5 and AaHD1 were found. It is proposed that AaYABBY5 might activate the promoter of AaHD1, which regulates trichome development. This research opened paths for future research where the molecular mechanism of the regulation of trichome development by AaYABBY5 can be found.
