Roy Upton explores if, and how, we can integrate the knowledge from traditional herbal medicine with modern science.
In most countries, traditional health practitioners such as in Ayurveda, Naturopathy, and TCM are required to be trained in Western anatomy, physiology, and pathology. While many Western medical schools have programs introducing students to traditional healing systems, they are presented as overviews, not as fully developed medical systems, but rather as non-evidenced-based therapies to which patients may be exposed. While traditional health care practitioners benefit greatly from the integrated traditional-western knowledge and integrate this into patient care to greater or lesser degrees, most western physicians cannot provide any guidance regarding traditional practices except to recommend they not be used.
Similarly, applying western standards of either chemical or pharmacological assessment to traditional herbal medicines can inform and increase our knowledge base of the botanical. However, it should be remembered that one of the reasons traditional herbal medicine is growing in popularity despite ready access to modern medicine is due to the failings of modern medicine. Traditional healing systems offer a different paradigm of understanding health and disease that increases the therapeutic options to those in need. Thus, attempting to regulate or restrict traditional herbal healing systems because they do not fit into the typical western paradigm is an antithesis to the healing system itself and a great detriment to patients. Froma traditional herbal medicine perspective, preservation, development, and acceptance of morphological and organoleptic assessment skills are critical in preserving traditional medicine practices, while the analytical tools of modern pharmacognosy enriches the knowledge base.
Regarding traditional medicine practice, there is some protection of traditional herbal medicine principles internationally. In most countries that regulate herbal medicines (e.g., much of Asia, the European Union, and India), herbal medicine practitioners are exempt from standard manufacturing GMPs. This is not the case in the US, where FDA maintains the authority to require practitioners to be in full compliance with standard dietary supplement manufacturing GMPs, an authority that threatens the perpetuation of traditional herbal healing in that country. As recognized by the WHO and European Union, the whole crude plant part is considered to be the active ingredient, not a specific amount of a compound(s) that can be quantified. Macroscopic and sensory evaluation of herbal ingredients, followed by therapeutic experience, are the sole means by which traditional herbal practitioners can continue practicing their traditional healing systems. Not recognizing traditional herbal assessment principles in formal pharmacopoeia limits the expression and evolution of traditional herbal medicine, and instead, pushes herbal medicines solely into a western pharmaceutical paradigm.
Histology and identification (microscopic characterization)
Prior to the advent of modern analytical chemistry, microscopic examination of crude herbal drugs, along with gross morphological and organoleptic assessment as described above, were the primary tools used for crude herb assessment.
European and American pharmacognosists put tremendous emphasis on the ability to identify plants to species microscopically, including those that that are closely related, detect adulterations, and, in some cases, even assess relative quality. According to noted American pharmacognosist Henry Kraemer (1908);
“The microscope furnishes the surest means of determining the identity of a powdered drug at our command…the microscope also furnishes the most reliable means for detecting and determining adulterants in powdered drugs… [and] detecting the presence of worm-eaten drugs or powders of certain classes of drugs which have been exhausted in whole or in part…”
Kraemer went so far as to say that even the time of gathering, method of drying, and length of time for which a botanical had been stored could “be judged in many instances by the use of the microscope”. Additionally, there is great sensitivity in being able to detect adulterating species. For example, many years ago there were numerous reports of the Chinese herb stephania (Stephania tetrandra) being mixed up with a different species of plant (Aristolochia fangchi), which contains the renal toxin and potential carcinogen aristolochic acid (AA). This adulteration was responsible for from several hundred to a few thousand deaths. With microscopic examination, the addition of as little as 0.3% of Aristolochia fangchi in a stephania sample can be detected by observation of the differing oxalate crystals that occur within the species (Upton 2006). Discerning crude drug species microscopically reached a very high level of refinement and is reflected in numerous seminal texts on the subject in the middle 19–20th centuries, most notably Berg (1865) and Meyer (1892) of Germany; Moeller of Austria (1890); Tschirch and Oesterle of Switzerland (1900); and microscopists of the UK and US such as Sayre (1917), Mansfield (1937), Greenish (1904), Kraemer (1920), and Youngken (1930), to name a few.
With advancements and evolution of analytical chemical techniques, first with paper chromatography followed by thin-layer chromatography (TLC) and then more quantitative techniques (liquid chromatography [LC], gas chromatography [GC], and finally molecular technologies (e.g., DNA), microscopy, like morphology and sensory assessment, as a scientifically valid analytical tool was regarded as outdated in the face of these more recent techniques. However, seldom is one analytical technique inherently superior to another. The superiority or applicability of one analytical method over another is dependent upon the desired analytical endpoint. From this perspective, microscopy is as scientifically valid of an analytical tool as any other technique and is currently enjoying a resurgence in the US where it had fallen into almost complete neglect in previous decades.
Chromatographic technologies: Quantitation and pattern recognition
The first of the chemical assessment techniques in most pharmacopoeias, Eastern and Western, is thin layer chromatography (TLC), and more recently, high performance TLC (HPTLC), and is primarily used as an identification assay.TLC and HPTLC analysis provide a snapshot of the constituent profile of crude plant drugs and has been a standard entry in pharmacopoeias for decades. In previous decades, TLC as an analytical technique was very crude, had a limited degree of reproducibility, was cumbersome, and not very GMP-compliant. In more recent decades, the introduction of HPTLC has addressed many of these deficiencies. As a chemical analytical tool, HPTLC is extremely versatile and sensitive for the identification of many crude herbal drugs and is especially useful for the detection of adulterations, often with a very high degree of sensitivity.
However, it must be recognized that any chemical testing is a surrogate for identification as, in contrast to identifying the actual plant material botanically, morphologically, organoleptically, or microscopically; it is the chemical profile that is being identified not the plant. Closely related species of plants may be very similar chemically. Additionally, the constituent profile of the plant can vary dramatically across the range of commercial materials that may be traded, variations occurring due to growing, harvesting or post-harvesting conditions, age of the plant, or extracting parameters. Identification of plants chemically is therefore limited and is most accurate when coupled with appropriate physical tests. For compliance with pharmacopoeia monographs, TLC/HPTLC or other quantitative assays are conducted in tandem with the physical tests outlined above.
Critical to any chemical testing methodology of medicinal plants is to look at the suite of constituents from the perspective of chromatographic fingerprinting and not only individual markers. The concept of chromatographic fingerprinting was developed primarily by Chinese phytochemists who recognized that the activity of a Chinese herb is in its collection of compounds not just the single active constituent approach typical of modern drugs. Numerous papers have been published on the subject and provide guidance on appropriate ways to apply modern chemistry to the analysis of traditional herbal drugs (Fan et al. 2006; Liang et al. 2004; Xie et al. 2006, among others).
A critical starting point to chromatographic fingerprinting is to obtain multiple samples of the desired botanical that is grown, harvested, and dried in a manner that is optimal for the plant part. Such criteria are traditionally determined organoleptically by evaluating the color, smell, taste, texture, relative purity, and other physical characteristics of the plant. Optimal picking times can be informed by traditional literature, such as harvesting the flowering tops of St. John’s wort on the eve of St. John’s day (June 24). While the specificity of a single day may not be required, in most growing areas, St. John’s day reflects a harvest time when the plant is in full bloom, the traditional time to harvest most flowers; roots and barks are typically harvested in the Spring or Fall; leaves, prior to flowering before the energy of the plant goes to flower, seed, and fruit production. In Chinese herbal medicine, specific herbs must be gathered at a specific age, such as the roots of dang gui (Angelica sinensis) at a minimum of four years of age; or Chinese ginseng (Panax ginseng) at a minimum of five years of age. Optimal harvest times can be greatly informed by chemical analysis. Specifically regarding St. John’s wort, the highest concentration of the suite of St. John’s wort constituents including flavonoids and napthodinathrones are yielded during the budding and flowering stage (during blooming), the concentrations dropping dramatically after flowering and being higher in upper leaves than lower leaves (Tsitsina 1969).
Another critical aspect in developing chromatographic fingerprinting methodologies for medicinal plants is to obtain samples of closely related and potentially adulterating species that may be inadvertently traded for the authentic material. Having multiple species of the same appropriate quality material, ideally from multiple regions for multiple years, allows the analyst to observe the variation inherent in natural products due to changing environmental conditions. Having closely related or adulterating species helps the analyst know if the chromatographic fingerprint is robust enough to identify the authentic from substitute species. In recent decades, chromatographic fingerprinting techniques have been coupled with chemometric programs (Mok and Chau 2006, among others). These programs are designed to store and analyze multiple datapoints from multiple samples and provide statistically relevant comparative fingerprints that allow for developing relatively objective criteria for what constitutes an acceptable chromatographic fingerprint. Ideally, such determinations should be made based on the available clinical data that establishes the efficacy of the herbal drug being analyzed; without a clinically relevant correlation, even the chemical fingerprint is only a surrogate marker of quality.
It is important to recognize that much of the basic work in developing appropriate chromatographic fingerprints is best done in academic settings and is a very expensive and time-consuming process that does not allow for the high throughput testing often required by industry. Once done in academia, such testing criteria can then be codified into pharmacopoeia as a way to promote a consistent standard.
Quantitive assays
In addition to identity tests, most Western pharmacopoeial monographs require quantitative assays, many of which will correlate directly with the pharmacopoeial definition of the botanical drug ingredient and may or may not directly correlate with activity. Quantitative assays are often lacking in Eastern pharmacopoeias (e.g., Ayurvedic Pharmacopoeia; PPRC). In most cases, quantitative assays will quantify a specific compound that is correlated with a known active constituent or class of constituents. Other compounds assayed are considered surrogate marker compounds for activity, other chosen compounds may reflect a constituent or class of constituents that provide a baseline of “quality” based on harvest or processing practices, which may or may not correlate with activity, and yet others will reflect other measures of “quality”, such as an organoleptic bitterness value for a particular botanical such as gentian root (Gentiana spp.) or a swelling index as discussed previously regarding mucilage content of slippery elm bark or flax seed.
As noted, very seldom is a single constituent correlated with the total activity of a particular botanical. Rather, the total extract or profile of the crude drug is considered the active ingredient or substance. This is reflected in WHO documents (WHO 1991) and acknowledged by renowned medicinal plant researchers (e.g. see Mukherhee 2011; Wagner et al. 2009; Xie et al. 2013). For example, valerian root (Valeriana officinalis) is a sedative herb used at least since the 1st century (Pickering 1879). The essential oil was long considered to represent the active fraction. However, early pharmacological work demonstrated that the essential oil fraction was only associated with approximately 1/3 of the total activity of the extract (Gstirmer and Kind 1951). Later, research demonstrated depressant activity of the dichloromethane extract but activity could not be attributed to either the valepotriates, valerenic acid, valeranone, or the essential oil (Krieglstein and Grusla 1988). To date, the total activity of valerian has not been fully articulated though its efficacy in reducing the time required to fall asleep and to stay asleep is supported. Today, Western pharmacopoeias require quantitation of the essential oil fraction of valerian as the primary marker of quality. St. John’s wort (Hypericum perforatum) provides another example of a plant whose efficacy as an antidepressant has been documented over a period of more than 2000 years, but for which the pharmacological mechanisms in humans have yet to be determined (Cott 2005). Thus, quantification of a single marker or group of markers should be viewed as representing only a portion of the activity of the whole plant or extract made therefrom.
Regarding quantitative assays, varying pharmacopoeias take different approaches in determining what compound(s) are to be assayed. For example, the pharmacopoeias of Germany and the US tend towards specificity of compounds favoring relatively sophisticated quantitative technologies such as HPLC in contrast to more generic technologies such as spectrophotometry. Whereas HPLC has greater accuracy in quantifying individual compounds, spectrophotometry groups like compounds together, often resulting in higher quantitative values as compared to HPLC. Both technologies have strengths and weaknesses. HPLC offers greater accuracy in quantitation; this is its primary advantage. If used appropriately, HPLC can also be used for purposes of identification if the total chromatographic fingerprint of the plant is considered and thus has a greater chance of detecting adulterants. Spectrophotometric techniques, in grouping like compounds, may give a more accurate representation of the totality of the plant or extract, but has the distinct disadvantage of being easily fooled by purposeful admixtures of adulterants. In the US, there are no requirements to follow any monograph standards for herbal supplements. Analysts and botanical dietary supplement manufacturers often attempt to perform a single test for all or most identity and quality testing, often times choosing the most economically feasible assay, but not always the most appropriate. Such testing may or may not provide scientifically valid results for the desired analytical endpoint. In the EU, where there is a legal requirement to follow all aspects of the European Pharmacopoeia monographs, there appears to be a greater use of spectrophotometric methods, which are only employed after appropriate identity of the botanical ingredient has been confirmed. In the USP, there appears to be greater utilization of specific methods, requiring multiple analytes, greater time, and greater expense.
Another limitations of current pharmacopoeial works include the fact that many perceived independent standards are based on single proprietary products. In such cases, a sponsor submits analytical data, often of a proprietary extract, and the characterization of that proprietary extract becomes the standard for the botanical generically, but may not be widely applicable to generic products creating a false sense of independent verification, but also codification of a commercial proprietary product as a generic standard providing an unfair advantage to the submitting manufacturer as all competitors then must either submit their own data for consideration of revision of the monograph (a lengthy process) or be encouraged to follow the standards of the their competitor, which they may not be able to do due to the proprietary nature of the original extract used. This is more of an issue in countries where the process is driven more by commercialism than science.
Some Chinese researchers propose a tiered approach to traditional Chinese herbal drug quality control development, recommending “elementary, intensive, and advanced” levels of testing. The “elementary” level reflects current standard requirements of pharmacopoeias that focus on basic identity, qualitative tests, and analysis of a single marker compound. The “intensive” level includes multiple component analysis, including differentiation between “high” and “low” quality materials based on quantitative assays of a particular compound(s). The “advanced” level proposes that the multiple component herbal medicine be subjected to formal pharmacological study but acknowledges the inherent lack of Western pharmacological models for assessing traditional Chinese actions of herbs such as to “expel wind” and “nourish kidney” (Xie and Li 2007). Similarly, it has been proposed that the comparison of a crude herbal drug to a properly made well characterized extract is a more appropriate way to assess the quality of a Chinese herbal drug than to simply assay a single constituent (Xie et al. 2013). While these represent more appropriate ways to ensure the integrity of multi-component ingredients and formulas than the typical Western pharmacopoeial approach, they continue to depend on chemical analysis versus traditional herbal assessment skills and therefore, do not take into consideration the full suite of traditional assessment techniques.
Other qualitative parameters not considered in pharmacopoeias
Botanical identification
Botanical identification remains the primary technique for the identification of plants worldwide. While chemotaxonomy and DNA analysis have made great strides in altering the classifications of traditional taxonomy, neither technology has supplanted botany as the primary means of identifying medicinal plants. However, in virtually all official pharmacopoeias, botanical identification is not included as a required test with which to conform. The reason for this is that oftentimes the botanically unique characters of a specific plant are not intact in the crude plant drug.
Historically, different species of a specific genus of plant, and sometimes, different genera of plant, were used as a specific botanical drug. Rather than being referred to by its generic and specific name e.g., Salix purpurea, medicinal plants were described according to their Galenic names e.g., Cortex salicis, the Galenic name referring to the bark of varieties of willow that were historically used interchangeably without differentiation of species. Such a practice is represented in most pharmacopoeias but is increasingly moving towards greater speciation. For example, in the European Pharmacopoeia, the bark of several species of Salix is allowed as long as the barks conform to the identity tests given and possess a minimum quantity of salicin (1.5%), the primary putative active constituent, and precursor to the development of acetyl salicylic acid (aspirin).
Conversely, in the US, where herbs are predominantly traded as ‘dietary supplements’ not ‘medicines’, for purposes of supplement labeling, the names of botanical ingredients must conform to Herbs of Commerce of the American Herbal Products Association (AHPA). The current edition of this text, rather than recommending use of Galenic or more common names (e.g. “willow bark”), mostly requires speciation such as crack willow (Salix fragilis), purple willow (Salix purpurea), white willow (Salix alba), etc. Unfortunately, while conformity to the identification tests of pharmacopoeias for these barks can be assured, most of these barks cannot be identified to species when in their crude form as they are morphologically, organoleptically, chemically, and medicinally very similar, the very reason why the multiple species are used interchangeably. Thus, greater levels of specificity are often required for either academic or arbitrary regulatory reasons creating unnecessary impediments, to continued use of traditional herbal medicines.
Daodi (region specificity)
Daodi is a philosophy that is exclusively applied to the development of medicinal substances. Dao is an ancient Chinese unit of measure, which became applied to the development of districts. Di means earth or land and is applied to specific geographical regions. According to Zhao et al. (2012), daodi medicinal materials are generally defined as medicinal materials produced in specific geographic regions under natural ecological and environmental conditions that are optimum for the growing of the plant, with particular attention to cultivation, harvesting, and processing techniques. Daodi lead to quality and clinical effects that are considered to surpass those of the same botanical produced in a different region. The concept of daodi was used by early Chinese herbalists and codified by Sun Simiao in his Qian Jin Yi Fang (Supplement to the Formulas of a Thousand Gold Worth). According to Sun:
“When ancient doctors used medicinals they depended on the earth, therefore when the treated ten people they achieved results in nine. Although contemporary doctors understand the pulse and prescriptions, they discard the timing for harvesting medicinals. They are not familiar with the originating land or the freshness, aged nature, emptiness or fullness; therefore they only achieve results in five or six cases out of ten.”
Sun’s teachings underscore the importance of daodi, organoleptic assessment, and a quality control system that integrates the proper harvest time and source of the originating plant material. Furthermore, in the Ben Cao Pin Hui Jing Yao (Essentials of Chinese Materia Medica) of China’s Liu Wentai (1488–1505), specific attention was given to medicinal plant development that encompassed the sprouting of the plant, the land on which it was grown, the timing of planting and harvest, and potential substitutes. Thus, prior to the development of formalized good manufacturing practices and prior to the adherence to mandatory pharmacopoeial standards, Chinese physicians recognized the need for strict adherence to quality control in ways that ensured the quality of the medicinal plant from the field to the medicine long before the advent of chemical analytical techniques. This care is reflected in the many grades of herbs that can be observed in any Chinese herbal pharmacy today.
Such principles reflect common sense in ensuring the quality of herbal drugs. Similar principles are being applied internationally and in a myriad of Good Agriculture and Collection Practices (GACP) developed by varying national and international bodies. The People’s Republic of China (PPRC) is perhaps most active in establishing good agricultural practices (GAP) for medicinal herbs and integrating them in medicinal plant production.
Similar to Sun’s observation in the Tang Dynasty, modern herbal practitioners are often not aware of the quality control needs of herbal medicines, and rather rely on industry to produce quality medicines, when much of the industry lacks the skills necessary to do so. Daodi is an important concept to include in the development and selection of herbal medicines and there may be wisdom in codifying such principles in modern pharmacopoeias. According to Zhao et al. (2012), among the 500 most commonly used Chinese medicinal materials, approximately 200 are recognized as having daodi medicinal material specifications. These 200 medicinal materials account for 80% of the total consumption of medicinal materials in China, making daodi a very important concept in modern herbal medicine GMPs. Today, the source and manner in which the botanical was grown and harvested is seldom considered by industry and regulators.
Molecular identification
In animals, a single barcoding region from the mitochondrial cytochrome c oxidase-1 gene can identify most species. In plants, no single barcoding region has been identified with sufficient resolution to identify most species, therefore a combination of at least two regions (barcodes) is required and use of other techniques in tandem provide greater levels of accuracy. Molecular techniques of medicinal plant identification are not currently included in most pharmacopoeia but some attempts have been made to consider their inclusion.
Like all analytical technologies, molecular analysis has strengths and weaknesses in crude plant identification. First and foremost it is solely a identification tool. Whereas most all other methods provide both identity and quality assessments, molecular techniques is only for species identification. Among the technique’s strengths are its great sensitivity. If the unique identifying primers for the particular target species has been appropriately developed, and the technique used is robust enough to filter DNA from non-target organisms (e.g., other plants, non-target plant parts, etc.), then it is an exceptionally powerful tool for identifying plant material to species and beyond, including the same species that have different genetic lineage, such as wild American ginseng (Panax quinquefolium) grown in Kentucky, versus the same species from cultivated root stock in Wisconsin, versus that grown in Canada. A second advantage of molecular techniques is in detecting admixtures with other species with great sensitivity. A logistical advantage of the technique is that it uses very small amounts of material for analysis.
Among its weaknesses is also the technique’s great sensitivity. Molecular techniques have the ability to detect the presence of trace of insignificant amounts of non-target species, such as rice and pollen or a blade of grass, the results of which can be interpreted as ‘contamination’. In other cases, genetic material of a non-target species can be preferentially amplified over the genetic material of the target species, that can be inaccurately interpreted as an ‘adulteration’. A significant disadvantage is its inability to discern different plant parts, a requirement of all medicinal plant GMPs. As with chemical analysis, DNA analysis will not detect the presence of non-organic contaminants that are readily detected organoleptically or microscopically.
Perhaps the greatest strength of genetic testing is in forensic analyses of plant material that is difficult to identify with standard testing methodologies or to detect the presence of difficult to find adulterations and contaminations. Perhaps its greatest weakness, unlike virtually all other standard techniques from sensory assessment to chemistry, is its inability to offer any meaningful data regarding quality. It is purely an identification tool that is limited in its ability to discern plant parts, that works best in intact plant material, where most other techniques provide data points for both identification and quality of a crude plant or extract. However, its utility in routine quality control is limited as was poignantly highlighted by researchers in Australia who applied various methods of genetic barcoding to various stages of the manufacture of medicinal plant products. The researchers attempted to determine if DNA barcoding was a useful routine test for the quality control of botanicals and genetically analyzed the ingredients throughout the manufacturing process from gathering of botanical vouchers to drying, to pulverizing, to extracting, to preparation of the finished product. Even for vouchered specimens, a high of only 40% of specimens were able to be identified with DNA using one set of markers and a low of about 18% with another set of markers, determining that substantial degradation of DNA occurred in properly dried materials.
