New research conducted by Integria Healthcare/MediHerb in conjunction with Australian Genome Research Facility, the Medicinal Plant Herbarium at Southern Cross University and the University of Queensland has identified major shortcomings in the use of DNA barcoding to authenticate herbal materials in finished products.
The study, was presented by Dr Hans Wohlmuth in April at the 16th Annual Oxford International Conference on the Science of Botanicals, held at the University of Mississippi in the USA. Questions about the veracity of DNA barcoding have been in the news since early 2015, when the New York Attorney General forced several major retailers to stop selling a variety of dietary supplements containing herbal ingredients on the basis that DNA barcoding had failed to identify the presence of the labelled herbal species or had identified the presence of unlabelled plant species.
The study examined four widely used universal DNA barcodes in 61 samples. The samples, which were also the subject of phytochemical analysis, included 17 dried herb raw materials, 17 MediHerb liquid extracts made from the same batches of raw materials, and 6 MediHerb tablets containing some of the dried liquid extracts. The results showed that while the phytochemical composition (assessed by chemical profile and the presence of active compounds) was preserved in both extracts and tablets, the DNA barcodes were not. DNA barcoding also performed poorly with respect the dried raw materials, with only just over half being correctly identified by at least one of the four barcodes used.
The researchers concluded that the most likely explanation for this is that DNA strands start to break up when plant material is dried and extracted, making it difficult to successfully amplify and sequence the universal barcode regions.
“With this work we have clearly demonstrated that DNA barcoding is not always appropriate for herbal extracts and finished products containing extracts,” said Dr Wohlmuth.
“Further, we have shown that the absence of DNA barcodes does not indicate the absence of phytochemicals and active compounds. Based on these results, we do not believe DNA barcoding using universal barcodes would make a valuable addition to our existing routine quality control program, which is extensive and based on specific morphological and chemical tests prescribed by pharmacopoeial monographs.”
“It is also important to realise that no DNA technique can provide information about which plant part was used or which phytochemicals are present,” added Dr Wohlmuth.
“When it comes to assessing the quality of herbal medicines, the important questions are:
- Is the product derived from the correct species?
- Was the correct plant part used?
- Does it contain the active compounds in meaningful amounts? No DNA based method is able answer the last two, very important questions.”
“I have no doubt that various DNA techniques have the potential to make valuable contributions to the quality assurance of herbal medicines, but our study highlights what most experts agree on, namely that no single method can provide all the answers when it comes to the complex task of authenticating herbal medicines. Clearly, morphological and chemical tests, as prescribed by pharmacopoeial monographs, continue to form the basis for rigorous quality assurance of herbal medicines,” Dr Wohlmuth said.
What is DNA barcoding?
DNA barcoding is a specific DNA technique, first described in 2003, that aims to identify the species from which biological material is derived using a relatively short sequence from a standard part of the genome, i.e. the exact same part of the genome in all species. The term DNA barcoding is sometimes used more loosely to refer to different, more specific DNA approaches; this is unfortunate and confusing.
In animals, DNA barcoding generally works very well, and the universal animal barcode, which is a sequence from the cytochrome c oxidase 1 gene, is being used successfully to identify various animal tissues and products, including fish species entering the food supply chain.
Finding a single universal barcode region that can be used to identify most plant species has proven elusive, and a combination of at least two barcoding regions is necessary to achieve a good identification rate for most plant groups.