Molecular authentication and differentiation of Dendrobium species by rDNA ITS region sequence analysis

Owing to their significant medicinal and edible values, the natural Dendrobium species have underdone over-collection and habitat destruction, and cultivated species emerged for candidates. However, these Dendrobium plants are similar in shape to be easily confused, leading to extreme difficulties for identification based on their morphological and chemical features. In this study, the rDNA ITS region sequence analysis was developed for rapid and accurate identification of thirteen wild and cultivated Dendrobium species belonging to two sections Formosae and Chrysotoxae. By cloning and sequencing the rDNA ITS region genes from 13 Dendrobium species, the phylogenetic relationships among them were analyzed. Results showed that the variation of the ITS region, together with the lengths and Guanine and Cytosine contents of ITS, 5.8s rDNA, ITS1 and ITS2 sequences occurred in the tested Dendrobium species, and which from section Chrysotoxae was higher than that from section Formsae. Phylogenetic analysis based on neighbor-joining and maximum p-arsimony trees indicated that the Dendrobium species of sections Formosae and Chrysotoxae could be well divided into two groups. A majority of Dendrobium species exhibited distinctive ITS2 secondary structures, while for those with close genetic relationships were similar. Therefore, the ITS2 region sequence analysis is simple, quick, and highly reliable that can be used as an effective tool for molecular identification and classification, as well as the reconstruction of the phylogeny of wild and cultivated Dendrobium species belonging to different sections.


Introduction
The genus Dendrobium is one of the largest genera in the family Orchidaceae including approximately 1600 species (Chiang et al. 2012;Wu et al. 2009) that is mainly distributed in tropical and subtropical Asia, and northern and eastern Australia (World Checklist of Selected Plant Families 2015; Zhu et al. 2009). Dendrobiums species have been thought as one of the most valuable medicinal genus in folk medicine and industrial crop with excellent medicinal merits, such as clearing unhealthy heat, nourishing Yin, benefiting the stomach, enhancing the body's immunity, resisting cancer, and prolonging life (Ge et al. 2018;Zha et al. 2017;Zhao et al. 2017). And they have long been extensively used as the popular tonic and traditional Chinese medicines (TCMs) in many Asian countries (Chinese Pharmacopoeia Committee 2015), as well as a high-quality health food in South and Southeast Asia for hundreds of years (Kanlayavattanakul et al. 2018;Zhu et al. 2009).
However, the slow growth rate and excessive harvesting of Dendrobium species has left the wild and natural sources critically endangered and looking for the proper substitutes is in urgency. In this case, three cultivated species including D. nobile, D. chrysotoxum, and D. fimbriatum are officially allowed for use (Chinese Pharmacopoeia Committee 2015). Whereas, motivated by the medicinal values and economic benefits, a few of other species, such as D. officinale, D. longicorna, D. aphyllum, D. bellatulum, etc., are also widely used in various herbal Open Access  (Wu et al. 2009;Zeng and Hu 2004). In China, about 80 species of this genus and two varieties can be found on the market. And in the traditional product region, cultivated Dendrobiums have already become the major resource. Owing to different environmental and developmental factors during the plant and growth processes, the constituent compositions and their contents in these Dendrobium species vary significantly among the raw herbs, even if the same species from different locations (Liang et al. 2004). But, the similar appearance and tissue structure among these species have led to notorious difficulties to identify them. The drug markets of Dendrobium species from one place are always complex with those from other places (Niu et al. 2018;Xu et al. 2014). Therefore, it is critical and important to precisely differentiate the wild and cultivated Dendrobiums plants at specie level for guaranteeing their quality and safety in practical use, as well as genetic resource conservation.
Although some methods, such as gas chromatographymass spectrometry fingerprint (Chen et al. 2016), fourier transform infrared spectroscopy , capillary electrophoresis (Zha et al. 2009) have been applied for the authentication of Dendrobium species, they are complicated needing professional operators, cost-and time-consuming, and hard to distinguish the commercial samples that share similar textures, chemical and microscopic characteristics after processing and preparation as decoction pieces. In recent decades, molecular techniques, such as DNA barcoding (Wilson et al. 2019) and real-time PCR (Sobrino-Gregorio et al. 2019) have been used to differentiate many closely related species with significant advantages in the fields of taxonomy, phylogeny, evolution and breeding (Teixeira da Silva et al. 2016), especially for the morphological analogues or complexes including Dendrobium species. Currently, nuclear sequences [including internal transcribed spacer (ITS), intergenic spacer (IGS)] and plastid genes (matK, rbcL, psbA-trnH, trnL intron, etc.) based multiple specific genomics and DNA barcoding method is mainly applied for the discernment of Dendrobium species (Xiang et al. 2013;Xu et al. 2014). Among them, the ribosomal DNA (rDNA) ITS regions exhibited relatively high mutation and evolution rates, and high authentication efficiency regarding the length and sequence, which have been the most frequently utilized barcodes for rapid identification of Dendrobium at the genus and species levels (Ye et al. 2014;Feng et al. 2015). However, from then on, the application of rDNA ITS region sequences for the differentiation of wild and cultivated Dendrobium species has not been reported.
So, based on our previous findings and the reported ITS sequence information of Dendrobium specie plants, this study attempted to first amplify ITS sequences from the plants by using the optimized extraction methods for DNA to obtain the rDNA ITS region sequences through gene cloning and sequencing, which were finally to be used for distinguishing and differentiating 13 wild and cultivated Dendrobium species of sections Formosae and Chrysotoxae in China. Our results have shown that rDNA ITS region sequences could be used as valuable markers for rapid and reliable authentication study of Dendrobium species.

Plant materials
A total of 13 species of wild and cultivated Dendrobium plants were collected from the main distribution areas-Yunan province of China with Nos. 1-7 of section Formosae and Nos. 8-13 of section Chrysotoxae. The detailed information of these plants was listed in Table 1 with corresponding pictures in Fig. 1. The voucher samples were deposited in National Traditional Chinese medicine Germplasm Database of Chengdu University of Traditional Chinese medicine, Chengdu, China.

DNA extraction
Approximately 100 mg flesh leaves of each sampled specimen were randomly collected for genomic DNA extraction using a CW0531A NuClean Plant Gen DNA Kit (Hefei, China) as the instructions of the manufacturer. DNA extraction from the 13 Dendrobium plants was performed using the two-step cetyltrimethylammonium bromide (CTAB) method. The DNA samples in the quality requirements (A260/280 ratio = 1.8-2.0, A260/230 ratio > 1.7, and DNA concentration > 100 ng/mL) were selected for sequencing. The obtained genomic DNA was stored at − 20 °C for further analysis.

DNA amplification and sequencing
The sequences of the following pair of universal primers: P1, 5′-CGTAA CAA GGT TTC CGT AGG TGA AC-3′ and P2, 5′-TTA TTG ATA TGC TTA AAC TC AGC GGG -3′, that were synthetized by Tsing Ke Biotechnology Co., Ltd. (Chengdu, China), were used to amplify the rDNA ITS of Dendrobium plants. The following polymerase chain reaction (PCR) steps were conducted: The mixture containing 25 μL of 2 × Es Taq Master Mix polymerase, 2 μL of each primer, 1 μL of genomic DNA, and 20 μL of distilled water was denatured at 94 °C for 3 min, then it underwent 35 cycles of 40 s at 94 °C, 40 s at 56.4 °C, 1.5 min at 72 °C, and then a final extension for 5 min at 72 °C. The PCR products in TE (1.0%, w/v) were detected by using the agarose gel electrophoresis method for the concentration and quality of DNA, and were finally photographed under UV light exposure. The remaining PCR products were stored at − 20 °C until use. The amplified product was directly sequenced according to the dideoxy nucleotide chain termination method (Tsing Ke Biotechnology Co. Ltd., Chengdu, China).

Data analysis
The DNA sequences were compared and aligned using MAGA 5.0. The genetic distance matrix was calculated by the two-parameter method. Phylogenetic trees were constructed by Neighbor-joining (NJ) and Maximum p-arsimony (MP) method with interior branch tests of 1000 replicates based on the alignments of ITS1, 5.8S rDNA, ITS2. The secondary structure of rDNA was predicted by using the bioinformatics software based on ITS2 sequence. (http://its2.bioap ps.bioze ntrum .uniwuerz burg.de/).

PCR amplification and sequence characteristics
The sequencing characteristics were aligned with BLAST in National Center for Biotechnology Information (NCBI). All Dendrobium species can match with corresponding species by identity value of 99% or 100%. The ITS (ITS1-5.8S-ITS2) region from 13 Dendrobium species were aligned in Table 2.
As for the 7 Dendrobium plants of section Formsae: it could be found that the lengths of ITS sequence among the 13 Dendrobium species varied from 630 to 646 bp for the ITS region, from 228 to 240 bp for the ITS1 region, and from 243 to 247 bp for the ITS2 region, while was 163 bp for the 5.8s rDNA gene. The polymorphic numbers of variation sites were 119 bp (18.5%) in the ITS region, 65 bp (27.1%) in the ITS1 region, 8 bp (4.9%) in the 5.8s rDNA, and 46 bp (18.6%) in the ITS2 region. In addition, the contents of Guanine and Cytosine (GC) were from 52.7 to 54.5% in the ITS region, from 55.8 to 58.9% in the 5.8s rDNA, from 51.3 to 53.7% in the ITS1 region, and 49.8% to 52.7% in the ITS2 region. The average GC contents of ITS, ITS1, 5.8s rDNA and ITS2 were 53.6%, 52.7%, 57.8% and 51.8%, respectively.
While, for the 6 Dendrobium plants of section Chrysotoxae: It was shown that the lengths of ITS sequence were from 641 to 651 bp for the ITS region, from 163 to 164 bp for the 5.8s rDNA gene, from 230 to 243 bp for the ITS1, and from 245 to 248 bp for the ITS2 region. The numbers of the variation sites were 196 bp (30.1%) in the ITS region, 109 bp (44.9%) in the ITS1 region, 11 bp (6.7%) in the 5.8s rDNA, and 76 bp (30.6%) in the ITS2 region. In addition, the GC contents varied from 50.5 to 54.8% in the ITS region, 47.4% to 51.9% in the ITS1 region, 57.1% to 59.5% in the 5.8 s rDNA region, and 47.0% to 55.1% in the ITS2 region. And the average GC contents of ITS, ITS1, 5.8 s rDNA and ITS2 regions were 52.5%, 49.4%, 58.2% and 51.7%, respectively.
Then, it could be concluded that the polymorphic sites of 5.8s rDNA were less than that of ITS1 and ITS2, and the variation rates were very low. Previous studies also exhibited that 5.8s rDNA are highly conserved, and the ITS1 and ITS2 regions were more variable. Thus, the diversity of the ITS region could be used widely as a molecular marker for species authentication and polygenetic analysis. The lengths of the ITS2 region were a bit longer than that of the ITS1 region. The mutation rates of the ITS2 region were less than ITS1 region, which indicated that the diversity in the ITS1 region is much higher than that in the ITS2 region, and both the two regions would provide more molecular evidence for the accurate identification of Dendrobium species. In addition, the results showed that the variation from distinct groups of Dendrobium was different, and the variation rate of Dendrobium in section Chrysotoxae was higher than that from section Formsae.
In summary, the lengths and GC contents of ITS, 5.8s rDNA, ITS1 and ITS2 sequences from the collected Dendrobium species were relatively variable, which was satisfactory for next analysis.

Genetic distance and phylogenetic analysis
With Pholidota yunnanensis obtained from NCBI as an out-group, the genetic distance matrix and phylogenetic tree were established based on ITS sequences (ITS1-5.8S-ITS2) of the 13 Dendrobium species and Pholidota yunnanensis, which were shown in Table 3. It could be observed that the genetic distances among the 14 species were within the range of 0.003-0.283 by the Kimura 2-parameter (K2P) model. Among the seven species of section Formsae, the range of genetic distances was from 0.003 to 0.135, and from 0.007 to 0.184 for the six Dendrobium species of section Chrysotoxae. D. williamsonii and D. cariniferum were relatively close with a genetic distance of 0.003. D. chrysotoxum and D. lindleyi were more divergent with a genetic distance of 0.184. Pholidota yunnanensis, as the out-group, exhibited the highest genetic distance of 0.227 to 0.283 and the largest differences from the 13 Dendrobium species. These findings were in agreement with the previous report (Tsai et al. 2004). Then, the phylogenetic trees were constructed by the NJ and MP methods, which were in agreement. It could  Table 3 Genetic distances of the ITS sequences among Dendrobium species (Nos. 1-13) and Pholidota yunnanensis (No. 14) 1,D. longicorna;2,D. infundibulum;3,D. trigonopus;4,D. bellatulum;5,D. williamsonii;6,D. cariniferum;7,D. sinense;8,D. jenkinsii;9,D. chrysotoxum;10,D. thyrsiflorum;11,D. densiflorum;12,D. lindleyi;13,D. sulcatum;14,Pholidota yunnanensis  be found from the NJ and MP trees in Fig. 2 that the 13 Dendrobium species could be clearly grouped in six clusters: D. longicorna, D. infundibulum, D. sinense, D. bellatulum, D. williamsonii and D. cariniferum that all belong to section Formsae were grouped together into cluster I with 99% similarity. However, D. trigonopus was separated in a single cluster (cluster III) from section Formsae species. The results were in accordance with previous researches (Xiang et al. 2013;Xu et al. 2014). These findings provided more evidences for the Dendrobium species in section Formsae as monophyletic except D. trigonopu, which showed that cluster analysis of section Formsae based on molecular characteristics was in match with the classification systems based on morphological analysis. D. jenkinsii was grouped together with D. lindleyi into cluster II (NJ tree) and III (MP tree) with 100% similarity. D. sulcatum, D. thyrsiflorum and D. densiflorum were grouped together into cluster IV with 98% (NJ tree) and 92% (MP tree) similarity, respectively. Samples of the Dendrobium species in cluster II (NJ tree) or III (MP tree) and cluster IV belong to section Chrysotoxae. In addition, D. chrysotoxum was separated from the other 12 Dendrobium species with the interior branch test of 89%. From the NJ tree, the Dendrobium species of section Chrysotoxae except D. chrysotoxum were divided into three branches. Thus, it can be concluded that the phylogenetic tree for section Chrysotoxae species was not completely matched with the results of the classification systems based on morphological characters.

Secondary structure of ITS2 sequences
It is known that secondary structure is a guide to align the nucleotide positions of the ITS2 sequences with advantages of convenience and a wider range of taxonomy comparison (Coleman 2003). The secondary structure can be vividly regarded as a four-fingered hand and consists of Helix I, II, III and IV. The difference of secondary structure are mainly divided into the following three types: (1) the difference of angle among Helix I, II, (2) the difference of length of Helix I, II, III and IV; and (3) the difference of the number and shape of Stem and Loop in Helix I, II, III and IV, which illustrate the morphological characteristics of ITS2 between species. Currently, all the secondary structures of genus Dendrobium have not been established. Hence, a secondary structure may have matched several Dendrobium species. Here, the secondary structures of the 13 Dendrobium species have obtained and were shown in Fig. 3. Then, based on the measured angle among Helix I, II, III and IV and the length of Helix I, II, III and IV was, as well as the counted number of Loop in Helix I, II, III and IV, the dendrogram of hierarchical cluster (Fig. 4) was constructed by means of between-groups linkage, and the distance was calculated by the Block method.
It could be seen that D. longicorna, D. infundibulum, D. bellatulum and D. sinense were in one cluster and shared the highest similarity based on the ITS2 secondary structure mode. This finding was supported by the previous result of a close genetic relationship among the four species based on the genetic distance analysis. The four Dendrobium species are clustered into a class with a relatively high bootstrap supporting rate in phylogenetic trees. D. williamsonii and D. cariniferum hold the same secondary structure, which also occurred between D. thyrsiflorum and D. densiflorum. The genetic distances between them were also extremely low and they were Fig. 3 Diagram of secondary structure of ITS2 transcript of Dendrobium species gathered together with a 100% bootstrap supporting rate in the phylogenetic trees. D. trigonopus and D. jenkinsii were clustered into the same group with far relationship in the phylogenetic trees. Thus, it could be concluded that the dendrogram based on secondary structure was partially matched with the results of phylogenetic tree. These studies indicated that the secondary structure of the ITS2 regions could enhanced the clustering patterns of the taxonomic levels, and sequence + structure based phylogenetic analysis was more reliable than single sequence analysis for the Dendrobium species regarding their molecular authentication and differentiation.

Discussion
A rapid and accurate method for molecular identification and differentiation of Dendrobium species is essential to ensure their quality and safe use, and to preserve Dendrobium germplasm resources. From then on, the ITS region sequences are the most popular for evaluating molecular evolution and inter specific level divergences in plants. The ITS regions of 13 Dendrobium species exist their unique ITS sequences, therefore, ITS can be used as a characteristic fragment to distinguish Dendrobium species. To our knowledge, this is the first report that the ITS region sequences have been applied for molecular identification of wild and cultivated Dendrobium species of sections Formosae and Chrysotoxae.
This study showed that the rDNA ITS sequences in the Dendrobium species are powerful for phylogenetic analysis of rapid molecular identification and differentiation to establish the genetic relationships among the wild and cultivated plant belonging to sections Formosae and Chrysotoxae. Both the genetic distance and phylogenetic analysis demonstrated the rDNA gene in Dendrobium species of section Formosae was more similar than that in section Chrysotoxae. Phylogenesis classification of Dendrobium species in section Formosae is more close to the finding of morphological observation. The secondary structure of the ITS2 regions could enhanced the clustering patterns of the taxonomic levels, and sequence + structure based phylogenetic analysis was more reliable for the molecular authentication and differentiation of the Dendrobium species.
In summary, this study demonstrated that the ITS region sequence analysis is simple, quick, and highly reliable that can be used as an effective tool for identification