Proper identification and authentication of medicinal plants are vital for the quality and safety of herbal medications. Morphological and phytochemical-based traditional methods of identification may not suffice in discriminating close relatives, especially in dried, ground and processed herbs. DNA barcoding has emerged as a molecular identification technique that uses genomic loci in differentiating plant species as well as detecting botanical adulteration. This paper reviews the relative performance of three common DNA barcodes; nuclear ribosomal internal transcribed spacer 2 (ITS2) and chloroplast rbcL and matK genes. A structured narrative review was carried out on published literature on amplification and sequencing feasibility, ability of these three DNA barcodes to discriminate plant species, genetic variability, barcode gap properties and the use of the three barcodes in medicinal plant identification. According to published data, the ITS2 is more effective than the other two loci in providing high species discrimination, especially among medicinal plant species while rbcL provides successful amplification and sequencing but with low discrimination power. On the other hand, matK locus provides extra sequence variation with amplification efficiency being dependent on primer compatibility and taxonomy of plants. The findings in the study of medicinal plant species, such as Lamiaceae, provide additional proof that barcoding results can differ greatly depending on the quality of samples, taxonomic groups, database completeness, and identification process. Furthermore, the review stresses the importance of applying species-specific PCR amplification, quantitative PCR amplification, high-resolution melting analysis, and DNA metabarcoding in the detection of adulteration and misidentification in herbal products. However, the problems related to degraded DNA, hybridization, intragenomic polymorphism, primers' preference, and incompleteness of database cannot be overlooked either. Thus, the use of complementary DNA barcodes along with reference samples, validated sequence databases, and phytochemistry is the key.

Lamiaceae contains many medicinal and aromatic plants for which accurate identification can be problematic due to morphological resemblance of related species, environmental plasticity, renaming (e.g., Rosmarinus officinalis, currently considered Salvia rosmarinus), and substitution/adulteration of herbal material dried and processed. This narrative review, based on laboratory and in Silico data as examples, provides a workflow for the identification and molecular authentication of some species of Lamiaceae, starting from their field collection. Literature regarding plant barcoding (e.g., rbcL, matK, ITS2, trnH-psbA) and additional methods such as species-specific PCR, high-resolution melting, digital PCR and metabarcoding was synthesised. The workflow includes field collection and botanical identification of voucher specimens; CTAB DNA extraction procedure, optionally supplemented with spin-column DNA purification; assessment of DNA quality and amplicon size using agarose gel electrophoresis; SYBR Green real-time qPCR (targeting GAPDH and ITS2); GenBank and BOLD search followed by MUSCLE multiple alignment and phylogenetic analysis in MEGA 12.1.3 (1,000 bootstrap replicates). However, no one marker is considered to be able to discriminate all taxa, rbcL and matK markers form the basis, while ITS2 and trnH-psbA can be used as complementary markers. For example, in the case of the rbcL tree based on the alignment of ten reference sequences, the maximum level of support (100%) could be seen for the Origanum majorana- Ocimum tenuiflorum clade only, while the other nodes have low levels of support (less than 50%). Taxonomic identification thus needs to involve voucher information, sequence comparison, phylogenetic placement and, when necessary, chemical composition.