Abstract

Bletilla striata polysaccharide (BSP) is the main component of Bletilla striata, which has important pharmacological and pharmacological effects; however, due to the lack of genetic data, the metabolic pathways of BSP remain unclear. For this study, 11 representative resources of B. striata were analyzed, and the BSP contents of the different samples were significantly different; however, the monosaccharide composition of BSP was glucose and mannose. The representative samples were selected to observe their life history in situ, which were then divided and cultured in a greenhouse. Finally, samples from various organs of different plants were combined for transcriptome sequencing using the Illumina system. Our results summarized the BSP metabolic pathway, and we found that there were eight enzyme genes involved in biosynthesis, but these genes showed tissue specificity. Following qRT-PCR validation and comparative analysis, manA showed the highest expression; however, there were significant differences between the two germplasm resources in which the BSP content was significantly different, while UGP2, GPI, PMM, and GMPP had significant differences between the two samples. In summary, this study lays the foundation for further research into BSP metabolism and other physiological processes at the molecular level.

Keywords:
Bletilla striata; transcriptome; germplasm resources; polysaccharide biosynthesis; qRT-PCR

Introduction

Bletilla striata, typically known as an important species of the Bletilla genus, is a significant medicinal compound for the treatment of many diseases, as recorded by Chinese pharmacopoeia (Chen et al., 2018Chen Z, Cheng L, He Y and Wei X (2018) Extraction, characterization, utilization as wound dressing and drug delivery of Bletilla striata polysaccharide: A review. Int J Biol Macromol 120:2076-2085.; Zhang et al., 2019aZhang C, Gao F, Gan S, He YN, Chen ZJ, Liu XW, Fu CM, Qu Y and Zhang JM (2019a) Chemical characterization and gastroprotective effect of an isolated polysaccharide fraction from Bletilla striata against ethanol-induced acute gastric ulcer. Food Chem Toxicol 131:110539.). It is an ornamental flower but belongs to the Orchidaceae family (He et al., 2017He X, Wang X, Fang J, Zhao Z, Huang L, Guo H and Zheng X (2017) Bletilla striata: Medicinal uses, phytochemistry and pharmacological activities. J Ethnopharmacol 195:20-38.). Wild B. striata sources are primarily distributed across Yunnan, Shaanxi, Gansu, Hubei, and Zhejiang Provinces of China at elevations from 110-3,200 m (He et al., 2017He X, Wang X, Fang J, Zhao Z, Huang L, Guo H and Zheng X (2017) Bletilla striata: Medicinal uses, phytochemistry and pharmacological activities. J Ethnopharmacol 195:20-38.), as well as the Korean Peninsula and Japan (Chung et al., 2013Chung MY, López-Pujol J and Chung MG (2013) Low genetic diversity in marginal populations of Bletilla striata (Orchidaceae) in southern Korea: Insights into population history and implications for conservation. Biochem Syst Ecol 46:88-96.). At present, various constituents have been extracted and isolated from dried B. striata pseudobulbs, including polysaccharides, benzyl, phenanthrax, astragalus, terpenes, ethers, and anthocyanins (Guo et al., 2016Guo JJ, Dai BL, Chen NP, Jin LX, Jiang FS, Ding ZS and Qian CD (2016) The anti-Staphylococcus aureus activity of the phenanthrene fraction from fibrous roots of Bletilla striata. BMC Complement Altern Med 16:491.; Li et al., 2018Li JY, Kuang MT, Yang L, Kong QH, Hou B, Liu ZH, Chi XQ, Yuan MY, Hu JM and Zhou J (2018) Stilbenes with anti-inflammatory and cytotoxic activity from the rhizomes of Bletilla ochracea Schltr. Fitoterapia 127:74-80.).

However, polysaccharides are one of the main components, which plays a variety of biological functions (Wang and Meng, 2015Wang W and Meng H (2015) Cytotoxic, anti-inflammatory and hemostatic spirostane-steroidal saponins from the ethanol extract of the roots of Bletilla striata. Fitoterapia 101:12-18.; Zhang et al., 2019bZhang N, Tang ZF, Zhang J, Li X, Yang ZQ, Yang C, Zhang ZF and Huang ZX (2019b) Comparative transcriptome analysis reveals the genetic basis underlying the biosynthesis of polysaccharides in Hericium erinaceus. Bot Stud 60:15.). Research has revealed that BSP not only facilitates coagulation by inducing the expression of vascular endothelial growth factor (VEGF), but also prevent gastric ulcers by forming a protective film that coats the stomach wall, and is a common immune modulator. (Luo et al., 2018Luo L, Zhou Z, Xue J, Wang Y, Zhang J, Cai X, Liu YQ and Yang F (2018) Bletilla striata polysaccharide has a protective effect on intestinal epithelial barrier disruption in TAA-induced cirrhotic rats. Exp Ther Med 16:1715-1722., 2019Luo L, Liu YQ, Cai X, Wang Y, Xue J, Zhang J and Yang F (2019) Bletilla striata polysaccharides ameliorates lipopolysaccharide-induced injury in intestinal epithelial cells. Saudi J Gastroenterol 25:302-308.; Zhang et al., 2019aZhang C, Gao F, Gan S, He YN, Chen ZJ, Liu XW, Fu CM, Qu Y and Zhang JM (2019a) Chemical characterization and gastroprotective effect of an isolated polysaccharide fraction from Bletilla striata against ethanol-induced acute gastric ulcer. Food Chem Toxicol 131:110539.). The structure and function of BSP is closely related to its biological activity, which is often modified or cross-linked with other substances for use as a biomaterial in wound healing and drug delivery applications (Wang et al., 2018Wang LC, Wu YJ, Li JS, Qiao HZ and Di LQ (2018) Rheological and mucoadhesive properties of polysaccharide from Bletilla striata with potential use in pharmaceutics as bio-adhesive excipient. Int J Biol Macromol 120:529-536.). Although B. striata has numerous functions, data on its genetic makeup remains scar3ce, and the polysaccharide metabolism pathway is not clear. Therefore, it is necessary to conduct in-depth genetic research to discover its genes and their functionalitiess.

With the development and implementation of advanced DNA sequencing technologies, the biological metabolisms of active components may be discerned in medicinal plants (Fu et al., 2019Fu SH, Lei M, Zhang YQ, Deng ZM, Shi J and Hao DD (2019) De novo transcriptome analysis of Tibetan medicinal plant Dysphania schraderiana. Genet Mol Biol 42:480-487.; Liang et al., 2019Liang TT, Zou LQ, Sun SJ, Kuang XJ, Wei JH, Wang LZ, Li Y and Sun C (2019) Hybrid sequencing of the Gynostemma pentaphyllum transcriptome provides new insights into gypenoside biosynthesis. BMC Genomics 20:632.; Niazian, 2019Niazian M (2019) Application of genetics and biotechnology for improving medicinal plants. Planta 249:953-973.). There are few studies on the biosynthesis of polysaccharides (this research has focused primarily on fungal species) due to their unique structures and large molecular weights (Ribeiro et al., 2012Ribeiro DA, Cota J, Alvarez TM, Brüchli F, Bragato J, Pereira BM, Pauletti BA, Jackson G, Pimenta MT, Murakami MT et al. (2012) The Penicillium echinulatum secretome on sugar cane bagasse. PLoS One 7:e50571.; Zhang et al., 2019bZhang N, Tang ZF, Zhang J, Li X, Yang ZQ, Yang C, Zhang ZF and Huang ZX (2019b) Comparative transcriptome analysis reveals the genetic basis underlying the biosynthesis of polysaccharides in Hericium erinaceus. Bot Stud 60:15.). The biosynthesis pathways of Ganoderma lucidum polysaccharides provide a preliminary study of the various synthesis pathways of medicinal plants (Gao et al., 2015Gao JP, Wang D, Cao LY and Sun HF (2015) Transcriptome sequencing of Codonopsis pilosula and identification of candidate genes involved in polysaccharide biosynthesis. PLoS One 10:e0117342.).

The pharmacopoeia of the People's Republic of China has entered the genus Polygonatum cyrtonema and Polygonatum sibiricum, both of which employ polysaccharides as their key components, and their polysaccharide metabolic pathways have been elucidated (Wang et al., 2017Wang SQ, Wang B, Hua WP, Niu JF, Dang KK, Qiang Y and Wang ZZ (2017) De Novo assembly and analysis of Polygonatum sibiricum transcriptome and identification of genes involved in polysaccharide biosynthesis. Int J Mol Sci 18:1950.; Wang et al., 2019aWang CK, Peng DY, Zhu JH, Zhao DR, Shi YY, Zhang SX, Ma KL, Wu JW and Huang LQ (2019a) Transcriptome analysis of Polygonatum cyrtonema Hua: Identification of genes involved in polysaccharide biosynthesis. Plant Methods 15:65.). It has been verified, via hot water extraction and ion exchange chromatography purification, that BSP is composed of glucose and mannose; however, the genetic data related to its polysaccharide biosynthesis is still unknown (Kong et al., 2015Kong LS, Yu L, Feng T, Yin XJ, Liu TJ and Dong L (2015) Physicochemical characterization of the polysaccharide from Bletilla striata: Effect of drying method. Carbohyd Polym 125:1-8.).

For this study, comparative transcriptomes were analyzed using the roots, stems, leaves, flowers and seeds of B. striata. After determining the candidate genes involved in BSP biosynthesis, we selected the highest and lowest BSP germplasm contents, respectively, and verified the quality of the database and related genes involved in the BSP metabolic pathway by quantitative real-time PCR (qRT-PCR). The results of this study provided a theoretical basis for the study of B. striata.

Materials and Methods

Ethics statement

All experimental materials for this study were collected in China, but did not cause the species to be threatened or endangered. This research was conducted at the National Engineering Laboratory for Resource Development of Endangered Chinese Crude Drugs in Northwest China, Xi'an, China.

Plant materials

The transcriptome-sequenced plants were collected from Zhen'an County, Shaanxi Province, in China on July 20, 2013. Once the plants were transplanted with soil at the greenhouse of National Engineering Laboratory for the Development of Endangered Medicinal Materials Resources in Northwest China, we separated apart the pseudobulbs of B. striata one by one for individual propagation, and finally the sample was divided into 11 plants. We chose to extract samples from the flowering to fruiting stage. At this time, the upper portion of the peduncle has a flower bud, whereas the lower part has a capsule that is formed by fertilization. Six samples were randomly selected for sampling. The roots, stems, leaves, flowers and fruits of different organs from all of the specific plants were collected equally and mixed together. After the samples (each about 2.0 g) were collected, they were flash-frozen with liquid nitrogen and stored at −80 °C.

Eleven regionally representative populations of B. striata were collected from 2014 to 2015, all of which were identified by Prof. Yaping Xiao (College of Life Science, Shaanxi Normal University). The origins of the rhizome samples collected from different germplasms are specified in Table S1, and all materials were grown in the laboratory germplasm field.

Isolation, purification and detection of polysaccharides

The method of extracting the polysaccharides from B. striata rhizomes drew from the relevant articles previously published by our research group (Niu et al., 2019Niu JF, Wang SP, Wang BL, Chen LJ, Zhao GM, Liu S, Wang SQ and Wang ZZ (2019) Structure and anti-tumor activitity of a polysaccharide from Bletilla ochracea Schltr. Int J Biol Macromol 154:1548-1555.). Once the fresh plant material was dried at 60 °C, it was ground to a powder and passed through a 60 mesh screen. The removal of lipids from the dry powder sample (100g) was achieved mixing it with petroleum ether, followed by twice extracting with distilled water (80 °C) for three hours (1:30, w: v). Once the extract was concentrated on a rotary evaporator, 95% alcohol (about a four-fold volume) was added and left overnight at 4 °C, after which the polysaccharide was precipitated and filtered. Following the removal of proteins using a repeated freeze-thaw method, the crude BSP polysaccharides were obtained via dialysis and lyophilization.

BSP monosaccharide composition

GC-MS was employed to determine the composition of the BSP monosaccharides. The BSP sample (2.0 mg) was initially added to 2 M trifluoroacetic acid (TFA), and following hydrolyzation at 100 °C for 90 min, it was converted to alditol acetates. Subsequently, the acetylated structure was evaluated using an Agilent GC-MS QQQ7000 with a HP-5 fused silica capillary flame ionization detector.

The temperature gradient of the column was increased from a starting temperature of 160 °C at a rate of 4 °C min⁻¹ to 250 °C, which was held for five min. The temperature of the injector and detector of the instrument was set to 250 °C, N₂ was selected as the carrier gas and the flow rate was 1 mL min⁻¹. The H₂ flow rate and air flow rate were 30 mL min⁻¹, and 400 mL min⁻¹, respectively. The following standard monosaccharides were used as references: rhamnose, fucose, arabinose, xylose, mannose, galactose, and glucose.

Establishment of library and sequencing (mRNA-Seq)

Transcriptome sequencing plants were collected from Zhen'an County, in Shaanxi Province. Mixed samples of collected roots, stems, leaves, flowers, and fruits were separately sequenced. The total RNA for each sample was extracted following the instructions of the total RNA Isolation Kit (TIANGEN, Beijing, China). High quality RNA was evaluated for library building using the HiSeq 2500 platform.

De novo assembly and functional annotation

The Unigene library for this species was obtained by filtering rRNA and low-quality reads (two-terminal sequences) by data evaluation. Trinity software was used for the mixed assembly of sample data (Grabherr et al., 2011Grabherr MG, Haas BJ, Yassour M, Levin JZ, Thompson DA, Amit I, Adiconis X, Fan L, Raychowdhury R, Zeng Q et al. (2011) Full-length transcriptome assembly from RNA-Seq data without a reference genome. Nat Biotechnol 29:644-652.). The most important transcripts were selected as unigenes (Mortazavi et al., 2008Mortazavi A, Williams BA, McCue K, Schaeffer L and Wold B (2008) Mapping and quantifying mammalian transcriptomes by RNA-Seq. Nat Methods 5:621-628.), and Blastx (Altschul et al., 1997Altschul SF, Madden TL, Schäffer AA, Zhang J, Zhang Z, Miller W and Lipman DJ (1997) Gapped BLAST and PSI-BLAST: A new generation of protein database search programs. Nucleic Acids Res 25:3389-3402.) comparison was carried out with four common protein databases: NR (non-redundant protein database) (Pruitt et al., 2005Pruitt KD, Tatusova T and Maglott DR (2005) NCBI Reference Sequence (RefSeq): A curated non-redundant sequence database of genomes, transcripts and proteins. Nucleic Acids Res 33:D501-D504.), Swiss-Prot Database (Apweiler et al., 2004Apweiler R, Bairoch A, Wu CH, Barker WC, Boeckmann B, Ferro S, Gasteiger E, Huang H, Lopez R, Magrane M et al. (2004) UniProt: the Universal Protein knowledgebase. Nucleic Acids Res 32:D115-D119.), KEGG Database (Kyoto Encyclopedia of Genes and Genomes protein Database) (Kanehisa et al., 2004Kanehisa M, Goto S, Kawashima S, Okuno Y and Hattori M (2004) The KEGG resource for deciphering the genome. Nucleic Acids Res 32:D277-D280.), and COGs (Clusters of Orthologous Groups) (Tatusov et al., 2000Tatusov RL, Galperin MY, Natale DA and Koonin EV (2000) The COG database: A tool for genome-scale analysis of protein functions and evolution. Nucleic Acids Res 28:33-36.). The functional annotation of the relevant unigene coding proteins was performed by comparing the data from proteins with similar results in the database.

Analysis of differential gene expression

Firstly, we compared the sequencing results with the published Unigenes databas, using Bowtie software (version 1.1.2, JHU, Baltimore, MD, USA) (Langmead et al., 2009Langmead B, Trapnell C, Pop M and Salzberg SL (2009) Ultrafast and memory-efficient alignment of short DNA sequences to the human genome. Genome Biol 10:R25.). Secondly, RSEM (Li and Dewey, 2011Li B and Dewey CN (2011) RSEM: Accurate transcript quantification from RNA-Seq data with or without a reference genome. BMC Bioinformatics 12:323.) was performed to reflect the expression levels of different genes, and the abundance of corresponding transcripts was determined by calculating the values of RPKM (Reads Per Kilobase of transcript per Million mapped reads) (Trapnell et al., 2010Trapnell C, Williams BA, Pertea G, Mortazavi A, Kwan G, Van Baren MJ, Salzberg SL, Wold BJ and Pachter L (2010) Transcript assembly and quantification by RNA-Seq reveals unannotated transcripts and isoform switching during cell differentiation. Nat Biotechnol 28:511-515.). Finally, DESeq software (Anders and Huber, 2010Anders S and Huber W (2010) Differential expression analysis for sequence count data. Genome Biol 11:R106.) was used to further compare the gene expression patterns from different germplasm resources, SXZA and AHXC, and five different organs: roots, stems, leaves, flowers and fruits.

Identification of genes related to BSP biosynthetic pathway

Through the annotation of unigenes in the KEGG database, the genes involved in the polysaccharide metabolism pathway were summarized as candidate genes. Candidate genes for further reference included KEGG metabolic pathways of starch and sucrose metabolism, as well as amino sugar and nucleotide sugar metabolism on the metabolic pathways involved in the gene's comments. As a result, further screening participated in the reaction of related enzymes, where all of the enzymes and Arabidopsis thaliana annotated information in the database, combined with Swiss-Prot, and a further annotated database relating to enzyme glycosyl transferase (GTs, Glycosyltransferases).

Real-time PCR

We selected GAPDH as the reference gene, and a total of eight candidate genes were screened, where the related primers being designed using Premier 3.0, with all of the above presented in Table S2. The total RNA was extracted from two germplasm groups, SXZA and AHXC, as well as five different organs, roots, stems, leaves, flowers, and fruits. Reverse transcription to cDNA was performed according to the instructions of Takara's Prime-Script^TM RT Master Mix (TaKaRa, Kyoto, Japan).

Real-time PCR experiments were performed on a Light Cycler 96 Instrument (Roche, Germany) according to the 2×Sybr Green qPCR Mix instructions (Aidlab, Beijing, China). The relative expression was calculated using the 2^−ΔΔCt method (Schmittgen and Livak, 2008Schmittgen TD and Livak KJ (2008) Analyzing real-time PCR data by the comparative C(T) method. Nat Protoc 3:1101-1108.). All qRT-PCR experiments involved in this study underwent three biological and three technical replications.

Statistical analysis

All experimental results in this study were expressed as a mean standard deviation (SD), using one-way analysis of variance (ANOVA) of DPS, and Duncans multivariate test was performed. Values were considered statistically significant, when p < 0.05.

Results

BSP composition and B. striata germplasm (SXZA) content

Water-soluble polysaccharides were extracted from the dried rhizomes of the B. striata germplasm (SXZA), which were collected from Zhen'an County, in Shaanxi Province, China. As shown in Figure 1, the content of BSP was 22.85 ± 0.04%, which was comprised of glucose and mannose, and the mole ratio was 0.745:0.255 by gas chromatography. We observed the life history of the whole plant from 2012 to 2013 on the original land (E109°9’11″, N33°25’33″). The plants were segmented into multiple plants by pseudobulbs, and the flowering to fruiting stage was further selected. Transcriptome sequencing was performed by mixing the different organs of different plants.

Polysaccharide yields from different germplasm resources

A total of 11 germplasm resources of B. striata were collected from Shaanxi, Anhui, Hubei, Sichuan, and Henan Provinces from 2013 to 2015, and the polysaccharide content were analyzed between the different samples. As shown in Table 1, the highest polysaccharide content was 36.48 ± 0.27% in Shangzhou, of Shaanxi Province, whereas the lowest sample was 20.56 ± 0.04% in Xuancheng, of Anhui Province.

Transcriptome sequencing, de novo assembly, and assessment of assembly program

Using the Illumina sequencing system, we obtained 11.4 Gb of clean data from five mixed samples. The clean data and Q30 values of each sample were greater than 2.04 Gb and 93.26% (Table 2). Following the removal of the adapter sequence, ambiguous reads, and low-quality reads in order, the remaining data was compiled using Trinity software, where after 4,784,551 contigs were obtained and the N50 was 48 bp. With clustering contigs based on pairing information and the similarity of overlapping sequences, we obtained 115,004 transcripts and the N50 was 1,927bp. The most important transcripts were selected as unigenes, where not only were 58,397 unigenes obtained, but their N50 attained 1,385 bp. The lengths greater than 1,000 bp and 500 bp accounted for 23.97% and 42.54%, respectively (Table 3).

Functional annotations

All unigenes generated by the Illumina platform were compared with NR, Swiss-Prot, GO, Subsequently, amino acid sequences were predicted and, unigenes were compared with the Pfam database to obtain annotation information. In these public databases, a total of 33,344 unigenes (57.10%) were consistent with the homologous sequences. Among them, 13,232 (22.66%) were 300 to 1,000 bp long while 13,224 (22.64%) were longer than 1,000 bp (Table 4).

When compared with the NR database, approximately 56.92% (33,239) of the unigenes were successfully annotated. Among these unigenes, 17.80% had the best matches to the Piriformospora indica sequences.

Unigenes that completed the NR database comparison were further compared with the GO database, and 21,036 (36.02%) unigenes were annotated and assigned to 56 functional groups based on the GO terms annotations (Figure 2A). Among these unigenes related to “cell components”, 12,799 (60.84%) were classified as “cell” (GO: 0005623), followed by “organelles” (GO: 0043226; 11,166 or 53.08%). However, among the unigenes related to “molecular function”, 47.69% were classified as catalytic activity (GO: 0003824; 10,032) and 46.73% were classified as binding activity (GO: 0005488; 9,830). The GO terms of biological processes were divided into “cellular process” (GO: 0009987; 58.58% or 12,322) and “metabolic process” (GO: 0008152; 63% or 13,254).

By comparing the transcriptome data of B. striata with the public COG database, the function and classification of unigenes were predicted. In general, 11,282 unigenes (19.33%) were divided into 24 categories (Figure 2B). Among these, the largest functional group was “General function prediction only” (2,898, 25.29% of the COG annotations), followed by “Replication, recombination and repair” (1,280, 11.35%) and “Translation, ribosomal structure and biogenesis” (1,206, 10.69%).

Finally, through comparisons with four public protein databases, we obtained 58,063 CDSs, which accounted for 99.46% of all identified unigenes. The length distribution of CDSs revealed that 21,443 were from 200 to 1,000 bp long, and 25.82% were longer than 500 bp. In addition, 5,847 (10.07%) were from 1,000 to 2,000 bp long, 1,353 (2.33%) were 2,000 to 3,000 bp long, and 557 (0.96%) were longer than 3,000 bp (Figure S1).

Identification of transcription factors

When compared with the plant transcription factor database (Jin et al., 2014Jin JP, Zhang H, Kong L, Gao G and Luo JC (2014) PlantTFDB 3.0: A portal for the functional and evolutionary study of plant transcription factors. Nucleic Acids Res 42:D1182-D1187.), we obtained 13,764 unigenes that had a good matching coefficients (identity >80%). These unigenes belonged to 59 families of plant transcription factors (Table 5), of which the most abundant unigenes were bHLH, NAC, MYB-related, MYB, WRKY, FAR1, B3, ERF, C2H2, and bZIP families.

KEGG pathway analysis

To further elucidate the biological metabolic processes of B. striata, we obtained 7,736 unigenes which were involved 112 metabolic pathways based on a comparison with the KEGG database (Table S3).

Among these unigenes, 529 were related to “ribosome pathways”, 364 played the “Oxidative phosphorylation pathways” function, 276 were involved in “Protein processing in endoplasmic reticulum”, 254 for “Protein processing in endoplasmic reticulum Spliceosome”, and 253 for “Glycolysis/Gluconeogenesis” (Table 6). Through analysis, a total of 1,364 unigenes were mapped to related metabolic pathways of carbohydrates, which mainly involved metabolic pathways such as glycolysis/gluconeogenesis, starch and sucrose metabolism, pyruvate metabolism, and amino sugar and nucleotide sugar metabolism (Table 7).

Candidate genes involved in BSP biosynthesis

To better understand the metabolic pathways of BSP, referring to the annotation information on the KO 00500 and KO 00520 metabolic pathways in the KEGG database, we determined the key enzyme genes involved in these pathways (Table 8).

The precursors of all plant polysaccharides were nucleoside-bisphosphate-sugar (NDP-sugar). Under the action of different glycosyltransferases (GTs), NDP-sugars were catalyzed to increased sugar chains to form new polysaccharides (Gutmann et al., 2017Gutmann A, Lepak A, Diricks M, Desmet T and Nidetzhy B (2017) Glycosyltransferase cascades for natural product glycosylation: Use of plant instead of bacterial sucrose synthases improves the UDP-glucose recycling from sucrose and UDP. Biotechnol J 12:1600557.; Yuan et al., 2018aYuan S, Yin S, Liu M and Kong JQ (2018a) Isolation and characterization of a multifunctional flavonoid glycosyltransferase from Ornithogalum caudatum with glycosidase activity. Sci Rep 8:5886.).

The results indicated that the polysaccharides of B. striata were derived from glucose and mannose; therefore, the metabolic processes of UDP-glucose (UDP-Glc) and GDP-mannose (GDP-Man) were the key metabolic pathway of BSP. UDP-Glc and GDP-Man were derived from glucose-6-phosphate (Glc-6P) and fructose-6-phosphate (Fru-6P), respectively. Under the action of NDP-sugar interconversion enzymes (NDP-sugar conversion enzymes), they carried out the next reaction.

Referring to previous research progress on polysaccharide metabolism (Liepman et al., 2010Liepman AH, Wightman R, Geshi N, Turner SR and Scheller HV (2010) Arabidopsis - a powerful model system for plant cell wall research. Plant J 61:1107-1121.; Wang et al., 2019aWang CK, Peng DY, Zhu JH, Zhao DR, Shi YY, Zhang SX, Ma KL, Wu JW and Huang LQ (2019a) Transcriptome analysis of Polygonatum cyrtonema Hua: Identification of genes involved in polysaccharide biosynthesis. Plant Methods 15:65.; Zhang et al., 2019bZhang N, Tang ZF, Zhang J, Li X, Yang ZQ, Yang C, Zhang ZF and Huang ZX (2019b) Comparative transcriptome analysis reveals the genetic basis underlying the biosynthesis of polysaccharides in Hericium erinaceus. Bot Stud 60:15.), the plant polysaccharide pathway consisted of three steps. Firstly, sucrose was converted to Glc-6P and fructose under the action of β-fructofuranosidase (encoded by sacA) and phosphotransferase system (encoded by scrA). Fructokinase (encoded by scrK) and hexokinase (encoded by HK) were also involved in the biosynthesis of fructose to Fru-6P. Under the action of related enzymes, Glc-6P and Fru-6P were formed into UDP-Glc and GDP-Man respectively (Zhang et al., 2017Zhang J, Yu L, Lin M, Yan Q and Yang ST (2017) n-Butanol production from sucrose and sugarcane juice by engineered Clostridium tyrobutyricum overexpressing sucrose catabolism genes and adhE2. Bioresource Technol 233:51-57.; Geiger et al., 2019Geiger J, Doelker R, Salö S, Roitsch T and Dalgaard LT (2019) Physiological phenotyping of mammalian cell lines by enzymatic activity fingerprinting of key carbohydrate metabolic enzymes: A pilot and feasibility study. BMC Res Notes 12:682.), which were glucose phosphotransase (encoded by pgm), uridine diphosphate glucose pyrophosphorylase (UGP2), glucose-6-phosphoisomerase (GPI), mannose-6-phosphoisomerase (manA), mannose phosphate enzyme (PMM), and mannose-1-phosphotransferase (GMPP).

Secondly, under the action of a series of enzymes, UDP-Glc formed NDP-sugars. The main pathways were UDP-Glc, which formed NDP-sugars through a series of enzymes. Thirdly, NDP-sugars were added to various sugar residue chains to form polysaccharide complex units under the action of glycosyltransferases (Figure 3).

In the first step reaction, glucose and fructose were converted under the action of a series of kinases, and the related enzymes included hexokinases (HKs) and fructokinases (scrKs). By comparison, we concluded that 537-bp Unigene 33226 was a HK homologue with the highest amino acid homology (93%) to HK in Dendrobium catenatum (Sequence ID: XM_020834402.2) (Figures S2, S3). In addition, the 1,488-bp Unigene 44091 was consistent in homology with the scrK from D. catenatum (Sequence ID: XM_020847040.2) (Figures S2, S4).

The results of the comparative transcriptome in Table 8 were found in the stem of the main accumulation site of BSP, and directed from Glc-6P through the intermediate product Fru-6P, Man-6P, and Man-1P. Eventually, GDP-Man, and related enzymes GPI, manA, PMM, and GMPP exhibited extremely strong activity, which means that the biotransformation process, from Glc-6P to GDP-Man may have been its primary metabolic pathway.

Unigenes involved in BSP metabolism were expressed by the RPKM value in transcriptome database. Under general conditions, the unigene with the maximum RPKM value was selected as the gene expression quantity (Table 9).

Depending on the abundance of the transcribed RNA-seq assay, we selected eight candidate unigenes; GMPP, PMM, manA, HK, scrK, GPI, pgm and UGP2. In the database, the least consistent with their homologs was scrK (Unigene 44,091; 98% identity), followed by PMM (Unigene 36,348; 99%), manA (Unigene 45,816; 99%), and GPI (Unigene 32,627; 99%). The remaining identity of the unigenes was 100%, which were GMPP (Unigene 39,307), HK (Unigene 33,226), pgm (Unigene 46,080) and UGP2 (Unigene 31559). All these details were aligned as shown in Table S4, which confirmed that the selection of unigenes related to BSP metabolism was quite conservative.

It can be seen from Table 8 that the expression levels of related enzyme genes in the roots, stems, leaves, and flowers of B. striata were different, among which GMPP, manA and GPI had the highest expression levels in stems. Both pgm and UGP2 were the most expressed in flowers, while PMM, HK, and scrK were most expressed in seeds.

Analysis of BSP biosynthetic pathway

In combination with the database annotation of the polysaccharide metabolic pathway and the experimental results of the monosaccharide composition of B. striata, we obtained the potential biosynthesis pathways for BSP formation from sucrose.

The BSP synthesis pathway primarily consisted of three steps. Firstly, under the action of related enzymes such as UGP2, pgm, GPI, manA, PMM, and GMPP, Glc-6-p and Fru-6-p formed UDP-Glc and GDP-Man, respectively. Secondly, the NDP-sugars were added to sugar residues to form complex polysaccharide units. Therefore, in the BSP metabolic pathway, UDP-Glc and GDP-Man were formed mainly by the key enzyme genes in step one, and finally complex units were formed through the NDP-sugars by NSEs in step two. Thirdly, under the action of related GTs, the NDP-sugars were added to the polysaccharide. The logarithms of the RPKM values of different tissues in the metabolic pathway were expressed as different colors (from left to right, the colors represent leaves, stems, roots, flowers, and seeds). Here, a normalized data processing method was used and the drawing was completed on a heat map generated by the R software (Figure 3).

Quantitative real-time PCR analysis

Quantitative real-time PCR can reflect the expression abundance of related genes in the BSP metabolic pathway, and we selected GAPDH as the experimental reference. We screened the expression patterns of GMPP, PMM, manA, HK, scrK, GPI, pgm, and UGP2 genes in the BSP metabolic pathways for different organs of B. striata (SXZA). According to the expression values of related genes in various tissues, HK, pgm, and UGP2 had the largest content in flowers, while GMPP, manA, and GPI had the highest content in stems. These genes may play an important role in the biosynthesis of BSP (Figure 4).

We further compared the expression patterns of different transcripts with the RPKM values of related genes in the transcriptome sequencing data group. GMPP, scrK, GPI, and UGP2 had the same expression patterns in the two groups of data, while the expression of the remaining four genes was inconsistent. Futher, according to the polysaccharide yield analysis of different germplasm resources, we selected the fresh plant pseudobulb (SXSZ-82 and AHXC-218) germplasm resources for experiments to verify the expression levels of related genes in different tissues and parts. Compared with the reference gene, it was concluded that there were no significant differences in expression level quantities between HK, scrK, and pgm. It was observed that manA had the highest expression between the two species; however, there were significant difference between them, while UGP2, GPI, PMM, and GMPP had considerable differences between the two samples (Figure 5).

Discussion

To date, this is the first time that De novo sequencing of the different organs of B. striata using the Illumina HiSeq 2500 platform explored the metabolic pathways of polysaccharides. After completing the assembly, we obtained 11.4 Gb of clean data and 58,397 unigenes, which made it possible to study the BSP pathway at the molecular level. Compared to the published databases of other medicinal orchid plants, B. striata had more unigenes than did Dactylorhiza hatagirea (37,371) and Gastrodia elata (56,884), but fewer than Dendrobium huoshanense (478,361), Cymbidium goeringii (85,868), and Dendrobium officinale (145,791) (Yuan et al., 2018bYuan YD, Yu MY, Jia ZH, Song XE, Liang YQ and Zhang JC (2018b) Analysis of Dendrobium huoshanense transcriptome unveils putative genes associated with active ingredients synthesis. BMC Genom 19:978.; Dhiman et al., 2019Dhiman N, Sharma NK, Thapa P, Sharma I, Kumar Swarnkar M, Chawla A, Shankar R and Bhattacharya A (2019) De novo transcriptome provides insights into the growth behaviour and resveratrol and trans-stilbenes biosynthesis in Dactylorhiza hatagirea - An endangered alpine terrestrial orchid of western Himalaya. Sci Rep 9:13133.; Wang et al., 2019bWang Y, Shahid MQ, Ghouri F, Ercisli S and Baloch FS (2019b) Development of EST-based SSR and SNP markers in Gastrodia elata (herbal medicine) by sequencing, de novo assembly and annotation of the transcriptome. 3 Biotech 9:292.).

The average unigene length was 777.11 bp, which was longer than D. hatagirea (551.10 bp) and D. huoshanense (764.73 bp), but less than C. goeringii (1,194 bp) and G. elata (844.34 bp). In our experimental results, 23.97% of the sequence length was greater than 1,000bp, which indicated that the use of the Illumina technology platform had a higher assembly quality and stability in the research of medicinal plants.

There have been published studies on the overall transcriptome of the B. striata plant over its entire developmental stage. The transcriptome sequencing of samples from the protocorm stage to root, stem, leaf, and pollenated capsule stages were performed, and 127,261 overlapping unigenes were identified with mean unigene lengths of 612 bp. The authors successfully screened five markers using EST-SSR technology for their identification with their relatives (Xu et al., 2018Xu DL, Chen HB, Aci M, Pan YC, Shangguan YN, Ma J, Li L, Qian G and Wang QY (2018) De novo assembly, characterization and development of EST-SSRs from Bletilla striata transcriptomes profiled throughout the whole growing period. PLoS One 13:e0205954.).

With the advent of the big data era and the development of omics technology, increasing numbers of studies on polysaccharide metabolic pathways have been conducted. Through the transcriptome analysis of medicinal plants Codonopsis pilosula using high-throughput sequencing technologies, the biosynthetic pathways of codonopsis polysaccharides were summarized and verified, which initially elaborated the polysaccharide metabolic pathways of medicinal plants (Gao et al., 2015Gao JP, Wang D, Cao LY and Sun HF (2015) Transcriptome sequencing of Codonopsis pilosula and identification of candidate genes involved in polysaccharide biosynthesis. PLoS One 10:e0117342.).

Similarly, polysaccharide is the main medicinal component of the Polygonatum genus. The polysaccharide metabolism pathways of P. cyrtonema and P. sibiricum were thoroughly investigated, which were consistent and comprised of three steps. However, due to the differences in monosaccharide composition and content, the expression of some key enzyme genes in the polysaccharide metabolic pathway may be different (Wang et al., 2017Wang SQ, Wang B, Hua WP, Niu JF, Dang KK, Qiang Y and Wang ZZ (2017) De Novo assembly and analysis of Polygonatum sibiricum transcriptome and identification of genes involved in polysaccharide biosynthesis. Int J Mol Sci 18:1950.; Wang et al., 2019aWang CK, Peng DY, Zhu JH, Zhao DR, Shi YY, Zhang SX, Ma KL, Wu JW and Huang LQ (2019a) Transcriptome analysis of Polygonatum cyrtonema Hua: Identification of genes involved in polysaccharide biosynthesis. Plant Methods 15:65.). The genome sequencing of D. candidum was completed, which provided ample support for the study of polysaccharide metabolism pathways in other orchids (Zhang et al., 2016aZhang GQ, Xu Q, Bian C, Tsai WC, Yeh CM, Liu KW, Yoshida K, Zhang LS, Chang SB, Chen F et al. (2016a) The Dendrobium catenatum Lindl. genome sequence provides insights into polysaccharide synthase, floral development and adaptive evolution. Sci Rep 6:19029.; Zhang et al., 2016bZhang JX, He CM, Wu KL, Teixeira da Silva JA, Zeng SJ, Zhang XH, Yu ZM, Xia HQ and Duan J (2016b) Transcriptome analysis of Dendrobium officinale and its application to the identification of genes associated with polysaccharide synthesis. Front Plant Sci 7:5.).

Through related bioinformatics software analysis, the candidate unigenes of the BSP metabolic pathway were identified. According to previous experimental results, the BSP composition was glucose and mannose; however, the molar ratio was different (Figure S5). Therefore, what was of more interest was the conversion between glucose and mannose, and the further formation of polysaccharide coincidence units. We further analyzed the expression levels of related genes in the polysaccharide pathway in different B. striata tissues and concluded that the expression levels of manA and UGP2 were the highest (Figure 4).

As is well recognized, UGP2 is the key enzyme gene for glucose-1-phosphate to form UDP-Glc, while manA is the key enzyme for fructose-6-phosphate to form mannose-6-phosphate in plants. The expression level of manA in plants was 3.36 times that of UGP2, while the molar ratio of mannose to glucose in the same material of B. striata was 2.91:1. The expression levels of the two key genes mentioned above were approximately consistent with the composition ratio of the two monosaccharides, and it is predicted that there may be some relationship between them, which requires verification through further experimentation.

The main component polysaccharide of B. striata accumulates in the pseudobulb (Liao et al., 2019Liao ZC, Zeng R, Hu LL, Maffucci KG and Qu Y (2019) Polysaccharides from tubers of Bletilla striata: Physicochemical characterization, formulation of buccoadhesive wafers and preliminary study on treating oral ulcer. Int J Biol Macromol 122:1035-1045.). We can observe from Figure 5 that manA has the highest expression in the stems of SXSZ-82 and AHXC-218 sample; however, the expression of the former is four times that of the latter. According to the experimental results in Table 1, the polysaccharide yield of the SXSZ-82 sample was 36.48%, which was much larger than 22.77% in the AHXC sample. While manA is a key gene inducing Glc-1P to the Man-6P pathway, we can understand that there should be a correlation between high-activity samples and highly expressed genes. The further focus would be the gene expression of key enzymatic polysaccharide pathways, and the study of their correlations between enzyme activity and the BSP content.

GMPP also showed strong biological activity in the B. striata polysaccharide metabolic pathway, as a key enzyme gene in the pathway, GMPP plays an important role in the accumulation of GDP-Man in plants. Similarly, in A. rabidopsis thaliana, the establishment of mutants and a yeast hybrid experiment also confirmed that GMPP was the rate-limiting enzyme synthesized by GDP-Man (Sawake et al., 2015Sawake S, Tajima N, Mortimer JC, Lao J, Ishikawa T, Yu X, Yamanashi Y, Yoshimi Y, Kawai-Yamada M, Dupree P et al. (2015) KONJAC1 and 2 are key factors for GDP-mannose generation and affect L-ascorbic acid and glucomannan biosynthesis in Arabidopsis. Plant Cell 27:3397-3409.). The same conclusion was also drawn in the study of the polysaccharide metabolic pathway of P. sibiricum (Wang et al., 2017Wang SQ, Wang B, Hua WP, Niu JF, Dang KK, Qiang Y and Wang ZZ (2017) De Novo assembly and analysis of Polygonatum sibiricum transcriptome and identification of genes involved in polysaccharide biosynthesis. Int J Mol Sci 18:1950.). In the process of studying the polysaccharide metabolism pathway, a complex network structure is formed due to the variety of monosaccharide components. It is possible to further implement such work by selecting species that are similar in type to BSP, have a small variety of structures and a well-defined overall structure.

Although the transcriptomes of different organs of B. striata were sequenced in this study, its bioinformatics database was established, and the enzyme genes in relevant BSP metabolic pathways were summarized. However, the research on B. striata is primarily focused on pharmacology, and there are relatively few studies on its molecular biology. The complexity of the BSP structure makes it difficult to study its monosaccharide composition. Therefore, future research should focus on verifying the functional prediction of related genes through modern molecular biology and proteomics techniques.

Conclusions

As a common Chinese medicinal plant, B. striata has a variety of pharmacological effects. In this study, a bioinformatics database was established by performing transcriptome analysis on different organs of B. striata. Our experimental results provided useful information for studying the BSP metabolic pathway of this species, and the functions of several genes were confirmed by qRT-PCR. These research results not only contribute to laying the foundation for further elaborating the biosynthesis of related polysaccharides, but also provide candidate genes for metabolite synthesis, which can be employed for the evaluation of germplasm resources in B. striata.

Acknowledgments

This study was financially supported by the Key R&D Program of Shaanxi Province (2019SF-307) and the Key R&D Program of Shanxi Province (201603D3111003), the Fundamental Research Funds for the Central Universities (GK201906008 and GK201806006), the National Natural Science Foundation of China (31670299) and the National Key Technologies R & D Program for Modernization of Traditional Chinese Medicine (2017YFC1701300 and 2017YFC1700706).

References

Supplementary Material

The following online material is available for this article:

Figure S1

Distribution of lengths for CDSs predicted from B. striata unigenes.

Figure S2

Trees produced from Neighbor-joining phylogenetic analysis of homologs for HK and scrK in various plant species.

Figure S3

Alignment of HK between B.letilla striata and D.endrobium catenatum.

Figure S4

Alignment of scrK between B.letilla striata and D.endrobium catenatum.

Figure S5

GC-MS map of monosaccharide composition of BSP in different germplasm resources.

Table S1

Origins of rhizome samples collected from different germplasms of B.letilla striata.

Table S2

B.letilla striata genes and primers used in real-time PCR.

Table S3

Number of unigenes from B. striata assigned to KEGG reference pathways.

Table S4

Identities of candidate unigenes related to BSP biosynthesis, analyzed at nucleotide level.

Publication Dates

History