Laser capture microdissection transcriptome (LCM RNA-seq) reveals BcDFR is a key gene in anthocyanin synthesis of non-heading Chinese cabbage

Background Purple non-heading Chinese cabbage [ Brassica campestris (syn. Brassica rapa ) ssp. chinensis ] has become popular because of its richness in anthocyanin. However, anthocyanin only accumulates in the upper epidermis of leaves. Further studies are needed to investigate the molecular mechanisms underlying the specific accumulation of it. Results In this study, we used the laser capture frozen section method (LCM) to divide purple (ZBC) and green (LBC) non-heading Chinese cabbage leaves into upper and lower epidermis parts (Pup represents the purple upper epidermis, Plow represents the purple lower epidermis, Gup represents the green upper epidermis, Glow represents the green lower epidermis). Through transcriptome sequencing, we found that the DIHYDROFLAVONOL 4-REDUC-TASE-encoding gene BcDFR , is strongly expressed in Pup but hardly in others (Plow, Gup, Glow). Further, a deletion and insertion in the promoter of BcDFR in LBC were found, which may interfere with BcDFR expression. Subsequent analysis of gene structure and conserved structural domains showed that BcDFR is highly conserved in Brassica species . The predicted protein-protein interaction network of BcDFR suggests that it interacts with almost all functional proteins in the anthocyanin biosynthesis pathway. Finally, the results of the tobacco transient expression also demonstrated that BcDFR promotes the synthesis and accumulation of anthocyanin. Conclusions BcDFR is specifically highly expressed on the upper epidermis of purple non-heading Chinese cabbage leaves and regulates anthocyanin biosynthesis and accumulation. Our study provides new insights into the functional analysis and transcriptional regulatory network of anthocyanin-related genes in purple non-heading Chinese cabbage.


Introduction
Anthocyanins are natural, water-soluble flavonoid pigments that provide a variety of colors to plant leaves, petals, and fruit.Bright colors help plants attract animals for pollination.It also has antioxidant activity and enhances the antioxidant capacity of plants.Studies have shown that anthocyanins have pharmacological effects such as anti-aging, anti-allergy, cardiovascular protection, anti-inflammatory, obesity prevention, antioxidant, free radical scavenging, vision improvement, and cancer prevention [1].Anthocyanin-rich fruits and vegetables have also become more attractive because of their beneficial effects on human health [2].
Many genes involved in anthocyanin biosynthesis in plants have already been cloned and studied.The discovery of COLORED1 (C1), the first anthocyanin-related R3R3-MYB gene found in maize (Zea mays), has revealed the intricate regulatory mechanisms behind anthocyanin production [9].This pioneering research reveals the secrets of nature's vibrant palette.Subsequently, AtPAP1 (AtMYB75), the first anthocyanin-related R2R3-MYB transcription factor in Arabidopsis, was identified.Interestingly, AtMYB75 together with the bHLH proteins AtGL3, AtEGL3, and AtTT8, as well as the WDR protein TTG1, it regulates anthocyanin accumulation by forming the MYB-bHLH-WDR (MBW) complex [10].Vibrant hues are indispensable for showcasing fruits.In apples, the MdMYB1/MYBA transcription factor was found to be associated with apple peel color, while MdMYB10 is the key player responsible for the production of anthocyanin in the fruit [11][12][13].PpMYB10 promotes anthocyanin accumulation in pear by regulating genes encoding anthocyanin pathway enzymes [14].In tomatoes, the colorful fruits are also attributed to LeANT1 and LeAN2 [15,16].In rice, three genes have been revealed as essential regulators of leaf anthocyanin synthesis, including OsC1, OsRb and OsDFR [17].
Brassica contains many vegetables rich in anthocyanins.In Brassica species, many studies have been devoted to the discovery of genes associated with anthocyanin accumulation and their functional analysis.In rapidcycling B. rapa, an anthocyaninless (anl) locus, which inhibits anthocyanin synthesis, had been mapped to R9 [18].Two AtEGL3 homologs, BrEGL3.1 and BrEGL3.2, had been identified on chromosome A09, encoding bHLH transcription factors, which are candidate genes associated with anthocyanin accumulation in zicaitai (B.rapa L. ssp.chinensis var.purpurea) [19].The BoDRF gene, which regulates the synthesis of anthocyanins in the stems of Chinese kale (Brassica oleracea var.alboglabra), had been fine-mapped to chromosome C09 [20].BrPur, a single dominant gene controlling endophytic leaf anthocyanin accumulation in Chinese cabbage, had been mapped to linkage group A07 [21].In B. napus, the BnAPR2 gene was mapped on chromosome A03 as a candidate for regulating leaf purple color [22].Also, many genes related to anthocyanin accumulation have been cloned.For example, low temperature treatment induced the expression of BoTT8 and BoMYB2/ BoPAP1 and regulated the accumulation of anthocyanins in the purple kale [23].In purple mustard, up-regulation of anthocyanin biosynthesis genes induced the expression of BjTT8, which promoted anthocyanin synthesis [24].Meanwhile, the purple cabbage MYB homolog BoMYBL2-1 was predicted to be a negative regulator of anthocyanin synthesis [25].In all, these studies provide insight into the coloration mechanisms of Brassica species.However, the discovery and functional analysis of anthocyanin-related genes in Brassica vegetables still need to be further explored.
Non-heading Chinese cabbage [Brassica campestris (syn.Brassica rapa) ssp.chinensis] is an important leafy vegetable cultivated in Asia [26].The leaf color is generally green.In recent years, purple non-heading Chinese cabbages have received more attention for their vibrant colors and high content of anthocyanins.Interestingly, anthocyanin in the purple variety is mainly accumulated in the upper epidermis and veins of the leaves, with almost no anthocyanin in the lower epidermis.In this study, two varieties (purple ZBC and green LBC) of nonheading Chinese cabbage were used to investigate the key genes involved in anthocyanin regulatory pathway.The upper and lower epidermis of leaves were collected from both varieties at the same growth period using a laser capture frozen section method (LCM) to perform RNAseq.Finally, we found that BcDFR may be a key gene regulating the specific accumulation of anthocyanin on the upper epidermis of purple non-heading Chinese cabbage leaves.Our study provides new insights into transcriptional regulatory network of anthocyanin-related genes in purple non-heading Chinese cabbage.

Plant materials
Two varieties of non-heading Chinese cabbage were used in our study, purple ZBC and green LBC.F1 is a cross between ZBC and LBC.Seeds were placed in pots with substrate (soil matrix and vermiculite 1:1) and grown in a climate chamber with long-day conditions (16 h of light, 22 °C/8 h of darkness 18 °C).In addition, robust ZBC at the 7-leaf stage were selected for low-temperature treatment (4 °C, 30 days).All materials used in this study were provided by Nanjing Agricultural University's Chinese Cabbage System Biology Laboratory.

Total anthocyanin analysis
For anthocyanin measurement, 0.1 g of fresh leaf tissue was collected.The leaves were thoroughly crushed in liquid nitrogen, followed by the addition of 2 ml of acidified methanol (99 CH3OH:1HCl, v/v) and extraction for 24 hours at low temperature in the dark.Then 12,000×g centrifugation was performed for 5 min, and the absorbance of supernatants was determined using a UV-visible spectrophotometer from optical density (OD) at 530 and 650 nm.Total anthocyanin content (mg/g) = (OD530-0.25× OD650) × V/0.0462/M.Where V was total volume of extract, M was leaf mass [27].

Laser-capture microdissection
Tissue specimens were collected from the purple and green leaves.Samples were wrapped using (OCT, Sakura, USA) compound and subsequently cut samples into 20 μm sections using a cryomicrotome (Leica CM3050S, Germany) at − 20 °C.The tissue sections were then uniformly adhered to PEN membrane slides (Leica).Segmentation of the tissue sections into upper and lower parts using an automated laser microdissection system (Leica, LMD7000) [28].

RNA extraction and transcriptome analysis
Total RNA was extracted from LCM leaf samples using the PicoPureTM RNA Isolation Kit (Thermo Fisher Scientific).Total RNA quantity and purity were analyzed using an Agilent 2100 Bioanalyzer and NanoPhotometer spectrophotometer (Agilent, CA, USA).High quality RNA samples with RIN > 7.0 were used for sequencing library construction.mRNA was purified from total RNA (5 μg) using Dynabeads Oligo (dT) (Thermo Fisher, CA, USA).mRNA was subsequently fragmented into short pieces and reverse transcribed to cDNA using SuperScript ™ II reverse transcriptase (Invitrogen, cat.1,896,649, USA).The average insert size for the final cDNA libraries was 300 ± 50 bp.Finally, 2 × 150 bp pairedend sequencing (PE150) was performed on an Illumina Hiseq2000 Truseq SBS Kit v3-HS (200 cycles) following the vendor's recommended protocol.After performing quality filtering on the raw sequencing data, we ended up with clean data of 461,721,426 reads.The cultivar NHCC001 v1.0 genome (https:// www.tbirs.cn/ NHC-CDB/ Genome.jsp) was used as the reference genome.Differentially expressed genes (DEGs) were identified using DEseq2 with screening criteria of FDR < 0.05, |log-2FoldChange| ≥ 1.00.All DEGs were annotated and classified using the GO (http:// www.geneo ntolo gy.org/) and the KEGG (http:// www.kegg.jp/ kegg/).

Quantitative real-time PCR
cDNA was synthesized using Hifair ® AdvanceFast One step RT gDNA Digestion SuperMix for qPCR (Yessen Biotech Co. Ltd., China).Subsequently, qPCR analyses were performed on the StepOnePlus system (Applied Biosystems, USA) using Hieff ® qPCR SYBR Green Master Mix (High Rox) (Yessen Biotech, Co. Ltd., China) according to the instructions.The relative expression of genes was analyzed using the 2 −∆∆CT method, and the expression of BcPP2A (BraC07g034860.1)and BcActin (BraC05g034900.1)was used as a reference.All primers used for qRT-PCR were listed in Supplementary Data Table S1.

Cloning and analysis of BcDFR
We amplified the BcDFR sequences of ZBC and LBC lines and sequenced them by Tsingke Biotechnology Company (Nanjing, China).Primers used for cloning were shown in Supplementary Data Table S1.The MEME (https:// meme-suite.org/ meme/) was used to analyze the conserved motif sequence of the DFR genes.DFR promoter sequences for Arabidopsis and six Brassica species were obtained from BRAD (http:// brass icadb.cn/#/).Ciselements were predicted using the PlantCARE database (https:// bioin forma tics.psb.ugent.be/ webto ols/ plant care/ html/).STRING (https:// cn.string-db.org/) was used to predict BcDFR protein interaction network.Anthocyanin-related DEGs specifically expressed on the upper surface of leaves constructed the interaction network of BcDFR.We set the minimum required score for the interaction to medium confidence (0.40).

BcDFR transient expression assay
The BcDFR gene was cloned into the pRI101 vector.A tumefaciens strain GV3101 containing 35S:BcDFR construct was independently inoculated into LBC cotyledons for transient expression [29].

Statistical analysis
SPSS software was used for the statistical analyses.All data are presented as the mean ± standard deviation of three biological replicates.One-way ANOVA at 0.05 significance level was used to evaluate differences between samples.

Phenotypic analysis of two non-heading cabbage varieties
The purple and green non-heading Chinese cabbage varieties (ZBC and LBC) were used in this study (Fig. 1a, b).Total anthocyanin content of the two varieties was measured and was 0.32 mg/g for ZBC while only 0.04 mg/g for LBC (Fig. 1f ).As the leaves developed, anthocyanins accumulated mainly on the upper leaf surface of ZBC, with almost no anthocyanin accumulation in the lower epidermis (Fig. 1)de.Previous studies have reported that anthocyanin accumulation was associated with low temperature [30].To investigate whether anthocyanin accumulation in non-heading Chinese cabbage responds to low temperature, we treated the purple ZBC with 4 °C for 30 days.Compared with the control, ZBC after low temperature treatment showed internal leaf discoloration (Fig. 1c).In addition, the total anthocyanin content also decreased by 43.75% compared with control, only 0.18 mg/g (Fig. 1f ).In short, the upper epidermis of the purple ZBC accumulates a large amount of anthocyanin and is affected by low temperature.

Laser-capture microdissection and RNA-Seq
To investigate genes specifically expressed on the upper epidermis of purple leaves, upper and lower epidermis tissues of leaves from the same growth period were collected from ZBC using a laser capture microdissection frozen section method [28].Meanwhile, to avoid the difference other than anthocyanins between the upper and lower epidermis, the green LBC was used as a control.Hence, four sequencing libraries were constructed with three replicates.Gup, LBC leaf upper epidermis (Fig. 2a), Glow, LBC leaf lower epidermis (Fig. 2b), Pup, ZBC leaf upper epidermis (Fig. 2c), Plow, ZBC leaf lower epidermis (Fig. 2d).S2).
DEG analysis was performed to identify the DEGs in each sample.The transcript levels of the genes were expressed in FPKM values.Four samples were compared to each other, for a total of six groups.In groups Pup_Plow, Pup_Gup, Plow_Glow, and Gup_Glow,  3a).The number of DEGs in the upper and lower epidermis of the leaves was similar, suggesting that these genes were expressed in the same location in ZBC or LBC leaves.We also found that the number of DEGs in groups Pup_Glow and Plow_Gup was much higher than that in other groups, which indicates that the expression of genes in the upper and lower epidermis of different color leaves is remarkable.Anyway, the expression of genes in leaves is spatially variable, while the differences become more significant in different materials.

Functional annotation and classification of DEGs
We screened DEGs in groups Pup_Gup and Pup_Plow to identify which genes were expressed mainly on the upper epidermis of ZBC leaves.This analysis allowed us to further investigate these specific DEGs.Finally, we obtained 1234 genes that were differentially expressed only on the upper epidermis of ZBC leaves (Fig. 3b).Then, GO functional enrichment analysis was performed on 1234 DEGs (p < 0.05).The results showed that the DEGs were mainly enriched in terms of response to UV-B (GO:0010224), biosynthesis process of anthocyanin-containing compounds (GO:0009718), response to karrikine (GO:0080167), response to sucrose (GO:0009744) and  S3).
In addition, we also analyzed the DEGs for KEGG pathway enrichment.Metabolic pathways and biosynthesis of secondary metabolites were the most significant among all KEGG classifications, followed by flavonoid biosynthesis, phenylpropanoid biosynthesis, and glyoxylate and dicarboxylate metabolism (Fig. 3d; Table S4).

Verification of RNA-Seq data by qRT-PCR
To verify the quality of the RNA sequencing results, we selected nine genes involved in the anthocyanins synthesis for verification by quantitative reverse transcription PCR (qRT-PCR).Anthocyanin biosynthetic genes are usually divided into two subgroups: early biosynthetic genes (EBG) and late biosynthetic genes (LBG).EBGs are also flavonoid biosynthesis genes [5].In our results, the expression levels of both EBGs and LBGs were significantly up-regulated in Pup, such as CHS, CHI, DFR, ANS, and so on (Fig. S1).The anthocyanin-regulated genes MYB111, TT8 and EGL3 were also highly expressed in Pup (Fig. S1).RNA-Seq and qRT-PCR analyses showed similar expression patterns of these genes, indicating the reliability of the transcriptomic data in our study.

Expression patterns of BcDFR in ZBC and LBC
Previous studies have shown that DFR, which is the first enzyme in the biosynthetic pathway, exhibits the highest specificity for anthocyanin production and is regulated by multiple transcription factors [31,32].In our work, notably, the expression of BcDFR (BraC09g018850) was most significantly up-regulated in Pup, consistent with the qRT-PCR results (Fig. S1).These findings indicate that BcDFR may play a critical role in regulating the specific enrichment of anthocyanins on the upper epidermis of ZBC leaves.
To understand why BcDFR is highly expressed in Pup but barely expressed in the other three parts, we cloned full-length BcDFR from ZBC and LBC.The gene sequences of the two cultivars, including exons, introns, and UTR structures, showed a high degree of consistency (Fig. S2a).This suggests that the difference in expression is not caused by variations in gene sequence, prompting us to further analyze the promoter region.Compared with ZBC, BcDFR of LBC showed an 11 base pair deletion at − 585 and a 16 base pair insertion at − 629 in the promoter region (Fig. 5a, S2b).Two additional primers were developed based on the specific deletions and insertions sequence of BcDFR promoter, to accurately differentiate the genotypes of ZBC, LBC and F 1 (hybrid generation of ZBC and LBC).In contrast to ZBC, LBC showed 16 bp insertion and 11 bp deletion at these two loci and F 1 showed heterozygous banding patterns (Fig. 5) bc, S3.The insertion and deletion potentially be responsible for the low expression of BcDFR in LBC.

Analysis of regulatory elements and gene structure of DFR
To further understand the evolution of the DFR gene, we analyzed the gene structure of it in Arabidopsis and six Brassica species, including B. napus, B. rapa, B. carinata, B. nigra, B. juncea, and B. oleracea.Eight DFR genes are highly conserved, containing six exons and five introns.Domain analysis revealed 10 conserved domains within the DFR homologous gene, with only the absence of motif 7 in BjuB001305 (Fig. S4).
Cis-elements in the promoter region of the DFR gene indicate that these regulatory elements were associated with light responsiveness, MYB binding sites, anaerobic induction, salicylic acid responsiveness, meristem expression and a variety of phytohormone responses (https:// bioin forma tics.psb.ugent.be/ webto ols/ plant care/ html/).These findings indicate that DFR was involved in various aspects of plant growth and development and exerts a significant influence.In addition, there are more binding sites for light responsiveness and anaerobic induction elements in the DFR promoter (Fig. 6), suggesting that some stress-related genes probably play a critical role in the anthocyanin biosynthesis of plants.In addition, we also found many MYB binding sites on DFR promoter, suggesting that MYB transcription factors may act as upstream of DFR and then regulate its expression.

Prediction of BcDFR-mediated protein-protein interaction networks
To better understand the function of BcDFR in the anthocyanin synthesis pathway, we used STRING to predict the protein interaction network of BcDFR (Fig. S5).The results showed that BcDFR directly interacted with C4H, CHS, CHI, F3'H, FLS, UGT75C1, and TT8 proteins, while it also indirectly interacted with ANS, F3H, PAL, MYB111, and MYBL2.Interestingly, MYBL111, TT8, together with MYBL2, directly or indirectly affect catalysis proteins for the whole anthocyanin biosynthesis network.Further, RNA-Seq analysis showed that BcMYBL111, BcTT8, and BcMYBL2 were substantially up-regulated in Pup (Fig. S6).The specific expression of them also impacts the elevated expression of BcDFR in Pup, which may result in the anthocyanins only accumulating in Pup.

Transient expression of BcDFR promotes anthocyanin accumulation
To further study BcDFR function in anthocyanin accumulation, we inject 35S: BcDFR into LBC cotyledon for the transient expression assay.We used the empty vector as a control.Compared to the control, the lower epidermis of green cotyledons injected with 35S: BcDFR showed anthocyanin accumulation (Fig. 7a).
The total anthocyanin content of the control was 0.028 mg/g, while that of the 35S: BcDFR cotyledons was 0.044-0.063mg/g (Fig. 7b).This result indicates that BcDFR expression promotes anthocyanin synthesis and accumulation.

Discussion
Non-heading Chinese cabbage [Brassica campestris (syn.Brassica rapa) ssp.chinensis] is a popular leafy vegetable.The accumulation of anthocyanins gives the leaves a purple color and also a higher nutritional value.In this study, we used laser capture frozen section method (LCM) to divide purple ZBC and green LBC leaves into upper and lower epidermis (Pup, Plow, Gup, Glow).RNA-seq analysis combined with functional verification indicates that BcDFR is a key gene regulating anthocyanin accumulation in non-heading Chinese cabbage leaves.

DFR is essential for anthocyanin synthesis and accumulation
Many enzymes are involved in anthocyanin biosynthesis [33].DFR plays an essential role as the first catalytic protein in the anthocyanin biosynthetic pathway.Under the help of ANS, UFGT, and NADPH, DFR selectively catalyzes dihydroquercetin, dihydrokaempferol, and dihydromyricetin formation of cyanidin-3glucoside, pelargonidin-3-glucoside, and delphinidin-3-glucoside [34].O'Reilly et al. [35] used the translocon labeling method to first isolate the DFR gene from maize and later obtain it from Petunia [36], and other species such as Gerbera [37], Ginkgo [38], and Scutellaria [39].DFR plays a crucial role in plant coloration.Previous studies have shown that BoDFR1 is a candidate gene for controlling red or purple leaves in Kale [39].The mutations in BoDFR1 render it non-functional, disrupting anthocyanin synthesis and accumulation [40].In our study, we combined RNA-seq and qRT-PCR to identify BcDFR, which is the most up-regulated gene in the upper epidermis of purple leaves.Sequence analysis of BcDFR in ZBC and LBC showed a deletion and an insertion fragment in the promoter region.It remains to be investigated how these variants affect BcDFR expression.

Anthocyanin specific accumulation of non-heading Chinese cabbage leaves
Anthocyanin is a key factor in determining the purple color phenotype of leaves.The color is determined by the accumulation of anthocyanin, which is primarily regulated by genes in the anthocyanin pathway [41].These genes have been relatively well studied, and we have a relatively accurate understanding of their function and expression patterns [42,43].In this work, the anthocyanins of purple ZBC only accumulated on the upper epidermis of the leaves, while the lower epidermis were green (Fig. 1) de.Traditional quantitative detection often targets the whole leaf and cannot differentiate the differences between the upper and lower epidermis of the leaf.Therefore, in this study, we used the LCM method for the first time to divide the leaves into upper and lower parts.Combined with transcriptome sequencing, it provides more comprehensive and accurate gene expression data for in-depth research on the specific accumulation of anthocyanins.
To determine whether anthocyanin enrichment is caused by the specific expression of the genes, we analyzed the gene expression of the anthocyanin synthesis pathway.We found that almost all of these anthocyaninrelated genes were up regulated in Pup (Fig. 4, S1), consistent with the phenotype of ZBC leaves.The predicted protein-protein interaction network of BcDFR suggested that BcDFR has the ability to interact with most all genes associated with the anthocyanin biosynthetic pathway, in addition to its association with MYB111, MYBL2, and TT8 (Fig. S5).Further, RNA-Seq analysis showed that MYBL111, TT8, and MYBL2 were substantially up-regulated in Pup (Fig. S6).Thus, specific expression of anthocyanin synthesis pathways and regulatory genes may be the reason for specific accumulation of anthocyanins.Interestingly, MYBL2 had been reported as a negatively regulated gene for anthocyanins [44], but was upregulated in Pup.These results need to be further explored.
In conclusion, we used LCM method to study DEGs that are specifically expressed in the upper epidermis of purple non-heading Chinese cabbage leaves.Finally, our study identified BcDFR, a key gene in the anthocyanin synthesis pathway.Our study provides new insights into the functional analysis and transcriptional regulatory network of anthocyanin-related genes in purple nonheading Chinese cabbage.
Four samples were sequenced by Illumina technology and aligned to the reference genome.After removing adapters and unknown or poor-quality reads, 40,441,648-44,345,740; 41,387,390-9,937,566; 38,134,220-44,624,234 and 39,410,282-42,125,688 clean reads remained in Pup, Plow, Gup and Glow respectively.The Q30 score for each sample was 93.40-94.29%and the mean GC content was 48.44% (Table

Fig. 1
Fig. 1 Phenotype and anthocyanin content of ZBC and LBC.a-b Phenotypes of ZBC and LBC.(c) Phenotype of ZBC after low temperature treatment.d-e Phenotypes of upper (d) and lower (e) epidermis of leaves.f Total anthocyanin content of ZBC, LBC and LTZ.Bar charts represent mean values and scatter plots represent individual data values.Z, ZBC; L, LBC; LTZ, ZBC after low temperature treatment.The data are from three independently repeated experiments.Error bars represent ± SD.Different letters indicate significant differences (p < 0.05)

Fig. 2
Fig. 2 Samples harvested by Laser capture microdissection method.a Gup, green LBC leaf upper epidermis.b Glow, green LBC leaf lower epidermis.c Pup, purple ZBC leaf upper epidermis.d Plow, purple ZBC leaf lower epidermis

Fig. 3
Fig. 3 Analysis of differentially expressed genes in transcriptome.a The number of DEGs in Pup/Plow, Pup/Gup, Plow/Glow, Gup/Glow, Pup/Glow, Plow/Gup.b Venn diagram of DEGs between Pup/Gup and Pup/Plow.c GO enrichment analysis of 1234 DEGs based on Pup/Gup and Pup/Plow results, showing the 10 most enriched GO terms.High and low P values are indicated in blue and red, respectively.d KEGG enrichment analysis of 1234 DEGs based on Pup/Gup and Pup/Plow results, showing the 14 most enriched KEGG terms.High and low P values are indicated in blue and red, respectively

Fig. 7
Fig. 7 Phenotype and anthocyanin content of BcDFR transient expression in LBC cotyledon.a Phenotypes of upper and lower epidermis LBC cotyledons after injection for 7 days.b Total anthocyanin content of CK and 35S: BcDFR LBC cotyledons.CK, empty vector controlling; 35S: BcDFR, BcDFR overexpression.The data are from three independently repeated experiments.Error bars represent±SD.Different letters denote significant differences (p < 0.05)