Introduction

Over 1.1 billion people smoke tobacco worldwide, which causes approximately 8 million deaths each year, as reported by the World Health Organization (WHO)1. According to the report, smoking-related deaths will increase if no effective smoking cessation policies are implemented. Currently, nicotine replacement and non-nicotine replacement products are utilized for smoking therapy. Various dosage forms of smoking cessation aids are available, such as teas, gums, inhalers, lozenges and nasal sprays. However, approximately 80% of smokers worldwide are living in low- and middle-income countries1, where cessation aid products are expensive, leading to unsuccessful attempts to quit. In addition, side effects such as dry mouth, nausea and sedation are associated with using cessation aids2.

In recent years, traditional and complementary medicines have received growing interest worldwide. In Thailand, an herbal species in the Asteraceae family, Cyanthillium cinereum (L.) H.Rob. or Vernonia cinerea (L.) Less. (Thai: Ya dok khao) has been applied as a complementary medicine for the cessation of smoking (Fig. 1). C. cinereum tea has been included in the National List of Essential Medicines, Ministry of Public Health, Thailand, since 20123. According to a previous report, C. cinereum may act as nicotine replacement therapy because the leaves and flowers contain tiny amounts of nicotine (0.154% and 0.123%, respectively)4. Furthermore, an abundance of nitrate salt is found in stem and leaf extracts of C. cinereum (21% and 19%, respectively)4. A study on the C. cinereum extract revealed that nitrate salt can cause tongue numbness and results in reduced cigarette smell and taste5. Moreover, supplementation of exercise with C. cinereum juice can reduce oxidative stress and the number of cigarettes smoked per day in volunteer smokers6.

Figure 1
figure 1

Plants and smoking cessation products. (A) Cyanthillium cinereum field, (B) raw material and smoking cessation products.

Full size image

Emilia sonchifolia (L.) DC. (Thai: Hu pla chon), also in the Asteraceae family, shares similar flower and fruit characteristics with C. cinereum. Moreover, the plant is often found in the same habitat as C. cinereum. Careless collection practices and lack of standard quality control lead to contamination by E. sonchifolia and result in substandard quality of the smoking cessation aid products; therefore, appropriate markers for quality control of herbal raw materials are necessary. In addition, products should be processed with good agricultural practices (GAPs) and good manufacturing processes (GMPs) to ensure sustainable agriculture and ecological safety. Concrete policy or guideline frameworks for herbal products, including agricultural and manufacturing practices, product labeling, product registration, marketing and trade, would be seriously regulated.

Authentication of raw materials can be performed by different methods. The standard methods are micro- and macroscopic characterization and phytochemical analysis. To complement traditional identification methods, innovative technologies such as DNA barcoding are also used. Macroscopic and microscopic methods for botanical identification are recommended as basic tools in the Thai Herbal Pharmacopoeia (THP)7. The methods are simple, rapid and inexpensive, but they must be performed by trained experts, and reliable references are required8. Chemical detection methods may produce uncertain results due to environmental factors that affect the chemical composition of herbal species and biological activities of the substances. Approaches such as thin-layer chromatography (TLC) or high-performance thin-layer chromatography (HPTLC) are versatile and specific for the identification of phytochemical constituents in herbal products. However, standard substances are required for analytical methods to yield quantitative benefits9. Although DNA testing is a high-cost technique, it yields precise and reliable results because herbal genetics do not vary with environmental factors10. DNA testing is also applicable for limited sources of samples. As mentioned above, each analysis method has advantages and limitations, and combined methods should be applied to control the quality of herbal raw materials for consumer safety and to meet international standards. Therefore, this study aimed to differentiate C. cinereum, a smoking cessation herb, from its adulterant, E. sonchifolia, using macroscopic and microscopic examination, HPTLC profiling and DNA barcoding methods.

Results

Macroscopic characteristics

C. cinereum and E. sonchifolia plants are slender-stemmed herbs in the Asteraceae family. Flowers of C. cinereum look very similar to E. sonchifolia flowers. Individual plants produce capitula with purple or white florets. However, distinct characteristics can be observed by macroscopic examination. C. cinereum flower-heads are tubulate with pink to purple florets (Fig. 2A), while long vase-shaped flower bracts with petals tinted pink to purple are observed in E. sonchifolia (Fig. 3A). The fruit of C. cinereum is a subcylindrical linear achene covered with short hairs (Fig. 2B), while that of E. sonchifolia is an achene with longitudinal ribs, short hairs and surmounted by white pappus (Fig. 3B). Serrated leaves with attenuated bases are found in C. cinereum (Fig. 2C), while auriculate leaf bases and triangulate or pinnate-lobed lower leaves and arrow-shaped upper leaves with bases encircling the stem are present in E. sonchifolia (Fig. 3C). Dried herbs of C. cinereum are greenish brown, while those of E. sonchifolia exhibit a light green color. Grass-like odor is observed in both dried herbs, but C. cinereum express a stronger odor than that of E. sonchifolia. The taste of both dried C. cinereum and E. sonchifolia is slightly bitter.

Figure 2
figure 2

Macroscopic and microscopic characteristics of C. cinereum. (A) Inflorescence, (B) infructescence, (C) leaves, (D) pappus with trichomes, (E) transverse section of stem; 1: T-shaped trichome; 2: epidermis; 3: chlorenchyma; 4: collenchyma; 5: phloem; 6: xylem; 7: pith, (F) T-shaped trichome, (G) lower epidermis of leaf; 8: anomocytic stoma; 9: epidermal cell; 10: cicatrix of T-shaped trichome; 11: glandular trichome.

Full size image
Figure 3
figure 3

Macroscopic and microscopic characteristics of E. sonchifolia. (A) Inflorescence, (B) infructescence, (C) leaves, (D) pappus with trichomes, (E) transverse section of stem; 1: epidermis; 2: chlorenchyma; 3: collenchyma; 4: phloem; 5: xylem; 6: pith, (F) lower epidermis of leaf; 7: anomocytic stoma; 8: epidermal cell.

Full size image

Microscopic features

Microscopic studies showed the anatomical features of C. cinereum and E. sonchifolia. The fruit pappi of both plants were examined under a microscope. The pappi resembled fine feathery hairs of calyx consisting of one to several rows of thin-walled elongate cells and unicellular trichomes in both plants. Pappi of C. cinereum exhibited more than 3 rows of thin-walled elongated cells, while 2 to 3 rows with unicellular trichomes were found in E. sonchifolia. The unicellular trichomes of C. cinereum pappi were longer than those of E. sonchifolia (Fig. 2D and Fig. 3D). Transverse sections of stems showed epidermis with a layer of rectangular cells in both C. cinereum and E. sonchifolia (Fig. 2E). Two types of trichomes, T-shaped and glandular trichomes, were observed in transverse sections of the stem of C. cinereum (Fig. 2E,F), while long, multicellular, uniseriate trichomes were found in E. sonchifolia. The cortex contained 2–4 layers of collenchyma cells in C. cinereum and 1–3 layers in E. sonchifolia. Pith cells of C. cinereum were composed of large ordinary parenchyma cells and rosette aggregates of calcium oxalate crystals (Fig. 2E). The stele of E. sonchifolia showed phloem and xylem, including pith cells with large ordinary parenchyma cells (Fig. 3E). Investigation of leaves showed that the lower epidermis of C. cinereum was composed of wavy-walled cells, including numerous anomocytic stomata (Fig. 2F), similar to E. sonchifolia (Fig. 3F). The base of T-shaped trichomes and glandular trichomes were also observed in C. cinereum leaves (Fig. 2G). Microscopic study revealed T-shaped trichome, glandular trichomes, pappus and pollen grain in C. cinereum tea product (Fig. 6).

HPTLC fingerprint profiles

HPTLC profiles of ethanolic extracts of C. cinereum and E. sonchifolia were obtained. The mobile phase was a mixture of hexane and acetone (9:1, v/v). In general, the HPTLC patterns of C. cinereum and E. sonchifolia were similar. However, a distinct band at Rf = 0.48 was detected in only C. cinereum extract after plates were sprayed with anisaldehyde-sulfuric acid detecting reagent (Fig. 4) and the band was subjected to GC–MS analysis. A combination of the four triterpenoid compounds β-amyrin, taraxasterol, lupeol, and betulin was identified in C. cinereum.

Figure 4
figure 4

High-performance thin-layer chromatography (HPTLC) chromatograms of representative samples of C. cinereum (tracks 1–3) and E. sonchifolia (tracks 4–6) extracts. (A) Under white light, (B) under UV at 254 nm, (C) under UV at 366 nm, and (D) under white light after postchromatographic derivatization with anisaldehyde-sulfuric acid reagent. Hexane:acetone mixture (1:9, v/v) was used as the mobile phase. The arrows in C and D show the presence of bands at Rf = 0.48 only in C. cinereum.

Full size image

DNA barcoding analysis

To distinguish C. cinereum from its adulterant E. sonchifolia, the nucleotide sequences of the four core DNA barcode regions, rbcL, matK, ITS and psbA-trnH intergenic spacer, were analyzed. PCR amplicons of each barcode region were successfully amplified and sequenced. All plant specimens in the same species but from different collection places showed identical DNA sequences. There is no intraspecific divergence for all four barcode regions of both species. The length of the amplified product of each barcode region of the two species varied (Table 1). GenBank accession numbers were obtained (Table 2). The sequences of all DNA barcode areas exhibited indels (Fig. 5 and Supplementary Data S1–S4). The percentage of sequence variation among the four barcode regions was ranked as follows: ITS > psbA-trnH intergenic spacer > matK > rbcL. The percent GC content among the same barcode areas of C. cinereum and E. sonchifolia was similar. The highest percent GC content was observed in the ITS region, followed by rbcL, matK and psbA-trnH intergenic spacer. The percentage of variation was calculated, and the highest percent variation was found in the ITS region (29.24%), followed by the psbA-trnH intergenic spacer (19.90%), matK (6.55%) and rbcL (5.13%). The highest interspecific divergence was found in the ITS region (32.38%), while the psbA-trnH intergenic spacer, matK and rbcL exhibited values of 9.51%, 5.86% and 4.21%, respectively.

Table 1 Sequence analysis comparing core DNA barcode regions of C. cinereum and E. sonchifolia.
Full size table
Table 2 Authentic plant species used in this study and the accession numbers of their DNA barcode loci deposited in GenBank.
Full size table
Figure 5
figure 5

Illustration comparing core DNA barcodes of C. cinereum and E. sonchifolia. Polymorphic sites and indels in each DNA barcode area are shown; (A) rbcL, (B) matK, (C) ITS and (D) psbA-trnH intergenic spacer. The nucleotide positions of the DNA barcode regions are indicated.

Full size image

Discussion

Several approaches have been applied to help smokers quit smoking. Smoking cessation aids can be medications or non-medicinal treatments. Nicotine replacement therapy (NRT) products, varenicline and bupropion are FDA-approved medications to help users manage withdrawal3. However, the cost of medications remains expensive, especially for smokers who live in low- or middle-income countries. Another disadvantage of using medications is their side effects, such as nausea, dry mouth, increased appetite, insomnia and sedation11. Therefore, non-medicinal approaches, such as the use of local herbs for smoking cessation, have gained increasing attention because they provide a low cost of treatment with a high percentage of successful quitting of smoking12. It has been suggested that regular drinking of herbal teas containing high antioxidant levels may protect smokers from oxidative damage caused by smoking13. A report revealed that Syzygium aromaticum (L.) Merr. & L.M.Perry and Astragalus mongholicus Bunge teas, which have high antioxidant activity, can reduce smoking withdrawal symptoms13. High antioxidant activity is observed in many parts of C. cinereum, as reported by Ketsuwan et al.4. Moreover, a study of C. cinereum and lime in Thailand demonstrated that the herb can reduce smoking urges in two weeks and decrease the number of cigarettes smoked12. Furthermore, the abstinence rate tends to increase in users who drink C. cinereum tea compared with those who take a placebo14. In addition to smoking cessation, C. cinereum has been used as a traditional medicine to cure skin diseases, including cough, bronchitis, asthma, malaria, cancer, gastrointestinal disorder, diuresis, pain, and diabetes15,16. A wide range of biological, pharmacological, antimicrobial, anti-inflammatory and anticancer activities have been reported for C. cinereum in numerous studies17,18. Letha et al.19 showed that C. cinereum extract exhibits no significant toxicity in mice and brine shrimp.

It needs to be emphasized that C. cinereum is a common weed, it is not cultivated, and the plant is often found in the same habitat as E. sonchifolia. Species analysis based on morphological characteristics, including aerial parts, flowers and fruits, revealed several similar features between the two species. Careless herbal collection will lead to mixing the two species. In 2018, a local herbal company successfully developed a smoking cessation lozenge from C. cinereum. One large-scale delivery of C. cinereum raw material received from local collectors, however, was contaminated with E. sonchifolia. This contamination resulted in the rejection of the material (unpublished data). E. sonchifolia has been reported to contain pyrrolizidine alkaloids (a class of hepatotoxic and tumorigenic compounds) and has caused food poisoning in several countries20. Eleven pyrrolizidine alkaloid compounds have been identified from E. sonchifolia aqueous extract, and senkirkine was found as the major pyrrolizidine alkaloid in dry herbs21. Moreover, up to 0.2% of doronine and senkirkine from E. sonchifolia may cause significant intoxication22. Therefore, to prevent the adulteration and contamination of C. cinereum materials by E. sonchifolia, a reliable differentiation tool is needed. In addition, the tool should facilitate herbal production of C. cinereum that complies with GAPs and GMPs.

As mentioned above, C. cinereum and E. sonchifolia exhibit similar morphological features, but there is an obvious differential characteristic at the leaf base. An auriculate leaf base is the main characteristic of E. sonchifolia23, while C. cinereum has a simple, alternating and variable leaf shape. As dry herbs and powder of C. cinereum are utilized for smoking cessation products, it is difficult to identify their origin when the adulterant is mixed in. According to the THP, various techniques, including macroscopic, microscopic and TLC identification, are recommended for medicinal plant authentication. Therefore, in this study, several markers of C. cinereum were identified by following THP guidelines. Microscopic authentication revealed that the anomocytic type of stomata is common in both species. Interestingly, C. cinereum contained two types of trichomes: T-shaped and glandular. An abundance of glandular and T-shaped trichomes was observed in C. cinereum leaves, while there was no specific type of trichome present in E. sonchifolia leaves. This finding suggests that the trichomes may be a promising characteristic for the differentiation of C. cinereum from E. sonchifolia raw materials by microscopic observation. Furthermore, our developed macro- and microscopic identification methods for the differentiation of the two species was applied to test whether a commercial smoking cessation tea product purchased from a local market contained the authentic plant species or a substitution. Greenish brown herbal materials with grass-like odor and slightly bitter taste from the tea product were observed. It is difficult to identify the plant species of samples by observing only their macroscopic and organoleptic characteristics. However, under microscopic observation, T-shaped and glandular trichomes, which are unique characters of C. cinereum, were clearly visible in the commercial product (Fig. 6), confirming the authenticity of the C. cinereum tea product. These results are supported by previous studies showing that trichome characteristics can be useful for the differentiation of species with similar ecological or geographical features24.

Figure 6
figure 6

C. cinereum smoking cessation tea product and its photomicrographs. (A,B) C. cinereum tea product, (C,D) overall cell components; 1: fragment of lamina in surface view; 2: lamina in surface view; 3: parenchyma of stem; 4: T-shaped trichome; 5: pappus; 6: fragment of multicellular uniseriate trichome of petal; 7: pollen grain, (E) T-shaped trichome, (F) pappus, (G) glandular trichome and (H) pollen grain.

Full size image

Phytochemical substances such as steroids, saponins, alkaloids, carbohydrates, flavonoids, phenols, tannins, nitrate and proteins have been reported in C. cinereum extracts25. In this study, a distinct band from C. cinereum extract was identified by GC–MS (Supplementary Figure S1), and its mass spectrum was run against reference mass spectra in the NIST2011 library via spectrum matching, resulting in its identification as a mixture of four triterpenoid compounds: β-amyrin, taraxasterol, lupeol, and betulin. Our finding agrees with those of Rao and Bose (1962), who found β-amyrin, lupeol and their acetate derivatives in whole-plant extracts of C. cinereum26. In addition to the four triterpenoid compounds obtained in this study, two other triterpenoid compounds, stigmast-5,17(20)-diene-3β-ol and 24-hydroxytaraxer-14-ene, are also present in extracts of C. cinereum according to a previous report27. This is the first study to identify taraxasterol and betulin from C. cinereum. Therefore, the four triterpenoids could be used as chemical markers for the differentiation of C. cinereum from E. sonchifolia.

In addition to the traditional methods recommended in the THP, DNA barcoding with selected gene candidates, such as matK, rbcL, ITS, and psbA-trnH spacer, has been applied for discrimination of plant species28,29,30. In this study, DNA barcoding analysis was performed for comparison of C. cinereum and E. sonchifolia sequences. Nucleotide polymorphisms and indels were found in all DNA barcode areas. The rbcL gene revealed the lowest nucleotide variation (5.13%) among the DNA barcode regions, which correlated with results from previous work28. Recently, Veldman et al.29 showed that rbcL exhibited very low discriminatory power, making it suitable for identification of plants at the family level. In this study, the full-length matK sequence was obtained even though it has been reported that matK has a low sequencing success rate in medicinal plants because of low primer universality30. The highest nucleotide variation in the two plant species was observed in the ITS region. The ITS sequence contains highly variable sites with the highest interspecific divergence (32.38%) among the four core barcode areas. This finding is consistent with the fact that the ITS region shows greater discriminatory power than the plastid regions and is useful for the authentication of plants at the species level31,32,33. The psbA-trnH spacer, a noncoding region, has a high number of nucleotide variations between C. cinereum and E. sonchifolia. This spacer is found to be the best candidate for discrimination at the species and subspecies levels34. In a previous report, a two-tiered approach to DNA barcoding was suggested for species identification35. Herbal species, including plant fillers composed of herbal supplements, were successfully identified with a two-tiered approach using both the rbcL and ITS2 regions36. A two-locus combination of the psbA-trnH intergenic spacer and ITS2 regions was also a powerful tool for herbal authentication, and this strategy was recommended for the identification of closely related species37.

Since adulterant problems frequently occur due to the morphological similarity of raw materials, efficient and effective methods to identify and monitor the origins of plants are needed. In this study, integrative analysis by macroscopic, microscopic, chemical and molecular markers is proposed for the first time to differentiate raw materials of C. cinereum, a smoking cessation herb, from the adulterant E. sonchifolia. In herbal products or mixed raw materials, however, it is challenging to distinguish C. cinereum from E. sonchifolia. Mixed chemical profiles and specific characteristics of those two plants will be observed if the adulterant is mixed in. Combination of DNA barcoding with advance technique such as Next-Generation Sequencing (NGS) would be beneficial for examination of the admixture. As herbal authentication is a crucial step, confirmation of plant identity with integrative methods should be exploited as a part of quality control for the safety and efficacy of the herb or its products. To protect consumers, policy or guideline frameworks involving good practices for collecting herbal raw material and manufacturing should be seriously considered by regulatory agencies.

Methods

Sample collections

C. cinereum and E. sonchifolia were collected from different natural places in Thailand. Raw C. cinereum materials were kindly provided by an herbal company. Sample details along with their voucher numbers and places of collection were recorded (Table 2). All plant samples were identified by the taxonomist Assoc. Prof. Thatree Phadungcharoen at Rangsit University, Pathumthani, Thailand. Each voucher specimen was assigned a specific number and deposited at the Museum of Natural Medicine, Chulalongkorn University, Thailand.

Macroscopic and microscopic characterization of C. cinereum and E. sonchifolia

The macroscopic and microscopic diagnostic features of C. cinereum and E. sonchifolia were established. Transverse sections of stems and leaves were prepared using sharp razor blades. Sections were mounted with glycerin and subsequently examined under a light microscope (Olympus CH, Japan). Stem and leaf features were observed by 400 × lens magnification under a microscope, and photographs were taken. Slides were prepared in triplicates. Key microscopic features of C. cinereum product were diagnostic under the same conditions as described above.

High-performance thin-layer chromatography of C. cinereum and E. sonchifolia

Whole plants of C. cinereum and E. sonchifolia were dried at 50 °C for 3 days before being crushed into powder and subjected to HPTLC analysis. Subsequently, 20 g of fine powder was extracted in 20 ml of ethanol by immersing the powder in ethanol at room temperature for 48 h. Centrifugation was performed at 10,000 rpm for 10 min at 25 °C, and the supernatant was collected. Extraction solution (3 µL) was sprayed on an HPTLC plate (10 × 10 cm, silica gel 60 F254, Merck, Germany) using an Automatic TLC Sampler 4 (AST4, CAMAG, Muttenz, Switzerland). Each band was 4 mm in length, while the distance between lanes was 2 mm. The distance from the left side was 15 mm, and the distance from the bottom edge was 15 mm. A hexane:acetone mixture (9:1, v/v) was used as the mobile phase. The HPTLC chamber was saturated with 20 ml of the mobile phase for 20 min before development. HPTLC plates were sprayed with anisaldehyde-sulfuric acid reagent and heated at 105 °C to detect phytochemical constituents. Plates were visualized and imaged under short (254 nm) and long (366 nm) ultraviolet (UV) wavelengths.

Phytochemical markers to differentiate C. cinereum from E. sonchifolia

In HPTLC analysis, an unambiguous band unique to C. cinereum was subjected to GC–MS experiments to identify the chemical. The analysis was performed on a GC/MS instrument (triple quadrupole GC/MS (GC-QQQ), Agilent Technologies, Agilent 7890B GC system and Agilent 7000C GC/MS Triple Quad) equipped with an HP-5 ms (length 30 m × inner diameter (ID) 0.25 mm, film thickness 0.25 µm) porous-layer open-tubular (PLOT) column. The split ratio was 1:10. The temperature of the injection port was 280 °C. The flow rate of the helium carrier gas was maintained at a constant 1.0 mL/min. The column temperature was held at 40 °C for 2 min to concentrate the hydrocarbon at the head of the column; subsequently, the column temperature was increased to 150 °C, ramped at 25 °C/min to 300 °C, and held at 300 °C for 15 min. The MS analyses were performed in full scan mode in electrospray ionization (EI) mode with a scan range of m/z 33–600. The ion source was maintained at 70 °C, and an ionization energy of 230 eV was used for each measurement. Compounds were identified by query mass spectrum matching with the NIST2011 reference library (National Institute of Standards and Technology, Gaithersburg, MD, USA).

DNA barcoding of C. cinereum and E. sonchifolia

Genomic DNA extraction

Young leaves of authentic C. cinereum and E. sonchifolia were collected and ground into fine powder by mortar and pestle. Genomic DNA from both plant species was extracted from 80 to 100 mg powder samples using a DNeasy Plant Mini Kit (Qiagen, Germany) following the manufacturer’s instructions. Subsequently, the genomic DNA was further purified using a GENECLEAN Kit (MP Biomedicals, USA) according to the manufacturer’s protocol. The quantity and quality of the purified genomic DNA were determined using a NanoDrop One UV–Vis spectrophotometer (Thermo Scientific, USA) and agarose gel electrophoresis, respectively. Nucleic acids were visualized under UV light (Bio-Rad Gel Doc XR Imaging system, USA) and imaged by using Analysis Quantity One version 4.6.8. Purified genomic DNA was kept at −20 °C for further analysis.

DNA barcoding of authentic C. cinereum and E. sonchifolia

Four core DNA barcode regions, ribulose-1,5-bisphosphate carboxylase/oxygenase large subunit (rbcL), maturase K (matK), internal transcribed spacer (ITS) and trnH-psbA intergenic spacer, were assessed by PCR. The DNA barcode regions were amplified by universal primers (see Supplementary Table S1 online). PCR amplifications were performed in 25 µL of reaction mixture consisting of 1X PCR buffer with 1.5 mM MgCl2, 0.2 mM dNTP mix, 0.5 µM each forward and reverse primer, 0.5 U of Platinum Taq DNA polymerase (Invitrogen, USA) and 50 ng of genomic DNA. PCR was carried out in a T100 Thermal Cycler (Bio-Rad, USA) using 94 °C for 2 min; followed by 30 cycles of 94 °C for 30 s, 57 °C for 30 s, and 72 °C for 1 min (for matK and rbcL) or 30 s (for ITS and psbA-trnH spacer); and a final extension at 72 °C for 10 min. Successful PCR amplifications were observed by 1.2% agarose gel electrophoresis in 1X TAE buffer and imaged using a Gel Doc XR system (Bio-Rad, USA). PCR amplicons were purified and subsequently bidirectionally sequenced by Sanger sequencing on an ABI 3730XL DNA analyzer. The same primer sets used for DNA barcode generation were used for sequencing steps. Nucleotide sequence results were edited by BioEdit software version 7.2.6. DNA sequences were aligned by MUSCLE. The alignment parameters were set as follows: gap open = -400, gap extend = 0, clustering methods = UPGMB and min diag. length = 24. The DNA barcode sequences were deposited in GenBank at the National Center for Biotechnology Information (NCBI). Polymorphisms, nucleotide variation and interspecific divergence of all DNA barcode regions between C. cinereum and E. sonchifolia were analyzed.