The Heterologous Expression of the Chrysanthemum R2R3-MYB Transcription Factor CmMYB1 Alters Lignin Composition and Represses Flavonoid Synthesis in Arabidopsis thaliana

Plant R2R3-MYB transcription factor genes are widely distributed in higher plants and play important roles in the regulation of many secondary metabolites at the transcriptional level. In this study, a chrysanthemum subgroup 4 R2R3-MYB transcription factor gene, designated CmMYB1, was isolated through screening chrysanthemum EST (expressed sequence tag) libraries and using rapid application of cDNA ends (RACE) methods and functionally characterized. CmMYB1 is expressed in the root, stem, leaf and flowers, but most strongly in the stem and most weakly in the root. Its heterologous expression in Arabidopsis thaliana reduced the lignin content and altered the lignin composition. The heterologous expression also repressed the flavonoids content in A. thaliana. Together, these results suggested that CmMYB1 is a negative regulator of genes involved in the lignin pathway and flavonoid pathway, it may be a promising gene for controlling lignin and flavonoids profiles in plants.


Introduction
Transcription factors play critical roles in regulating plant development and response to environmental stress. The structure of the DNA binding sequences of transcription factors has allowed the recognition of several distinct gene families [1]. In plants, MYB family is one of the most abundant transcription factor classes [2]. Among approximately 1500 transcription factor genes identified in the Arabidopsis thaliana genome, 130 belong to the MYB members [3,4]. The MYB DNA binding domain consists of one to three imperfect 50-53 residue repeats [5], and the number of these repeats present has been used to categorize the family into the subclasses 3R-MYB, R2R3-MYB and a MYB-related group; of these, R2R3-MYB is the most commonly occurring type [6,7]. Based on variation within certain conserved C-terminal motifs, the R2R3-MYBs have been classified into 22 sub-groups [8], with members within each sub-group predicted to share similar or identical functions [9].
The complex polyphenol molecule lignin is deposited in the secondary cell walls of all vascular plants and is synthesized through the lignin biosynthesis pathway, a major branch of the phenylpropanoid pathway [10]. Its accumulation improves the efficiency of water transport, increases the stiffness of mechanical tissues, and forms a physical barrier against pathogens and wounding [10][11][12]. The lignin synthesis is heavily regulated by R2R3-MYB transcription factors [13,14]. Several R2R3-MYB transcription factors belonging to different subgroups such as Arabidopsis AtMYBPAP1 [15], Pinus taeda PtMYB1 and PtMYB4 [16,17], poplar PttMYB21a [18], Eucalyptus gunnii EgMYB2 [19] and Vitis vinifera VvMYB5a [20] have already been shown to either positively or negatively control the lignin biosynthesis through their in vitro interaction with ACI, ACII, and ACIII cis-elements. So far, all the repressors of lignin synthesis appear to be confined to subgroup 4. For example, the Antirrhinum majus AmMYB308 and AmMYB330 were the first R2R3-MYB factors associated with the down-regulation of lignification. When either of the MYB proteins was heterologously expressed in tobacco, the vascular tissue lignin content was markedly reduced, and transcript abundance of the genes encoding 4-coumarate:CoA ligase (4CL1), cinnamate-4hydroxylase (C4H) and cinnamyl alcohol dehydrogenase (CAD) was reduced [21]. Similarly, C4H expression was enhanced in the A. thaliana knock-out mutant Atmyb4 while that of CCoAOMT (caffeoyl-CoAO-methyltransferase) was reduced [22]; meanwhile in the Atmyb32 mutant, the gene encoding caffeic acid O-methyl-transferase (COMT) was up-regulated [23]. The two Zea mays subgroup 4 R2R3-MYB factors ZmMYB31 and ZmMYB42 also act as repressors of lignin synthesis [24,25], and the heterologous expression of ZmMYB42 in A. thaliana appears to reduce lignin content, alter lignin composition and repress flavonoids biosynthesis [26]. Similarly, the over-expression of ZmMYB31 significantly reduces lignin content with alter polymer composition, and enhances CHI, F3H, F39H and DFR gene expression in A. thaliana [25]. Finally, EgMYB1 expression reduces lignin synthesis in both transgenic A. thaliana and P. trichocarpa [27,28].
Chrysanthemum (Chrysanthemum morifolium) is a commercially important ornamental plant worldwide. Here, we describe the isolation of the subgroup 4 R2R3-MYB transcription factor CmMYB1. Its participation in the regulation of lignin synthesis and flavonoid synthesis was demonstrated by its heterologous expression in A. thaliana. A series of gene expression experiments demonstrated that several genes in the lignin synthesis pathway and flavonoid biosynthesis were down-regulated by the presence of the CmMYB1 transgene, leading to a reduction in the lignin content, lignin compositions as well as a decrease in the flavonoids content in the transgenic plants.

Plant Materials and Plant Growing Conditions
The chrysanthemum variety 'Zhongshanzigui' was obtained from the Chrysanthemum Germplasm Resource Preserving Centre, Nanjing Agricultural University, China. Plants were grown in a 1:1 mixture of garden soil and vermiculite without any additional fertilizer, and were maintained in a greenhouse under standard growing conditions. Young plants were watered daily, and fertilized weekly with half strength Hoagland's nutrient solution. The expression profile of CmMYB1 was determined in the root, stem, young leaf and ray floret of each of three chrysanthemum plants, using RNA extracted from snap-frozen fresh plant material. A. thaliana ecotype Col-0 plants were grown in a 1:1:1 mixture of perlite:vermiculite:soilrite in a growth chamber set to deliver a 16 h photoperiod at 23uC during the lit period (80-100 mmol m 22 s 21 illumination) and 18uC during the dark period. The plants were watered every 4 days. To determine the expression level of various lignin synthetic genes, RNA was

Full-length cDNA Isolation and Sequencing
Total RNA was isolated from chrysanthemum leaves using the RNAiso reagent (TaKaRa, Japan), following the manufacturer's instructions. The cDNA first strand was synthesized from 1 mg total RNA using SuperScript III reverse transcriptase (Invitrogen, USA), according to the manufacturer's instructions. A genespecific primer pair (F/R) was designed to amplify a fragment of CmMYB1 based on the sequence of a chrysanthemum EST (DK942906) [29], and RACE PCR was then used to obtain the full length cDNA. For the 39 RACE reaction, the first strand was synthesized using an oligo (dT) primer incorporating the sequence of the adaptor primer. Then a nested PCR was employed, using the gene-specific primer pair GSP39-1/22 and the adaptor primer (dT-AP). For the 59 RACE reaction, the nested PCR was based on the 59 RACE adaptor primer (Abridged Anchor Primer, AAP), the Abridged Universal Amplification Primer (AUAP) provided with the 59 RACE System kit v2.0 (Invitrogen) and the gene-specific primers (GSP59-1, GSP59-2 and GSP59-3). PCR products were purified using a Biospin Gel Extraction kit (Bio Flux) and cloned into the pMD19-T easy plasmid (TaKaRa) for DNA sequencing. Finally, a pair of gene-specific primers (Full-F/Full-R), designed from the putative 59 and 39 UTR sequences, was used to amplify the complete CmMYB1 open reading frame. The sequences of all the above primers are given in table S1 in file S1. Chrysanthemum genomic DNA was isolated from young leaves using a CTAB method [30], and the genomic sequence of CmMYB1 was amplified using the primer pair Full-F/Full-R. The resulting product was purified using a Biospin Gel Extraction kit and cloned into pMD19-T easy for sequencing.

Sequence Alignment and Phylogenetic Analysis of CmMYB1
The CmMYB1 peptide sequence was aligned with that of its presumed homologues using ClustalW [31]. A neighbour-joining based phylogenetic tree was constructed using MEGA5 [32].

Vector Construction and A. thaliana Transformation
The CmMYB1 coding sequence was amplified using a forward primer (MYB1-1301-F) incorporating a BamH I restriction site and a reverse primer (MYB1-1301-R) with a Sac I site (table S1 in file S1). BamH I-Sac I digested amplicons were inserted into pCAMBIA1301 to generate a 35S::CmMYB1 construct, which was introduced into Col-0 via an Agrobacterium tumefaciens EHA105mediated oral dip method [37]. Transformed progeny were selected by germination on a standard medium containing 50 mg mL 21 hygromycin, and confirmed by subsequent RT-PCR analysis.

Lignin Analysis
Stem cross-sections of ten 35S::CmMYB1 transgenic A. thaliana plants and ten wild type plants were prepared using a scalpel blade, respectively. Sections of thickness 70 mm were mounted on a microscope slide and kept moist by the addition of distilled water. After removal of all excess water, a few drops of 6 mol . L 21 HCl were placed on the sections and left for 3 min. Thereafter, a few drops of 5% (w/v) phloroglucinol-HCl were added and the sample covered with a cover slip for 10 min [38]. Phloroglucinol-HCl reacts with the hydroxycinnamaldehyde and benzaldehyde groups of lignin, and the intensity of the red stain generated by this reaction roughly reflects total lignin content [39]. The samples were inspected by light microscopy.
Total lignin content was determined by the spectrophotometric acetyl bromide lignin method with modifications [40]. The samples were determined for three biological replicates, each with 60 mg of drying stem. The cell walls of samples were isolated and then extracted in 100 ml acetyl bromide (25% v/v acetyl bromide in glacial acetic acid) at 50uC for 2 h, and cooled on ice to room temperature. Subsequently, add 400 ml 2 mol . L 21 NaOH and 70 ml of freshly prepared 0.5 mol . L 21 hydroxylamine hydrochloride to the cooled samples, vortex volumetric flasks. Fill up volumetric flask exactly to the 2 ml with glacial acetic acid, cap  and invert several times to mix and the content was measured through spectrophotofluorometer at 280 nm. The lignin monomer composition from A. thaliana mature stems was measured with the CuO oxidation method by HPLC [41].The HPLC separation procedures followed protocol of Sonbol FM et al [26].

Cellulose Content Analysis
Method for measuring the content of cellulose is essentially described by Updegraf [42]. Briefly, 60 mg dried cell wall was added to 1 ml Updegraff reagent (Acetic acid: nitric acid: water, 8:1:2 v/v) and heated at 100uC for 30 min. The pellets were washed with water and acetone and added to 72% sulfuric acid to hydrolyze completely. The glucose content was read at 625 nm.

Flavonoids Analysis by HPLC
HPLC analysis was performed using method of Burbulis [43] with minor revision. For flavonoids analysis, 100 mg fresh mature stems were harvested, and then extracted in 500 ml methanol with thoroughly vortexing, and centrifuged for 10 min at 10,000 rpm. The supernatant was hydrolyzed with 2 mol . L 21 HCl at 70uC for 40 min. After being cooled in room temperature, the hydrolysis were terminated by adding 500 ml methanol, the mixture was dried under N 2 stream and re-dissolved with 500 ml methanol for the HPLC analysis. The separation and quantification of flavonoids were determined in following conditions. Injected volume: 20 ml. Column: ZORBAX Eclipse Plus C18 of 5 mm, 4.66250 mm. Eluents: (a) 0.05% acetic acid, (b) methanol. Flux: 1 ml/min. Temperature: 25uC. The gradient started with 10% methanol, increasing to 50% in 10 min, 90% in 20 min, 100% in 5 min. Flavonoids (kaempferol, quercetin and isorhamnetin) were detected at 254 nm.

Isolation of CmMYB1, a new Subgroup 4 R2R3-MYB Factor from chrysanthemum
The full length cDNA CmMYB1 sequence (JF795917) was isolated by RT-PCR and RACE based on an EST created by Chen [29]. It consisted of 1, 237 nucleotides, of which 846 bp represented an open reading frame encoding 281 residues. The predicted gene product is a protein of molecular mass 31.7 kDa and a pI of 8.77, containing an R2R3 MYB domain with two DNA binding sites, one localized between residues 9 and 61, and the other between residues 62 and 116. XLEMK, and CX 1-2 CX 7-12 CX 1-2 C; these are, respectively, putative C1, C2, C4 and zing finger motifs, unique to subgroup 4 R2R3-MYB genes [8]. C2, which contains the core EAR-motif, plays a major role in gene repression [44,45]. The alignment of the CmMYB1 sequence with those of homologous R2R3-MYB proteins showed levels of similarity ranging between 55% and 62% (Fig. 1a), and suggested a close phylogenetic relationship with the well known subgroup 4 lignin repressors AmMYB308, AmMYB330, EgMYB1 and ZmMYB31 [21,25,27] (Fig. 1b). Comparison of the CmMYB1 genomic DNA and cDNA sequences identified the presence of a single 307 bp intron, the position of the intron (264 bp downstream of the ATG start codon) is well conserved among R2R3-MYB subgroup 4 genes (data not shown).

CmMYB1 Expression Profiles in chrysanthemum
To gain the expression pattern of CmMYB1, qRT-PCR was carried out to analyze samples from various organs of the chrysanthemum plant. The results showed that CmMYB1 was ubiquitously expressed with the strongest expression level in stems. CmMYB1 was also expressed at a lower level in leaves, as well as in flowers, and the weakest level in roots (Fig. 2).

The Effect of Heterologous Expression of CmMYB1 on A. thaliana
CmMYB1 was introduced under the control of the cauliflower mosaic virus 35S promoter into binary vector pCAMBIA1301 containing the hygromycin B and gusA as selectable markers. Transformation was performed using floral dip method [37]. Two independent transgenic A. thaliana lines (35S::CmMYB1-1, 35S::CmMYB1-2) which showed higher levels of transgene expression were selected from several transgenic lines for further experiments (Fig. 3a). There was no observable phenotypic difference between wild type and transgenic plants grown under long days (16 h light and 8 h dark). The phloroglucinol staining generated red coloration uniformly distributed throughout the lignified tissues of both the wild type and the transgenic plants (Fig. 3b), but the staining intensity was noticeably greater in the wild type stems than in those of either of the two transgenic lines.

CmMYB1 Affects Lignin Biosynthesis
The expression level of a set of key lignin synthesis genes (AtPAL1, AtC4H, At4CL1, AtHCT, AtC3H, AtCCoAOMT1, AtCCR1, AtF5H, AtCOMT1 and AtCAD6) was monitored in the transgenic lines using qRT-PCR, respectively. This experiment showed that the expression of AtCOMT1 and AtCAD6 had been decreased to 10% of the wild type level and that of AtC4H, At4CL1, AtHCT, AtCCR1, and AtF5H to between 25% and 50%. However, the level  of AtPAL1, At4CL3, AtC3H and AtCCoAOMT1 expression appeared to have been less affected (Fig. 4).
The expression level of genes involved in the biosynthesis of cellulose was also affected by heterologous expression of CmMYB1. qRT-PCR analysis showed that the expression of three secondary wall-associated cellulose synthase genes (CesA4, CesA7, and CesA8) [46,47] had been decreased. But that of xylan biosynthetic genes IRX8 [48] had been less down-regulated. The level of another gene of xylan biosynthetic IRX9 expression, however, was increased (Fig. 5).
To better determine the changes of lignin in two independent transgenic A. thaliana (35S::CmMYB1-1, 35S::CmMYB1-2), we assayed the content of acetyl bromide lignin in stems. The analysis revealed that the acetyl bromide lignin content decreased in stems of both the two transgenic plants. The lignin content of transgenic plant CmMYB1-1 and CmMYB1-2 had been decreased by 10% and 20% of the wild type plant, respectively (Table 1).We also analysed the lignin monomer composition. An increase in H subunits was observed in two independent transgenic A. thaliana plants. The content of G subunits increased significantly by 40% of the wild type while the content of S subunits decreased significantly by 40% of the wild type (Table 1).
Cellulose is another important secondary wall component. We also performed analysis of cellulose content, however, no significant reduction in cellulose was observed ( Table 2).

CmMYB1 Suppresses Flavonoid Biosynthesis
We performed qRT-PCR assays to determine the effect of CmMYB1 on flavonoids biosynthesis. The analysis indicated that the expression of CHS, CHI, FLS1 and DFR was suppressed, while F3H and F39H increased slightly in transgenic plants compared to that of the wild type plants (Fig. 6).
Flavonoids content in wild type and two independent transgenic plants of A. thaliana (35S::CmMYB1-1, 35S::CmMYB1-2) were determined via HPLC analysis. The contents of quercetin, kaempferol and isorhamnetin of transgenic plant CmMYB1-1 and CmMYB1-2 decreased by 60% and 50% of the wild type plant, respectively, which inferred that CmMYB1 might suppress the flavonoid synthesis (Table 3).

Discussion
MYB genes are a particularly abundant class of plant transcription factors, and the R2R3-MYB subfamily is prominent in higher plant species. Here we have described the isolation of the chrysanthemum R2R3-MYB transcription factor CmMYB1. Its sequence alignment and phylogeny strongly indicated that it belongs to R2R3-MYB subgroup 4, which includes a number of members involved in the repression of lignin synthesis [21,22,24,28]. It includes the C2 motif known to confer transcriptional repressor activity [45], which supports the idea that CmMYB1 functions as a negative regulator in chrysanthemum. CmMYB1 was ubiquitously expressed in the plant, although most strongly in the stem, which implies that it may play a role in the lignifying tissue of chrysanthemum.
A number of studies have shown that the over-expression of R2R3-MYB factors known to act as repressors of lignin synthesis produce alternations of the leaf morphology, with the appearance of white lesions on mature leaves and a reduction in growth rate when heterologous-expressed in tobacco and over-expressed in Arabidopsis [21,22]. However, in the present study the heterologous expression of CmMYB1 in A. thaliana did not produce any  observable phenotype, which is different from previous researches.
To determine whether CmMYB1 gene was related to the synthesis of lignin, we employed a direct histochemical staining method to detect the lignin content in stem, where the color intensity generated in the reaction with phloroglucinol was enhanced with the increase of lignin content. By this method, the strong reduction of lignin content was observed in inflorescence stems of CmMYB1 transgenic A. thaliana plants. We also employed the acetyl bromide lignin method to determine the lignin content in stems. By this quantitative method, we could confirm the lignin content of transgenic A. thaliana plants was decreased (Table 1). In addition, CmMYB1 overexpression in A. thaliana plants altered the lignin composition, showing higher content of H and G subunits while lower content of S subunits characterized by decreased S/G ratio of the lignin. Lignin synthesis was known to interact with cellulose and other secondary cell wall synthesis [26], however, cellulose synthesis was less affected in CmMYB1 transgenic plants ( Table 2). PAL is the first enzyme of the phenylpropanoid pathway. Overexpression of PtMYB4 in tobacco plants could reduce the expression level of PAL [17]. 4CL genes play key roles in the metabolic pathway of lignin synthesis, catalyzing the reaction for lignin synthesis. Most vascular plants (including A. thaliana) possess three distinct 4CL forms [49]. In A. thaliana, At4CL1 and At4CL2 are largely responsible for controlling the branching of the growing lignin molecule, while At4CL3 controls flavonoid branching. A common feature between tobacco heterologous expressing Am-MYB308 or AmMYB330 and A. thaliana over-expressing AtMYB4 is the down-regulation of 4CL1 expression, as also displayed in the transgenic A. thaliana lines expressing CmMYB1 (Fig. 4); this commonality indicates that CmMYB1 is involved in lignin synthesis. CAD is another key gene in lignin synthesis of A. thaliana stems [50], involved in the final step of the reduction reaction of lignin monomer. The lignin content in the stem of double mutant plant (cad-c, cad-d) was reduced by 40%, compared to that of normal species, so the stems of A. thaliana double mutant were soft and easy to lodge. Therefore, heterologous expression of CmMYB1 on A. thaliana reduced the expression of key enzymes AtCAD and affected lignin synthesis (Fig. 4). In addition to 4CL1 and CAD, subgroup 4 R2R3-MYB factors have also been reported to regulate other key genes in the lignin pathways. AmMYB308 and AmMYB330 both negatively regulate the expression of C4H and CAD when heterologously over-expressed in tobacco [21], and AtMYB4 is proved to be function as a repressor, particularly of C4H [22]. On the other hand, ZmMYB31 marginally enhances CAD expression and decreases COMT expression, but does not affect the expression of either C4H or CCoAOMT, while ZmMYB42 down-regulates the expression of C4H, CAD and COMT but does not affect the expression of CCoAOMT [24]. Here, similar to the behavior of both AmMYB308 and ZmMYB42, the heterologous expression of CmMYB1 suppressed the expression of C4H, COMT and CAD, while that of CCoAOMT was unaffected (Fig. 4).It has been reported that the down-regulated expression of C3H, 4CL1, F5H or COMT could alter the final lignin composition [51,52]. CmMYB1 involved in the expression of lignin biosynthetic genes suggests the process that A. thaliana biosynthetic pathways lead to the biosynthesis of cellulose, xylan, and lignin may be cogitated by pathway-specific transcription factors. CesA genes play an important part in cellulose biosynthesis [46,47] and IRX8 and IRX9 genes involved in xylan biosynthesis [48]. These results suggest that CmMYB1 is a transcription factor influencing cellulose biosynthesis, but appeared to have been largely unaffected in xylan biosynthesis (Fig. 5).
CmMYB1 seems to play an important role in regulating flavonoid pathway which is another branch of phenylpropanoid pathway. VvMYBF1, a R2R3-MYB transcription factor, is a functional regulator of flavonol synthesis [53]. ZmMYB42 repress the flavonol biosynthesis [26]. AtMYB4 of R2R3-MYB factor also affected the expression level of 4CL3 and CHS [22]. Overexpression of CmMYB1 in A. thaliana reduced the expression level of 4CL3, CHS, CHI, FLS and DFR. The contents of quercetin, kaempferol and isorhamnetin contents decreased significantly in consistence with the suppression of genes expression in two independent transgenic A. thaliana. Overall, the data indicate that CmMYB1 negatively regulates flavonoids synthesis.

Supporting Information
File S1 Supporting information file containing the following files. Table S1.Primer sequences used in this study. Table S2.Primer sequences used in qRT-PCR. (DOC) Author Contributions