Molecular barcoding of Melissa officinalis L. (badranjboye) in Iran and identification of adulteration in its medicinal services

Medicinal plants are an important source for treatment of diseases in many countries. Today, controlling the quality of the products of medicinal plants is an important task. Customer health may be in danger due to the fraud and misconduct by the sales associates in the sales centers. Melissa officinalis L. (Lamiaceae) is an important medicinal plant used for treatment of several diseases. In Iran, the species of Dracocephalum, Hymencrater, Nepeta and Stachys are mistakenly sold under the name of Badranjboye that have different pharmaceutical properties. For avoiding this mistake, this investigation was performed with the following aims: 1) Checking for the cheating and identifying the Badranjboye in the Iran’s market of medicinal plants, 2) Providing the molecular barcode for the medicinal species of Melissa. For this purpose, Market-sold plant samples (leaves) and original reference plant species were compared by morphology, odor as well as Internal transcribed spacer (ITS) and chloroplast DNA ((psbA–trnH) and (trnL–trnF)) sequences. Various molecular analyses, such as sequencing, determination of genetic distance, and construction of phylogenetic tree were performed. These reports have shown that internal transcribed spacer (ITS) and chloroplast DNA (psbA–trnH) sequences are an efficient molecular marker to produce barcode gap and differentiating Melissa officinalis from other species.


Introduction
Nowadays, identifying and applying the proper curing approaches versus harms threatening the human wellbeing including the presence of harmful chemicals and microorganisms in the food, water contamination, and illnesses is of great importance [1]. Traditional medicine is an important and historical treatment option in many countries including Iran [2].
Adulteration is referred to the deliberate replacement or addition of another plant or external material in order to increase the weight, power, or decrease the cost of the product, generally considered as deliberate action.
Thus, the forgery could occur due to the lack of knowledge about the original plant, the similarity in the name, morphology, and the aroma of species associated with low-cost plant material, so the wrong kinds should be identified from the beginning because of improper use of medicinal plants can cause irreparable damage to the body [3].
In this regard, in a study conducted in China, 1-hundred female patients were detected to have renal disease because of incorrect application of Stephania tetrandra S. Moore instead Aristolochia fangchi Y.C.Wu ex L.D.Chow & S.M.Hwang due to the similarity in the name and morphology of it. Other similar cases can mention the skin of the Cinnamomum verum J. Presl that causes poisoning because of mixing with toxic Cinnamomum cassia (L.) J.Presl species [1].
The morphological, anatomical, chemical, and molecular techniques are used as a general method for identification of the plants. Therefore, the traditional classification requires the expertise of the professionals with professional experience. In some cases, it may be difficult for the experts to identify the samples without diagnostic parts [3][4][5].
Microscopic mass spectrometry and more recently, DNA barcoding, genomics, proteomics, and metabolomics techniques have been used to differentiate the original medicinal plants from wrong ones [5,6].
DNA barcoding is a sequence of DNA that can be useful in the fast and precise identification of species. It has been applied for identifying the medication-related herbs and has differentiated the real and authentic goods from their counterfeit ones [7,8]. The main principles of DNA barcoding are standardization, minimalism, and scalability, which means selection one or a few standard loci that can be sequenced routinely and reliably in very large and diverse sample sets, and obtaining a reliable and conveniently comparable data to differentiate the species in question from one another [9]. These methods provide suitable tools for analysis of the medicinal plants in the markets, requiring precise identification of the plant species, such as sample sequences, comparing the obtained sequence with recorded ones in the reference database, and various nucleotide sequences to differentiate the similar species [6,10].
Identification of the medicinal plants based on the DNA sequence, multiple, and its variation with sister species is currently used. Different researches resulted in general agreement that several different marker combinations produce equivalent performance, and that none of the proposed barcodes is perfect in every respect [11]. Utilizing a multiple approach for a better species differentiation has been suggested by several authors (see for example, Fazekas [12]).
In most of the studies, researchers use of a common, easily amplified and aligned region such as rbcL, trn L-F spacer regions, mat K, trnHG-psbA, nrITS1, nrITS2 or the full ITS1-5.8S-ITS2 (nrITS), as suggested by the CBOL Plant Working Group and BOLD [13,14].
In Iran, there are around 8000 species of plants, among which 2300 species are aromatic and medicinal, and 450 species are sold in the traditional herbal shop [2].
Melissa officinalis L. is one of the medicinal species known as Badranjboye in Iran. This species is well known as a medicinal plant and is used by the people in a variety of forms such as essential oils, oily extract, oil, and infusion and has various properties including the treatment of stomach disorders with neural origin. Badranjboye as an essential drug has been prescribed by Avicenna for enhancement of the heart and the expansion of the soul. This plant has been used as an antidepressant until the seventeenth century [15][16][17][18].
In Iran, Dracocephalum L., Hymenocrater Fisch. & C.A.Mey., Nepeta L. and Stachys L. are wrongly named as Badranjboye. Thus, none of these species will have the impacts of the aforementioned plant if they are applied. They are different such that, Badranjboye possesses heart -formed leaves, whilst Dracocephalum moldavica L. possesses lengthy leaves with a longitude of 3-5 cm and width of 1-5 cm. Moreover, it is different from Hymenocrater having odorless leaves [19].
Considering that the identification of plants is sometimes not possible through their appearance due to their powdery form, in this study, the following objectives will be followed: 1) Determining the integrity of medicinal plants in the country's pharmaceutical market and detecting any fraud through molecular approaches and 2) Developing a molecular barcode to identify the species Melissa officinalis from other items identified as Badranjboye.

Material and methods
For determining the DNA barcode, leaves of the medicinal plants identified under the name of Badranjboye were collected from 22 traditional herbal shops in eight provinces in Iran. Voucher specimens are deposited in Herbarium of Shahid Beheshti University (HSBU) ( Table 1). For determining the appropriate parts for the barcodes, the sequences were recorded at the National Center for Biotechnology Information (NCBI) (https:// www. ncbi. nlm. nih. gov/). for ITS and chloroplast region (trnH-psbA, trnL-trnF) of Melissa, Dracocephalum, and Hymenocrater were first obtained ( Table 2).
The PCR reaction mixture consisted of 20 ng genomic DNA and 3 U of Taq DNA polymerase (Bioron, Germany); 50 mM KCl; 10 mM Tris-HCl buffer at pH 8; 1.5 mM MgCl2; 0.2 mM of each dNTP (Bioron, Germany); 0.2 l M of each primer in a total volume of 25 pi.
DNA amplification was performed on a BIO RAD (T100 Thermal cycler) with the following program: 5 min at 94 C°, 35 cycles of 1 min at 94 C°, 1 min at 50 C° in trnH-psbA and 54C° at ITS primers and 1 min at 72 C and a final cycle of 7 min at 72 C°.
The PCR amplified products were separated by electrophoresis on 2% agarose gels (Merck). The gels were stained with ethidium bromide and visualized under UV light or silver stained for added sensitivity.  Fragment size was estimated by using a 100 base pair (bp) molecular size ladder (Fermentas, Germany).

Data analyses
ITS and chloroplast (trnH-psbA, trnL-F) sequences obtained from the NCBI ( Table 2) were analyzed after alignment and curation using molecular evolutionary genetics analysis (MEGA) ver. 7 software [24]. Maximum likelihood, maximum parsimony, (Neighbor-Joining) NJ and (Unweighted Pair Group Method with Arithmetic mean) UPGMA tree were plotted to show the delimitation between the genera and their species. After clustering and measuring the distance between Dracocephalum, Hymenocrater, and Melissa species based on ITS, trnH-psbA and trnL-F markers, it was concluded that ITS and trnH-psbA are the most appropriate markers for the Melissa barcode. Therefore, the Iran's market samples were amplified by these markers. The unknown barcodes from Iran's market leaves were basically identified by the Basic Local Alignment Search Tool (BLAST) data with a minimum BLAST identity cut off of 97% for a top match. These results were verified by clustering and phylogenetic analyses where the branches of unknown specimens were compared with the sequences of reference species using the MEGA software ver. 7 [23].
Two different approaches were used for barcode gap analysis. Firstly, aligned sequences of both reference plants (Melissa officinalis) and potential adulterants (Dracocephalum, Hymenocrater, and the other Melissa species) were compared in order to identify barcodes and differentiating sequences between reference plants and the potential adulterants (commonly mixed products), as well as common sequences in the reference plant that differ with the adulterant plant species were searched. Secondly, the species discrimination power of the investigated loci was also assessed using the genetic distance approach, to evaluate whether the amount of variation displayed was sufficient to discriminate sister species without affecting their correct assignation through intraspecific variation. This approach is the basis of the "barcoding gap" definition, i.e. the assumption that the amount of sequence divergence within species is smaller than that between species. For this purpose, Kimura 2-parameter model of genetic distance in different sequences was used within and among the congeneric species by the molecular evolutionary genetic analysis using the MEGA7 software [2,24].
The true medicinal plant species presenting a minimum interspecific distance value higher than their maximum intraspecific distance were considered successfully discriminated from potential adulterant plant species [25].

Investigation of delimitation between medicinal genera and species using ITS marker
ITS, trnH-psbA, and trnL-F sequences ( Table 2) were analyzed after alignment and curation using the MEGA software ver. 7. In the maximum likelihood plot developed based on ITS and and trnH-psbA Sequences (Figs. 1 and 2), Melissa officinalis samples were separated from Hymenocrater, Dracocephalum and Melissa axillaris (Benth.) Bakh.f. but these genera could not be separated Fig. 1 Maximum likelihood phylogenetic tree of NCBI taxa based on ITS Sequences from the others based on trnL-F sequences (Fig. 3). Therefore, this region was not suitable for DNA barcoding in Melissa officinalis.
Based on Kimura 2-parameters genetic distance in ITS and trnH-psbA regions, there was no genetic distance within Melissa officinalis (0.00), This result revealed the absence of sequence variability within ITS and trnH-psbA sequences within Melissa officinalis (high uniformity). A high value of genetic distance occurred between Melissa officinalis with other species (Melissa axillar) and genera (Dracocephalum and Hymenocrater) (0.08 to 0.18). Kimura 2-parameters genetic distance based on trnL-F region showed a high value of genetic distance within Melissa officinalis.
Common and specific sequences were evaluated in the ITS and trnH-psbA genetic regions. It was attempted to create the barcode gaps between the studied species and genera. Accordingly, in ITS region, six nucleotides with numbers of (117-131-227-522-650-655) were identified as barcode gap for Melissa officinalis and Melissa axillaris while trnH-psbA region could only identify two barcodes to differentiate. Thirty nucleotides were used as a barcode gap for separation of Melissa genus from Hymenocrater and Dracocephalum in both regions. ITS and trnH-psbA region can be considered as a barcode gap in Melissa officinalis (Figs. 4 and 5). Therefore, it is suggested to use the ITS and trnH-psbA sequences as barcode gaps for barcoding gap and differentiating between Melissa officinalis and other species and genera.

Morphology and odor of the medicinal products
Morphologically, the leaves of the samples sold in the market were similar to Melissa, Dracocephalum, and Hymenocrater. It should be noted that, there were leaves belonging to the Boraginaceae and Poaceae families in the collected specimens. Two samples were in powder form and could not be identified by their leaves. For checking the aroma in the samples, we used Sheidai [2]. Melissa officinalis smells like lemon while other genus does not have this smell. The studied samples were mostly odorless and occasionally had a smell of lemon (Table 3).

Blast results of the market products/samples studied based on ITS and trnH-psbA regions
Sequencing of ITS and trnH-psbA regions, produced about 750 nucleotides. PCR gel photos and BLAST analysis of the market products/samples studied based on ITS and trnH-psbA regions showed similar results; therefore, only ITS results are presented in Table 4 and Fig. 6. The samples sold in the market as Badranjboye (Melissa officinalis) showed sequence similarity with various species

Genetic distance results within and between market samples and the reference species based on ITS and trnH-psbA regions
Kimura 2-parameter model of genetic distance was provided to determine the genetic distance both within the studied market samples and between the samples and the reference species based on ITS and trnH-psbA regions. Results showed that, the market samples have high genetic distance with each other for example, the genetic distance between the samples obtained from Shiraz (No 2 based on Table 1) and samples from Isfahan and Ardabil provinces (No 2 based on Table 1) was from 0.6 to 0.10, while they were highly similar with the samples having the highest percentage of similarity in the BLAST (Table 4) (0.00-0.02). This gap indicates that the samples are not identical in the market implying the fraud in their sales. Table 4 The BLAST results of the sample products Table 1 Blast of species of NCBI Accession Identity

Phylogenetic analyses of the market products/samples studied based on ITS and trnH-psbA regions
Phylogenetic analyses of the market products/samples studied based on ITS and trnH-psbA regions were done. Phylogenetic analyses of the studied market products/ samples were done based on ITS trnH-psbA regions. Different methods like Maximum parsimony, Maximum likelihood, and UPGMA produced similar results by MEGA software ver.7; therefore, only maximum likelihood plot is presented here (Figs. 7 and 8). In this plot, the market samples were placed next to the samples with the highest percentage of similarity in the BLAST (Table 4). For example, samples 1, 2, 3, and 4 that showed the highest similarity to Asperugo procumbens and Mertensia virginica were placed next to them (Table 4) showing the adulteration in marketing of this medicinal species.

Discussion
In the current century, people face various kinds of threats such as food contamination, air pollution, diseases, and so on. Humans rely on different types of treatments to get healthier and better quality of life. Traditional medicine, mainly based on the prescription of the plants is considered as an important and historical treatment options in many countries. Therefore, the highest care and supervision should be carried out to maintain safe and real marketing of the medicinal plants and consumer products. In this regard, it is important to use the best and most effective modern techniques. For this purpose, the barcoding and molecular techniques must be used as the available tools [2]. Despite the high Medicinal importance of Lamiaceae, there are relatively few DNA barcodes available in this family [2]. Accurate and effective identification of gene sequences may vary in different plant species and should be investigated separately in each case. Has yielded relatively high levels of species resolution. The ITS and chloroplast gene region trnH-psbA has been proposed in many DNA barcode evaluation studies. For example, Laiou [26] investigated "core barcode" for land plants (rbcL, matK, and trnH -psbA) in 24 taxa. They sought to identify the right species based on the sequences, presence of DNA, gaps, and differentiation. The highest genetic diversity was observed in the trnH -psbA region; however, DNA barcoding was found in most cases using a sequence divergence in such a species. In general, species were successfully identified by 66.7%. Armenise [27] also reported successful barcoding in Pinaceae with rbcL + trnH -psbA.
In the present study, we showed that based on ITS and trnH-psbA sequence analysis Melissa officinalis samples were separated from Hymenocrater, Dracocephalum, and Melissa axillaris but these samples could not be separated from the others based on trnL-F sequences which indicate, this region was not suitable for DNA barcoding in Melissa officinalis. In addition, the same PCR bands were observed for the entire ITS and trnH -psbA gene region in our study therefore, we suggest that these region is a suitable barcode for the samples.
Adulteration of the medicinal plants has been reported throughout the world. For example, Newmaster [10] studied the plant product of a firm to determine the probability of cheating and succeeded in improving the DNA barcode in most plant products (91%) and all leaf samples (100%) with a resolution of 95% in the species. They found that, most of the studied products (59%) had DNA barcode of plant species that were not marked on the tag. Almost half of the products (48%) were confirmed through the analysis.
Sheidai [2] using ITS marker revealed that Kakoti (ziziphora) is adulterated with Thymus species in Iran. In this study, a low value of genetic distance within Thymus and Ziziphora samples (0.00-0.02) revealed a low degree of sequence variability within ITS sequences (high uniformity) whereas, a higher value of genetic distance was  Table 1) observed between Ziziphora and Thymus samples (0.10-0.20) that can consider as a barcode gap between the two groups. In the other study, Mohebi anabat [28] used trnH-psbA region to separate the Iranian saffron from the world's saffron with a high genetic difference. We also reported that based on Kimura 2-parameters there was no genetic distance within Melissa officinalis (0.00) and a high value of genetic distance occurred between Melissa officinalis with other species (0.08 to 0.18).
In Iran, Badranjboye is one of the medicinal plants sold throughout the country. This study shows that this important product is introduced with the genera Dracocephalum and Hymenocrater as badranjboye. The results of morphology, odor, BLAST, phylogenetic tree, genetic  Table 1) and reference species identified by BLAST query, based on ITS sequences. (Numbers above branches are bootstrap value) distance, and barcode gap showed that at least 5 genera of Hymenocrater, Dracocephalum, Asperugo, Mertensia, Trigonella, and Melilotus were sold in the 22 markets studied as Melissa Officinalis which could have very negative effects on human health.

Conclusion
Due to the medicinal value of Melissa officinalis, this species is sold in the Iranian markets, and some genera like Dracocephalum, Hymenocrater, Asperugo, Mertensia, Trigonella, and Melilotus have been sold instead of this species  Table 1) and reference species identified by BLAST query, based on trnH-psbA sequences. (Numbers above branches are bootstrap value)