Adulteration of Botanical Raw Materials in India, Part 1: An Estimation of Herbal Product Adulteration Based on Identity Determination of 93 Commercial Materials Using DNA Barcoding

 

Reviewed: Shanmughanandhan D, Ragupathy S, Newmaster SG, Mohanasundaram S, Sathishkumar R. Estimating herbal product authentication and adulteration in India using a vouchered, DNA-based biological reference material library. Drug Safety. 2016;39(12):1211-1227.

 

Keywords: Medicinal plant, India, adulteration, DNA barcoding

 

This project was set up to develop a reference material DNA library for plants used in herbal medicine in India using a DNA barcode approach and to assess herbal product adulteration in the marketplace. In addition, the sequencing results obtained during the investigation were compared with sequence data deposited on the widely used database GenBank.

 

A total of 187 plant species were identified by an expert based on morphological features, and sequenced using the chloroplast gene region rbcL and the nuclear internal transcribed spacer region ITS2 to build the DNA barcode library. The gene region rbcL was chosen because it is easily amplified and aligned, and because the abundance of available sequence data, while ITS2 provides high species resolution and is a much shorter sequence, allowing a higher DNA recovery from processed plant materials. Cycle sequencing was used to determine the DNA sequences of the materials.

 

Commercial samples included 89 single-ingredient and four herbal mixtures representing 50 different plant species obtained from local markets throughout India. According to the authors, most products were in form of capsules or fine powders, and none of the products contained an extract. The commercial products included popular herbs like neem (Azadirachta indica, Meliaceae), turmeric (Curcuma longa, Zingiberaceae), licorice (Glycyrrhiza glabra, Fabaceae), ashwagandha (Withania somnifera, Solanaceae), or tribulus (Tribulus terrestris, Zygophyllaceae).

 

Barcode recovery from individual gene regions of the commercial samples was 68% and 90% for ITS2 and rbcL, respectively. For 11 samples, no DNA of sufficient quality could be obtained. According to the authors, nine samples (10%) were adulterated, meaning that the samples did not provide sequences of the labeled species. However, a closer look at the results suggests that only seven of these samples (7.5%) did not contain the labeled species (Table 1). Another 50% included plant material in addition to the labeled species, considered as “contaminated” samples by the authors, although the extent of the contamination was not determined. Interestingly, fenugreek (Trigonella foenum-graecum, Fabaceae) was found in 14% of all commercial products.

 

Table 1: List of adulterated plant materials and adulterating species

 

Labeled species	Scientific name	Number of samples adulterated	Adulterant(s)
Aloe vera	Aloe vera (Xanthorrhoeaceae)	1	Chelidonium majus (Papaveraceae)
Bael tree	Aegle marmeols (Rutaceae)	1	Tectona grandis (Lamiaceae) Medicago sativa (Fabaceae)
Chebulic myrobalon	Terminalia chebula (Combretaceae)	1	Quercus spp. (Fagaceae)
Gotu kola	Centella asiatica (Apiaceae)	1	Merremia spp. (Convolvulaceae)
Guggul	Commiphora mukul (Burseraceae)	1	Vitex negundo (Lamiaceae)
Guggul	Commiphora wightii (Burseraceae)	2	Tectona grandis Indigofera australis (Fabaceae) Bauhinia spp. (Fabaceae) Trigonella foenum-graecum

 

Overall, DNA sequences from herbal samples in the study matched the author’s library in 100% of the cases, while only 64% matched the GenBank data, prompting the authors to comment that “using an uncurated and vouchered DNA library such as GenBank is not recommended as it will likely result in some misidentifications of herbal species that are not well-represented within GenBank.” [emphasis ours]

 

Comment: The authors have presented an alarming picture of the herbal trade in India, with 10% of the analyzed samples allegedly adulterated, and 50% of samples contaminated. The self-made definitions of substitution, contamination, and use of fillers are not consistently used throughout the manuscript and in the manuscript’s Table 1 where each of the samples is listed; thus, the reader of this paper is left wondering just exactly how the authors came up with the numbers of adulterated and contaminated samples. Since the amount of “contaminant” was not determined, it is not possible to know if the samples were indeed of low quality, or if the findings represent acceptable levels of foreign organic matter. United States Pharmacopeia, for example, generally allows 2% of “foreign organic matter” in a sample of botanical raw material.¹

 

Also, the results need to be put into context. Many of the herbal products were purchased as bulk material or powder at local stores in India, where piles of herbs are exposed next to each other, so accidental mixing and contamination can easily happen. As an example, guggul (C. mukul) was found to be adulterated with Chinese chaste tree (Vitex negundo). Chinese chaste tree is a widely used medicinal plant. It may have been placed next to the guggul in the store where it was purchased, and small amounts may have been mixed with the guggul. It is also worthy of note that some vernacular plant names in India are used for more than one species; so, it is possible that this may have led to some confusion with regard to the requested plant.

 

The accuracy of results when DNA barcoding is used to identify tree gums is questionable. According to experts in the genetic identification of plant materials, the extraction of DNA from gums, which contain substantial amounts of terpenes but little cell material, is challenging. In addition, plant resins are sticky, and finding pollen or fragments from other plants growing nearby (or transported by insects) on/in the resin is common. Therefore, the reported DNA barcoding method may not have been the optimal way to determine the authenticity of these materials, and the inclusion of chemical methods might have been helpful to confirm the results. An issue of the experimental section is the use of Sanger sequencing, which provides only one sequence and therefore is a best used for clean samples. However, sequences derived from DNA mixtures (=mixed PCR products) result in overlaying nucleotide peaks, which makes accurate determination of a species difficult. As such, the results of this paper should be interpreted with caution.

 

Reference

1.      Foreign Organic Matter. In: USP 39-NF 24, General Chapters: <561> Articles of Botanical Origin. Rockville, MD: United States Pharmacopeial Convention; 2016.