Integrated Metabolome and Transcriptome Analysis Uncovers the Role of Anthocyanin Metabolism in Michelia maudiae.
Flower coloration is one of the most attractive characteristics and quality traits of ornamental plants and has the ability to attract pollinators and seed distributors. Moreover, a broad spectrum of colors also provides an important aesthetic function for plants, especially in horticulture applications. Flower color is largely determined by the production of pigments, usually anthocyanins, betalains, or carotenoids. Anthocyanins are major contributors to flower colors. The primary shade of a flower color is mainly determined by the ratio of three different classes of anthocyanidins, namely, pelargonidin (orange to brick red), delphinidin (purple to blue), and cyanidin (red to pink to blue), and subsequent modifications to structure such as glycosylation, methylation, and acylation . Anthocyanin biosynthesis has been extensively studied in various horticultural plants, for example, Chrysanthemum grandiflorum , Begonia semperflorens , and Matthiola incana . Two major classes of genes are required for anthocyanin biosynthesis: structural genes and transcription factors (TFs). The structural genes encode the enzymes that are responsible for the biochemical reactions of anthocyanin synthesis. Anthocyanin biosynthesis starts from phenylalanine and produces colorless secondary intermediate metabolites that are sequentially catalyzed by a number of enzymes, including phenylalanine ammonia-lyase (PAL), cinnamic acid 4-hydroxylase (C4H), 4-coumarateCoA ligase (4CL), chalcone synthase (CHS), chalcone isomerase (CHI), flavanone 3-hydroxylase (F3H), flavanone 3'-hydroxylase (F3'H), flavanone 3'5'-hydroxylase (F3'5'H), and dihydroflavonol 4-reductase (DFR) . The colorless leucoanthocyanidins are then transformed into stable colored anthocyanins by anthocyanidin synthesis (ANS) and UDP-glucose flavonoid 3-O-glucosyltransferase (UFGT) resulting in a series of flower color polymorphisms . The TFs control the expression of the structural genes during anthocyanin biosynthesis. There are evidences that TFs MYB, basic helix-loop-helix (bHLH), and WDR play an important role in regulating the activity of the structural genes [7-9]. The coordinated expression of the structural genes and TFs lead to the accumulation of anthocyanins resulting in color variation.
The color mutants provide important materials to unravel the complex mechanism of flower pigmentation. The detailed knowledge of the events of the biosynthesis pathway is based on the research with color mutation forms of Zea mays, Petunia hybrida, and Arabidopsis thaliana [10-12]. In Arabidopsis, MYB75, MYB90, and bHLH have been reported to regulate the expression of DFR and ANS genes, thus affecting the biosynthesis of anthocyanin . In Vaccinium myrtillus, PAL, CHS, F3H, DFR, and ANS are all involved in color mutant forms . In Muscari armeniacum, flavonol synthase (FLS) and DFR are responsible for blue pigmentation . In recent years, combined large-scale sequencing analyses have been used to lucubrate on the pigmentation of flowers or fruits [3, 14, 15]. Integrated metabolome and transcriptome analyses have elucidated the expression of differentially expressed genes (DEGs) and the respective metabolic products [13, 16]. It is possible to gain a comprehensive analysis relating metabolites to gene expression profiles and get an insight into the variation mechanisms, which would not be possible from transcript analysis or metabolomic analysis alone.
Magnoliaceae, a primitive family of flowering plants, plays an important role in the studies of biogeography and evolution . Moreover, these plants are widely appreciated as ornamental shrubs and trees because the trees have a graceful appearance and produce beautiful flowers and they attract a great deal of attention from horticulturalists. Michelia maudiae Dunn., one species of Magnoliaceae, is a broad-leaved evergreen tree that has a fleshy stem that profusely branches and bears white flowers; it is popular as an ornamental plant in East and Southeast Asia, Central America, Southeast North America, and South America . The natural flower color of M. maudiae is pure white (Figure 1(a)). In 2009, a M. maudiae var. rubicund with a rubellis flower was found by Fan et al. in the wild  (Figure 1(b)). The variation of flower color in M. maudiae provides a higher ornamental value in the family of Magnoliaceae. However, the mechanism of flower pigmentation in the family Magnoliaceae is still far from conclusive.
In this study, metabolome and transcriptome analyses were performed using the early and later stages of the white and rubellis flowers of M. maudiae. Beyond identifying many differentially altered metabolites, the corresponding transcriptional changes of structural genes and TFs in the phenylpropanoid and flavonoid biosynthesis pathways have also been revealed. Our research provides valuable information to further elucidate the molecular basis of flower color formation in M. maudiae and to further elucidate the novel gene resources for Magnolia plant breeding.
2. Materials and Methods
2.1. Plant Materials. M. maudiae flowers were divided into seven stages according to the length of the tepals and the main characters of the flowers (Figure 1(c), Table S1). The tepals of M. maudiae from the early stage (stage 3) were collected at 10:00 a.m., November 2015, and the tepals from the later stage (stage 6) were collected at 10:00 a.m., in March 2016. Tepals from both stages were taken from the Xi'an Botanical Garden, Shaanxi Province, China. Three biological replicates were collected per sample, each with 30 representative tepals randomly selected from several trees in the garden. All samples were immediately frozen in liquid nitrogen and stored at -80[degrees]C for further studies.
2.2. Identification and Quantification of Anthocyanin Compositions. The anthocyanin compositions of the two stages of rubellis and white tepals were determined using high-performance liquid chromatography (HPLC) with three biological replicates. The freeze-dried tepals were crushed and 100 mg powder was extracted overnight at 4[degrees]C in 1.0 ml 70% methanol. After centrifugation at 10,000g for 10 min, the supernatant was filtrated through 0.22 [micro]m pores and awaited HPLC and mass spectrometry (LC-MS) analysis. A quality-control sample was made by equal blending of all samples and was run after every 10 injections to ensure the stability of the analytical conditions.
The filtered supernatant was injected into a HPLC system (Shim-pack UFLC SHIMADZU CBM30A, Kyoto, Japan) equipped with a Waters ACQUITY UPLC HSS T3 C18 column (1.8 [micro]m, 2.1 mm x 100 mm; Waters, Milford, MA, USA). The solvent system was ultrapure water (0.04% acetic acid) and acetonitrile (0.04% acetic acid). The gradient program was 100:0 at 0 min, 5:95 at 11.0 min, 5:95 at 12.0 min, 95:5 at 12.1 min, and 95:5 at 15.0 min. The flow rate was set at 0.40 ml/min. The temperature of the column was maintained at 40[degrees]C. The effluent was then scanned using an ESI-triple quadrupole-linear ion trap-MS/MS system (Applied Biosystems 4500 Q TRAP) as described previously [16, 20]. The results were further analyzed by annotating against public databases, including MassBank (http://www.massbank.jp/), HMDB (http://www.hmdb.ca/), KNAPSAcK (http://www.knapsackfamily.com/KNApSAcK/), METLIN (http://metlin.scripps.edu/index.php), and MoToDB (http://www.ab.wur.nl/moto/) to identify metabolites. Metabolite quantification was carried out using MRM (multiple reaction monitoring) . Partial least squares-discriminant analysis (PLS-DA) was performed with the identified metabolites . Compositions with fold change [greater than or equal to] 2 or [less than or equal to] 0.5 and the thresholds of variable importance in project (VIP) [greater than or equal to] 1 were defined as significant-difference metabolites.
2.3. RNA Extraction, Library Construction, and Transcriptome Sequencing. Total RNAs from the stage 3 and stage 6 of rubellis and white tepals were extracted using a plant RNA kit following the manufacturer's protocol (Omega Bio-tek Inc., Doraville, GA, USA). To confirm the reliability of the RNA-sequencing experiments, three replicates of the samples per developmental stage of the same color tepals were equally blended to obtain RNA extracts. The quality of RNA was evaluated using Bioanalyzer 2100 (Agilent Technologies Inc., CA, USA). Those samples with an RNA integrity number (RIN) above 6.6 were used for the preparation of cDNA libraries.
A total amount of 1 [micro]g RNA per sample was used to input material for library construction. Sequencing libraries were generated using Illumina's NEBNext[R]Ultra[TM] RNA Library Prep Kit (NEB, USA) according to the manufacturer's instructions. Briefly, mRNA was purified from total RNA using oligo (dT) magnetic beads. And then, mRNA was fragmented into small pieces by the addition of NEBNext First-Strand Synthesis Reaction Buffers (5x) under high temperature. Secondly, the fragments served as templates to synthesize first-strand cDNA using random hexamer primers and M-MuLV reverse transcriptase. Second-strand cDNA was subsequently synthesized using DNA polymerase I and RNaseH. Lastly, to preferentially select cDNA fragments 150~200 bp in length, the library fragments were purified with the AMPure XP system (Beckman Coulter, Beverly, USA). The size-selected, adaptor-ligated cDNA was used as templates for PCR amplification and assessed using the Bioanalyzer 2100 system. Four libraries (early and later tepals of rubellis flowers and early and later tepals of white flowers) were sequenced on the Illumina HiSeq[TM] 2000 platform. The stage 3 and stage 6 in M. maudiae Dunn are termed as "white flower" at early stage (WE) and later stage (WL), respectively. The stage 3 and stage 6 in the mutant are noted as "rubellis flower" at early stage (RE) and later stage (RL), respectively.
2.4. Transcriptome Assembly and Functional Annotation. Clean reads were obtained by removing adaptor sequences and low-quality reads from raw data and using these for all the subsequent analyses. Q20 and Q30 values and GC content of the clean data were calculated. Clean reads from all libraries were pooled together and then assembled using the software Trinity v2.8.0 [23, 24] with min_kmer_cov set to 3. To annotate the M. maudiae transcriptome, assembled sequences were aligned against the public dataset of non-redundant (Nr) databases, nucleotide sequence (Nt) database, the Swiss-Prot protein sequence database (SWISS-PROT), the Kyoto Encyclopedia of Genes and Genomes (KEGG) database, Gene Ontology (GO), and the Clusters of Orthologous Groups of proteins (COG) databases, respectively, using BLASTX with a threshold E value of <1x [10.sup.-5] [25-27].
2.5. Differential Expression Analysis of Unigenes. Clean reads of each sample were mapped back to the assembled transcriptome by expectation maximization . And then for FPKM, the fragments per kb per million reads of each unigene was calculated for estimating gene expression levels based on the mapping results . Differential expression analysis of two samples was performed using the DEGseq R package . DEGseq estimates differential expression from sequencing data based on the negative binomial distribution. The resulting Q values were adjusted using the Benjamini-Hochberg method for controlling the false discovery rate (FDR) . Genes with the criterion of Qvalue <0.01 (cut-off at 5% FDR) screened by DEGseq and an absolute value of log2 ratio [greater than or equal to] 2 were assigned as differentially expressed. GO and KEGG functional enrichment analysis of the DEGs was implemented using the GOseq R package based on the Wallenius noncentral hypergeometric distribution .
2.6. Phylogenetic Analysis. Putative R2R3-MYB sequences were extracted and aligned with the closely related R2R3-MYB protein using the program ClustalW. MEGA version 7 (MEGA7) was applied for phylogenetic inference. Maximum likelihood phylogenies were estimated using amino acid sequence alignments. The following GenBank accession numbers were used: AAS68190 (VvMYB5a), Q58QD0 (VvMYB5b), AAG42001 (AtMYB75), AM259485 (VvMYBA1), ACK56131 (VvMYBPA2), Q9FJA2 (AtMYB123), AAK84064 (FaMYB1), ACX50288 (VvMYBC2-L2), AHX24372 (PhMYB27), ABW34393 (vvMYBC2-L1) P81395 (AmMYB330), NP_849749 (AtMYB8), NP_192684 (AtMYB6), NP_179263 (AtMYB7), NP_195225 (AtMYB32), AAC83582 (AtMYB4), NP_ 001106009 (ZmMYB42), ADX33331 (PhMYB4), CAE09058 (EgMYB1), AEM17348 (PvMYB4a), ABL61515 (VvMYB4a), ACN94269 (VvMYB4b), and NP_001105949 (ZmMYB31).
2.7. Gene Validation and Expression Analysis. To validate the transcriptional abundance results from sequencing analysis, nine color-related unigenes were subjected to real-time quantitative PCR analysis. A total of 1 [micro]g RNA was reverse transcribed to first-strand cDNA using the HiScript[R] III RT SuperMix for qPCR kit (Vazyme Biotech Co., Ltd., China) according to the manufacturer's instructions. The reaction product was diluted 20-fold with sterile water for RT-qPCR analysis using a ChamQ[TM] Universal SYBR[R] qPCR Master Mix (Vazyme Biotech Co., Ltd., China). Real-time PCR was performed on a CFx96[TM] instrument (Bio-Rad, USA). The actin gene was used as the internal control for the normalization of gene expression. Primers were designed using Primer Premier 5 and listed in Table S2. The RT-qPCR reactions were carried out in 20 [micro]l volumes containing 0.4 [micro]l of each primer (concentration of 10 [micro]M), 10 [micro]l of ChamQ Universal SYBR qPCR Master Mix, and 2 [micro]l cDNA. The RT-qPCR cycling conditions were as follows: 95[degrees]C for 30 s, followed by 40 cycles of 95[degrees]C for 10 s, and 60[degrees]C for 30 s. A melting-curve analysis was performed with the default setting of the instrument. Relative expression was calculated by the [2.sup.-[DELTA][DELTA]CT] method. Values for mean expression and standard deviation were calculated from the results of three biological replicates.
3.1. Developmental Stage Division and Morphological Description of M. maudiae Flowers. For the convenience of further study, we divided M. maudiae flowers into seven developmental stages according to the length of the tepals and the main characteristics of the flowers (Table S1, Figure 1(c)). The pigmentation of the tepals begins at stage 3 and reaches the maximum level at stage 6, just ready to open (Figure 1(c)). At stage 3, both the tepals of the white flower and its rubellis mutant have a green appearance, with a slight rubellis at the bottom of the tepals in the rubellis mutant (Figure 1(d)). At stage 6, the flowers are about to bloom for a very short time, with superne rubellis and inferne atrophoeniceis in the tepals of the rubellis mutant (Figure 1(d)).
3.2. Metabolome Composition Analyses in M. maudiae. To examine the biochemical mechanism of the appearance of the rubellis color phenotype in M. maudiae, the metabolomic profiles of the tepals were compared. Firstly, the total contents of anthocyanins in the tepals of two strains were detected by a spectrophotometer (Figure S1). As expected, total anthocyanins were highly accumulated in RE and RL but not in WE, and only a bit was detected in WL. The result corresponded well with the pigmentation of flowers in M. maudiae.
Furthermore, to gain additional information on the coloration of M. maudiae flowers, the general secondary metabolites were identified in the samples WL and RL. The metabolite profiles of WL and RL showed significant differences (Figure S2). A total of 149 metabolites were identified from these two samples with three biological replicates, among which 48 differentially altered metabolites (DAM) were identified, including 7 anthocyanins, 5 catechin derivatives, 2 flavanone, 17 flavones, 4 flavonols, 3 isoflavones, 8 flavone C-glycosides, and 2 proanthocyanidins (Table 1).
Almost differentially altered anthocyanins demonstrated significantly higher contents in the later stage of rubellis tepals. Compared with white tepals, peonidin O-hexoside was found with 3,381-fold increments; cyanidin 3-O-glucoside and cyanidin 3,5-O-diglucoside were detected with 157-fold and 322-fold increments, respectively; and cyanidin O-syringic acid and pelargonidin 3-O-glucoside were detected at significantly higher concentrations in rubellis flowers but at barely detectable levels in white flowers, which could explain the rubellis color of the mutant. Surprisingly, one derivative of cyanidin, rosinidin O-hexoside, was detected at significantly higher levels in white tepals but was detected at a level so low as to barely be detectable in rubellis tepals (Table 1).
The derivatives of catechin are also widely distributed in plants. Catechin and epicatechin were the two major catechin derivatives presented in white tepals, while the rubellis-flowered strain contained epigallocatechin, protocatechuic acid, and protocatechuic aldehyde with significantly higher levels (Table 1).
Compared to the color components in rubellis flowers, the colorless A- and B-type procyanidins were markedly upregulated in white flowers (Table 1). Procyanidin B3 in WL was upregulated over 25-fold compared to that in RL. The content of procyanidin A2 was far greater in white flowers, but the content was barely detectable in rubellis tepals.
In addition, the rubellis tepals also contain several flavones, flavonols, and isoflavones. These components are either colorless or slightly yellow. For the flavones, several derivatives of tricin, including tricin O-vanilloyl hexoside, tricin di-O-hexoside, tricin O-rhamnoside, tricin O-hexosylO-syringin alcohol, and tricin 5-O-hexoside, were remarkably upregulated in the rubellis tepals vs. the white tepals. In addition, an intermediate in the tricin biosynthesis, tricetin O-rutinoside, was also markedly upregulated in the rubellis tepals. Hence, tricin may play an essential role in rubellis tepals.
3.3. De Novo Assembly and Functional Annotation. To further elucidate the matter on M. maudiae discussed above at the molecular level, four libraries from stage 3 and stage 6 of white and rubellis tepals (WE, WL, RE, and RL) were conducted, resulting in approximately ~41 million clean reads for each sample (Table S3). The Q20 and Q30 of each sample were greater than or equal to 96.9% and 93%, respectively. The GC contents accounted for ~47% of these reads. De novo assembly of these clean reads resulted in 109,729 nonredundant unigenes with a mean length of 925 bp (Figure S3). The N50 value was 1,611 bp and 90% had a length of at least 352 bp. All these data indicated that the sequencing quality was high enough for further study.
Nearly 37.64% of unigenes had top matches to sequences from Nelumbo nucifera. The other five top-hit species were Vitis vinifera, Elaeis guineensis, Phoenix dactylifera, Amborella trichopoda, and Gossypium raimondii (Figure S4). Based on GO annotation, 96,164, 58,324, and 35,013 unigenes were successfully assigned into three main GO categories: biological process, cellular component, and molecular function (Figure S5). A search against the COG database resulted in the classification of 39,811 unigenes into 25 COG categories (Figure S6). The maps with the highest unigene representation were "General function prediction only" (6,540 unigenes), followed by "Transcription" (3,898 unigenes), "Replication, recombination, and repair" (3,566 unigenes), "Signal transduction mechanisms" (2,811 unigenes), and "Posttranslational modification, protein turnover, and chaperones" (2,837 unigenes).
Clean data from the above four libraries were normalized to FPKM values to calculate the expression of unigenes in M. maudiae. Unigenes that are differentially expressed in pairs of two stages of different color strains are shown in Figure 2. As the flower develops from the closed bud of stage 3 to stage 6, 19,581 and 28,002 unigenes are differentially expressed in white and rubellis flowers, respectively, indicating significant variation in gene expression profiles during the blooming of flower development. Specifically, the comparison of WE and RE and WL and RL resulted in 29,799 and 28,858 DEGs, respectively.
3.4. Analysis of DEGs Related to Phenylpropanoid, Flavonoid, and Anthocyanidin Biosynthetic Pathways in M. maudiae. The increase in the contents of anthocyanins in rubellis flowers let us focus on the transcriptional changes in the phenylpropanoid, flavonoid, and anthocyanidin biosynthesis pathways that are related to flower pigmentation. Most of the biosynthesis pathways were strengthened by the upregulation of gene expression at the later stage of rubellis flowers (Figure 3). For the comparison of WE and RE, the transcript abundance of genes PAL (unigene20142, unigene35278), 4CL (unigene48319), CHS (unigene35471), FLS (unigene56358), DFR (unigene 16496), and UFGT (unigene42116) were markedly upregulated in rubellis tepals than in white tepals, while one CHS-like gene (unigene49321), one PAL (unigene29365), and one anthocyanidin reductase (ANR, unigene28164) were downregulated in the rubellis tepals. At the later stage, the levels of transcripts encoding the secondary metabolite biosynthesis enzymes, including three PAL (unigene20142, unigene29365, and unigene35278), two 4CL (unigene62335, unigene973), one CHS (unigene49321), one DFR (unigene16496), and one ANS (unigene16058), were markedly abundant in rubellis tepals compared with those in white tepals (Figure 3). It is noteworthy that the transcripts of DFR-like sequences presented higher levels of gene transcripts in rubellis tepals during the blooming of flowers (Figure 3). Notably, unigene28164 with an ANR-like sequence was downregulated 2.43-fold and 5.80-fold in the rubellis flower in early and later stages, respectively.
3.5. Transcription Factor Analyses. The TFs play an important role in regulating the expression of the structural genes during anthocyanin biosynthesis [33, 34]. Putative TFs were searched in the M. maudiae transcripts by HMMER software using the "Plant Transcription Factor Database" (http://planttfdb.cbi.pku.edu.cn/). Setting FDR <0.0001 together with the absolute value of [log.sub.2] ratio [greater than or equal to] 2 as thresholds for significant differences, 120 and 105 genes were identified as TFs in WE vs. RE and WL vs. RL, respectively. And there were 105 and 161 differentially expressed TFs identified in WE vs. WL and RE vs. RL, respectively (Table S4). The differentially expressed TFs here contained the family MYB, bHLH, AP2, ERF, WRKY, and HSF.
Nine and five R2R3-MYBs were differentially expressed in WE vs. RE and WL vs. RL, respectively (Figure 4(a)). Among the MYBs in the comparison of WE and RE, three genes were identified upregulated and six downregulated in RE. In the later stage of flower development, two genes were found upregulated and three downregulated in RE. We found one MYB DEG (unigene48479) expressed at very high levels with FPKM [greater than or equal to] 470 and FPKM [greater than or equal to] 360 in RE and RL, respectively, but with low FPKM values in WE and WL. R2R3MYBs also operate as positive and negative regulators. Six R2R3-MYBs were closely related to the C2 repressor motif clade in phylogenetic analyses (Figure 4(b)), which might play a key role in negatively regulating the level of anthocyanins. These R2R3-MYB proteins from the C2 repressor motif clade all have the EAR repression domain (Figure S7).
Twenty-five bHLH were found in WE vs. RE (13 upregulated, 12 downregulated, Table S4). Twenty-six bHLH were markedly different in WL vs. RL (15 upregulated, 11 downregulated). For the comparison of WE vs. WL, 31 bHLH revealed differences, including 8 upregulated and 23 downregulated. During flower blooming, 51 bHLH DEGs were screened in RE vs. RL with 32 downregulated and 19 upregulated. Among these bHLH DEGs, 6 (unigene47744, unigene49005, unigene55012, unigene55013, unigene9769, and unigene54696) transcripts had markedly higher levels in both WE vs. RE and WL vs. RL. 4 (unigene95320, unigene6259, unigene41244, and unigene10054) bHLH DEGs had decreasing expression in both WE vs. RE and WL vs. RL.
3.6. RT-qPCR Validation of the Transcriptomic Data. To validate the transcriptome results, nine color-related DEGs were selected for RT-qPCR analysis. The expression patterns of these genes were very similar to the sequencing results (Figure 5). A significant correlation (R > 0.84) was found between the two methods, indicating that the relevance of the RNA-seq data and RT-qPCR showed good consistency (Figure S8).
Color mutations play an important role in plant breeding and horticulture. The complex metabolic mechanism of the mutants is beginning to be revealed using metabolomic and high-throughput sequencing methods . In this study, the combined analysis of metabolites and transcriptomes of two M. maudiae that differ in color was carried out, providing large-scale information on metabolic product profiles and the underlying modifications in gene expression profiles.
Flower development and pigmentation are tightly linked with anthocyanin biosynthesis in most cases [35, 36]. The primary shade of flower color is mainly determined by three biosynthesis reactions: pelargonidin, delphinidin, and cyanidin . Herein, most derivatives of anthocyanidins were highly significantly accumulated in the rubellis flowers, while the colorless procyanidin and other catechin derivatives were highly accumulated in white flowers. This finding provides a global view of the large-scale secondary metabolite changes for color maturation in M. maudiae. Although the level of anthocyanins was limited, tepal extracts of white flowers contained all of the most core anthocyanidin compositions that had been detected in rubellis flowers, including the derivatives of pelargonidin, delphinidin, and cyanidin. This suggested that the lack of anthocyanin in white tepals cannot be due to any blockage of the corresponding reactions. There might be multiple reasons for the loss of pigmentation. The cyanidin anthocyanins mainly tend to exist as a form of 3,5-O-diglucoside, O-syringic acid, and 3-O-glucoside and peonidin O-hexoside in rubellis tepals, while they were mainly present in the white-flowered strain with rosinidin O-hexoside, the colorless procyanidin, and other catechin derivatives. Cyanidin derivatives have been identified as being the major anthocyanins present in the skin of red apples, and also in the purplish-red, bronze, and pink Chrysanthemum inflorescences [37, 38]. Hence, we inferred that cyanidin and cyanidin derivatives might be the target compound for determining the pigmentation of the rubellis tepals in M. maudiae. Surprisingly, the levels of one derivative of cyanidin, rosinidin O-hexoside, were markedly higher in white tepals but were barely detectable in the rubellis flower strain. Rosinidin has been found in Primula rosea  and Catharanthus roseus  as a pigment. Further studies are needed to determine the role of rosinidin O-hexoside in M. maudiae. Catechin and epicatechin were increasingly expressed in white tepals. The units of catechin and epicatechin can compose proanthocyanidins, an important quality component of many fruits . The levels of all of the proanthocyanidins that were identified in rubellis tepal extracts were markedly higher in white tepals. It could be concluded that the middle-stream metabolites in the leucoanthocyanidins or biosynthetic anthocyanidins were also present in the white tepals but were mainly converted to colorless proanthocyanidins.
In the synthesis of pelargonidin, the color compound peloargonidin-3-D-glucoside was abundant in rubellis tepals. In the synthesis of delphinidin, the increasing expression of epigallocatechin in rubellis tepals might partially block the synthesis of color delphinidin derivatives and cause the production of other colorless compounds. It has been explained why blue is not the predominant color hue in M. maudiae. It is the conclusion that the metabolism of cyanidin plays a vital role in the flower coloration system of M. maudiae, whereas the metabolism of pelargonidin and delphinidin is less significant, resulting in the appearance of rubellis pigments.
For other metabolite compounds, several forms of tricin were markedly upregulated in rubellis flowers. Tricin is widely recognized as a valuable health compound due to its anti-inflammatory, antioxidant, and cardioprotective potentials . The biosynthesis of tricin starts from naringenin and produces apigenin, which requires further modifications to generate tricin. Apigenin is converted to luteolin by F3'H and then to chrysoeriol by an OMT. Chrysoeriol generates selgin and then sequentially leads to tricin formation . The lower expression of F3'H was found in white tepals and might result in the low content of the derivatives of tricin. Recent studies mainly focused on the role of tricin in lignification [43,44]. Further studies are needed to determine the role of tricin O-linked conjugates in pigmentation.
Genes PAL, C4H, 4CL, CHS, F3H, DFR, F3'H, ANR, FLS, and ANS in phenylpropanoid and flavonoid biosynthesis pathways were identified in the transcriptome. Most of these genes were upregulated in rubellis tepal, except ANR, F3'H, and UFGT. It is notable that the level of gene transcripts of DFR-like sequences was higher in rubellis flowers not only in the early stage but also in the later stage. DFR reduces dihydroflavonols to colorless leucoanthocyanidins, which are catalyzed by ANS to colored anthocyanidin. The suppression of the DFR expression in Dianthus caryophyllus L. or the low expression in Arabidopsis restricted their pigmentation [45, 46]. The transcripts of DFR were abundant in rubellis tepals, but only a trace of expression was detected in the white tepals, suggesting that DFR may be important for pigmentation probably through a quantitative regulatory system rather than through the absolute blockage of anthocyanin biosynthesis, similar to the coloration of Chrysanthemum flowers . Recently, studies have shown that the overexpression of McDFR increased anthocyanin production, resulting in red-leaf and red fruit peel phenotypes in crabapples (Malus spp.) . The upregulation of DFR could increase the production of anthocyanins in rubellis tepals. Another enzyme, ANS, was encoded by unigene16058_All, which shared the highest identities (99%) with the ANS gene of Magnolia sprengeri in GenBank. ANS, one of the dioxygenases, can catalyze the formation of colored anthocyanidins from leucoanthocyanidins, which might result in the accumulation of anthocyanin. So if ANS reactions are strongly enhanced, the reaction of colored anthocyanin production is effectively positive. One ANR (unigene28164) was increasingly expressed in white flowers. The introduction of ANR genes into tobacco (Nicotiana tabacum cv. Petite Havana SR1) resulted in the loss of anthocyanins and in the accumulation of catechin and epicatechin contents . Transcript accumulation of ANR genes and the abundance of the content of catechin and epicatechin in white flowers in our result correspond well with this phenomenon in tobacco transgenic flowers. Proanthocyanidins are phenolic polymers of condensed flavan-3-ols and are the major compounds in higher plants. Overexpression of ANR genes in tobacco have been reported to promote the biosynthesis of proanthocyanidins in flowers . What's more, a negative correlation between anthocyanin and proanthocyanidin biosynthesis was also found during leaf and fruit peel development in Malus crabapple . Similarly, the levels of two types of proanthocyanidins (Procyanidins A2 and B3) were significantly higher in white flowers than those of the rubellis flowers.
Plants require complex regulatory mechanisms to make sure that the level of anthocyanin pigmentation is appropriate for environmental changes. The degree of anthocyanins is transcriptionally controlled by positive and negative regulators from the MYB, bHLH, and WD repeat (MBW) families. Three R2R3-MYB members (unigene62902, unigene75793, and unigene75794) that clustered with VvMYBC2-L1 and PhMYB27 were expressed at lower levels in RE. In V. vinifera, VvMYB4a and VvMYB4b showed a severe reduction in phenolic compounds, and VvMYBC2-L1 and VvMYBC2-L3 may play a key role in negatively regulating the flavonoid biosynthesis, balancing the inductive effects of positive regulators . Experiments with P. hybrida indicate that PhMYB27 can act as an anthocyanin repressor and repress transcription through its EAR repression domain . Thus, unigene62902, unigene75793, and unigene75794 might be closely related to color morph regulation in M. maudiae. Unigene48479 was closely related to AmMYB330, AtMYB6, and AtMYB8. Overexpression of AmMYB330 from the flowers of Antirrhinum has been proven to inhibit phenylpropanoid metabolism in transgenic tobacco (Nicotiana tabacum cv Samsun NN) plants . Unigene49636 and unigene16157 clustered with ZmMYB31, which showed a significantly reduced lignin content and negatively regulated several genes in the phenylpropanoid pathway . This illustrates that unigene48479, unigene49636, and unigene16157 might be related to the regulation of the phenylpropanoid biosynthesis in M. maudiae.
Our metabolomic and transcriptomic data provide novel insights in understanding the modulated metabolites and gene expression in the rubellis mutant. Differentially altered metabolites associated with flower pigmentation were identified between white and rubellis tepals. Putative homologues of color-related genes involved in the phenylpropanoid and flavonoid biosynthesis pathways were identified in the transcriptome. The changes of anthocyanins and the discovery of genes associated with their biosynthesis and metabolism in M. maudiae are interesting and necessary for further studies.
All sequence data were deposited at the NCBI in the Sequence Read Archive (SRA) database under the accession numbers SRR8173854, SRR8173855, SRR8173856, and SRR8173857 for WE, WL, RE, and RL, respectively.
Conflicts of Interest
The authors declare no conflict of interest regarding the publication of this paper.
Lingfei Li and Shouzhou Zhang planned and designed the research; Xiaoan Lang, Na Li, and Lingfei Li collected plant materials and performed experiments; Na Li, Lingfei Li, and Xiaoan Lang analyzed data; Na Li and Lingfei Li wrote the manuscript; and Shouzhou Zhang supervised the research. All authors read and approved the final manuscript. Xiaoan Lang and Na Li contributed equally to this work.
We thank Liangsheng Wang (Institute of Botany, Chinese Academy of Sciences, China) for kind and helpful comments and discussion. This work was supported by funds from the Public Welfare Forestry Industry Project of State Forestry Administration, China (no. 201504322), National Natural Science Foundation of China (no. 31601784), and the Scientific Project of the Shenzhen Urban Administration (no. 201917).
Supplementary 1. Figure S1: total content of anthocyanins detected by a spectrophotometer.
Supplementary 2. Figure S2: the metabolite profiles of a white flower and a rubellis flower at the later stage.
Supplementary 3. Figure S3: length distribution of sequencing reads and contigs of M. maudiae.
Supplementary 4. Figure S4: the top-hit species distribution of M. maudiae.
Supplementary 5. Figure S5: histogram representing Gene Ontology (GO) classification of M. maudiae.
Supplementary 6. Figure S6: histogram representing Clusters of Orthologous Groups of proteins (COG) classification of all unigenes in M. maudiae.
Supplementary 7. Figure S7: consensus sequences of C2 repressor motif clade identified by MEME Suite.
Supplementary 8. Figure S8: correlation of the expression levels of nine differentially expressed genes by RNA-seq and RT-qPCR analysis.
Supplementary 9. Table S1: developmental stages of flowers in M. maudiae.
Supplementary 10. Table S2: primers used in quantitative real-time PCR.
Supplementary 11. Table S3: summary of RNA-seq results.
Supplementary 12. Table S4: number of differentially expressed transcription factors in comparison groups.
 Y. Tanaka, S. Tsuda, and T. Kusumi, "Metabolic engineering to modify flower color," Plant and Cell Physiology, vol. 39, no. 11, pp. 1119-1126, 1998.
 S. -M. Chen, C. -H. Li, X. -R. Zhu et al., "The identification of flavonoids and the expression of genes of anthocyanin biosynthesis in the chrysanthemum flowers," Biologia Plantarum, vol. 56, no. 3, pp. 458-464, 2012.
 J. Wang, M. Guo, Y. Li, R. Wu, and K. Zhang, "High-through-put transcriptome sequencing reveals the role of anthocyanin metabolism in Begonia semperflorens under high light stress," Photochemistry and Photobiology, vol. 94, no. 1, pp. 105-114, 2018.
 V. Hemleben, A. Dressel, B. Epping, R. Lukacin, S. Martens, and M. Austin, "Characterization and structural features of a chalcone synthase mutation in a white-flowering line of Matthiola incanaR. Br. (Brassicaceae)," Plant Molecular Biology, vol. 55, no. 3, pp. 455-465, 2004.
 L. Jaakola, K. Maatta, A. M. Pirttila, R. Torronen, S. Karenlampi, and A. Hohtola, "Expression of genes involved in anthocyanin biosynthesis in relation to anthocyanin, proanthocyanidin, and flavonol levels during bilberry fruit development," Plant Physiology, vol. 130, no. 2, pp. 729-739, 2002.
 Y. Tanaka, N. Sasaki, and A. Ohmiya, "Biosynthesis of plant pigments: anthocyanins, betalains and carotenoids," Plant Journal, vol. 54, no. 4, pp. 733-749, 2008.
 I. Hichri, S. C. Heppel, J. Pillet et al., "The basic helix-loop-helix transcription factor MYC1 is involved in the regulation of the flavonoid biosynthesis pathway in grapevine," Molecular Plant, vol. 3, no. 3, pp. 509-523, 2010.
 A. M. Takos, F. W. Jaffe, S. R. Jacob, J. Bogs, S. P. Robinson, and A. R. Walker, "Light-induced expression of a MYB gene regulates anthocyanin biosynthesis in red apples," Plant Physiology, vol. 142, no. 3, pp. 1216-1232, 2006.
 Y. Pang, J. P. Wenger, K. Saathoff et al., "A WD40 repeat protein from Medicago truncatula is necessary for tissue-specific anthocyanin and proanthocyanidin biosynthesis but not for trichome development," Plant Physiology, vol. 151, no. 3, pp. 1114-1129, 2009.
 K. Saito, K. Yonekura-Sakakibara, R. Nakabayashi et al., "The flavonoid biosynthetic pathway in Arabidopsis: structural and genetic diversity," Plant Physiology and Biochemistry, vol. 72, pp. 21-34, 2013.
 C. Hu, Q. Li, X. Shen et al., "Characterization of factors underlying the metabolic shifts in developing kernels of colored maize," Scientific Reports, vol. 6, no. 1, article 35479, 2016.
 S. Tsuda, Y. Fukui, N. Nakamura et al., "Flower color modification of Petunia hybrida commercial varieties by metabolic engineering," Plant Biotechnology, vol. 21, no. 5, pp. 377-386, 2004.
 Q. Lou, Y. Liu, Y. Qi et al., "Transcriptome sequencing and metabolite analysis reveals the role of delphinidin metabolism in flower colour in grape hyacinth," Journal of Experimental Botany, vol. 65, no. 12, pp. 3157-3164, 2014.
 S. Bai, T. Saito, D. Sakamoto, A. Ito, H. Fujii, and T. Moriguchi, "Transcriptome analysis of Japanese pear (Pyrus pyrifolia Nakai) flower buds transitioning through endodormancy," Plant & Cell Physiology, vol. 54, no. 7, pp. 1132-1151, 2013.
 C. Feng, M. Chen, C. J. Xu et al., "Transcriptomic analysis of Chinese bayberry (Myrica rubra) fruit development and ripening using RNA-Seq," BMC Genomics, vol. 13, no. 1, p. 19, 2012.
 Z. Wang, Y. Cui, A. Vainstein, S. Chen, and H. Ma, "Regulation of fig (Ficus carica L.) fruit color: metabolomic and transcriptomic analyses of the flavonoid biosynthetic pathway," Frontiers in Plant Science, vol. 8, article 1990, 2017.
 D. Cicuzza and A. Newton, The Red List of Magnoliaceae, IUCN, 2007.
 Y. Sun, Y. F. Liu, J. Wang, Y. Guo, and H. W. Huang, "Ten polymorphic microsatellite markers in Michelia maudiae (Magnoliaceae)," American Journal of Botany, vol. 97, no. 12, pp. e157-e158, 2010.
 J. C. Fan, T. P. Yi, and Z. Z. Luo, "A new mutation of Michelia maudiae species of Magnoliaceae in China," Journal of Sichuan Forestry Science and Technology, vol. 30, no. 4, p. 76, 2009.
 W. Chen, L. Gong, Z. Guo et al., "A novel integrated method for large-scale detection, identification, and quantification of widely targeted metabolites: application in the study of rice metabolomics," Molecular Plant, vol. 6, no. 6, pp. 1769-1780, 2013.
 C. G. Fraga, B. H. Clowers, R. J. Moore, and E. M. Zink, "Signature-discovery approach for sample matching of a nerve-agent precursor using liquid chromatography-mass spectrometry, XCMS, and chemometrics," Analytical Chemistry, vol. 82, no. 10, pp. 4165-4173, 2010.
 U. Lutz, R. W. Lutz, and W. K. Lutz, "Metabolic profiling of glucuronides in human urine by LC-MS/MS and partial least-squares discriminant analysis for classification and prediction of gender," Analytical Chemistry, vol. 78, no. 13, pp. 4564-4571, 2006.
 M. G. Grabherr, B. J. Haas, M. Yassour et al., "Full-length transcriptome assembly from RNA-Seq data without a reference genome," Nature Biotechnology, vol. 29, no. 7, pp. 644-652, 2011.
 B. J. Haas, A. Papanicolaou, M. Yassour et al., "De novo transcript sequence reconstruction from RNA-seq using the Trinity platform for reference generation and analysis," Nature Protocols, vol. 8, no. 8, pp. 1494-1512, 2013.
 J. Ye, L. Fang, H. Zheng et al., "WEGO: a web tool for plotting GO annotations," Nucleic Acids Research, vol. 34, Supplement_2, pp. W293-W297, 2006.
 A. Conesa, S. Gotz, J. M. Garcia-Gomez, J. Terol, M. Talon, and M. Robles, "Blast2GO: a universal tool for annotation, visualization and analysis in functional genomics research," Bioinformatics, vol. 21, no. 18, pp. 3674-3676, 2005.
 M. Kanehisa, M. Araki, S. Goto et al., "KEGG for linking genomes to life and the environment," Nucleic Acids Research, vol. 36, Supplement_1, pp. D480-D484, 2008.
 B. Li and C. N. Dewey, "RSEM: accurate transcript quantification from RNA-Seq data with or without a reference genome," BMC Bioinformatics, vol. 12, no. 1, p. 323, 2011.
 G. P. Wagner, K. Kin, and V. J. Lynch, "Measurement of mRNA abundance using RNA-Seq data: RPKM measure is inconsistent among samples," Theory in Biosciences, vol. 131, no. 4, pp. 281-285, 2012.
 L. Wang, Z. Feng, X. Wang, X. Wang, and X. Zhang, "DEGseq: an R package for identifying differentially expressed genes from RNA-Seq data," Bioinformatics, vol. 26, no. 1, pp. 136-138, 2010.
 Y. Benjamini and Y. Hochberg, "Controlling the false discovery rate: a practical and powerful approach to multiple testing," Journal of the Royal Statistical Society: Series B (Methodological), vol. 57, no. 1, pp. 289-300, 1995.
 M. D. Young, M. J. Wakefield, G. K. Smyth, and A. Oshlack, "Gene ontology analysis for RNA-seq: accounting for selection bias," Genome Biology, vol. 11, no. 2, article R14, 2010.
 H. Chu, J. C. Jeong, W. J. Kim et al., "Expression of the sweet-potato R2R3-type IbMYB1a gene induces anthocyanin accumulation in Arabidopsis," Physiologia Plantarum, vol. 148, no. 2, pp. 189-199, 2013.
 W. Xu, C. Dubos, and L. Lepiniec, "Transcriptional control of flavonoid biosynthesis by MYB-bHLH-WDR complexes," Trends in Plant Science, vol. 20, no. 3, pp. 176-185, 2015.
 S. G. Shi, M. Yang, M. Zhang, P. Wang, Y. X. Kang, and J. J. Liu, "Genome-wide transcriptome analysis of genes involved in flavonoid biosynthesis between red and white strains of Magnolia sprengeri pamp," BMC Genomics, vol. 15, no. 1, p. 706, 2014.
 G. F. Tsanakas, M. E. Manioudaki, A. S. Economou, and P. Kalaitzis, "De novo transcriptome analysis of petal senescence in Gardenia jasminoides Ellis," BMC Genomics, vol. 15, no. 1, p. 554, 2014.
 K. E. Schwinn, K. R. Markham, and N. K. Giveno, "Floral flavonoids and the potential for pelargonidin biosynthesis in commercial chrysanthemum cultivars," Phytochemistry, vol. 35, no. 1, pp. 145-150, 1993.
 C. Honda, N. Kotoda, M. Wada et al., "Anthocyanin biosynthetic genes are coordinately expressed during red coloration in apple skin," Plant Physiology and Biochemistry, vol. 40, no. 11, pp. 955-962, 2002.
 T. Iwashina, "The structure and distribution of the flavonoids in plants," Journal of Plant Research, vol. 113, no. 3, pp. 287-299, 2000.
 K. Toki, N. Saito, Y. Irie, F. Tatsuzawa, A. Shigihara, and T. Honda, "7-O-Methylated anthocyanidin glycosides from Catharanthus roseus," Phytochemistry, vol. 69, no. 5, pp. 1215-1219, 2008.
 J. Bogs, M. O. Downey, J. S. Harvey, A. R. Ashton, G. J. Tanner, and S. P. Robinson, "Proanthocyanidin synthesis and expression of genes encoding leucoanthocyanidin reductase and anthocyanidin reductase in developing grape berries and grapevine leaves," Plant Physiology, vol. 139, no. 2, pp. 652-663, 2005.
 J. M. Duarte-Almeida, G. Negri, A. Salatino, J. E. de Carvalho, and F. M. Lajolo, "Antiproliferative and antioxidant activities of a tricin acylated glycoside from sugarcane (Saccharum officinarum) juice," Phytochemistry, vol. 68, no. 8, pp. 1165-1171, 2007.
 P. Y. Lam, H. Liu, and C. Lo, "Completion of tricin biosynthesis pathway in rice: cytochrome P450 75B4 is a unique chrysoeriol 5'-hydroxylase," Plant Physiology, vol. 168, no. 4, pp. 1527-1536, 2015.
 W. Lan, F. Lu, M. Regner et al., "Tricin, a flavonoid monomer in monocot lignification," Plant Physiology, vol. 167, no. 4, pp. 1284-1295, 2015.
 K. Stich, T. Eidenberger, F. Wurst, and G. Forkmann, "Enzymatic conversion of dihydroflavonols to flavan-3,4-diols using flower extracts of Dianthus caryophyllus L. (carnation)," Planta, vol. 187, no. 1, pp. 103-108, 1992.
 D. N. Feyissa, T. L0vdal, K. M. Olsen, R. Slimestad, and C. Lillo, "The endogenous GL3, but not EGL3, gene is necessary for anthocyanin accumulation as induced by nitrogen depletion in Arabidopsis rosette stage leaves," Planta, vol. 230, no. 4, pp. 747-754, 2009.
 J. Tian, Z.-y. Han, J. Zhang, Y. J. Hu, T. Song, and Y. Yao, "The balance of expression of dihydroflavonol 4-reductase and flavonol synthase regulates flavonoid biosynthesis and red foliage coloration in crabapples," Scientific Reports, vol. 5, no. 1, article 12228, 2015.
 Y. Han, S. Vimolmangkang, R. E. Soria-Guerra, and S. S. Korban, "Introduction of apple ANR genes into tobacco inhibits expression of both CHI and DFR genes in flowers, leading to loss of anthocyanin," Journal of Experimental Botany, vol. 63, no. 7, pp. 2437-2447, 2012.
 H. Li, J. Tian, Y. Y. Yao et al., "Identification of leucoanthocyanidin reductase and anthocyanidin reductase genes involved in proanthocyanidin biosynthesis in Malus crabapple plants," Plant Physiology and Biochemistry, vol. 139, pp. 141-151, 2019.
 E. Cavallini, J. T. Matus, L. Finezzo et al., "The phenylpropanoid pathway is controlled at different branches by a set of R2R3-MYB C2 repressors in grapevine," Plant Physiology, vol. 167, no. 4, pp. 1448-1470, 2015.
 N. W. Albert, K. M. Davies, D. H. Lewis et al., "A conserved network of transcriptional activators and repressors regulates anthocyanin pigmentation in eudicots," Plant Cell, vol. 26, no. 3, pp. 962-980, 2014.
 L. Tamagnone, A. Merida, A. Parr et al., "The AmMYB308 and AmMYB330 transcription factors from Antirrhinum regulate phenylpropanoid and lignin biosynthesis in transgenic tobacco," Plant Cell, vol. 10, no. 2, pp. 135-154, 1998.
 S. Fornale, X. Shi, C. Chai et al., "ZmMYB31 directly represses maize lignin genes and redirects the phenylpropanoid metabolic flux," Plant Journal, vol. 64, no. 4, pp. 633-644, 2010.
Xiaoan Lang, Na Li, Lingfei Li [ID], and Shouzhou Zhang [ID]
Key Laboratory of Southern Subtropical Plant Diversity, Fairy Lake Botanical Garden, Shenzhen & Chinese Academy of Sciences, Shenzhen 518004, China
Correspondence should be addressed to Lingfei Li; firstname.lastname@example.org and Shouzhou Zhang; email@example.com
Received 7 March 2019; Revised 10 July 2019; Accepted 23 August 2019; Published 3 November 2019
Academic Editor: Antonio Ferrante
Caption: Figure 1: Flower morphology of Michelia maudiae. (a) "White" flower of M. maudiae Dunn. (b) "Rubellis" flower mutation of M. maudiae Dunn var. rubicund. (c) Different developmental stages of the two types of M. maudiae. (d) Definition and characterization of experimental materials for transcriptome and metabolome data. WE: "white flower" at an early stage (stage 3); WL: "white flower" at a later stage (stage 6); RE: "Rubellis flower" at an early stage (stage 3); RL: "Rubellis flower" at a later stage (stage 6).
Caption: Figure 2: Differentially expressed genes between early and later stages of "white flowers" and "rubellis flowers." (a) Numbers of DEGs. (b) Venn diagram of DEGs. WE: "white flower" at early stage (stage 3); WL: "white flower" at later stage (stage 6); RE: "Rubellis flower" at early stage (stage 3); RL: "Rubellis flower" at later stage (stage 6).
Caption: Figure 3: Metabolomic and transcript profiling in the phenylpropanoid and flavonoid biosynthetic pathways in "white flower" and "rubellis flower." Grids with a color-scale from yellow to black represent FPKM values 1-10,10-20, 20-40,40-80, 80-160, 160- 320, 320-640,640-1,280, 1,280-2,560, and over 2,560, respectively. The numbers in square brackets after the grid represent [log.sub.2]- fold values of unigenes WE/RE and WL/RL, respectively. Compared with WL, in RL upregulated metabolites were in red boxes and downregulated metabolites were in blue boxes. PAL: phenylalanine ammonia-lyase; C4H: cinnamic acid 4-hydroxylase; 4CL: 4-coumarate-CoA ligase; CHS: chalcone synthase; CHI: chalcone isomerase; F3H: flavanone 3-hydroxylase; F3'H: flavanone 3'-hydroxylase; F3'5'H: flavanone 3'5'- hydroxylase; DFR: dihydroflavonol 4-reductase; ANS: anthocyanidin synthesis; UFGT: UDP-glucose flavonoid 3-O-glucosyltransferase; FLS: flavonol synthase; ANR: anthocyanidin reductase.
Caption: Figure 5: Real-time quantitative PCR validation of transcript profiles for representative color-related genes. Values for mean expression and standard deviation were calculated from the results of three independent replicates.
Table 1: Differentially altered metabolite compounds in the tepals of the later stage of M. maudiae. Class Compounds Average content WL RL Anthocyanins Peonidin O-hexoside 9 3.04E + 04 Rosinidin O-hexoside 1.42E + 05 9 Cyanidin 3-O-glucoside 3.37E + 05 5.32E + 07 Cyanidin O-syringic acid 9 1.58E + 07 Cyanidin 3, 3.99E + 03 1.29E + 06 5-O-diglucoside Pelargonidin 9 5.06E + 06 3-O-glucoside Cyanidin 6.58E + 04 1.32E + 06 Catechin Catechin-catechin 2.01E + 05 9 derivatives catechin Epicatechin 3.96E + 04 9 epiafzelechin Epigallocatechin 9 8.58E + 04 Protocatechuic aldehyde 9 1.82E + 05 Protocatechuic acid 9 2.99E + 04 Proanthocyanidins Procyanidin A2 1.53E + 05 9 Procyanidin B3 1.03E + 07 3.88E + 05 Flavonol Kumatakenin 2.65E + 04 9 Dihydrokaempferol 8.96E + 06 3.85E + 05 Rhamnetin 7.23E + 03 1.66E + 06 Fustin 1.15E + 07 5.61E + 05 Calycosin 2.72E + 03 1.88E + 05 Isoflavone Prunetin 1.61E + 04 3.39E + 05 Sissotrin 1.05E + 04 9.42E + 05 Flavone Selgin 5-O-hexoside 6.45E + 04 2.19E + 06 O-Methylchrysoeriol 8.55E + 03 1.96E + 06 5-O-hexoside Tricin O-vanilloyl 9 2.70E + 04 hexoside O-Methylchrysoeriol 1.03E + 04 2.09E + 06 7-O-hexoside Tricetin O-rutinoside 9 1.80E + 04 Luteolin 8.44E + 05 3.68E + 04 O-hexosyl-O-pentoside Chrysoeriol 9 6.60E + 04 O-hexosyl-O-hexoside Tricin 3.94E + 03 6.41E + 05 5-O-hexosyl-O-hexoside Tricin di-O-hexoside 9 3.17E + 06 Tricin O-rhamnoside 1.68E + 03 1.85E + 06 Tricin 9 3.30E + 04 O-hexosyl-O-syringin alcohol Chrysoeriol 7-O-hexoside 1.90E + 06 9 Luteolin O-eudesmic 1.85E + 03 7.03E + 04 acid-O-hexoside Tricin 5-O-hexoside 9 4.21E + 06 Chrysoeriol 5.00E + 05 1.33E + 04 Acacetin 1.55E + 04 3.25E + 05 Sakuranetin 4.68E + 03 2.07E + 05 Flavone C-Pentosyl-C-hexosyl- 1.03E + 05 5.27E + 03 C-glycosides apigenin C-Hexosyl-apigenin 2.89E + 04 9 O-caffeoylhexoside Luteolin 9 1.66E + 04 O-feruloylhexoside 8-C-Hexosyl-apigenin 9 9.75E + 04 O-hexosyl-O-hexoside C-Hexosyl-apigenin 4.06E + 04 2.38E + 03 O-pentoside 8-C-Hexosyl-luteolin 9 2.91E + 04 O-hexoside Chrysoeriol C-hexoside 9 1.90E + 04 Isovitexin 1.78E + 05 6.36E + 03 Flavanone Naringenin 2.22E + 03 1.77E + 04 O-malonylhexoside 7-O-Methyleriodictyol 1.61E + 03 8.68E + 05 Class Compounds VIP Fold change Anthocyanins Peonidin O-hexoside 1.65 3381.48 Rosinidin O-hexoside 1.81 6.34E-05 Cyanidin 3-O-glucoside 1.31 157.61 Cyanidin O-syringic acid 2.21 1.76E + 06 Cyanidin 3, 1.40 322.20 5-O-diglucoside Pelargonidin 2.12 5.62E + 05 3-O-glucoside Cyanidin 1.01 20.01 Catechin Catechin-catechin 1.84 4.48E-05 derivatives catechin Epicatechin 1.68 2.27E-04 epiafzelechin Epigallocatechin 1.76 9533.33 Protocatechuic aldehyde 1.83 2.03E + 04 Protocatechuic acid 1.66 3318.52 Proanthocyanidins Procyanidin A2 1.82 5.90E-05 Procyanidin B3 1.05 0.04 Flavonol Kumatakenin 1.61 3.40E-04 Dihydrokaempferol 1.03 0.04 Rhamnetin 1.35 229.24 Fustin 1.01 0.05 Calycosin 1.20 69.20 Isoflavone Prunetin 1.01 21.06 Sissotrin 1.23 89.94 Flavone Selgin 5-O-hexoside 1.09 34.02 O-Methylchrysoeriol 1.36 229.63 5-O-hexoside Tricin O-vanilloyl 1.65 3000.00 hexoside O-Methylchrysoeriol 1.34 202.46 7-O-hexoside Tricetin O-rutinoside 1.60 2000.00 Luteolin 1.03 0.04 O-hexosyl-O-pentoside Chrysoeriol 1.74 7337.04 O-hexosyl-O-hexoside Tricin 1.61 162.72 5-O-hexosyl-O-hexoside Tricin di-O-hexoside 2.08 3.52E + 05 Tricin O-rhamnoside 1.78 1105.81 Tricin 1.67 3670.37 O-hexosyl-O-syringin alcohol Chrysoeriol 7-O-hexoside 2.04 4.73E-06 Luteolin O-eudesmic 1.40 38.03 acid-O-hexoside Tricin 5-O-hexoside 2.10 4.68E + 05 Chrysoeriol 1.11 0.03 Acacetin 1.01 20.99 Sakuranetin 1.12 44.26 Flavone C-Pentosyl-C-hexosyl- 1.01 0.05 C-glycosides apigenin C-Hexosyl-apigenin 1.65 3.12E-04 O-caffeoylhexoside Luteolin 1.59 1840.74 O-feruloylhexoside 8-C-Hexosyl-apigenin 1.77 1.08E + 04 O-hexosyl-O-hexoside C-Hexosyl-apigenin 1.05 0.06 O-pentoside 8-C-Hexosyl-luteolin 1.65 3229.63 O-hexoside Chrysoeriol C-hexoside 1.61 2107.41 Isovitexin 1.06 0.04 Flavanone Naringenin 1.19 7.96 O-malonylhexoside 7-O-Methyleriodictyol 1.54 538.13 Note. WL: "white flower" at later stage; RL: "rubellis flower" at later stage. Number 9 represents a level so low as to barely be detectable. Differentially accumulated compounds were identified by threshold VIP (variable importance in projection) [greater than or equal to] 1. Figure 4: Changes in transcript abundance of predicted R2R3-MYB transcription factors. (a) Differentially expressed R2R3-MYBs in comparison groups. (b) Phylogenetic tree showing selected plant R2R3- MYB members. Accession numbers are listed in Materials and Methods. (c) Consensus sequences of R2R3 motif of MYB members. [Log.sub.2] fold Transcript Annotation WE/RE WL/RL Unigene16157 Myb-related protein 308-like - 1.43 Unigene29273 Myb-related protein P 1.78 - Unigene42467 Transcription repressor MYB5 -2.05 - Unigene48479 Myb-related protein 330 3.68 1.61 Unigene49636 Myb-related protein Hv1 - -3.86 Unigene62902 Transcription repressor MYB4-like -1.35 - Unigene69351 Myb-related protein 12 -1.36 -2.20 Unigene69647 Transcription factor MYB39 - - Unigene69995 Myb transcription factor -1.34 - Unigene71205 Myb-related protein 305 - 10.89 Unigene75793 Transcription repressor MYB4-like -1.56 - Unigene75794 Transcription repressor MYB4-like -3.30 - Unigene77569 Myb-related protein 108 2.21 -
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||Research Article|
|Author:||Lang, Xiaoan; Li, Na; Li, Lingfei; Zhang, Shouzhou|
|Publication:||International Journal of Genomics|
|Date:||Nov 1, 2019|
|Previous Article:||Cell-Free DNA Plasma Levels Differ in Age-Specific Pattern in Healthy Rats and Castrates with Testosterone-Induced Benign Prostatic Hyperplasia.|
|Next Article:||Genotyping for Blast (Pyricularia oryzae) Resistance Genes in [F.sub.2] Population of Supa Aromatic Rice (Oryza sativa L.).|