Atriplex Canescens: A New Host For Cistanche Deserticola

Contact: tina.xiang@wecistanche.com

Fangming Wang a, Bingyu Zhuo b, Shuai Wang c, Jin Lou c, Yuan Zhang b, Qingliang Chen a, Ziyi Shi a, Yuelin Song b, Pengfei Tu a,* 

a State Key Laboratory of Natural and Biomimetic Drugs and Department of Natural Medicines, School of Pharmaceutical Sciences, Peking University Health Science Center, Beijing, 100191, China 

b School of Chinese Materia Medica, Beijing University of Chinese Medicine, Beijing, 102488, China 

c Quheng Foundation, Asia Sci-Tech Center, 4760 Jiangnan Avenue, Binjiang, Hangzhou, Zhejiang, 310053, China

A B S T R A C T

Cistanche deserticola has been historically used in traditional Chinese medicine for supplementing kidney (yang) function, benefiting blood and essence, and moistening intestines in order to pass stool. Its host. Haloxylon ammodendron is an important pioneer plant used for windbreaks and sand dune fixation, which are strategies used the control desertification. For a long time, it has been considered that C.deserticola can only parasitize H. ammodendron. In this study, morphological identification, gene barcoding identification, and inoculation experiments were carried out, we finally found that C. deserticola can also parasitize Atriplex canescens. A.canescens is a species of Chenopodiaceae with a wide range of adaptability. Compared with H. ammodendron, it has more biomass and a wider range of ecological adaptability, making it more suitable for the industrial production of C. deserticola. In addition, we also found that the concentration of active components was higher in C. deserticola parasitized on A. canescens than in those parasitized on H. ammodendron; this finding further suggests that the application of C. deserticola on a larger scale warrants further exploration.

Keywords: Cistanche deserticola Parasitism DNA barcode-based identification Traditional Chinese medicine Cistanche salsa

1. Introduction

The use of Cistanche in traditional Chinese medicine was first recorded in the Shennong Herbal Scripture, for its effects of toning kidney yang, boosting the essence of blood, and moistening the intestines to facilitate stool passage. It has also been recorded in works of ancient herbal medicine as 'desert ginseng'. The dry fleshy stems and scaly leaves of Cistanche deserticola Y.C. Ma and Cistanche tubulosa (Schenk) Wight have been the first joint described in 2005 in the Chinese Pharmacopoeia. Cistanche is mainly grown in Xingjiang, Inner Mongolia, and Gansu of China, and globally, it is found in semi-arid and arid areas throughout the European Iberian Peninsula, North Africa, Arabia, Iran, Afghanistan, Pakistan, North India, Mongolia et al. [1]. It is resistant to harsh environmental conditions, such as extremely arid climates, severe variations of temperatures, and depauperate soils [2]. According to the Taxonomical Index of Chinese Higher Plants, there are six Cistanche species in China. However, a further study confirmed the existence of only four species and one variant of Cistanche, namely, C. deserticola Y.C. Ma, C. tubulosa (Schenk)R. Wight, C. salsa(C.A.Mey.)G.Beck, C.sinensis G.Beck and C. salsa var. albiflora P.F. Tu et al,[3].

C. deserticola is regarded as the only traditional source of Cistanche and has a long history of use in medicine, since the Eastern Han Dynasty (25 AD-220 AD)[4].In the Compendium of Materia Medica (written by Li Shizhen. Ming Dynasty), it was documented to smoothly tonify yang (in contrast to other herbs that have a more vigorous action). An array of efficacious chemical constituents, including phenylethanoid glycosides, iridoids, lignans, alditols, oligosaccharides, polysaccharides, and alkaloids, have been isolated from C. deserticola by modern phytochemical methods [5]. Pharmacological studies have shown that phenethyl glycoside is the main active component, and it has been reported to improve sexual function, exert neuroprotective effects, improve learning and memory, and protect the liver. It also exerts therapeutic effects against dementia, Alzheimer's disease, Parkinson's disease, fatigue, and tumors along exhibiting anti-inflammatory and immunomodulatory properties [6, 7].

C. deserticola is an obligatory parasitic plant that lives exclusively on the roots of Haloxylon ammodendron [8]. A study reported that C.deserticola is not even found on Haloxylon persicum [9]. In recent years, an increasing amount of attention has been paid to C, deserticola. since it not only is a source of components of medicinal value but also contributes greatly to the control of desertification [10].H. ammodendron is the only host that has been used in studies involving C. deserticola. In April 2017, Wang Shuai, an employee of the Zhejiang Ouheng Public Welfare Fund, inoculated seeds of C. deserticola on Atriplex canescens in the Desert Botanical Garden of Mingin, Gansu Province, and C. deserticola was found to bloom in May 2018 and continued to flower till May 2019. However, the seeds were purchased from the market, and it is doubtful whether they were truly seeds of C. deserticola. In addition, this phenomenon breaks traditional knowledge and needs to be studied further.

A. canescens is a C4 perennial shrub native to the deserts of Southwestern America and rapidly adapts to conditions of salinity, heavy metals, drought, and high temperatures [11]. Since it is highly palatable and rich in nutrients, it is used as fodder for most livestock and large animals [12]. Moreover, it is especially useful for erosion control and reclamation of marginal lands on account of its excellent adaptability and extensive root system. It was first introduced in China from the United States of America in 1989 and has been widely used for soil and water conservation, sand-fixing, and saline land restoration [13]. Although the study reporting the growth of C.deserticola on A. canescens overturns the exclusive parasitic understanding of C. deserticola, this might turn out to be a revolutionary finding, since A. canescens is more suitable for the growth of C. deserticola because it has more biomass and a wider range of ecological adaptability compared with H. ammodendron.

In order to ensure the accuracy of the accidental discovery, plant identification and artificial inoculation experiments were carried out. Traditional plant identification includes organoleptic evaluation(such as touch, smell, sight, and taste), analysis of morphological characteristics (such as microscopic and macroscopic), and chemical profiling (such as high-performance liquid chromatography, thin-layer chromatography, and gas chromatography)[14]. It is relatively simple to exclude C. tubulosa and C. Sinensis due to the difference in size, color, and arrangement of vascular bundles in the stem. The real challenge is to distinguish between C. deserticola and C. salsa. According to the Flora of China, the length of the bract of C. salsa is about 1/3 of the corolla, while it is equal in C. deserticola. The transection of the fleshy stems is similar between C. deserticola and C. salsa and is composed of the epidermis, cortex, vascular bundles, and pith. The main difference is in the vascular bundle sheath since it is caudate for C. deserticola and triangular or semi-circular for C. salsa.

In recent years, DNA barcoding technology has frequently been used for species identification. It is a process that uses a short DNA sequence from the standard genome, which is generally conserved and is not affected by external factors, such as age and type of plant tissue. The popular candidate sequences for plant DNA barcodes are rbcL, matK, psbA-trnH, ITS, and ITS2 [15]. Several studies have indicated that ITS/ITS2 is the most effective identification tool for plants. It has also been suggested that the ITS2 region should be incorporated into the core barcodes because of its higher discriminatory power than that of plastid barcodes. It has been accepted that ITS2 could be used as a novel universal barcode for the identification of a broad range of plant taxa [16, 17,18, 19,20,21]. Although many studies have tried to identify a uni-versal plant barcode, none of the available loci work across all species, so a multi-locus method is necessary to discriminate between plant species [22,23,24,25,26,27,28].In this study, ITS2, rbcL,psbA-trnL were used as barcodes.

In addition to morphological and molecular identification techniques, direct evidence comes from inoculation experiments. Inoculation experiments need to be carried out in order to demonstrate that C. deserticola can parasitize A. canescens. In addition to identification, quality control becomes a primary consideration. Further investigations are needed in order to ascertain the difference between the quality of C.deserticolaparasitized on H. ammodendron root and that parasitized on A. canescent.

2. Materials and methods

2.1. Plant materials

Cistanche grows on soft sandy soils with mild salinization, generally parasiting on 30-100 cm deep side roots of the host. The climate in the suitable growing area is arid, less rainy, has large evaporation, long sunshine hours, and large temperature difference between day and night. Minqin county and Baiving city are the collection locations for these samples, They are geographically close, and have a temperate continental arid climate, with an average annual precipitation of 113.2mm and an average annual relative humidity of 44%. Specific and detailed sample collection information is shown in Table 1. All samples were frozen and preserved at -20°C in the State Key Laboratory of Natural and Bio-mimetic Drugs Lab, Beijing, China.

2.2. Tissue staining and observation

Fresh samples were obtained and stored in a solution consisting of 70% ethanol, glacial acetic acid, and formaldehyde in a ratio of 90:5:5 and were dehydrated using an ethanol gradient (75%, 95%,100%, 100%)for 1 h. The dehydrated sections were subjected to a xylene gradient (25%,50%,75%,100%,100%)for 1 h in order to obtain transparent sections. The transparent sections were subjected to paraffin infiltration, wherein a volume of paraffin equal to the volume of xylene was added to the xylene containing the sample, half of the resultant solution was then aspired out, and an equal volume of paraffin was added again. This process was repeated 10 times, and finally, all of the solutions were aspired out and replaced with an equal volume of paraffin; this final step was repeated twice, and the resulting solution after each step was incubated for 1 h at 75°C. After paraffin infiltration, the sections were subjected to embedding, wherein samples were placed in an iron tank containing liquid paraffin and additional liquid paraffin was rapidly added to fill the whole tank and left to solidify. The resultant wax block was trimmed and sectioned. Embedded sections were placed in warm water, fished out, placed on a slide, and incubated at 45°C for 30 min. Sections on the slide were dewaxed by serially soaking in 100% xylene, 100% xylene, 50% xylene,50% xylene, 100% ethanol, 100% ethanol, 95% ethanol, and 75% ethanol and then soaked in safranine O for 40 min. This was followed by another round of serial, rapid soaking in 75%ethanol and 95% ethanol, and then the slides are submerged in fast green for 1 min. Finally, the sections were subjected to one last serial soak in 95% ethanol, 95% ethanol,100% ethanol, 100% ethanol, 50% xylene, 50%xylene, and 100% xylene. After the sections were stained, a drop of resin glue was placed on the slide, and a cover glass was placed over it. The slides were left undisturbed for a week, and the tissue sections were observed using an Olympus optical microscope and imaged.

2.3. DNA extraction and PCR amplification

Total genomic DNA was extracted from flower specimens using a plant genomic DNA extraction kit(Solarbio Science & Technology Co., Ltd., Beijing, China)according to the manufacturer's protocols. The primers for gene amplification and sequencing and the reaction conditions are shown in Table 2. Each gene amplification was repeated thrice for each specimen.

2.4. Sequencing analysis

In order to obtain accurate sequences, the final PCR products, after purification using the Transgene Quick Gel Extraction Kits, were cloned separately into pEASY-Blunt Cloning Vectors, according to the manufacturer's instructions. After cloning, they were transformed into chemically competent Trans5a cells. Three colonies of each sample were randomly selected and sequenced using M13 primers. These colonies were sequenced bi-directionally by Sanger sequencing, using the BigDye Terminator V3.1 Cycle Sequencing Kits on the ABI Prism 3700 DNA Analyzers. Obtained sequences were aligned using Clustal X(v1.8.7)[29]and synchronized manually in BioEdit(v.1.3.0)[30.Utilizing the aligned sequence data, we reconstructed the phylogeny using the MEGA 7 software using the neighbor-joining (NJ) method. Kimura 2-parameter (K2P) model was used and the bootstrap was 1000 repetitions [31].

2.5. Inoculation of C. deserticola

Three grams of C. deserticola seeds were added to the pots(diameter ×height× bottom diameter=20 cm×20 cm×12 cm) containing sandy soil and stirred to ensure an even spread. The control group consisted of3 g of C. salsa seeds added to similar pots containing sandy soil. Finally, A. canescens was planted in each pot, and the pots were placed outdoors. When the soil moisture content is less than 13%(g/g), the pots were watered. The experiment was conducted at the Zhongguancun Life Science Park, Beijing, China (latitude 39°56'N, longitude 116°20'E;20m above sea level) from May to July. Day temperature varied between 16 and 35°C, night temperature between 12 and 16°C. The relative humidity of the air is greater than 50%. Sunlight is abundant. Approximately 80 days later, the soil was removed from the pots and the inoculation rate was determined.

2.6. Determination of concentration of medicinal components

The determination of the concentration of medicinal components include two parts, one is the procedure for liquid chromatography, and the other is the preparation of reference substance and test substance, more details are as follows:

i). Determination of echinacoside and verbascoside

Echinacoside and verbascoside were weighed and added in 50%methanol to yield a 0.2 mg/ml solution, which was used as the reference solution. The first is to grind dry C. deserticola into powder, then the powder was mixed into 50 ml of 50% methanol in a 100 ml brown volumetric flask, and the test liquid was obtained after subjecting the mixture to shaking, soaking, sonication, standing, and filtration. The chromatographic column was Agilent ZORBAX SB-C18 column (4.6 mm ×150 mm,5um),with Methanol(A)-0.1% Formic acid solution (B) as mobile phase, gradient elution (0-17 min,26.5% A;17-20 min,26.5%→29.5% A; 20-27 min,29.5% A), flow rate was 1.0 mL/min,column temperature was 35·C, detection wavelength was 330 nm, injection volume was 10μul.

ii). Determination of betaine, mannitol, fructose, glucose, and sucrose

Betaine, mannitol, fructose, glucose, and sucrose were weighed precisely and added to water to make a solution of 0.25 mg/ml, which was used as a reference solution. Five milliliters of the aforementioned Cistanche test solution were mixed with 50% methanol in a 25 ml volumetric flask, shaken well, and filtered with a 0.2um microporous membrane. The chromatographic column was SHODEXASHAIPAK NH2P-50 4E polymerized gel column （250 mm × 4.6 mm， 5μm）， the mobile phase was parasitized on A. canescens, we found that it had a caudate shaped vascular bundle sheath, resembling that of C. deserticola(Figure 2).

acetonitrile-water(77:23), the flow rate was 0.7 mL/min, the column temperature was 25°C, Using evaporative light scattering detector (ELSD), the drift tube temperature was 100°C, the carrier gas flow rate was 3 L/min, the injection volume of reference substance and sample were5μl. 

3. Results

3.1. Morphological identification of flowers

To confirm the species of Cistanche that parasitizes A. canescens, morphological analysis of flower specimens was carried out (Figure 1and Figure S1). The overall morphology of flowers of the parasitic plant was similar to that of C. deserticola. Further, the corolla was thicker than that of C.salsa on different hosts.

According to Flora of China, C. deserticola and C.salsa have obvious differences in the bract of flowers. In C. deserticola, bracts sub-equal the corolla, while the bract of C. salsa is 1/3 the length of the corolla. Based on our statistical analysis, the bracts of Cistanche parasitized on A. canescens and that of C. deserticola sub-equaled the corolla (Figure S2). The Cistanche on A. canescens exhibited morphological features of C. deserticola, suggesting that C. deserticola may be the parasite on A. canescens.

3.2. Microscopic identification of stained tissue specimens

The transection of the fleshy stem of C. deserticola is very similar to that of C. salsa, and both of them are composed of the epidermis, cortex, vascular bundle, and pith. The vascular bundles of both plants are arranged in wavy or deep wavy rings, and the piths are obviously visible. The main difference lies in the lateral shape of the vascular bundle sheath; it is caudate in C. deserticola and triangular or semi-circular in C. salsa. Upon performing a microstructure analysis on the Cistanche

3.3. Molecular identification

In addition to morphological identification, we also carried out a molecular identification and selected three gene fragments, namely, ITS2, rbcL, and psbA-trnL. An evolutionary tree was constructed using the sequence information of each fragment (Figure 3), and all three phylogenetic trees displayed that the Cistanche parasitized on A.canescens had a close phylogenetic relationship with C. deserticola. These results indicate that C. deserticola may be the parasite on A. canescens.

Detailed gene divergences among the different Cistanche species were observed upon multiple sequence alignment (Figure 4). We found three single nucleotide polymorphisms(SNPs) in the ITS2 gene body between C. deserticola and C.salsa, at bases 139,295, and 472. In the rbcL gene body, there were four gene divergences between C. deserticola and C. salsa, containing two SNPs and two insertions and deletion (indel)mutations. Compared with ITS2 and rbL, the differences in the psbA-trnL gene body between C. deserticola and C. salsa were more obvious, with seven sequence divergences, wherein four were SNPs and three were InDel mutations. In particular, a series of thymine repeats, starting at base 414 of the aligned sequence, could be used to develop simple sequence repeat (SSR) markers for distinguishing C. deserticola and C. salsa.

3.4. Inoculation of C. deserticola

To test whether C. deserticola or C. salsa could parasitize A.canescens, an inoculation experiment was carried out, and we found evidence of parasitism in all C.deserticola-inoculated pots with an inoculation rate of almost 100%(Figure 5). No parasitism was observed in the control groups. This result directly proves that C. deserticola easily parasitized A. canescens, whereas C. salsa could not.

3.5. Determination of concentration of significant medicinal components

We estimated the concentration of important medicinal components in C.deserticola parasitized on A. canescens. The specific chromatogram is shown in the supplementary material. To obtain accurate results, four independent experiments were set up. Based on our measurements (Table 3), we found that the concentration of verbascoside and echinacoside were 20 times higher than that reported in the Chinese Pharmacopeia (according to the Chinese Pharmacopoeia, the percentage of the sum of concentrations of echinacoside and verbascoside in C. deserticola should be less than 0.30%). The concentrations were also significantly higher than that in C. deserticola parasitized on H. ammodendron (generally 0.2-1.5%)[32]. The concentration of mannitol, betaine, fructose and other carbohydrate components was also very high, and the overall quality was better than that in C. deserticola parasitized on H. ammodendron. Thus, these results indicate that A. canescens can be used to grow C. deserticola at an industrial level and protect endangered wild resources.

4. Discussion 

Previously, C. deserticola has been considered to exclusively parasitize H. ammodendron. However, in this study, using morphological and molecular identification techniques, we demonstrated that C. deserticola can also parasitize A. canescens. Although H. ammodendron, A. canescens, and H. persicum all belong to the family Chenopodiaceae, it is interesting and peculiar that C. deserticola has species selectivity, possibly governed by the signaling molecules secreted by the host. A. canescens, originating from the United States of America, exhibits strong resistance to environmental perturbations and has relatively large biomass. A. canescens is a viable host for C. deserticola for a variety of reasons. First, it can survive in a wide range of environmental conditions. Second, the biomass and growth rate of C. deserticola may be larger and more rapid, respectively, on A. canescens than on H. ammodendron. Third, due to the wide range of adaptability of A. canescens, the planting area can be further expanded. Thus, A. canescens has distinctive advantages over H. ammodendron as the host and will aid the industrial production of C. deserticola. C. deserticola and C. salsa are difficult to distinguish, and morphological identification in the past has produced confounding results. With advances in the field of molecular biology, molecule-based identification techniques have been widely used in Chinese herbal medicine. Since most Chinese herbal medicines offer little genomic information, DNA barcoding technology has emerged as a breakthrough identification technique. In this study, morphological and DNA barcoding technology were comprehensively applied for identifying the unknown Cistanche species; this has not been attempted before, and our results demonstrate that this approach is feasible. Since C. deserticola parasitizes A. canescens, it is important to determine differences in the quality of C. deserticola on the roots of A. canescens and that on the roots of H. ammodendron. According to our results, the concentration of active components was higher in C. deserticola parasitized on A. canescens than in that parasitized on H. ammodendron. Thus, our results lay a solid theoretical foundation for the large-scale production of C. deserticola parasitized on A. canescens. 

5. Conclusions 

For a long time, C. deserticola has been considered to exclusively parasitize H. ammodendron. Previously, it was found that C. deserticola seeds purchased from the market can parasitize A. canescens, another Chenopodiaceae plant. Using morphological and molecular identification methods, we confirmed that the Cistanche species parasitizing A. canescens was C. deserticola. This result was further confirmed by the inoculation experiment. We determined the concentration of significant medicinal components, and our results suggest that the concentration and quality of components were greater in C,deserticola parasitized on A.canescens than in that parasitized on H. ammodendron. The discovery of new hosts can promote the industrial production of C. deserticola, and it can also effectively protect wild resources and the ecological environment.

Declarations

Author contribution statement

Fangming Wang: Conceived and designed the experiments; Performed the experiments; Analyzed and interpreted the data; wrote the paper.

Bingyu Zhuo, Yuan Zhang, Qingliang Chen, Ziyi Shi& Yuelin Song: Performed the experiments.

Shuai Wang & Jin Lou: Contributed reagents, materials, analysis tools, or data.

Pengfei Tu: Conceived and designed the experiments.

Funding statement

Fangming Wang was supported by the National Key Research and Development Program of China (2019YFC1710903).

Dr.Pengfei Tu was supported by the National Natural Science Foundation of China (8177140819).

Data availability statements

Data will be made available on request.

Declaration of interests statement

The authors declare no conflict of interest.

References 

[1] D.Y. Tan, Q.S. Guo, C.L. Wang, Study on the status quo of Cistanche deserticola and its exploitation and utilization in China, For. Resour. Manag. 33 (2004) 29–32. [2] X.Y. Qiao, H.L. Wang, Y.H. Guo, Study on conditions of seed germination of Cistanche, Zhongguo Zhong Yao Za Zhi 32 (2007) 1848–1850. 

[3] P.F. Tu, Y.P. He, Z.C. Lou, A survey on the origin and resource protection of Cistanche, Chin. Tradit. Herb. Drugs 25 (1994) 205–208. 

[4] L.D. Karalliedde, C.T. Kappagoda, The challenge of traditional Chinese medicines for allopathic practitioners, Am. J. Physiol. Heart Circ. Physiol. 297 (2009) 1967–1969. 

[5] Y. Jiang, P.F. Tu, Analysis of chemical constituents in Cistanche species, J. Chromatogr. 1216 (2009) 1970–1979. 

[6] T. Wang, X.Y. Zhang, W.Y. Xie, Cistanche deserticola Y.C. Ma, “desert ginseng”: a review, Am. J. Chin. Med. 40 (2012) 1123–1141. 

[7] Pharmacopoeia NCOC, Pharmacopoeia of the People’s Republic of China, The Chemical Industry Press, Beijing, 2020. 

[8] G.X. Meng, X.S. Cui, Y. Wu, Y.H. Guo, Effects of Leveillula saxaouli on growth, chlorophyll, and carbohydrate of the Haloxylon ammodendron, North. Hortic. 14 (2012) 141–143. 

[9] Y.C. Chen, M. Li, M.Z. Wu, Y.X. Song, Structure and composition of roots in two species of Haloxylon Bunge, Plant Physiol. J. 49 (2013) 1273–1276. 

[10] P.F. Tu, Y. Jiang, Y.H. Guo, Y.Z. Tian, et al., Developing ecological industry of Cistanches herba for promoting ecological civilization of the western desert region, Mod. Chin. Med. 4 (2015) 297–301. 

[11] S.C. Sanderson, H.C. Stutz, High chromosome numbers in Mojavean and Sonoran desert Atriplex canescens (Chenopodiaceae), Am. J. Bot. 81 (1994) 1045–1053. 

[12] J.L. Peterson, D.N. Ueckert, R.L. Potter, J.E. Huston, Ecotypic variation in selected four-wing saltbush populations in western Texas, J. Range Manag. 40 (1987) 361–366. 

[13] D.S. Kong, Morphological characteristics and eco-physiological adaptability of Atriplex canescens a review, Chin. J. Ecol. 32 (2013) 210–216. 

[14] M.A. Bashir, M.S. Faezah, S.S.O. Mohd, W. Alina, Review: DNA barcoding and chromatography fingerprints for the authentication of botanicals in herbal medicinal products. Evid. Based Complement, Alternat. Med. 2017 (2017) 1–28. 

[15] X.W. Li, Y. Yang, et al., Plant DNA barcoding: from gene to the genome, Biol. Rev. 90 (2015) 157–166. 

[16] S.L. Chen, H. Yao, J.P. Han, et al., Validation of the ITS2 region as a novel DNA barcode for identifying medicinal plant species, PloS One 5 (2010), e8613. 

[17] K. Luo, S.L. Chen, K.L. Chen, et al., Assessment of candidate plant DNA barcodes using the Rutaceae family, Sci. China Life Sci. 53 (2010) 701–708. 

[18] T. Gao, H. Yao, J.Y. Song, et al., Identification of medicinal plants in the family Fabaceae using a potential DNA barcode ITS2, J. Ethnopharmacol. 130 (2010) 116–121.

[19] T. Gao, H. Yao, J.Y. Song, et al., Evaluating the feasibility of using candidate DNA barcodes in discriminating species of the large Asteraceae family, BMC Evol. Biol. 10 (2010) 324. 

[20] X.H. Pang, J.Y. Song, Y.J. Zhu, et al., Using DNA barcoding to identify species within Euphorbiaceae, Planta Med. 76 (2010) 1784–1786. 

[21] X.H. Pang, J.Y. Song, Y.J. Zhu, et al., Applying plant DNA barcodes for Rosaceae species identification, Cladistics 27 (2011) 165–170. 

[22] P.D. Hebert, E.H. Penton, J.M. Burns, D.H. Janzen, W. Hallwachs, Ten species in one: DNA barcoding reveals cryptic species in the neotropical skipper butterfly Astraptes fulguration, Proc. Natl. Acad. Sci. U.S.A. 101 (2004) 14812–14817. 

[23] M.W. Chase, R.S. Cowan, et al., A proposal for a standardized protocol to barcode all land plants, Taxon 56 (2007) 295–299. 

[24] W.J. Kress, D.L. Erickson, A two-locus global DNA barcode for land plants: the coding rbcL gene complements the non-coding trnH-psbA spacer region, PloS One 2 (2007) e508. 

[25] D.L. Erickson, J. Spouge, A. Resch, et al., DNA barcoding in land plants: developing standards to quantify the maximize success, Taxon 57 (2008) 1304–1316. [26] N.C. Kane, Q. Cronk, Botany without borders: barcoding in focus, Mol. Ecol. 17 (2008) 5175–5176. 

[27] R. Lahaye, M. van der Bank, D. Bogarin, et al., DNA barcoding the floras of biodiversity hotspots, Proc. Natl. Acad. Sci. U.S.A. 105 (2008) 2923–2928. 

[28] N. Kane, S. Sveinsson, H. Dempewolf, et al., Ultra-barcoding in cacao (Theobroma spp.; Malvaceae) using whole chloroplast genomes and nuclear ribosomal DNA, Am. J. Bot. 99 (2012) 320–329. 

[29] J.D. Thompson, T.J. Gibson, F. Plewniak, F. Jeanmougin, D.G. Higgins, The CLUSTAL_X windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools, Nucleic Acids Res. 25 (1997) 4876–4882. 

[30] T.A. Hall, BioEdit: a user-friendly biological sequence alignment editor and analysis program for windows 95/98/NT, Nucl. Acids Symp. Ser. 41 (1999) 95–98. [31] S. Kumar, M. Nei, J. Dudley, K. Tamura, MEGA: a biologist-centric software for evolutionary analysis of DNA and protein sequences, Brief. Bioinform. 9 (2008) 299–306. 

[32] P.F. Tu, B. Wang, T. Deyama, Z.G. Zhang, Z.C. Lou, Analysis of phenylethanoid glycosides of Herba cistanche by RP-HPLC, Acta Pharm. Sinica. 32 (1997) 294–300.