The botanical supplement black cohosh surged in popularity about a decade ago. Women suffering through menopause turned to the supplement after questions surfaced about the safety of standard hormone replacement therapy. The plant, Actaea racemosa, is also known as bugbane or snakeroot and is native to North America. It had been used for pain relief and to treat gynecological conditions by Native Americans and 19th-century households, and there had been no reports of harmful side effects.
Yet as more women began using black cohosh in the 21st century, reports surfaced of liver damage in people taking the supplement. In response, government agency Health Canada and a product manufacturer analyzed the contents of four black cohosh products. None actually turned out to contain A. racemosa (Canadian Adverse Reaction Newsletter 2010, 20, 1).
Two studies, the most recent published last year, found Asian Cimicifuga in commercial black cohosh products (J. Agric. Food Chem., DOI: 10.1021/jf0606149; J. AOAC Int., DOI: 10.5740/jaoacint.11-261). This switch could account for the observed liver toxicity, say James Neal-Kababick, director of natural products testing company Flora Research Laboratories, and other experts. Called sheng ma in China and sometimes translated into English as black cohosh or bugbane, the Asian species are used in traditional Chinese medicine but at a lower dose than A. racemosa, Neal-Kababick says.
The tale of black cohosh is a cautionary one for a growing botanicals market. Consumers who see “natural” products as safer than pharmaceuticals drove herbal and botanical supplement sales to $5.3 billion in 2011, up $1 billion in a decade, according to an analysis by New Hope Natural Media, a market research firm.
But botanical products present substantial challenges to both researchers and manufacturers. They are inherently complex substances with a slew of chemical compounds that change with processing. Add in concerns about raw material identity in global supply chains and a regulatory push for better manufacturing practices and quality control, and analysis becomes an increasingly important task. To address the challenges, supplement makers rely on an array of analytical tools to ensure that the product dispensed is what’s on the label. These tools range from simple microscopy to chromatography coupled to tandem mass spectrometry or infrared spectrum profiling coupled to chemometrics.
As defined by the 1994 Dietary Supplement Health & Education Act, a dietary supplement is a product that is not represented as a conventional food and is meant to supplement the diet. Supplements can contain vitamins, minerals, herbs or other botanicals, amino acids, enzymes, organ tissues, or metabolites. They can be concentrates or extracts and can be formulated into various forms, including tablets, gelcaps, powders, or liquids.
The Food & Drug Administration largely regulates dietary supplements as food rather than pharmaceuticals. If a supplement was available as a food ingredient prior to 1994, a manufacturer or distributor does not need to demonstrate safety prior to marketing. FDA limits the health-related claims that manufacturers can make on a supplement label. In 2007, the agency also started requiring that manufacturers follow current Good Manufacturing Practices (cGMP) specific to dietary supplements, including instituting quality-control procedures and testing of ingredients and finished products.
Knowing the identity of the botanical ingredient and how it was prepared is critical for both safety and efficacy. Researchers studying the safety and efficacy of such products typically grow their own plants, standardize how products are prepared, and then profile the chemicals in the product to home in on key compounds for further study. Often those compounds are “obscure” secondary plant metabolites, says Ian Acworth, director of customer and application support for liquid chromatography (LC) products at Thermo Fisher Scientific. That makes botanical supplement testing very different than for foods, where quality control more often focuses on easier-to-measure qualities such as carbohydrate, sodium, or amino acid content, he says.
A particular problem in supplement research historically is inadequate description or analysis of the material investigated, scientists say. That renders reams of studies “just a few notches above worthless,” says Mark Blumenthal, founder and executive director of the American Botanical Council. Whether a research material was grown nearby and extracted in a lab or obtained commercially, different plant species or processing methods can alter the safety or efficacy of a product by eliminating beneficial compounds, retaining toxic ones, or altering the ratios of components that work synergistically.
Consequently, “you can’t reproduce a study if you don’t know exactly what species was used, what part of the plant was used, and how it was prepared,” says Richard B. van Breemen, a professor of medicinal chemistry and pharmacognosy at the University of Illinois, Chicago, and director of a National Institutes of Health-funded botanical research center focusing on supplements for women’s health. Lack of documentation and consequent inconsistency in material preparation are frequently why research into the efficacy of botanical products to treat health disorders tends to show mixed results, he says. That’s one reason that NIH’s National Center for Complementary & Alternative Medicine developed its natural product integrity policy, which outlines required identification, preparation, and characterization information for projects it funds.
Some researchers closely control the plants that they grow. That’s the approach at another NIH-funded botanicals center, a collaborative effort of researchers at Louisiana State University and Rutgers University. The researchers are trying to assess use of botanical products for insulin resistance and metabolic syndrome, a combination of disorders that together increase the risk of cardiovascular disease and diabetes. One of the center’s projects focuses on the use of Russian tarragon, Artemisia dracunculus, to treat these conditions (PLoS One, DOI: 10.1371/journal.pone.0057112). Researchers order seeds from a reputable commercial source, then grow the plants hydroponically in a greenhouse to carefully control parameters such as light, nutrients, and pesticide use, says David M. Ribnicky, a professor of plant biology and pathology at Rutgers.
The Artemisia researchers monitor six bioactive compounds for quality control. They were able to synthesize one marker compound, a chalcone, to use as a standard. “The structure is such that we were able to do that without too much effort,” Ribnicky says. “For some natural compounds, that would be impossible.”
Van Breemen’s Chicago center, in contrast, grows licorice—European Glycyrrhiza glabra and Asian G. uralensis, as well as some rare North American species—at a suburban field station to study cancer prevention and antioxidant effects of the plants’ roots. Several unique compounds in the Glycyrrhiza species distinguish them from other plants as well as from each other, van Breemen says. The flavonoid glabridin, for example, is found in G. glabra but not G. uralensis. The team is isolating and evaluating the compounds for bioactivity.
Van Breemen and his colleagues are also studying whether hops, Humulus lupulus, can address menopause symptoms and prevent cancer. They get hop cones from the brewing industry, which van Breemen says “is very well organized and follows good agricultural practices,” including the ability to trace crop varieties and originating farms. The team analyzes estrogenic prenylflavonoid compounds in hops for quality control and in bioassays. They recently developed an ultraperformance liquid chromatography-tandem mass spectrometry (UPLC-MS/MS) method to detect the compounds in human serum, using 8-isopentylnaringenin as an internal standard.
Other researchers and manufacturers may have less certain supply chains, especially manufacturers that gather ingredients around the world. The first order of business in these cases is to be precise when ordering the plant and parts desired, in particular by using Latin rather than common names. “It’s a problem when people are not specific enough with their suppliers,” says Sidney Sudberg, president of Alkemists Laboratories. “One might ask for ginseng but not Panax ginseng and instead get the American Panax quinquefolius.”
Botanical ingredient purchasers also have to watch for contamination, whether it’s from a harvester accidentally collecting weeds or mistakenly substituting similar-looking plants, or someone intentionally adulterating supply with a less expensive material. A supply of whole or cut botanical material is relatively easily identified, either macro- or microscopically, but employing a well-trained botanical analyst is advisable, says Flora Labs’ Neal-Kababick. He points to bear’s garlic, Allium ursinum, as an example. The plant is popular in Europe for indigestion or high blood pressure, but the leaves are easily confused with those of toxic lily of the valley, Convallaria majalis.
The real identification challenge, however, comes when plant material gets processed into powders or extracts. Grinding may destroy telltale cell structures that could be identified by microscope, making chemical analysis critical. In these cases, it is also essential that manufacturers do statistically rigorous sampling. A farmer may sell to a botanical broker, who then powders and bags the material. A lot might simply be everything that broker can process in a day, done in the order it arrives. “Maybe the guy at the first farm is good at controlling invasive weeds, while the guy at the last farm is not doing any abatement and one-third of his crop is foreign matter,” Neal-Kababick says. Depending on how the lot is sampled, the content could look entirely different.
The concerns for powders also hold true for extractions but with an added twist: Different extraction processes—in particular, those using different solvents—can yield products with different chemical compositions. One example is cinnamon, which is used for bronchitis or gastrointestinal problems. True cinnamon, Cinnamomum zeylanicum, is often substituted with cassia bark, C. cassia or Chinese cinnamon. The two can be distinguished by comparing volatile oil profiles using thin-layer chromatography or gas chromatography/mass spectrometry (GC/MS). “But there are products out there that are water extracts, so they have no oil in them,” Neal-Kababick says. One of his clients uses IR spectroscopy to profile organic acids and sugars instead.
And when products contain several botanical ingredients blended together, the analytical challenges increase exponentially. “cGMP requires tests to confirm the presence and identity of all those materials,” Neal-Kababick says. Overall, companies need to take a multipronged testing approach, identifying specific markers to track desired ingredients but also using assays that will pick up unwanted material.
Considerable effort in recent years has gone into developing and validating analytical methods and reference materials and making them publicly available to promote botanical identification and quality control. “Everybody used to hold their cards close,” Neal-Kababick says. “There was no sharing between companies or consensus about solving problems.”
Part of the push to get methods out in the open came from the NIH Office of Dietary Supplements, which established its Analytical Methods & Reference Materials program in 2002. AMRM supports method and reference standard development in collaboration with FDA, the National Institute of Standards & Technology, the Department of Agriculture, and AOAC International. Program Director Joseph M. Betz says he sees a critical need for methods to establish botanical identity in extracts.
Past botanicals targeted for AMRM development included Ginkgo biloba, used for memory; goldenseal, Hydrastis canadensis, used for infections; soy, Glycine max, used for a variety of health conditions; and Ephedra sinica, used for weight loss and energy but banned in the U.S. since 2004. Even where analytical methods existed, Betz’s team had to determine whether they were suitable for the intended use and could be validated. It was particularly challenging, Betz recalls, to develop a method to measure Ephedra as a raw material as well as in extracts, tablets, mixed supplements, and drink mixes. The six principal Ephedra alkaloids fall into three different classes—primary, secondary, and tertiary amines—and also occur across a broad concentration range. Eventually a team from Covance worked out an LC-MS/MS method (J. AOAC Int.2003,86, 471).
Putting analytical methods into the public domain is also part of the mandate of an FDA center for botanical dietary supplement research at the University of Mississippi, says center director Ikhlas A. Khan. Khan is a professor of pharmacognosy, assistant director of the university’s National Center for Natural Products Research, and a collaborator in the NIH-funded Botanical Estrogen Research Center based at the University of Illinois, Urbana-Champaign. Khan and his colleagues are trying to define chemical fingerprints to identify plant species. Recent work demonstrated the use of UPLC with ultraviolet and MS detection to evaluate samples of yohimbe bark (J. Nat. Med., DOI: 10.1007/s11418-012-0642-2). Yohimbe, Pausinystalia yohimbe, is an African evergreen tree; the bark is used for sexual dysfunction, particularly erectile dysfunction in men.
The U.S. Pharmacopeial Convention (USP), a group that sets analytical testing and release standards for pharmaceuticals, also has testing protocols and reference standards that manufacturers may use voluntarily for dietary supplements. USP’s Dietary Supplements Compendium now includes not just monographs outlining tests for individual botanical materials but also supporting information such as photographs, micrographs, and chromatograms to help support companies and quality-control analysts as they ramp up efforts to follow cGMP regulations.
If there’s one underlying principle to botanical analysis, it’s to use separation science, says Gabriel Giancaspro, USP’s vice president of foods, dietary supplements, and herbal medicines. “To obtain a true characterization of a material, both for identity and for content, we want to make sure that we can separate as many components as possible,” he says.
And if there’s one staple technique for botanical quality control, it’s high-performance thin-layer chromatography (HPTLC). With this method, researchers can quickly run multiple samples in parallel for easy comparison. Improved plate materials, standardized methodology, and environmental control chambers all make the method reproducible and easy to validate. One plate can also be subjected to multiple types of detection, such as UV light, fluorescence, and MS, says Eike Reich, director of Camag’s method research and development lab. Scientists can also add enzymes or microorganisms to a plate to check for biological activity. The International Association for the Advancement of High Performance Thin Layer Chromatography also develops and validates standards and methods for identification of plants and frequent adulterants.
Other common techniques include high-performance LC (HPLC), with UPLC methods starting to become more frequent along with capillary electrophoresis. Thermo Fisher’s Acworth likes to combine HPLC with charged aerosol or coulometric electrochemical detectors. Both work particularly well for complex products and yield detailed compound fingerprint profiles for comparison to reference materials. Acworth also sees companies using GC or GC/MS techniques to look for pesticide residue and inductively coupled plasma MS to check for heavy-metal contamination. “More and more we’re seeing demand for higher-end equipment,” he says.
IR spectroscopy can also be a useful tool for fingerprinting in a quality-control setting, even for a complex material. Now Foods worked with PerkinElmer to develop an approach that works on powders in high-throughput settings. The technique works well to screen for economically motivated adulteration when ingredients or products have relatively high amounts of unwanted material, says Jerry Sellors, IR product planning manager for PerkinElmer.
Regardless of the method used, the key with any kind of botanical testing is setting parameters. The same species grown in two different fields or harvested at different times can vary chemically because of the vagaries of nature and growing conditions. That makes developing reference standards particularly tricky and also makes chemometric models for data analysis essential. “The variation that these models should span is quite important,” Sellors says. “It’s necessary to carefully select your samples to build the model and the right mathematical parameters to ensure that you’ve got just the right specificity for whatever it is you want to screen.” Make the parameters too tight and good lots might fail, but make them too loose and a manufacturer could miss a new adulteration scheme.
Development and dissemination of analytical techniques such as these should help to get the botanical supplements field onto more rigorous scientific footing. The effort hasn’t been easy so far, Alkemists Labs’ Sudberg says. Nevertheless, he’s optimistic that people are coming around, and in the end consumers will benefit from improved products.