The Low Copy Nuclear Gene Region, Granule Bound Starch Synthase (GBSS1), as a Novel Mini-DNA Barcode for the Identification of Different Sage (Salvia) Species

• Abstract
• Introduction
• Results
• Discussion
• Materials and Methods
• Plant material
• Primer Design
• DNA extraction and PCR amplification
• Sanger Sequencing
• Species Identification using distance method
• Species Identification using phylogenetic tree
• Disclaimer
• Data Accessibility
• Contributorsʼ Statement
• References

Abstract

Morphological similarity within species makes the identification and authentication of Salvia species challenging, especially in dietary supplements that contain processed root or leaf powder of different sage species. In the present study, the species discriminatory power of 2 potential DNA barcode regions from the nuclear genome was evaluated in 7 medicinally important Salvia species from the family Lamiaceae. The nuclear internal transcribed spacer 2 and the exon 9 – 14 region of low copy nuclear gene WAXY coding for granule-bound starch synthase 1 were tested for their species discrimination ability using distance, phylogenetic, and BLAST-based methods. A novel 2-step PCR method with 2 different annealing temperatures was developed to achieve maximum amplification from genomic DNA. The granule-bound starch synthase 1 region showed higher amplification and sequencing success rates, higher interspecific distances, and a perfect barcode gap for the tested species compared to the nuclear internal transcribed spacer 2. Hence, these novel mini-barcodes generated from low copy nuclear gene regions (granule-bound starch synthase) that were proven to be effective barcodes for identifying 7 Salvia species have potential for identification and authentication of other Salvia species.

Introduction

Salvia is the largest genus of plants in the commonly known mint or sage family Lamiaceae, with more than 1000 species distributed throughout the subtropical and temperate areas of the Old and New Worlds [1]. Due to the production of aromatic oils and secondary metabolites, some of these plants are commonly used as spices for cooking, such as sage (Salvia officinalis L.), and many of the species have been used as medicinal herbs in traditional Chinese medicine (TCM) [2]. The important medicinal sage species are S. apiana Jeps. (white sage), S. divinorum Epling & Játiva (seerʼs sage or sage of the diviners), S. mellifera Greene (black sage), and S. miltiorrhiza Bunge (red sage). The commonly used medicinal herb in TCM is Danshen (red sage) [2]. The roots and rhizomes of S. miltiorrhiza have been used in TCM for treating coronary heart disease and stroke [3]. Recent Chinese Pharmacopeia (2015) listed the functions of Danshen as “activating blood circulation to remove blood stasis, promoting menstrual discharge to relieve menalgia, clearing heart heat and cooling blood to relieve restlessness and to disperse carbuncles” [2]. Danshen-containing products or dietary supplements are also sold in American and European markets to promote blood circulation and heart health [4], [5]. More than 70 Salvia species are endemic to China and, based on morphologic character-based numerical taxonomy, these are divided into 3 groups–high mountain Danshen, low-mountain Danshen, and non-Danshen [6]. Since the species are identified through morphological characters and root is the plant part used in herbal medicine, the adulteration of commercial products is common. For example, the roots of S. yunnanensis, S. przewalskii, and S. trijua are usually mistaken for Danshen (S. miltiorrhiza). Hence, the identification and authentication of S. miltiorrhiza are especially essential for the safety and efficacy of Danshen drugs or herbal supplements. Moreover, the species hybridize naturally and have many hybrids and varieties, making the identification of species through morphological and genetic methods more complex [7], [8]. Therefore, a reliable and rapid method that is not based on morphology is required to identify different Salvia sp. to ensure the botanical authenticity and safety of commercial products.

DNA barcoding is a molecular technique proposed by Hebert et al. [9] to identify biological species. The technique has been used for the identification of the plant species using short segments of DNA (also known as DNA barcode) from various gene regions, like rbcL, matK, rpoB, rpoC1, trnH-trnL, psbA-trnH, ndf, or atpB, from the chloroplast genome and internal transcribed spacer (ITS) from nuclear genomes by various groups [10], [11], [12], [13], [14]. With sequencing technology advancements, new techniques such as next-generation sequencing, meta-barcoding, barcoding high-resolution DNA melting analysis, and DNA-gold nanoparticle sensors have been used to identify medicinal plants and herbal products [15], [16], [17], [18].

Herbal products or dietary supplements mainly use medicinal plants or herbs as raw materials. In recent years, the demand for herbal products has increased worldwide, and so has adulteration and counterfeit products. Different chemical/analytical methods such as TLC, HPLC, MS, and NMR were being used to confirm the authenticity of these products [19]. However, the chemical detection methods were not always accurate due to variations in geographical location, storage conditions, and seasons [20]. Hence, there was a paradigm shift in the techniques used to authenticate and identify medicinal plants in the last decade, and DNA-based molecular markers called DNA barcodes were validated as additional identification tools. DNA barcodes have been efficiently useful for identifying raw plant materials in a fresh/dried state [12], [14]. However, most herbal products and dietary supplements undergo harsh treatments such as pulverization, extraction using organic solvents, drying in high heat, leaching, and granulation. All these processes can result in fragmented DNA from which the full-length barcodes (500 – 700 bp) are difficult to retrieve. Many research studies suggested short sequences (100 – 150 bp), called DNA mini-barcodes, for processed material to overcome this limitation [21], [22], [23]. The technique has also been tested for powdered or dried samples in which morphological characters are absent and DNA is highly fragmented. DNA mini-barcodes 100 – 300 bp in length for processed samples can be more successful because short sequences could be easily retrieved even from fragmented DNA [21], [22].

The applicability and effectiveness of various gene regions as suitable barcodes for Salvia sp. have been tested by Han et al. [24], De Mattia et al. [25], and Wang et al. [26]. All 3 studies reported different barcode regions from both chloroplast and nuclear genomes as suitable barcodes for identifying Salvia sp. For example, the ITS2 region was most suitable for identifying the tested species by Han et al. [24]. De Mattia et al. [25] tested 4 different barcode loci from the chloroplast genome from which the psbA-trnH region was found to be the most suitable barcode for identifying Lamiaceae species followed by matK. Wang et al. [26] tested 4 DNA regions (rbcL, matK, trnL-F, and psbA-trnH) from the chloroplast genome and ITS, ITS1, and ITS2 from the nuclear genome and 3 combinations (rbcL + matK, psbA-trnH + ITS1, and trnL-F + ITS1) for authentication of 27 Salvia species. Their results revealed that the trnL-F region possessed the highest identification efficiency among the various chloroplast loci, followed by rbcL. The nrDNA showed higher overall identification rates than cpDNA, and ITS1 had the highest authentication capability [26].

Most of the previously published reports have tested loci from the chloroplast genome to identify Salvia species except for the ITS region from the nuclear genome. Generally, the ribosomal DNA (rDNA) is highly repetitive, and the ITS region from rDNA is present in multiple copies in plants. The barcodes from nuclear ITS regions were more useful for identifying plant species, but the presence of paralogous copies of ITS within a given taxon makes specific identification difficult. Hence, more loci apart from ITS from the nuclear genome are required to identify and authentic Salvia sp. A few single-copy loci from the nuclear genome were recently tested for their species discrimination ability in plants [27]. The major advantage of single-copy genes from the nuclear genome is their high rate of variation that helped in easy species identification [27]. The granule bound starch synthase (GBSS1) region from any nuclear genome is a single copy gene that has been useful for the phylogenetic assignment of several plant taxa [28], [29], [30]. However, to the best of our knowledge, there have been no reports of the GBSS1 gene region being tested to identify species using DNA barcoding. Therefore, in the present investigation, the potential of the GBSS1 region for the identification of 7 different Salvia species was tested, and novel DNA mini-barcodes were developed from GBSS1 regions.

Results

The available sequences from low copy nuclear gene GBSS1 for different Salvia species were downloaded from NCBI GenBank and were aligned. The alignment showed a 200 bp variable region with flanking-conserved regions ([Fig. 1]). The genus-specific primers were then developed from the flanking-conserved region for GBSS1. Universal primers for ITS2 [31] were used to compare whether universal or genus-specific primers would work better for the identification of Salvia species,

Genomic DNA was extracted from 30 fresh/dried samples ([Table 1]), and the ITS2 and GBSS1 genomic regions were successfully amplified from various analyzed samples. The success of PCR amplification plays an important role in identifying suitable barcodes for a particular genus. The universal primers used for ITS2 were ITS-S2F/ITS4 [19], and genus-specific primers for the GBSS1 region were GBSS1-F1/R1 and GBSS1-F1/R2 ([Table 2]). The ITS2 genomic region was amplified from 17 of the 30 tested samples (56.7%). At the same time, the GBSS1 region was amplified from 22 out of 30 samples (73.3%). For both loci, most of the amplified products (22 for GBSS1 and 15 for ITS2) resulted in single bands and were directly sequenced in both directions after the clean-up. For ITS2 locus, 2 samples (7%) from S. mellifera (NCNPR#22 497, 22 498) resulted in 2 bands, which were loaded on agarose gel for separation. The band corresponding to the standard marker of the expected size (400 bp approx.) was excised from the gel, purified, and sent for sequencing. The length of the ITS2 region varied from approximately 300 – 350 bp between the analyzed samples. While the GBSS1 region varied from 220 – 300 bp for GBSS1-F1/R1 and for GBSS1-F1/R2 primer pair, the length varied from 150 – 200 bp. Since the retrieval of good quality sequence from the GBSS1-F1/R1 was more successful in the pilot experiment than GBSS1-F1/R2, the former primer pair was selected for further analysis. For ITS2, 14 good quality sequences were obtained out of 17 amplified products, and hence the sequencing success rate was 82.35%. While for GBSS1, all the 22 amplicons resulted in good quality sequences, and therefore the sequencing rate was 100%.

A suitable barcode region must exhibit high interspecific divergence and low intra-specific variations to distinguish species of a particular genus. Interspecific divergence is calculated in terms of average interspecific distance and minimum interspecific distance. The sequences from both the nuclear regions were analyzed, and the interspecific distances were estimated as 0.089 – 0.394 for GBSS1 and 0.045 – 0.170 for ITS2. Among the tested Salvia species, GBSS1 has the highest average interspecific distance (0.225), and the lowest minimum interspecific distance was 0.089. While ITS2 has the lowest minimum interspecific distance (0.045), the average interspecific distance was 0.108, almost half as GBSS1. The intraspecific distances for both ITS2 and GBSS1 were lower than the minimum interspecific distances, and there was a perfect barcode gap for both the tested loci. Hence, based on the comparisons of intraspecific, interspecific distances and interspecific divergences, the mini-barcode from the GBSS1 region was a more suitable barcode for the identification and authentication of Salvia species.

A phylogenetic tree-based method was used to test the efficiency of species identification rates of the tested DNA barcode loci using UPGMA, maximum likelihood, and Bayesian inferences. This species identification tree based on phylogenetic methods showed that all the tested species could be resolved into unique clades (a clade is a group of organisms that are monophyletic–that is, composed of a common ancestor and all its lineal descendants on a phylogenetic tree) for both ITS2 (6 species) and GBSS1 sequences (7 species) ([Fig. 2], [3]). Also, the different individuals of the same species formed monophyletic clades with bootstrap values > 65% ([Fig. 2], [3]). However, the GBSS1 tree showed the accession number# 4807 labeled as S. sclarea formed a clade with S. officinalis samples ([Fig. 2]). While for ITS2, the accession number labeled as #4807 did not result in a good quality sequence and hence was not included in the phylogenetic analysis.

The Basic Local Alignment Search Tool (BLAST) at NCBI GenBank was used to confirm the identity of the generated mini-barcodes. The BLAST analysis for GBSS1 sequences showed that the 6 S. officinalis matched the S. officinalis sequences available in NCBI GenBank with 99% certainty. The GBSS1 sequence of accession number #4807 labeled as S. sclarea also matched with S. officinalis with 98.6% homology. The rest of the sequences are novel for the GBSS1 region and matched with the Salvia genus only with 80 – 90% homology. BLAST analysis for ITS2 sequences matched corresponding species with 99 – 100% homology for 15 out of 17 samples. The 2 S. apiana samples (NCNPR# 13 096, 22 773) matched the common weeds.

Discussion

The widespread use of Salvia species including Danshen (S. miltiorrhiza), black (S. mellifera), and white sage (S. apiana) in TCM and other traditional medicine has led to adulteration/contamination of herbal products with other species, thus posing for a significant safety concern of herbal formulations. These species have distinct chemical profiles and pharmacological properties, but their roots are very similar at the morphological level. The safety and efficacy of the substituted or adulterated herbal component are often unknown, leading to herbal drug inefficiency and pose several deleterious effects such as herb-drug interactions. Hence, the correct identification and authentication of these species are imperative for the safe and effective application of herbal products derived from the plant material. In the present work, 2 nuclear genomic regions, ITS2 and GBSS1, were compared for their effectiveness in species identification at the genomic level for 7 different species of Salvia.

The applicability and effectiveness of various gene regions as suitable barcodes for Salvia sp. have been tested before by Han et al. [24], De Mattia et al. [25], and Wang et al. [26]. All 3 studies reported different barcode regions from both the chloroplast and nuclear genomes as suitable barcodes for identification of Salvia sp. For example, Han et al. [24] compared DNA barcoding and chemical fingerprinting methods to identify 42 medicinal plant species in the genus Salvia, while 6 species belonged to 3 different genera were outgroups. ITS2 region was found to be most suitable for the identification of all the tested species.

De Mattia et al. [25] tested 43 samples representing 16 different species and 21 commercial samples using 4 different barcode loci from chloroplast genome viz. rpoB, rbcL, matK, and psbA-trnH. Among the 4 loci, psbA-trnH was the most suitable barcode for identifying Lamiaceae species, followed by matK.

Wang et al. [26] tested 4 DNA regions–rbcL, matK, trnL-F, and psbA-trnH from the chloroplast genome and ITS, ITS1, and ITS2 from the nuclear genome and 3 combinations (rbcL + matK, psbA-trnH + ITS1, and trnL-F + ITS1) for authentication of 27 Salvia species. Fifty-six accessions from 27 species of Salvia were collected, and 77 Salvia sequences were downloaded from GenBank for analysis. Their results revealed that trnL-F possessed the highest identification efficiency, followed by rbcL. The nrDNA showed higher overall identification rates than cpDNA. ITS1 had the highest authentication capability, correctly identifying 81.48% of the species. In addition, the success rates of the combined barcodes did not increase for all single loci. These published reports have tested loci mostly from the chloroplast genome.

Interspecific hybridization among Salvia species is very common in nature. The chloroplast is mostly maternally derived in the hybrids, thus rendering the barcode regions unsuitable for identifying hybrids, as only the maternal lineage would be identified. The ITS region from rDNA often identifies species but is frequently not variable enough to distinguish closely related species. Moreover, for Salvia species, there is incongruence between various published reports. For example, Wang et al. [26] reported the ITS1 as more suitable for identification, while Han et al. [24] found ITS2 as a candidate barcode for Salvia species. In the present investigation, ITS2 was shown to 1) have lower amplification success and 2) extraneous material was often amplified due to the use of universal primers. Therefore, another gene region, especially from the nuclear genome with high variability and low-copy number, is required to correctly identify and authenticate Salvia species. Compared to highly repetitive rDNA and chloroplast genes, the low copy number genes are not often used for identification since fewer sequences are available. A high copy number is often seen as an asset in processed material, but there is often poor knowledge about the extent of concerted evolution among the various copies [28]. However, Kurian et al. [27] reported that the low copy nuclear gene RBP2 was variable enough to identify Calamus species and a potential DNA barcode for other Palm species.

This study found that the exon 9 – 14 region of the low-copy nuclear gene called WAXY gene (GBSS1) has been used for plantsʼ phylogenetic assignment [28], [29], [30] but not for DNA barcoding as a potential barcode region. For Salvia species, only 19 GBSS1 sequences were available in NCBI GenBank (as of November 15, 2020). Genus-specific primers were developed from the available sequences for the GBSS1 region, and universal ITS2 primers were used to generate DNA mini-barcodes for 7 different Salvia species. The novel DNA mini-barcodes generated from the GBSS1 region were then compared with DNA barcodes from ITS2 to identify Salvia species. The amplification and sequencing success rates for GBSS1 were approximately 20 – 30% higher than ITS2; hence the mini-barcodes from GBSS1 were easily retrievable from the tested Salvia sp. compared to ITS2. The ITS2 region was difficult to amplify in the tested samples compared to GBSS1, which could be attributed to the use of genus-specific primers developed in this study for the GBSS1 locus.

On the other hand, the universal primers were used for ITS2, which could bind to other plant species even if present in only small amounts. Hence, the retrieved sequences of the 2 samples labeled as S. apiana (NCNPR# 13 096, 22 773) matched with the common weeds with BLAST analysis. However, the chemical profile of these 2 samples matched with S. apiana (data not shown). Moreover, multiple bands were found in S. mellifera samples for the ITS2 locus. Though the ITS2 region was suitable for DNA barcoding of few plant species [12], [31], the presence of paralogous copies of ITS within a taxon and/or preferential amplification of endophytic fungal ITS makes this locus unsuitable for identification purposes for some plant species. The interspecific variations analyzed using the K2P model showed higher interspecific distances for GBSS1 than ITS2 (Supplementary File S1, S2). A perfect barcode gap was exhibited by GBSS1 locus for all 7 tested species. In almost all taxa, the intraspecific distances were lower than the minimum interspecific distances, indicating the tested locusʼs good discrimination power. This was also confirmed by the phylogenetic tree-based identification method, where the different samples of particular taxa formed a single clade that is distinguished from other clades ([Figs. 2], [3]). Sample #4807 labeled as S. sclarea formed a clade with S. officinalis in GBSS1 tree-based identification, thus indicating incorrect identification or mislabeling ([Fig. 2]). The discrimination power of the ITS2 locus was found to be comparable to GBSS1, and all the tested species formed monophyletic clades. However, due to intraspecific variations of ITS2 within S. mellifera, the individuals branched out within the same clade ([Fig. 3]).

There are several chemical and morphological methods for identifying Danshen [3], [4], [5], [6], but sometimes these methods fail due to close similarity between related species, or chemical analysis may be affected by storage and other physiological conditions. Chen et al. used ITS2 to successfully identify several medicinal plants, as well as S. miltiorrhiza [12]. However, ITS2 was unsuitable for identifying S. miltiorrhiza accessions tested by Wang et al. [26]. In the present investigation, the ITS2 amplified product also was difficult to retrieve from 5 out of 6 tested samples of S. miltiorrhiza. By contrast, the GBSS1 sequences were obtained from 3 tested samples, all grouped together in 1 clade ([Fig. 3]). The close genetic relationship between S. miltiorrhiza and its related species and the high intraspecific variation in ITS2 could account for the difficulty in distinguishing the different samples when the accession has high interspecific and low intraspecific variations within ITS2.

Different complex processing methods used in the natural herbal products (NHP) industry made traditional barcoding insufficient to identify and authenticate herbal products. Short-length DNA mini-barcodes were then suggested as a derived barcoding method to identify plant materials in NHP [20], [21], [22], [23]. For example, Little [21] developed mini-barcodes from the matK region to identify Ginkgo biloba. Lo et al. found that 88-bp DNA fragments could be retrieved after TCMs were boiled for 120 min, whereas 120 bp fragments were not amplified from the same [32]. The mini-barcodes have extended the validity of DNA barcoding as a rapid and reliable tool for authentication of botanicals/products in the NHP industry, which earlier relied on chemical/analytical methods for identification. The mini-barcodes developed using GBBS1 and ITS2 regions in the present study could also be successfully used to rapidly identify NHP made from Salvia sp in the industry or by the quality control agencies. However, the genus-specific primers pose a limitation for detecting adulteration/contamination by unknown plant species. Since the primers for amplifying mini-barcodes were developed from highly specific, inter-specifically conserved fragments of different species belonging to genus Salvia only, they cannot detect any adulterants or contamination present in NHP, especially in complex NHP that contains multiple species. As a result, Little [21] emphasized that the mini-barcodes could not be used as universal biomarkers for all the species but rather to identify specific target medicinal plant species.

Although mini-barcodes successfully identify processed products (up to an extent), the technique has limitations, such as specificity and length. The DNA mini-barcodes with genus/species-specific primers have different lengths (50 – 200 bp), and some genetic information at the ends of sequences might get lost when Sanger sequencing is used. The adapter primers (M13/P-JET) were added to specific primers to overcome the loss of important genetic information, and a 2-step PCR method was used in this investigation to attain maximum amplicon length. Another limitation is that the short DNA sequence may contain unstable mutation sites within the sequences, resulting in incorrect identification of closely related species. Hence, selecting the correct length and position of mini-barcode that could identify closely related species is essential [33].

In conclusion, DNA analysis showed that the novel mini-barcode from the low copy number GBSS1 gene region was more useful in amplification and sequencing success rates and species identification rate than ITS2 from the nuclear genome. Though low copy nuclear genes are difficult to amplify sometimes due to their less availability, their high variation rate makes them ideal for species identification. For Salvia species, there is a lack of congruence about the best locus/loci to be used as DNA barcodes for identification. Hence, the present work proposes the GBSS1 gene region as a suitable mini-barcode for accurately identifying and authentication of tested Salvia species.

Materials and Methods

Plant material

Salvia samples were collected at the Maynard W. Quimby Medicinal Plant Garden, University of Mississippi, or provided by the Missouri Botanical Garden ([Table 1]). All the samples were identified by botanists Dr. Vijayasankar Raman and Dr. John Sebastian at the University of Mississippi.

Primer Design

Initial in silico analysis was conducted, and 17 GBSS1 sequences for different Salvia sp. were retrieved from the NCBI GenBank (https://www-ncbi-nlm-nih-gov.accesdistant.sorbonne-universite.fr/nuccore/). The sequences were aligned using ClustalW in GeneDoc software version 2.7 [34], and an approximately 200 bp region with the greatest number of single nucleotide polymorphisms (SNPs) and insertions and deletions of bases among the Salvia species, along with the flanking conserved region, was identified ([Fig. 1]). For the exon 9 – 14 region of GBSS1, 1 forward and 2 reverse primers ([Table 2]) were designed manually to amplify from genomic DNA using the primers as a combination of GBSS1_F1 + GBSS1_R1 or GBSS1_F1 + GBSS1_R2 ([Table 2]). The primers contained 5′-pJET tails to achieve maximum sequence length from the GBSS1 region. The ITS2 genomic region was amplified using the universal primers, with the forward primer being ITS-S2F and ITS4 as the reverse primer [35]. The ITS2 primers were also flanked with either 5′pJET or 5′ M13 tails for obtaining more sequence length specific to ITS2 ([Table 2]).

DNA extraction and PCR amplification

DNA was extracted from 30 plant samples ([Table 1]) using DNeasy Plant Mini Kit (Qiagen Inc.), and the DNA quality was assessed using 0.8% agarose gel. PCR amplification was carried out in 2 steps with different primer sets and annealing temperatures. For GBSS1, the first PCR was carried out using a 50 °C annealing temperature, and the second PCR was done at 60 °C. The ITS2 genomic region was amplified using annealing temperatures of 50 °C and 56 °C for the first and second PCRs, respectively. The first PCR amplification was carried out in 25 or 50 µL reaction mixture containing 10 – 20 ng genomic DNA, 1× PCR reaction buffer, 0.2 mM dNTP mixture, 0.2 µM of each forward and reverse primers, 1.5 mM MgCl₂, and 2 U of Platinum Taq DNA Polymerase (Invitrogen). PCRs were run in a SimpliAmp Thermal Cycler (Life Technologies) and consisted of 2 steps. The first PCR program was for 8 cycles at low annealing temperatures and consisted of 1 initial denaturation step at 94 °C for 3 min followed by cycles of 94 °C for 30 sec, adequate annealing temperature for 30 sec, and extension at 72 °C for 30 sec. The second PCR program, using 1 µL of the first PCR as a template and standard primers (M13 or pJet), consisted of 30 cycles at 94 °C for 30 sec, adequate annealing temperature for 30 sec, 72 °C for 30 sec, with a final extension at 72 °C for 3 min.

After amplification, a 10 µL aliquot was analyzed by electrophoresis on a 1.5% TAE agarose gel stained with ethidium bromide, visualized under UV light. The PCR products were compared the molecular size standard 1 Kb plus DNA ladder (Invitrogen). The remaining PCR product aliquot with single bands was purified with MinElute PCR Purification Kit (Qiagen). In the case of multiple bands for S. mellifera, the lower band corresponding to the expected amplicon length was cut and purified with MinElute PCR Purification Kit (Qiagen).

Sanger Sequencing

Purified PCR products were directly sequenced in both directions at GeneWiz. Bidirectional sequences were assembled and trimmed using DNASTAR Lasergene version 15.3.0 (DNASTAR INC.) to remove the low-quality bases at the end of sequences. Some ITS2 sequences resulted in many background sequences and hence were not included in the final alignment. The length of contigs varied from 200 – 230 bp for GBSS1 and 350 – 400 bp for ITS2. The identity of sequences was confirmed using BLAST on NCBI GenBank. The ITS2 sequences have been submitted to NCBI GenBank; accession numbers obtained were MW589129 – MW589142. The GBSS1 sequences were submitted in this paper as Supporting Information Table S3.

Species Identification using distance method

The trimmed sequences were aligned using ClustalW [36], and the length of the alignment was 239 bp for GBSS1, while for ITS2, the alignment was 359 bp long. The intra-specific and inter-specific distances were estimated using Kimura-2-parameter in MEGAX [37], [38] (Supplementary files Tables S1, S2).

Species Identification using phylogenetic tree

Sequence data were aligned with the MUSCLE plugin for Geneious Prime 10.2.3 (www.geneious.com) using default parameters. Phylogenetic trees used for species identification were inferred in 3 ways, all analyzed in Geneious Prime. These included a UPGMA tree with a Jukes-cantor genetic distance model, resampled 1000 times; a maximum likelihood tree using the PHyml plugin [39] with default parameters including a GTR substitution model, 100 bootstraps, fixed proportion of invariable sites, 1 substitution rate categories and optimization for Topology/length/rate; and Bayesian analysis using the MrBayes 2.3.6 plugin [40] with default parameters: GTR substitution model, equal rate variation, MCMC settings: chain length of 1 100 000, 4 heated chains, 0.2 heated chain temp, a subsampling frequency of 200, 100 000 burn-in length. The branch lengths were unconstrained.

Disclaimer

This article reflects the authorʼs views and should not be construed to represent FDAʼs views or policies.

Data Accessibility

The ITS2 sequences generated in this study have been submitted to NCBI GenBank with accession numbers MW589129 – MW589142. GBSS1 sequences were submitted as Supporting Information Table S3 of this publication.

Contributorsʼ Statement

NT and IP developed the primers and IP performed all the experiments. IP wrote the manuscript with support from SH and NT. All authors discussed the results and contributed to the final manuscript. IK and AG helped supervise the project. Design of the study: I. Parveen, N. Techen, S. Handy, J. Li, C. Wu, A. G. Chittiboyina, I. A. Khan; data collection, experiments: I. Parveen; data analysis and interpretation: I. Parveen, S. Handy; drafting the manuscript: I. Parveen; critical revision of the manuscript: N. Techen, S. Handy, J. Li, C. Wu, A. G. Chittiboyina, I. A. Khan.

Conflict of Interest

The authors declare that they have no conflict of interest.

Acknowledgements

This research was supported by the “Holistic Approach for Potential Drug Interactions with Botanical Drugs – Importance of Chemical Fingerprinting and Biosimilarity” funded by the U. S. Food and Drug Administration grant HHSF223201810175.

• References

• 1 Standley P, Williams L. Labiatae. Fieldiana Bot 1973; 24: 237-317
  PubMedSearch in Google Scholar
• 2 Chinese Pharmacopoeia Commission. “Shuanghuanglian keli” in Pharmacopeia of the Peopleʼs Republic of China. China Medicine and Pharmacy. Vol. 1. China Medical Science and Technology Press; 2015: 739
  Search in Google Scholar
• 3 Li MH, Chen JM, Peng Y, Wu Q, Xiao PG. Investigation of Danshen and related medicinal plants in China. J Ethnopharmacol 2008; 120: 419-426
  CrossrefPubMedSearch in Google Scholar
• 4 Zhou LM, Zuo Z, Chow MSS. Danshen: An overview of its chemistry, pharmacology, pharmacokinetics, and clinical use [J]. J Clin Pharmacol 2005; 45: 1345-1359
  CrossrefPubMedSearch in Google Scholar
• 5 Mei X, Cao Y, Che Y, Li J, Shang Z, Zhao W, Qiao Y, Zhang J. Danshen: A phytochemical and pharmacological overview. Chin J Nat Med 2019; 17: 59-80
  PubMedSearch in Google Scholar
• 6 Xiao XH, Fang QM, Xia WJ. Numerical taxonomy of medicinal Salvia L. and the genuineness of Danshen. J Plant Res Env 1997; 6: 17-21
  PubMedSearch in Google Scholar
• 7 Hou YX, Tang HR, Dong XL, Xie WYZ, Chen Q. DNA extraction from Salvia and optimization of RAPD reaction condition. Chin Agric Sci Bull 2008; 24: 72-77
  PubMedSearch in Google Scholar
• 8 Wang M, Zhao HX, Wang L, Wang T, Yang RW, Wang XL, Zhou YH, Ding CB, Zhang L. Potential use of DNA barcoding for the identification of Salvia based on cpDNA and nrDNA sequences. Gene 2013; 528: 206-215
  PubMedSearch in Google Scholar
• 9 Hebert PDN, Cywinska A, Ball SL. Biological identifications through DNA barcodes. Proc R Soc Lond B Biol Sci 2003; 270: 313-321
  CrossrefPubMedSearch in Google Scholar
• 10 Kress WJ, Erickson DL. A two-locus global DNA barcode for land plants: the coding rbcL gene complements the non-coding trnH-psbA spacer region. PLoS One 2007; 2: e508
  CrossrefPubMedSearch in Google Scholar
• 11 CBOL Plant Working Group. A DNA barcode for land plants. Proc Natl Acad Sci U S A 2009; 106: 12794-12797
  CrossrefPubMedSearch in Google Scholar
• 12 Chen S, Yao H, Han J, Liu C, Song J, Shi L, Zhu Y, Ma X, Gao T, Pang X, Luo K, Li Y, Li X, Jia X, Lin Y, Leon C. Validation of the ITS2 region as a novel DNA barcode for identifying medicinal plant species. PLoS One 2010; 5: e8613
  CrossrefPubMedSearch in Google Scholar
• 13 Parveen I, Singh HK, Malik S, Raghuvanshi S, Babbar SB. Evaluating five different loci (rbcL, rpoB, rpoC1, matK, and ITS) for DNA barcoding of Indian orchids. Genome 2017; 60: 665-671
  CrossrefPubMedSearch in Google Scholar
• 14 de Boer HJ, Ichim MC, Newmaster SG. DNA barcoding and pharmacovigilance of herbal medicines. Drug Saf 2015; 38: 611-620
  CrossrefPubMedSearch in Google Scholar
• 15 Cheng X, Su X, Chen X, Zhao H, Bo C, Xu J, Bai H, Ning K. Biological ingredient analysis of traditional Chinese medicine preparation based on high-throughput sequencing: The story for Liuwei Dihuang Wan. Sci Rep 2014; 4: 5147
  CrossrefPubMedSearch in Google Scholar
• 16 de Boer HJ, Ghorbani A, Manzanilla V, Raclariu AC, Kreziou A, Ounjai S, Osathanunkul M, Gravendeel B. DNA metabarcoding of orchid-derived products reveals widespread illegal orchid trade. Proc Biol Sci 2017; 284: 20171182
  CrossrefPubMedSearch in Google Scholar
• 17 Soares S, Grazina L, Costa J, Amaral JS, Oliveira MBPP, Mafra I. Botanical authentication of lavender (Lavandula spp.) honey by a novel DNA-barcoding approach coupled to high resolution melting analysis. Food Control 2018; 86: 367-373
  CrossrefPubMedSearch in Google Scholar
• 18 Huang L, Zheng L, Chen Y, Xue F, Cheng L, Adeloju SB, Chen W. A novel GMO biosensor for rapid ultrasensitive and simultaneous detection of multiple DNA components in GMO products. Biosens Bioelectron 2015; 66: 431-437
  CrossrefPubMedSearch in Google Scholar
• 19 Li M, Hou XF, Zhang J, Wang SC, Fu Q, He LC. Applications of HPLC/MS in the analysis of traditional Chinese medicines. J Pharm Anal 2011; 1: 81-91
  CrossrefPubMedSearch in Google Scholar
• 20 Parveen I, Gafner S, Techen N, Murch SJ, Khan IA. DNA barcoding for the identification of botanicals in herbal medicine and dietary supplements: strengths and limitations. Planta Med 2016; 82: 1225-1235
  Thieme ConnectPubMedSearch in Google Scholar
• 21 Little DP. Authentication of Ginkgo biloba herbal dietary supplements using DNA barcoding. Genome 2014; 57: 513-516
  CrossrefPubMedSearch in Google Scholar
• 22 Song M, Dong G, Zhang Y, Liu X, Sun W. Identification of processed Chinese medicinal materials using DNA mini-barcoding. Chin J Nat Med 2017; 1: 481-486
  CrossrefPubMedSearch in Google Scholar
• 23 Gao Z, Liu Y, Wang X, Wei X, Han J. DNA mini-barcoding: A derived barcoding method for herbal molecular identification. Front Plant Sci 2019; 10: 987
  CrossrefPubMedSearch in Google Scholar
• 24 Han JP, Liu C, Li MH, Shi LC, Song LC, Yao H, Pang XH, Chen SL. Relationship between DNA Barcoding and chemical classification of Salvia medicinal herbs. Chin Herb Med 2010; 2: 16-29
  PubMedSearch in Google Scholar
• 25 De Mattia F, Bruni I, Galimberti A, Cattaneo F, Casiraghi M, Labra M. A comparative study of different DNA barcoding markers for the identification of some members of Lamiacaeae. Food Res Int 2011; 44: 693-702
  CrossrefPubMedSearch in Google Scholar
• 26 Wang M, Zhao H, Wang L, Wang T, Tang R, Wang X, Zhou Y, Ding C, Zhang L. Potential use of DNA barcoding for the identification of Salvia based on cpDNA and nrDNA sequences. Gene 2013; 528: 206-215
  CrossrefPubMedSearch in Google Scholar
• 27 Kurian A, Dev SA, Sreekumar VB, Muralidharan EM. The low copy nuclear region, RPB2 as a novel DNA barcode region for species identification in the rattan genus Calamus (Arecaceae). Physiol Mol Biol Plants 2020; 26: 1875-1887
  CrossrefPubMedSearch in Google Scholar
• 28 Mason-Gamer RJ, Weil CF, Kello EA. Granule-bound starch synthase: structure, function, and phylogenetic utility. Mol Biol Evol 1998; 15: 1658-1673
  CrossrefPubMedSearch in Google Scholar
• 29 Miller RE, Rausher MD. Phylogenetic and systematics of Ipomoea (Concolvulaceae) based on ITS and Waxy sequences. Syst Bot 1999; 24: 209-227
  CrossrefPubMedSearch in Google Scholar
• 30 Wang Y, Chen Q, Chen T, Zhang J, He W, Liu L, Luo Y, Sun B, Zhang Y, Tang HR, Wang XR. Allopolyploid origin in Rubus (Rosaceae) inferred from nuclear granule-bound starch synthase I (GBSSI) sequences. BMC Plant Biol 2019; 19: 303-316
  CrossrefPubMedSearch in Google Scholar
• 31 Parveen I, Techen N, Khan IA. Identification of species in the aromatic spice family Apiaceae using DNA mini-barcodes. Planta Med 2019; 85: 139-144
  Thieme ConnectPubMedSearch in Google Scholar
• 32 Lo YT, Li M, Shaw PC. Identification of constituent herbs in ginseng decoctions by DNA markers. Chin Med 2015; 10: 1
  CrossrefPubMedSearch in Google Scholar
• 33 Gao Z, Liu Y, Wang X, Song J, Chen S, Ragupathy S, Han J, Newmaster S. Derivative technology of DNA barcoding (nucleotide signature and SNP double peak methods) detects adulterants and substitution in Chinese patent medicines. Sci Rep 2017; 7: 5858
  CrossrefPubMedSearch in Google Scholar
• 34 Nicholas KB, Nicholas jr. HB, Deerfield DW. GeneDoc: Analysis and Visualization of Genetic Variation. Embnet news 1997; 4: 14
  PubMedSearch in Google Scholar
• 35 White TJ, Bruns T, Lee S, Taylor JW. Amplification and direct Sequencing of fungal ribosomal RNA Genes for Phylogenetics. In: Innis MA, Gelfand DH, Sninsky JJ, Whit TJ. eds. PCR Protocols: A Guide to Methods and Applications. New York: Academic Press, Inc.; 1990: 315-322
  Search in Google Scholar
• 36 Higgins D, Thompson J, Gibson T, Thompson JD, Higgins DG, Gibson TJ. CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res 1994; 22: 4673-4680
  CrossrefPubMedSearch in Google Scholar
• 37 Kumar S, Stecher G, Li M, Knyaz C, Tamura K. MEGA X: Molecular evolutionary genetics analysis across computing platforms. Mol Biol Evol 2018; 35: 1547-1549
  CrossrefPubMedSearch in Google Scholar
• 38 Tamura K, Nei M. Estimation of the number of nucleotide substitutions in the control region of mitochondrial DNA in humans and chimpanzees. Mol Biol Evol 1993; 10: 512-526
  PubMedSearch in Google Scholar
• 39 Guindon S, Dufayard J, Lefort V, Anisimova M, Hordijk W, Gascuel O. New algorithms and methods to estimate Maximum-Likelihood Phylogenies: Assessing the performance of PhyML 3.0. Syst Biol 2010; 59: 307-321
  CrossrefPubMedSearch in Google Scholar
• 40 Huelsenbeck JP, Ronquist F. MRBAYES: Bayesian inference of phylogenetic trees. Bioinformatics 2001; 17: 754-755
  CrossrefPubMedSearch in Google Scholar

Correspondence

Publication History

Received: 29 April 2021

Accepted after revision: 26 August 2021

Article published online:
20 September 2021

© 2021. Thieme. All rights reserved.

Georg Thieme Verlag KG
Rüdigerstraße 14, 70469 Stuttgart, Germany

• References

• 1 Standley P, Williams L. Labiatae. Fieldiana Bot 1973; 24: 237-317
  PubMedSearch in Google Scholar
• 2 Chinese Pharmacopoeia Commission. “Shuanghuanglian keli” in Pharmacopeia of the Peopleʼs Republic of China. China Medicine and Pharmacy. Vol. 1. China Medical Science and Technology Press; 2015: 739
  Search in Google Scholar
• 3 Li MH, Chen JM, Peng Y, Wu Q, Xiao PG. Investigation of Danshen and related medicinal plants in China. J Ethnopharmacol 2008; 120: 419-426
  CrossrefPubMedSearch in Google Scholar
• 4 Zhou LM, Zuo Z, Chow MSS. Danshen: An overview of its chemistry, pharmacology, pharmacokinetics, and clinical use [J]. J Clin Pharmacol 2005; 45: 1345-1359
  CrossrefPubMedSearch in Google Scholar
• 5 Mei X, Cao Y, Che Y, Li J, Shang Z, Zhao W, Qiao Y, Zhang J. Danshen: A phytochemical and pharmacological overview. Chin J Nat Med 2019; 17: 59-80
  PubMedSearch in Google Scholar
• 6 Xiao XH, Fang QM, Xia WJ. Numerical taxonomy of medicinal Salvia L. and the genuineness of Danshen. J Plant Res Env 1997; 6: 17-21
  PubMedSearch in Google Scholar
• 7 Hou YX, Tang HR, Dong XL, Xie WYZ, Chen Q. DNA extraction from Salvia and optimization of RAPD reaction condition. Chin Agric Sci Bull 2008; 24: 72-77
  PubMedSearch in Google Scholar
• 8 Wang M, Zhao HX, Wang L, Wang T, Yang RW, Wang XL, Zhou YH, Ding CB, Zhang L. Potential use of DNA barcoding for the identification of Salvia based on cpDNA and nrDNA sequences. Gene 2013; 528: 206-215
  PubMedSearch in Google Scholar
• 9 Hebert PDN, Cywinska A, Ball SL. Biological identifications through DNA barcodes. Proc R Soc Lond B Biol Sci 2003; 270: 313-321
  CrossrefPubMedSearch in Google Scholar
• 10 Kress WJ, Erickson DL. A two-locus global DNA barcode for land plants: the coding rbcL gene complements the non-coding trnH-psbA spacer region. PLoS One 2007; 2: e508
  CrossrefPubMedSearch in Google Scholar
• 11 CBOL Plant Working Group. A DNA barcode for land plants. Proc Natl Acad Sci U S A 2009; 106: 12794-12797
  CrossrefPubMedSearch in Google Scholar
• 12 Chen S, Yao H, Han J, Liu C, Song J, Shi L, Zhu Y, Ma X, Gao T, Pang X, Luo K, Li Y, Li X, Jia X, Lin Y, Leon C. Validation of the ITS2 region as a novel DNA barcode for identifying medicinal plant species. PLoS One 2010; 5: e8613
  CrossrefPubMedSearch in Google Scholar
• 13 Parveen I, Singh HK, Malik S, Raghuvanshi S, Babbar SB. Evaluating five different loci (rbcL, rpoB, rpoC1, matK, and ITS) for DNA barcoding of Indian orchids. Genome 2017; 60: 665-671
  CrossrefPubMedSearch in Google Scholar
• 14 de Boer HJ, Ichim MC, Newmaster SG. DNA barcoding and pharmacovigilance of herbal medicines. Drug Saf 2015; 38: 611-620
  CrossrefPubMedSearch in Google Scholar
• 15 Cheng X, Su X, Chen X, Zhao H, Bo C, Xu J, Bai H, Ning K. Biological ingredient analysis of traditional Chinese medicine preparation based on high-throughput sequencing: The story for Liuwei Dihuang Wan. Sci Rep 2014; 4: 5147
  CrossrefPubMedSearch in Google Scholar
• 16 de Boer HJ, Ghorbani A, Manzanilla V, Raclariu AC, Kreziou A, Ounjai S, Osathanunkul M, Gravendeel B. DNA metabarcoding of orchid-derived products reveals widespread illegal orchid trade. Proc Biol Sci 2017; 284: 20171182
  CrossrefPubMedSearch in Google Scholar
• 17 Soares S, Grazina L, Costa J, Amaral JS, Oliveira MBPP, Mafra I. Botanical authentication of lavender (Lavandula spp.) honey by a novel DNA-barcoding approach coupled to high resolution melting analysis. Food Control 2018; 86: 367-373
  CrossrefPubMedSearch in Google Scholar
• 18 Huang L, Zheng L, Chen Y, Xue F, Cheng L, Adeloju SB, Chen W. A novel GMO biosensor for rapid ultrasensitive and simultaneous detection of multiple DNA components in GMO products. Biosens Bioelectron 2015; 66: 431-437
  CrossrefPubMedSearch in Google Scholar
• 19 Li M, Hou XF, Zhang J, Wang SC, Fu Q, He LC. Applications of HPLC/MS in the analysis of traditional Chinese medicines. J Pharm Anal 2011; 1: 81-91
  CrossrefPubMedSearch in Google Scholar
• 20 Parveen I, Gafner S, Techen N, Murch SJ, Khan IA. DNA barcoding for the identification of botanicals in herbal medicine and dietary supplements: strengths and limitations. Planta Med 2016; 82: 1225-1235
  Thieme ConnectPubMedSearch in Google Scholar
• 21 Little DP. Authentication of Ginkgo biloba herbal dietary supplements using DNA barcoding. Genome 2014; 57: 513-516
  CrossrefPubMedSearch in Google Scholar
• 22 Song M, Dong G, Zhang Y, Liu X, Sun W. Identification of processed Chinese medicinal materials using DNA mini-barcoding. Chin J Nat Med 2017; 1: 481-486
  CrossrefPubMedSearch in Google Scholar
• 23 Gao Z, Liu Y, Wang X, Wei X, Han J. DNA mini-barcoding: A derived barcoding method for herbal molecular identification. Front Plant Sci 2019; 10: 987
  CrossrefPubMedSearch in Google Scholar
• 24 Han JP, Liu C, Li MH, Shi LC, Song LC, Yao H, Pang XH, Chen SL. Relationship between DNA Barcoding and chemical classification of Salvia medicinal herbs. Chin Herb Med 2010; 2: 16-29
  PubMedSearch in Google Scholar
• 25 De Mattia F, Bruni I, Galimberti A, Cattaneo F, Casiraghi M, Labra M. A comparative study of different DNA barcoding markers for the identification of some members of Lamiacaeae. Food Res Int 2011; 44: 693-702
  CrossrefPubMedSearch in Google Scholar
• 26 Wang M, Zhao H, Wang L, Wang T, Tang R, Wang X, Zhou Y, Ding C, Zhang L. Potential use of DNA barcoding for the identification of Salvia based on cpDNA and nrDNA sequences. Gene 2013; 528: 206-215
  CrossrefPubMedSearch in Google Scholar
• 27 Kurian A, Dev SA, Sreekumar VB, Muralidharan EM. The low copy nuclear region, RPB2 as a novel DNA barcode region for species identification in the rattan genus Calamus (Arecaceae). Physiol Mol Biol Plants 2020; 26: 1875-1887
  CrossrefPubMedSearch in Google Scholar
• 28 Mason-Gamer RJ, Weil CF, Kello EA. Granule-bound starch synthase: structure, function, and phylogenetic utility. Mol Biol Evol 1998; 15: 1658-1673
  CrossrefPubMedSearch in Google Scholar
• 29 Miller RE, Rausher MD. Phylogenetic and systematics of Ipomoea (Concolvulaceae) based on ITS and Waxy sequences. Syst Bot 1999; 24: 209-227
  CrossrefPubMedSearch in Google Scholar
• 30 Wang Y, Chen Q, Chen T, Zhang J, He W, Liu L, Luo Y, Sun B, Zhang Y, Tang HR, Wang XR. Allopolyploid origin in Rubus (Rosaceae) inferred from nuclear granule-bound starch synthase I (GBSSI) sequences. BMC Plant Biol 2019; 19: 303-316
  CrossrefPubMedSearch in Google Scholar
• 31 Parveen I, Techen N, Khan IA. Identification of species in the aromatic spice family Apiaceae using DNA mini-barcodes. Planta Med 2019; 85: 139-144
  Thieme ConnectPubMedSearch in Google Scholar
• 32 Lo YT, Li M, Shaw PC. Identification of constituent herbs in ginseng decoctions by DNA markers. Chin Med 2015; 10: 1
  CrossrefPubMedSearch in Google Scholar
• 33 Gao Z, Liu Y, Wang X, Song J, Chen S, Ragupathy S, Han J, Newmaster S. Derivative technology of DNA barcoding (nucleotide signature and SNP double peak methods) detects adulterants and substitution in Chinese patent medicines. Sci Rep 2017; 7: 5858
  CrossrefPubMedSearch in Google Scholar
• 34 Nicholas KB, Nicholas jr. HB, Deerfield DW. GeneDoc: Analysis and Visualization of Genetic Variation. Embnet news 1997; 4: 14
  PubMedSearch in Google Scholar
• 35 White TJ, Bruns T, Lee S, Taylor JW. Amplification and direct Sequencing of fungal ribosomal RNA Genes for Phylogenetics. In: Innis MA, Gelfand DH, Sninsky JJ, Whit TJ. eds. PCR Protocols: A Guide to Methods and Applications. New York: Academic Press, Inc.; 1990: 315-322
  Search in Google Scholar
• 36 Higgins D, Thompson J, Gibson T, Thompson JD, Higgins DG, Gibson TJ. CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res 1994; 22: 4673-4680
  CrossrefPubMedSearch in Google Scholar
• 37 Kumar S, Stecher G, Li M, Knyaz C, Tamura K. MEGA X: Molecular evolutionary genetics analysis across computing platforms. Mol Biol Evol 2018; 35: 1547-1549
  CrossrefPubMedSearch in Google Scholar
• 38 Tamura K, Nei M. Estimation of the number of nucleotide substitutions in the control region of mitochondrial DNA in humans and chimpanzees. Mol Biol Evol 1993; 10: 512-526
  PubMedSearch in Google Scholar
• 39 Guindon S, Dufayard J, Lefort V, Anisimova M, Hordijk W, Gascuel O. New algorithms and methods to estimate Maximum-Likelihood Phylogenies: Assessing the performance of PhyML 3.0. Syst Biol 2010; 59: 307-321
  CrossrefPubMedSearch in Google Scholar
• 40 Huelsenbeck JP, Ronquist F. MRBAYES: Bayesian inference of phylogenetic trees. Bioinformatics 2001; 17: 754-755
  CrossrefPubMedSearch in Google Scholar

• Supplementary Material