Methodologies of Identifying Medicinal Materials

3.2 Methodologies of Identifying Medicinal Materials

Assurance of the correct use of medicinal materials is fundamental for the development of traditional medicine industry. The traditional identification methods based on organoleptic and microscopic features, such as shape, color, texture, odor, tissue arrangement and cell components, are simple and inexpensive.

However, these methods are subjective and depend heavily on the experience and judgment of the inspector. Also, insufficient informative characters in processed materials may lead to low accuracy and limited resolution. Alternatively, chemical profiling has become a standard practice for species identification and quality control. However, chemical components vary with a number of factors including growing stage, harvest time, locality, storage condition, processing method and manufacturing procedure. The presence of large amounts of proteins, polysaccharides, resins, tannins and thousands of secondary metabolites makes chemical analyses difficult [ 16 ] .

Molecular authentication based on the variation of DNA sequences in different organisms provides an alternative approach. In principle, the genetic makeup is unique to a species independent to body parts, growing stage, and environment.

Therefore, DNA-based identification methods are less sensitive to biological, physiological, physical and environmental factors. In addition, benefited from the development of polymerase chain reaction (PCR), a small amount of sample is sufficient for carrying out the authentication process. These advantages are particularly important in identifying shredded materials or powder, not to mention expensive materials with limited supply [ 17 ]. Furthermore, DNA is relatively stable and may be extractable from herbarium specimens, processed food and commercial products. Therefore, DNA technique is applicable to a wide range of forensic issues.

Our group has pioneered in using molecular techniques to identify medicinal materials. In the mid-1990s, we applied arbitrarily primed polymerase chain reaction (AP-PCR) DNA fi ngerprinting to distinguish Oriental ginseng roots (Panax ginseng) from American ginseng (P. quinquefolius) [ 18, 19 ]. Thereafter, the application of molecular techniques has become popular. Various strategies, such as forensically informative nucleotide sequencing (FINS), DNA barcoding and isothermal amplification, have now been introduced to increase accuracy and efficiency [ 13, 20, 21 ] . There are three main molecular techniques being used, namely, DNA fingerprinting, DNA sequencing and DNA microarray. A general evaluation of the various identification methods is shown in Table 3.1.

3.2.1 DNA Fingerprinting

DNA fingerprinting explores the DNA polymorphism in the whole genome or in a specific region of the sample. The polymorphic patterns are usually visualized by agarose or polyacrylamide gel electrophoresis or capillary electrophoresis. Unlike fresh materials, the quantity and quality of DNA in a medicinal material may be poorly preserved due to post-harvest processing and storage. DNA fi ngerprints without DNA amplification, such as restriction fragment length polymorphism (RFLP), are less applicable because of the low yield of extracted DNA and poor integrity of genomic DNA. Consequently, PCR-based DNA fingerprinting is the preferred practice. These fingerprints include arbitrarily primed PCR (AP-PCR), random amplified polymorphic DNA (RAPD), amplified fragment length polymorphism (AFLP), direct ampli fi cation of length polymorphism (DALP), inter-simple sequence repeat (ISSR), PCR restriction fragment length polymorphism (PCRRFLP) and sequence-characterized ampli fi cation region (SCAR).

3.2.1.1 Arbitrarily Primed PCR (AP-PCR)

Arbitrarily primed PCR (AP-PCR), or arbitrarily chosen primer PCR (ACP-PCR), is a whole-genome fingerprint approach fi rst reported in 1990 [ 22 ] . Multiple loci are amplified using a single primer of approximately 20 nucleotides which anneals to the genomic DNA template at a number of sites and acts as both the forward and reverse primers. When two annealing sites are close enough, such as within two kilobases or less, the DNA in between can be successfully ampli fi ed under normal PCR conditions. The number of primer sites and the match of primers with the primer sites contribute to the polymorphic DNA fi ngerprints among samples. Since the primer anneals to the genomic DNA arbitrarily, AP-PCR does not require prior knowledge of the target genome, and multiple loci can be examined simultaneously.

There have been many publications using this approach to identify medicinal materials. For example, our group found that all the AP-PCR fi ngerprints generated using three primers (M13 forward, M13 reverse, and Gal-K primer) successfully differentiated the dried roots of Oriental ginseng (Panax ginseng) from American ginseng (P. quinquefolius) [ 18 ] . Similar approach was subsequently applied to the identification of other medicinal species, including Kudidan (Elephantopi Herba), Pugongying (Taraxaci Herba), and Dangshen (Codonopsis Radix) [ 19, 23– 25 ] .

3.2.1.2 Random Amplified Polymorphic DNA (RAPD)

Random amplified polymorphic DNA (RAPD) is another whole-genome fingerprint [ 26 ] . The principle of RAPD is quite similar to AP-PCR and differs only from the use of a single primer of 10 nucleotides under reduced stringent conditions. The polymorphic fingerprints are due to the number of primer sites, nucleotide polymorphism in the primer sites, and the distance between adjacent primer sites. RAPD also does not require prior knowledge of the genome and can be used to examine multiple loci simultaneously. In AP-PCR and RAPD, the quality and integrity of genomic DNA remain major concerns. Besides, AP-PCR and RAPD markers are dominant markers and are usually unable to distinguish homozygous loci fro heterozygous loci. Our group applied both RAPD and AP-PCR to differentiate medicinal Panax species from their adulterants [ 19 ] . It was shown that polymorphic fingerprints of Panax ginseng , P. quinquefolius, and P. notoginseng can be generated with appropriate RAPD and AP-PCR primers, and these fingerprints can differentiate Panax species from the adulterants derived from Mirabilis jalapa, Phytolacca acinosa , Platycodon grandi fl orum and Talinum paniculatum .

Furthermore, the degree of similarity among the RAPD and AP-PCR fingerprint suggested that P. ginseng is more closely related to P. quinquefolius than to P. notoginseng. Similar approach of RAPD fingerprinting was applied to identify Kudidan, Dangshen, and five medicinal Dysosma species [ 23, 25, 27 ] . Apart from the identification of medicinal materials, RAPD was also used to assess the genetic diversity of wild populations and cultivars [ 28– 30 ] .

3.2.1.3 Amplified Fragment Length Polymorphism (AFLP)

The principle of ampli fi ed fragment length polymorphism (AFLP) is to amplify a subset of DNA restriction fragments from the genomic DNA by restriction enzymes [ 31 ] . The genomic DNA is fi rst digested with restriction enzymes (e.g., Eco RI and Mse I) at various restriction sites in multiple loci to generate restriction fragments with sticky ends. Synthetic adaptors are then ligated to these ends which act as the annealing sites of specific primer for subsequent amplification by PCR under stringent conditions. The amplified fragments are separated by highly resolving polyacrylamide gel and visualized using autoradiography, fluorescence or silverstaining techniques. Similar to AP-PCR and RAPD, AFLP screens multiple loci of the whole genome randomly and simultaneously and does not require prior knowledge of the sequence information. AFLP can detect more loci and generate more polymorphic fragments than RAPD and can be used to differentiate closely related species [ 32 ] . However, DNA degradation in medicinal materials may affect the reproducibility of the polymorphic patterns. Our group used AFLP to differentiate closely related medicinal species such as the Oriental ginseng ( P. ginseng ) and American ginseng (P. quinquefolius) from various localities [ 33, 34 ]. Other examples of using AFLP include the identification of P. japonicus, medicinal Plectranthus species, and Cannabis sativa [ 35– 37 ] .

3.2.1.4 Direct Amplified Length Polymorphism (DALP)

Direct amplified length polymorphism (DALP) is a modi fied AP-PCR fi ngerprinting in which the 5’-end of the forward primer contains the core sequence of a universal primer (e.g., M13 sequencing primer), and thus the resultant fragments can be sequenced directly using the universal primer. DALP is an advanced fingerprinting method which allows simultaneous detection of a large number of polymorphic loci and simpli fi es the recovery and analysis of polymorphic fragments. Our group adopted this method to distinguish Oriental ginseng ( P. ginseng ) and American ginseng (P. quinquefolius) [ 38 ]. A 636 bp polymorphic DALP fragment ampli fi ed using primers DALP001 and DALPR1 was present in P. ginseng but absent in P. quinquefolius. This fragment was sequenced and specific primers were designed to allow rapid identification by amplifying this P. ginseng -specific fragment.

3.2.1.5 Inter-simple Sequence Repeats (ISSR)

Simple sequence repeats (SSR), also known as microsatellites, are tandem repeats of a few base pairs distributed throughout the genome. ISSR fingerprinting is a whole-genome scanning fi ngerprint which uses PCR primers designed based on the repeats found in other species [ 39 ] . As the PCR primers are based on the sequence repeats, such as (CA) n , or with a degenerate 3’-anchor, such as (CA) 8 RG or (AGC) 6 TY, this method does not require prior knowledge of sequence information to generate a large number of resultant fragments. ISSR fingerprinting is easy to use. It is useful to construct genetic maps and to study generic variation within populations of a species. Our group has recently employed ISSR to differentiate Huajuhong (Citri Grandis Exocarpium) derived from Citrus grandis “Tomentosa” from other Citrus varieties and cultivars [ 40 ] . A total of six ISSR primers ((CA) 8 G, (GT) 8 A, (AC) 8 G, (CA) 8 RG, (AC) 8 YT, and BHB(GA) 7 ) were used to reveal the relationship of 23 Citrus samples. The six primers generated 57 bands in which 52 (91.2%) of them were polymorphic across the 23 Citrus samples. Cladistic analysis based on the band polymorphism of the ISSR fi ngerprints showed that the cultivar Citrus grandis “Tomentosa” was clearly distinguished from C. grandis and other Citrus species. ISSR fi ngerprint was also applied to identify Cannabis sativa and Cistanche species [ 41, 42 ]. It was also used to study the genetic diversity of Salvia miltiorrhiza and Vitex rotundifolia [ 43, 44 ] .

3.2.1.6 PCR Restriction Fragment Length Polymorphism (PCR-RFLP)

PCR restriction fragment length polymorphism (PCR-RFLP) amplifies a specific region of the genome followed by restriction digestion to produce restriction polymorphic profiles. The specific region should be readily amplified using universal or specific primers. Standard DNA barcodes with high sequence variation, such as the internal transcribed spacer (ITS), are good candidate regions to start with. Restriction digestion of the ampli fi ed fragment (e.g., H inf I, T aq I and Sau 3A1) generates restriction fragments of different sizes. Mutations creating or disrupting a restriction site are the key to produce polymorphic fingerprints for sample discrimination. Although data interpretation of PCR-RFLP is simple, the discriminating ability of DNA polymorphism is less than that of ISSR and AFLP. Our laboratory successfully applied PCR-RFLP to discriminate various Panax species from the adulterants by amplifying the ITS region followed by restriction digestion using H inf I, T aq I, and Sau 3A1 [4 5 ] .

We also differentiated Dangshen derived from Codonopsis pilosula, C. tangshen , C. modesta, and C. nervosa var. macrantha from their adulterants by digesting the ITS region using Hinf 1 and Hha I [ 46 ] . Similar approach was applied to identify Alisma orientale , Sinopodophyllum hexandrum, and Artemisia species [ 47– 49 ] .

3.2.1.7 Sequence-Characterized Ampli fi cation Region (SCAR)

Sequence-characterized amplification region (SCAR) is a specific region fingerprinting based on the DNA sequences of polymorphic fragments obtained from a whole-genome fingerprint, such as RAPD or ISSR. The polymorphic fragment is cloned and sequenced for designing a pair of specific PCR primers to amplify the concerned polymorphic fragment. The amplification of the polymorphic fragment or the size difference of the fragments in different samples provides a means for differentiating the samples. This technique focuses on a single locus and is usually reproducible under high stringent PCR conditions. To increase the accuracy of differentiation, several SCAR of a sample are analyzed. SCAR requires prior information of the sequence of the polymorphic fragment for specific primer design. Degradation within the DNA fragment and the presence of PCR inhibitors may lead to false-negative results. We found a 25 bp insertion in a RAPD fragment of P. ginseng converted to a SCAR marker for differentiating P. ginseng and P. quinquefolius [ 50 ]. We also applied similar approach to identify medicinal snakes and crocodiles [ 51, 52 ] . Other similar work included the differentiation of Artemisia species, Phyllanthus emblica, and Lycium barbarum [ 53– 55 ] .

3.2.1.8 Isothermal Amplification

Conventional PCR ampli fi es DNA fragments through thermocycles for denaturing of double-strand DNA, annealing of primers, and synthesizing of new strand. Isothermal amplification is a technique allowing DNA amplification without thermocycling, and thus, DNA ampli fi cation can be achieved without PCR machines.

These techniques are mostly applied for on-site detection of viral and bacterial infections in undeveloped regions where laboratory equipment is limited. There are several ways to perform isothermal amplification. For example, strand displacement amplification (SDA) technique starts with an initial step of denaturing DNA template at 95 °C for 4 min followed by a 2 h incubation at 37 °C for primer annealing and DNA ampli fi cation [ 56, 57 ] . The ability of exonuclease-deficient Klenow DNA polymerase to extend the 3’-end and displace the downstream DNA strand leads to exponential amplification as the displaced single-strand DNA serves as the template for the synthesis of complementary strands. Double-strand DNA is digested with restriction enzymes Hinc II at the recognition site in the SDA primers to create nicks, and Klenow DNA polymerase extends the 3’-end and displaces the downstream strand, and therefore, single-strand DNA templates are continuously produced by strand displacement.

Loop-mediated isothermal amplification (LAMP) is another isothermal amplification technique with impressive specificity, efficiency, and rapidity [ 58 ]. Four special primers designed from six alleles (two alleles for the forward and reverse outer primers, respectively, and two alleles for the forward inner primer and two alleles for the reverse inner primer) are used to create “loops” at the end of DNA strands which significantly speed up the process of LAMP, and the whole process can befinished in 1 h. Ampli fi cation progress can be accelerated by additional loop primers to achieve amplification in 30 min [ 21 ]. Recently, LAMP was applied to identify herbal medicinal materials such as differentiating Curcuma longa from C. aromatica based on the trnK gene [ 59 ]. LAMP was also used to discriminate Panax ginseng from P. japonicus based on the 18S rRNA gene [ 21 ] . LAMP is efficient and sensitive when all the primers match the target DNA. However, primer design is dif fi cult because many combinations of primers are needed. The primer sites should be conserved regions with minimum intraspeci fi c variations. DNA degradation in dried or processed materials may give false-negative results. Integrity control of the amplifi ed region may be necessary to prove that negative amplification is independent to DNA degradation.

Helicase-dependent amplification (HDA) is an isothermal amplification technique that unwinds double-strand DNA by helicase in the presence of single-strand DNA-binding proteins [ 60 ]. HDA can be performed with or without an initial denaturation step at 95 °C. Helicase unwinds DNA duplex, and primers anneal to binding sites followed by amplification of complementary strand by DNA polymerase.

The double-strand DNA is separated by helicase, and the chain reaction repeats itself. HDA is relatively easy to set up because primer design is not as complicated as LAMP. This technique has been widely applied to detect virus and bacterial strains, but it has not yet been applied to identify medicinal species. The size of amplicon in HDA is restricted to around 100 bp, which is ideal for samples with degraded DNA content such as the processed medicinal materials. However, validation of DNA integrity of the studying region should be carried out to avoid false-negative results.

3.2.2 DNA Microarray

DNA microarray is a hybridization-based technology using labeled nucleotide probes to hybridize single or multiple loci in a target genome. The probes are short nucleotide fragments obtained either from restriction digestion or synthetic oligonucleotides.

They are fixed on a supporting matrix where hybridization of probes and tested DNA samples takes place. Our group amplified the internal transcribed spacer (ITS) of 16 Dendrobium species and used them as probes to identify medicinal Dendrobium species in a prescription with multiple herbs [ 61 ]. The ITS2 region of the tested samples were labeled with Cy3 fluorescent dye and allowed to hybridize to the ITS probes. Species-speci fi c fl uorescent signal was obtained to clearly identify the five medicinal Dendrobium species. We have also applied similar approach using 5S rDNA intergenic spacer as probes to differentiate D. officinale from other closely related Dendrobium species [ 62 ] .

3.2.3 DNA Sequencing

DNA sequencing is one of the most definitive means for identification as this technique can directly assess sequence variations on a defined locus. It also provides informative characters to reveal phylogenetic relationship. With the decrease of sequencing cost, identi fi cation of medicinal materials using DNA sequencing has become a routine practice. The commonly used DNA regions for medicinal materials identification include nuclear internal transcribed spacer (ITS) and 5S rDNA intergenic spacer (5S), chloroplast trnH-psbA intergenic spacer (trnHpsbA), large subunit of the ribulose-bisphosphate carboxylase (rbcL), maturase K gene (matK), trnL intron (trnL), trnL-trnF intergenic spacer (trnL-F), mitochondrial control region (CR), cytochrome c oxidase subunit 1 (COI), and cytochrome b gene (Cyt b). These regions have different evolutionary rates and therefore possess different variability. For example, the mitochondrial COI region is slowly devolved, and only a few variations were observed in the 1.4 kb COI sequences in fl owering plants [ 63 ] . However, this region evolve rapidly and is varied enough to discriminate most animal species. To differentiate medicinal materials from adulterants derived from closely related species, it is essential to search for DNA regions with high discriminative power. In 2003, the concept of barcoding global species by selected DNA regions was first proposed [ 14 ], and substantial effects have been put on the screening of appropriate DNA barcodes. Until recently, it is generally agreed that the chloroplast rbcL and matK regions are the standard DNA barcodes for higher plants, and the chloroplast trnH-psbA region and nuclear ITS region are the complementary DNA barcodes. For animals and fungi, the mitochondrial COI and nuclear ITS regions are the appropriate DNA barcodes, respectively [ 12, 14, 15 ] . These DNA barcodes have been proven to be useful not only in biodiversity and conservation studies but also in the identification of medicinal materials. For example, our group sequenced the ITS region to differentiate six Panax species from their adulterants derived from Mirabilis jalapa and Phytolacca acinosa [ 45 ] . We also used the ITS region to identify medicinal Dendrobium species [ 64 ], Muxiang (Aucklandiae Radix, Vladimiriae Radix, and Inulae Radix) [ 65 ], Baihuasheshecao (Hedyotii Herba) [ 66 ], Huajuhong (Citri Grandis Exocarpium) [ 40 ], and Leigongteng (Tripterygii Radix et Rhizoma) [ 67 ]. Chloroplast trnH-psbA region is another highly varied DNA barcode for identifying Madouling (Aristolochiae Fructus) [ 20 ] and Wutou (Aconiti Radix and Aconiti Kusnezoffii Radix) [ 68 ]. Apart from the standard DNA barcodes, a few regions are also useful for identifying medicinal materials. For example, the nuclear 5S region was used to identify Dangshen [ 69 ] , medicinal Swertia species [ 70 ], Muxiang [ 65 ] and Leigongteng [ 67 ]. Furthermore, the chloroplast trnL region was used to identify Baibu (Stemonae Radix) [ 71 ], and the trnL-F region was used to identify Madouling [ 20 ] .

3.2.3.1 Forensically Informative Nucleotide Sequencing

DNA sequencing is useful for differentiating groups of materials, but their identities are not known unless their DNA sequences are compared with reference sequences, and this kind of work is called forensically informative nucleotide sequencing (FINS) [ 72 ]. FINS was first applied to trace the origin of animals and their products, but its application has now been extended to the identifi cation of unknown medicinal samples [ 20, 66, 73 ]. To identify an unknown sample, a selected DNA region was ampli fi ed from the DNA extract and sequenced. The FINS approach emphasizes the comparison of the unknown sequence with the sequences of suitable reference species for revealing the identity of a sample. The resolution depends on the discrimination ability of the selected DNA region and the phylogenetic distance between the reference and the unknown sample. With the DNA barcode initiative, many DNA sequences of medicinal materials have now been deposited in public databases. Our group has recently constructed a freely access online database, Medicinal Materials DNA Barcoding Database, which contains approximately 20,000 DNA sequences from 1,300 medicinal materials [ 74 ]. Although the accuracy of many of these sequences has not been substantiated, these sequences are nevertheless valuable reference resources for FINS analysis. Our group applied FINS based on ITS region successfully revealed that four samples of Baihuasheshecao retailed in Hong Kong (PR China) and Boston (USA) were adulterants derived from Hedyotis corymbosa (Rubiaceae), and the three samples from Guangzhou (PR China) are genuine and derived from H. diffusa [ 66 ] . Similar approach was used to reveal the identities of the retailed samples of snake meat, Madouling and Leigongteng [ 20, 67, 75 ].