Introduction

Gene expression is regulated by various complex mechanisms, including modifications to DNA such as histone modification and DNA methylation, as well as various RNA-mediated processes. Transcription factors (TFs) regulate gene expression, and this is a well-known mechanism by which a TF binds to a specific nucleotide sequence upstream of target gene, ultimately controlling a range of biological processes1. MYB TFs exist widely in eukaryotes and are one of the largest and most diverse families of TFs in the plant kingdom, where they play an essential role in a wide range of physiological and biochemical processes2,3.

The first MYB gene (v-myb), which was isolated from avian myeloblastosis virus (AMV), encodes a MYB domain protein4. Ever since the first plant MYB gene (the Zea mays COLORED1 (C1) gene) was cloned5, numerous MYB genes have been identified from plants as an increasing number of plant genomic sequences became available. For example, 198, 256, 127, 231 and 122 MYB genes have been identified in Arabidopsis thaliana6, Brassica rapa3, Solanum lycopersicum7, Pyrus bretschneideri8 and Brachypodium distachyon9, respectively. MYB proteins share a highly conserved DNA-binding domain (the MYB domain), which ranges from one to four imperfect amino acid sequence repeats (R)10. Based on the number of adjacent repeats, the MYB genes in plants have been divided into four distinct groups: MYB-related (MYBR, only contains one R1- or R2-like repeat), R2R3-MYB (containing two R2/3R-like repeats), 3R-MYB (containing three R1/R2/R3-like repeats) and atypical MYB proteins (4R-MYB, four R1/R2-like repeats; CDC5-like)6,10. The atypical MYB group of proteins is the smallest class, and contains one or two genes in several higher plant genomes. For example, one and two atypical MYB genes were found in O. sativa and A. thaliana, respectively11. The 3R-MYB group is the second smallest class, and contains about five members in plants such as A. thaliana, Populus trichocarpa and Vitis vinifera3. The MYBR proteins contain only a single repeat and fall into five subclasses, CCA1/R-R-like, CPC-like, I-box-like, TBP-like and TRF-like, that contain 30 to 70 genes in plant genomes2. The R2R3-MYB group is the largest group in plants with more than 100 members that were extensively amplified approximately 500 million years ago after the appearance of land plants12. For example, 113, 126 and 244 genes encode R2R3-MYB proteins in O. sativa11, A. thaliana6 and Glycine max13, respectively. The A. thaliana R2R3-MYB proteins have been divided into 33 clades, while those in P. bretschneideri have been divided into 37 clades8, which suggests evolutionary diversity of the R2R3-MYB family.

All four groups of MYB proteins were found after genome-wide analyses of MYB TFs in plants and while the function of a number of MYB genes have been characterized in many plants, the function of 4R-MYB proteins remains unclear. The 3R-MYBs play a role in cell cycle control14. MYBR proteins are involved in cellular morphogenesis, secondary metabolism, organ morphogenesis, phosphate starvation, chloroplast development and circadian regulation15. More recently, two A. thaliana MYBR genes (MYBS1 and MYBS2) have been shown to have opposite roles in sugar signaling mechanisms16. Over the past two decades, the R2R3-MYB TFs have been extensively exploited and many R2R3-MYB proteins have been shown to play roles in several biological processes, such as development, response to biotic and abiotic stresses, and metabolism15,17. Among the metabolic processes, R2R3-MYB genes are acutely involved in phenylpropanoid metabolism18. Very recently, two R2R3-MYB genes from Marchantia polymorpha, MpMYB14 and MpMYB02, were found to act as essential regulators in the biosynthesis of riccionidins and marchantins, respectively19.

Secondary cell walls (SCWs), which have a critically important function by supporting plants and are a major source of plant biomass, are composed of cellulose, lignin and hemicellulose20. An increasing body of studies has demonstrated that R2R3-MYB proteins are critical for SCW biosynthesis21. In A. thaliana, AtMYB46 binds to the promoter of AtCSLA9, which is involved in glucomannan biosynthesis, and regulates its expression22. Testa mucilage, which is composed of polysaccharides, is regarded as a useful model for exploring the biosynthesis of cell wall polysaccharides23,24. Previous studies have demonstrated that R2R3-MYB genes, such as AtMYB5 and AtMYB61, are required for the production of seed mucilage, and the polysaccharide content of seed mucilage in the A. thaliana myb61 mutant was significantly reduced25,26. These results indicate that R2R3-MYB members play roles in the biosynthesis of plant polysaccharides.

Water-soluble polysaccharides (WSPs) play an important role in plants’ stress response. For example, tolerant genotypes of wheat (Triticum aestivum L.) seedlings accumulated more water-soluble carbohydrates, including glucose, fructose, sucrose, and fructan than sensitive genotypes under drought and salt stress27. Recent studies showed that WSPs isolated from plants may increase immunity28 and have an antitumor function29,30. The Orchidaceae is one of the largest plant families in the world and has about 25,000 species31. The Dendrobium genus, which belongs to the Orchidaceae, has several important species that are used as herbal medicines32. WSPs isolated from Dendrobium species such as D. huoshanense and D. officinale are regarded as major active ingredients and display immunomodulating activities33. Several genes involved in the biosynthesis of WSPs have been identified and characterized in D. officinale, whose stems contain an abundance of bioactive WSPs34,35,36. However, the TFs that regulate the biosynthesis of these WSPs are still unknown. In this study, MYB proteins were identified from two orchids, Phalaenopsis equestris and D. officinale, in a genome-wide process, and the putative R2R3-MYB genes related to SCW or mucilage biosynthesis were identified based on phylogenetic analysis. One R2R3-MYB gene involved in the biosynthesis of WSPs was characterized. This work provides novel information that would allow for a better understanding of the functional diversity of MYB genes in plants and would aid in revealing the molecular mechanisms underlying the biosynthesis of bioactive WSPs in D. officinale or in other plants.

Methods

Plant materials and treatments

D. officinale plants were grown as described previously34. The stems of five developmental stages were harvested to determine WSPs and for gene expression analysis. S1 is about 4 months after sprouting in April, while S2, S3, S4, and S5 are about 9, 10, 12, and 13 months after sprouting, respectively. Roots, leaves and stems that were collected from plants grown in a growth chamber when they were 10 cm in height, were used to analyze the expression pattern of different organs. D. officinale seed capsules were surface sterilized in 0.1% mercuric chloride (HgCl2), sown on half-strength Murashige and Skoog medium (half the macronutrients; ½MS)37 supplemented with 1 g/L activated charcoal, 20 g/L sucrose and 6 g/L agar and 0.5 mg/L 1-naphthalene-acetic acid (NAA), and cultivated at 26 ± 1 °C, 40 µmol m−2 s−1, and a 12-h photoperiod. Seedlings (1 cm in height and about 3 months after sowing) were transferred to liquid ½MS medium supplemented with 20 g/L sucrose and 0.5 mg/L NAA for three days. After adapting, 30 seedlings were used to perform abiotic stress bioassays in liquid medium containing 150 g/L polyethylene glycol (PEG) 6000 (Sigma-Aldrich, Shanghai, China), 300 mM mannitol (Sigma-Aldrich), or 250 mM NaCl (Guangzhou Chemical Reagent Factory, Guangzhou, China). Seedlings were transferred to fresh ½MS medium supplemented with 20 g/L sucrose and 0.5 mg/L NAA as the control. There were three replicates with 36 seedlings in each treatment. After 6 h, seedlings were harvested, frozen in liquid nitrogen and RNA was isolated immediately.

A. thaliana plants (Col-0) were grown in soil under a 16-h photoperiod at 22 °C. For screening resistant lines, A. thaliana seeds were surface sterilized and sown on ½MS medium supplemented with 15 g/L sucrose and 8 g/L agar, stratified in the dark at 4 °C for 2 d, and then cultivated under a 16-h photoperiod at 22 °C.

Identification of MYB transcription factors in orchids

The protein sequences of P. equestris and D. officinale were downloaded in a FASTA format from orchidbase (http://orchidbase.itps.ncku.edu.tw/EST/releaseSummary2012.aspx) and the National Center for Biotechnology Information (NCBI) provided by Zhang et al.38. HMMER 3.0 software (http://hmmer.janelia.org/) was used to identify the putative MYB TFs under default parameters. The putative MYB TFs were annotated by Pfam (Protein family)39, Swissprot40 and nr (NCBI non-redundant protein sequences)41. The putative MYB TFs, which were confirmed to be MYB TFs by annotation, were regarded as MYB TFs. The classification of MYBR, R2R3-MYB, 3R-MYB and atypical MYB proteins were based on the annotation and BLAST against the A. thaliana MYB TFs.

Phylogenetic analysis

MYB proteins from A. thaliana (At), P. equestris (Pe) and D. officinale (Do) were aligned using MAFFT software version 742. A phylogenetic tree of R2R3-MYBs was constructed using the Neighbor–Joining (NJ) method and 1,000 bootstraps with Clustalx43. The phylogenetic trees of MYBR and C2 (S6) R2R3-MYBs were constructed using MEGA 744 with the NJ method using 1,000 bootstraps.

Calculation of Ks and Ka of MYB family genes in the two orchids

Orthologous gene pairs of MYB genes between P. equestris and D. officinale were identified by Orthofinder v2.2.6 with the BLAST method under default parameters45. The orthologous gene pairs were used to calculate synonymous (Ks) and nonsynonymous (Ka) values using the KaKs_Calculator2.046.

Prediction of cis-responsive elements on the promoters of orchid MYB genes

The 2000 bp genomic DNA sequences upstream of the initiation codon (ATG) of orchid MYB genes were obtained and used for predicting cis-acting regulatory DNA elements (cis-responsive elements). The PlantCARE database47 and PLACE database48 were adopted to identify the putative cis-responsive elements.

Expression profiling of MYB genes from D. officinale under cold stress

For the expression profiles of D. officinale under cold stress (4 °C), the transcriptome sequencing data of the control condition (SRR3210630, SRR3210635 and SRR3210636) and cold stress treatment (SRR3210613, SRR3210621 and SRR3210626) were obtained from the NCBI Sequence Read Archive (SRA) database49. The clean reads were obtained by filtering out low quality reads and were mapped to the nucleotide sequences of MYB genes using TopHat version 2.0.850. The expression level of MYB genes was calculated by the fragments per kilobase of exon per million fragments mapped (FPKM) method using HTSeq51. The heatmap of expression profiling was draw by a green-red gradient in R version 3.4.1 (https://www.r-project.org/). The genes with a FPKM value >5 in the control or cold stress treatment were regarded as sense, then were used to calculate fold change (mean of FPKM cold/mean of FPKM control). Genes with a ≥1.5-fold change were defined as up-regulated genes, and those with a ≤0.66-fold change were regarded as down-regulated genes.

Quantitative RT-PCR (qRT-PCR) analysis

Total RNA from D. officinale organs (roots, stems and leaves) and seedlings, as well as A. thaliana seedlings, was extracted using an RNA extraction kit (Column Plant RNAout2.0, Tiandz, Inc., Beijing, China). RNA was purified by excluding genomic DNA using the DNase I digestion kit (Takara Bio Inc., Dalian, China). The integrity and content of purified RNA was determined by 1% agarose gel electrophoresis and a NanoDrop 2000c Spectrophotometer (Thermo Scientific, Wilmington, NC, USA), respectively. Total RNA was reversed transcribed into cDNA by M-MLV reverse transcriptase (Promega, Madison, WI, USA) according to the manufacturer’s protocol. The cDNA of each sample was diluted to 200 ng/mL and 1 μL was used as template for the qRT-PCR reaction. Three PCR reactions were performed using the SoAdvanced™ Universal SYBR® Green Supermix detection system (Bio-Rad, Hercules, CA, USA) in an ABI 7500 Real-time system (ABI, Foster City, CA, USA) with the following amplification regime: 95 °C for 2 min, and 40 cycles of 95 °C for 15 s and 60 °C for 30 s. Actin from D. officinale (NCBI accession number: JX294908) was used to normalize the expression of genes. The 2−ΔΔCT method52 was used to calculate the relative gene expression level. All the primers of DoMYB genes and actins for qRT-PCR were designed by an online web tool (http://www.idtdna.com/Primerquest/Home/Index) and are listed in Supplementary Table 1.

Generation of DoMYB75 transgenic lines

The coding sequence (CDS) of DoMYB75 without a termination codon was amplified using the KOD FX High Success-rate DNA polymerase Kit (Toyobo Biotechnology Co. Ltd., Shanghai, China) and cloned into the pCAMBIA 1302 vector (Cambia, Canberra, Australia) at the NcoІ site. The construct was verified by DNA sequencing at the Beijing Genomics Institute (Shenzhen, China). A. thaliana was transformed by the floral dip method53 using about 15 independent plants. Twenty five resistant lines were identified by screening in ½MS medium supplemented with 25 mg/L hygromycin B (Roche Diagnostics, Mannheim, Germany). Three resistant lines were randomly selected to extract genomic DNA and verified as transgenic lines using PCR. The 2 × BlueStar™ PCR Master Mix kit (Tingke Biotechnology Co. Ltd., Beijing, China) was used to perform PCR with 95 °C for 2 min and 35 cycles of 98 °C for 10 s, 58 °C for 30 s, and 72 °C for 30 s, followed by a final extension at 72 °C for 10 min. The primers used to construct the overexpression vector are listed in Supplementary Table 1.

Analysis of DoMYB75 transcript level in wild type and DoMYB75 transgenic A. thaliana plants

Total RNA from one-week-old A. thaliana seedlings were extracted, purified and reverse transcribed as indicated above. The 2 × BlueStar™ PCR Master Mix kit was used for semi-quantitative RT-PCR analysis. One microliter of cDNA sample (about 400 ng/µL) was used for each independent PCR reaction using the following thermocycling conditions: 95 °C for 2 min, followed by 40 cycles of 98 °C for 10 s, 55 °C for 30 s, and a final extension at 72 °C for 60 s. The DoMYB75 primer pair was the same as that used for vector construction. The primers UBQ10F/R for the A. thaliana ubiquitin gene (AtUBQ10) are listed in Supplementary Table 1.

Analysis of water-soluble polysaccharide content

The WSPs in D. officinale stems were extracted and determined as previously described34. Whole mature and dry A. thaliana seeds were ground to a fine powder using a tissue lyser (TL2020, Beijing Haoyuan Technology Co. Ltd., Beijing, China). Twenty mg of powder was weighed precisely, pre-extracted twice with 1 mL 80% (v/v) hot ethanol for 20 min in each extraction step, and centrifuged by a Centrifuge 5424 R (Eppendorf, Hamburg, Germany) at 10,000 rpm for 10 min at 16 °C. The supernatant was discarded. The pellet was suspended with 2 mL of distilled water, then incubated in an ultrasonic bath (VCX600, Sonics and Materials Inc., Newtown, CT, USA) for 2 h at 60 °C to extract the WSPs. After centrifugation at 10,000 rpm for 10 min at 16 °C, the supernatant was collected and used to analyze WSPs by the phenol-sulfuric acid method54, as described in He et al.34.

Statistical analyses

Data were analyzed using SigmaPlot12.3 software (Systat Software Inc., San Jose, CA, USA) by one-way analysis of variance (ANOVA) followed by Duncan’s multiple range test (DMRT) or Dunnett’s test. P < 0.05 was considered to be statistically significant.

Results

Identification of MYB superfamily genes in orchids

In this study, a total of 159 and 165 MYB TFs were identified in two orchids, P. equestris and D. officinale, respectively. In P. equestris, 40 MYBR genes, 115 R2R3-MYB genes, three 3R-MYB genes and one atypical MYB gene (CDC5-type) were found (Table 1). A similar number of MYB TF family members was found in D. officinale: 42 MYBR genes, 117 R2R3-MYB genes, four 3R-MYB genes and two atypical MYB genes (one 4RMYB and one CDC5-type) were identified in the D. officinale genome (Table 1). All the details of MYB genes in both orchids are listed in Supplementary Table 2.

Table 1 Number of members in the four groups of MYB transcription factors in A. thaliana, P. equestris and D. officinale.
Full size table

Classification of MYBR and R2R3-MYB proteins in two orchids

In plants, MYBR proteins can be divided into five subgroups: CCA1/RR-like, CPC-like, I-box-like, TBP-like and TRF-like2. Genes in these five subgroups were also found in P. equestris and D. officinale, 21 and 24 CCA1/RR-like, 4 and 3 CPC-like, 5 and 6 I-box-like, 8 and 7 TBP-like, and 2 and 2 TRF-like, respectively. The CCA1/RR-like subfamily is the largest of the five subfamilies, while the TRF-like family is the smallest, with just two members in each orchid (Fig. 1A,B).

Figure 1
figure 1

Phylogenetic trees of MYBR proteins. (A) Unrooted phylogenetic tree of P. equestris and A. thaliana MYBR proteins. (B) Unrooted phylogenetic tree of D. officinale and A. thaliana MYB proteins. The trees were generated by MEGA 744 using the Neighbor-Joining method and aligned by MAFFT42.

Full size image

Compared with the MYB-related family, the R2R3-MYB family contained nearly three times as many genes as MYBR proteins in both orchids. The R2R3-MYB proteins are classified into 25 clades based on conservation of the MYB domain and C terminal amino acid motifs in A. thaliana10. To survey the classification within the R2R3-MYB gene family, we conducted a phylogenetic analysis of A. thaliana (126 members), P. equestris (115 members) and D. officinale (117 members) R2R3-MYB proteins. Based on a phylogenetic tree, all the MYB proteins could be grouped into 35 clades (C1-C35) (Table 2 and Supplementary Fig. 1). The C3 (S15), C4 (S5) and C13 (S23) clade genes were absent in both orchids, while C11 subfamily genes were only found in P. equestris and C17 clade genes were only present in the two orchids (Table 2 and Supplementary Fig. 1). This result suggests that the C3 (S15), C4 (S5) and C13 (S23) proteins might have been lost in orchids after divergence from the most recent common ancestor.

Table 2 Classification and putative functions of R2R3-MYB transcription factors.
Full size table

Non-synonymous (Ka) and synonymous (Ks) substitutions in orthologous gene pairs between P. equestris and D. officinale

The Ka/Ks value is regarded as a pointer to assess selective pressure on a protein-coding gene. A Ka/Ks ratio less than 1 indicates a negative or purifying selection, a Ka/Ks ratio equal to 1 indicates neutral evolution, while a Ka/Ks ratio greater than 1 indicates positive or adaptive evolution. In total, 84 orthologous gene pairs between P. equestris and D. officinale were found (Table 3). In other words, about 50% of orchid MYB genes appeared to be duplicated. This suggests that most orchid MYB genes underwent functional diversity and expansion during evolution. In our research, most of these orthologous gene pairs were deduced to be under negative selection with a Ka/Ks ratio less than 1, except for DoMYBR22 and PeMYBR29, which had a Ka/Ks ratio greater than 1 (Table 3).

Table 3 Ka/Ks analysis and estimated selective pressure for orthologous gene pairs between P. equestris and D. officinale.
Full size table

Cis-responsive element analysis of MYB genes from P. equestris and D. officinale

All the 2000 bp upstream regions of the initiation codon of MYB genes from P. equestris and D. officinale were obtained from their respective genomes. The stress response elements, tissue-specific activation, hormone responsive elements, and other responsive elements were identified and analyzed (Fig. 2). Various stress responsive elements, including anaerobic induction, and response to antioxidant, dehydration, desiccation, drought, heat, low temperature, stress, and wound elements, were analyzed. Only anaerobic induction, low temperature response and wound response elements were widely present in the MYB gene promoters of both orchids (Fig. 2 and Supplementary Table 3). Hormone responsive elements such as ABA, ethylene, GA and MeJA response were abundant in the putative promoters of MYB genes, especially the ABA response element (Fig. 2). ABA, ethylene and MeJA are related to stress response in plants55. These results suggested that the MYB genes may play an important role in stress response in orchids.

Figure 2
figure 2

Average number of cis-responsive elements of orchid MYB genes from each group. The cis-responsive elements were analyzed in the 2 kb upstream promoter region of the initiation codon using the PlantCARE database and the PLACE database.

Full size image

Expression analyses of MYB genes under cold stress in D. officinale

MYB genes are involved in a plant’s response to stress, such as drought and cold stress56,57,58. To gain insight into D. officinale MYB proteins in stress responses, the expression of MYB genes were evaluated under the control condition (20 °C) and cold stress (4 °C) by comparing the FPKM values for each gene at 20 °C and 4 °C. Only nine of 117 R2R3-MYB genes were modulated by cold stress, consisting of three up-regulated genes (DoMYB07, −33, and −69) and six down-regulated genes (DoMYB01, −22, −37, −41, −51, and −96) (Fig. 3A and Supplementary Table 4). A total of 15 out of the 42 MYBR genes (DoMYBR02, −04, −06, −09, −12, −14, −18, −20, −27, −31, −34, −36, −39, −40, and −42) were modulated by cold stress, nine of which were up-regulated by at least two-fold while three genes were down-regulated (less than 0.5-fold, Fig. 3B). Four up-regulated genes (DoMYBR06, −18, −31, and −39) were from the I-box-like subfamily, seven up-regulated genes (DoMYBR02, −04, −20, −27, −34, −36 and −42) were from the CCA1/R-R-like subfamily, and one gene (DoMYBR14) was from the TBP-like clade (Fig. 3 and Supplementary Table 4). Three 3R-MYBs, all 4R-MYBs and only one CDC5-type genes of D. officinale showed no differences between the control and cold stress (Fig. 3 and Supplementary Table 4). One 3R-MYB (DoMYB3R4) gene was down-regulated (Fig. 3 and Supplementary Table 4).

Figure 3
figure 3

Expression patterns of the four groups of MYB genes from D. officinale under cold stress (4 °C). Expression profiles of R2R3-MYB genes (A), MYBR genes (B), and 3R-MYB and atypical MYB genes (C). The heatmap was generated using R version 3.4.1 with a color scale according to the gene expression level [log2 (FPKM + 1)]. Red indicates high gene expression level while green indicates a low level of expression. Each column indicates a discrete biological sample. All treatments consisted of three biological replicates.

Full size image

Identification and expression analysis of R2R3-MYB genes related to polysaccharide biosynthesis

Polysaccharides are abundant in the secondary cell wall and testa mucilage in A. thaliana24,59. Members of nine clades, namely C1 (AtMYB5), C2 (S6), C8 (S21), C26 (S2), C27 (S3), C30 (S8), C33 (S13), C34 (AtMYB26/AtMYB67/AtMYB103) and C35 (AtMYB46/AtMYB83), are involved in secondary cell wall or testa mucilage biosynthesis in A. thaliana (Table 2). The genes from these clades are probably involved in polysaccharide biosynthesis and are regarded as putative MYB genes of polysaccharide biosynthesis. A total of 32, 45 and 43 putative MYB polysaccharide biosynthesis genes were identified from A. thaliana, P. equestris and D. officinale, respectively.

To comparatively analyze their expression patterns, the mRNA steady state levels of the putative MYB genes related to polysaccharide biosynthesis were monitored in the roots, stems and leaves of A. thaliana, P. equestris and D. officinale. Interestingly, 13 of the putative MYB genes from A. thaliana (AtMYB20, −42, −43, −46, −52, −54, −55, −58, −63, −69, −83, −85, and −103) were highly expressed in stems, and most of these genes were also involved in secondary cell wall biosynthesis (Fig. 4A, Supplementary Table 5). Eight (PeMYB10, −11, −14, −26, −36, −42, −62, and −81) and nine (DoMYB02, −09, −14, −17, −31, −99, −105, −115, and −117) MYB genes were highly expressed in the stems of P. equestris and D. officinale, respectively (Fig. 4B,C, Supplementary Tables 5 and 6). All C35 (AtMYB46/AtMYB83) genes were abundantly expressed in stems (Fig. 4).

Figure 4
figure 4

Heat map displaying the expression pattern of nine clades of R2R3-MYB genes in roots, stems and leaves. (A) Expression profile of 32 R2R3-MYB genes in different tissues or organs of A. thaliana from microarray data sets available at Gene Expression Omnibus (GEO, https://www.ncbi.nlm.nih.gov/geo/). GEO accessions included: GSM131558, GSM131559 and GSM131560 (roots); GSM131655, GSM131656 and GSM131657 (stems); GSM131528, GSM131529 and GSM131530 (leaves). The color scale represents log2 of the mean of gene expression. (B) Expression profile of 36 R2R3-MYB genes from P. equestris. Expression data were downloaded from Orchidbase (http://orchidbase.itps.ncku.edu.tw). The color scale was based on the gene expression level, which was measured as log2 (gene expression valuen + 1). (C) Analysis of the 43 R2R3-MYB genes from D. officinale based on an analysis of qRT-PCR results, and visualized as a heatmap. The color scale represents log2 of the mean of gene expression. At least two biological replicates were performed.

Full size image

Screening R2R3-MYB genes involved in biosynthesis of water-soluble polysaccharides in D. officinale and their expression under water deficit stress

D. officinale is one of the most precious Chinese herbs with abundant WSPs in its stems33. The accumulation of WSPs in stems changes with plant growth34. To further screen the genes involved in the biosynthesis of WSPs, the expression of the 43 D. officinale genes belonging to the clusters C1 (AtMYB5), C2 (S6), C8 (S21), C26 (S2), C27 (S3), C30 (S8), C33 (S13), C34 (AtMYB26/AtMYB67/AtMYB103) and C35 (AtMYB46/AtMYB83) was analyzed across five plant developmental stages. In the S1 stage, plants were in the vegetative stage and had lowest WSP content (Fig. 5A,B). In stages S2-4, plants stopped growing and accumulated WSPs rapidly (Fig. 5A,B). In the S5 stage, plants started to undergo senescence and degradation of WSPs (Fig. 5A,B).

Figure 5
figure 5

Screening candidate R2R3-MYB genes related to the biosynthesis of WSPs in D. officinale by qRT-PCR. (A) Five stages of D. officinale stems were used to analyze gene expression. (B) Polysaccharide content in the stems of five stages. DW, dry weight. (C) Candidate genes identified using qRT-PCR. Details pertaining to S1-S5 can be found in the materials and methods and results sections. Bars represent mean ± SD (n = 3). Three biological replicates were performed. Different letters in each bar are significantly different at P < 0.05 (Duncan’s multiple range test).

Full size image

Ten genes, out of the 43 tested, showed an expression pattern that mirrored the accumulation pattern of WPS. Among these genes, six peaked at S2 (DoMYB28, −74, −75, −81, −97 and −111) and four at S3 (DoMYB27, −29, −54 and −78). All these genes were poorly expressed at S1, and most of them at S5 returned to the levels of S1 (Fig. 5C). The 10 MYB genes included two (DoMYB74 and −75) from C2 (S6), three (DoMYB27, −97, and −111) from C8 (S21), three (DoMYB28, −54, and −78) from C26 (S2) and two (DoMYB29 and −81) from C33 (S13). The remaining genes were either not detected or were inconsistent with changes in polysaccharide content (Supplementary Fig. 2).

In our previous studies in D. officinale, the genes of the WSP biosynthetic pathway were shown to be involved in abiotic stresses influenced by PEG and NaCl treatment34,36. Thus, the expression of candidate genes under 150 g/L PEG, 300 mM mannitol and 250 mM NaCl treatments was analyzed. Our results indicate that most of the candidate genes were up-regulated under salinity stress, i.e. DoMYB28, −29, −54, −75, −78, −81 and −111. Only two genes (DoMYB74 and −75) were up-regulated after exposure to PEG while only one gene (DoMYB81) was down-regulated after exposure to both PEG and mannitol (Fig. 6). DoMYB27 expression showed no significant difference between stress treatments and the control (Fig. 6). DoMYB75 was up-regulated in response to PEG and NaCl, similar with other WSP biosynthetic pathway genes such as DoPMM36 and DoCSLAs34.

Figure 6
figure 6

Analysis of the expression level of ten R2R3-MYB genes by qRT-PCR after D. officinale seedlings were subjected to PEG (150 g/L), mannitol (300 mM) and salt (NaCl, 250 mM) treatments. Control are seedlings treated with ½MS medium supplemented with 20 g/L sucrose (pH 5.4). Bars represent mean ± SD of three technical replicates. Three biological replicates of each treatment were performed. *indicates P < 0.05; **indicates P < 0.001 between control and stress treatments following Dunnett’s test.

Full size image

Overexpression of DoMYB75 in A. thaliana confirms its role in the biosynthesis of water-soluble polysaccharides

DoMYB75 may be involved in WSP biosynthesis, so it was used in further analyses. In a phylogenetic tree, the C2 (S6) clade was divided into two branches: cluster I included proteins of monocots (P. equestris and D. officinale) while cluster II only contained the proteins of A. thaliana, a dicot (Supplementary Fig. 3A). The complete CDS of DoMYB75 without a termination codon was cloned into an over-expression vector (pCABIA 1302 vector) driven by the CaMV-35S promoter (Supplementary Fig. 3B). DoMYB75 was detected and expressed in DoMYB75 transgenic Arabidopsis lines, but not in WT plants (Fig. 7A,B). Overexpression of DoMYB75 showed no difference in germinating seeds or seedlings (Fig. 7C). The average WSP content of seeds of all transgenic plants (around 79 mg/g DW) was significantly higher than WT plants (~69 mg/g DW, Fig. 7D). This indicates that DoMYB75 plays a role in WSP biosynthesis in A. thaliana seeds.

Figure 7
figure 7

Characterization of the DoMYB75 gene in the biosynthesis of WSPs. (A) Analysis of the DoMYB75 gene in wild type (WT) and transgenic lines by semi-quantitative PCR. (B) Analysis of the DoMYB75 gene in WT and transgenic lines by qRT-PCR. Expression levels were calculated relative to transgenic line 1. (C) Germinating seeds (one day after stratification) and seedlings (5 days after stratification) of WT and transgenic lines showed no obvious phenotypic changes. (D) Content of WSPs in mature dry seeds of A. thaliana. **Indicates P < 0.01 between WT and transgenic lines following Dunnett’s test.

Full size image

Discussion

Identification and classification of MYB proteins

In this study, we identified 159 and 165 MYB genes from P. equestris and D. officinale genomes, respectively. Although these orchid species belong to different genera, the number of MYB genes in their genomes is similar. Four groups of MYB proteins were found in D. officinale, similar to previous studies in other plant species such as A. thaliana6, G. max2,13 and P. bretschneideri8. However, no 4R-MYB gene was found in the genome of P. equestris, which had only one atypical MYB gene, namely the CDC5-type gene. Another monocot, rice, contained one CDC5-type gene but not a 4R-MYB gene in its genome11, similar to P. equestris in our study. This suggests that 4R-MYB genes are not found in all higher plants and might not play essential roles in all plants.

The MYBR proteins in both orchids could be divided into five subfamilies, with CCA1/RR-like as the largest subfamily, and containing two TRB-like genes in their genomes. Five MYBR subfamilies were also found in other higher plants such as A. thaliana, S. lycopersicum, V. vinifera, and B. distachyon, with the number of CCA1/RR-like genes ranging from 21 to 42, and with all plants just containing only one or two TRF-like genes in their genomes2. The R2R3-MYB group is largest group of the MYB family, with more than 100 members having been found in the genomes of both monocots and dicots.

The two orchids contain 115 and 117 R2R3-MYB genes in their genomes, respectively. Feller et al.12 predicted that the plant R2R3-MYB group underwent extensive amplification before the separation of monocots and dicots but after the separation of plants and animals. This explains why many members of R2R3-MYB genes have been found in orchids. The majority of clades of the R2R3-MYB family are present in both orchids and in the model plant A. thaliana (Supplementary Fig. 1), suggesting that they were present before the divergence between monocots and dicots. However, there is also an orchid-specific (C17) and a P. equestris-specific clade (C11), while in both orchids, C3 (S15), C4 (S5) and C13 (S23) are missing (Table 2). It can be deduced that several R2R3-MYB genes might have been lost in orchids during evolution, or experienced an amplification process that might have caused a change in their function in dicots.

The MYB genes play a role in abiotic stress responses

The MYB genes involved in abiotic stress responses have been widely investigated in plants and have mainly focused on the R2R3-MYB group. For example, R2R3-MYB genes were involved in high temperature stress by increasing the levels of cellular abscisic acid60, cold stress by regulating CBF genes or ascorbic acid synthesis56,61, drought stress62,63, and salt stress by regulating ABA signaling64,65. Among the 117 R2R3-MYB D. officinale genes, only nine that were modulated by low temperature were found in this study (Fig. 3). DoMYB28, −29, −54, −75, −78, −81, and −111 were up-regulated under salinity stress (Fig. 6). The DoMYB74 and DoMYB75 homologues, which were recognized as the C2 (S6) clade of R2R3-MYB genes, displayed different expression patterns in response to the two osmotic stresses, PEG and mannitol (Fig. 6). This may due to different cis-responsive elements among their transcriptional regulatory regions. For example, the putative promoter of DoMYB75 contained two ABA response elements, three dehydration response elements, one drought response element, three ethylene response elements, and six MeJA response elements, while DoMYB74 had only one ABA response element, one dehydration response element, two ethylene response elements, and four MeJA response elements, but no drought response element in its putative promoter (Supplementary Table 3).

MYBR genes are also involved in stress responses. For example, the MYB-related gene AQUILO improves cold tolerance in transgenic A. thaliana and caused the accumulation of oligosaccharides66, OsMYB48-1 is involved in salt stress by regulating the expression of stress-related genes67, and OsMYBR1 promoted drought stress in transgenic rice by increasing the free proline and soluble sugar content and up-regulated the expression of stress-related genes under drought treatment68. In this work, nine DoMYBR genes were up-regulated by cold stress, similar to the above studies, suggesting that the MYB genes in the MYBR or R2R3-MYB groups play roles in plant abiotic stress responses.

The involvement of R2R3-MYB genes in polysaccharide biosynthesis

Secondary cell walls (SCWs), which are mainly found in plant stems, are primarily composed of cellulose, lignin and hemicelluloses (xylan and glucomannan)20. Several TFs are involved in the regulation of SCW biosynthesis. MYB TFs make up the vast majority of TFs in transcriptional regulation of SCW biosynthesis21. The R2R3-MYB genes involved in plant SCW biosynthesis are thought to regulate the biosynthesis of SCW polysaccharides. For example, AtMYB46 acts as a regulator in SCW formation and directly regulates the expression of CSLA9, which encodes mannan synthase in A. thaliana22,69. AtMYB75 in the C2 (S6) group acts as a regulator in cell wall thickening, testa, as well as biosynthesis of anthocyanins70,71,72. Another MYB gene, AtMYB113, in the C2 (S6) subgroup increases pigment production and results in strongly visible anthocyanin pigmentation73. 35S::DoMYB75 transgenic Arabidopsis lines showed no anthocyanin, possibly due to the low levels of sucrose present in the medium, but increased WSPs in seeds of transgenic lines (Fig. 7).

In conclusion, 159 and 165 MYB genes were identified from P. equestris and D. officinale genomes, respectively. They could be classified into four groups in both orchids: MYBR, R2R3-MYB, 3R-MYB and atypical MYB proteins. Only three R2R3-MYB genes and 12 MYBR genes from D. officinale were up-regulated under low temperature, suggesting that MYB genes may play a role in the cold stress response in this orchid. Ten R2R3-MYB genes with an expression pattern corresponding to WSP accumulation were identified and regarded as the candidate genes involved in WSP biosynthesis. Over-expression of one candidate gene (DoMYB75) in A. thaliana caused the accumulation of WSPs in A. thaliana seeds.