DNA barcoding to trace Medicinal and Aromatic Plants from the field to the food supplement

Abstract

The global market of food supplements is growing, along with consumers demand for high-quality herbal products. Nevertheless,
substitution fraud, and adulteration cases remain a common safety
problem of global concern. In the last years, the DNA barcoding approach has been proposed as a valid identification method and it is
now commonly used in the authentication of herbal and food products. The objective of this study was to evaluate whether DNA
barcoding can be applied to trace the plant species from the starting raw material to the finished commercial products. We selected a
panel of 28 phytoextracts obtained through three different extraction
methods (i.e., maceration, percolation and sonication) with different
solvents (i.e., ethanol, deionized water and glycerol). Furthermore,
we chose six plant species for which we collected and analysed all
the intermediates of the industrial production. We sequenced and
analyzed the sequence variability at DNA barcoding (psbA-trnH,
ITS) and minibarcoding (rbcL 1-B) marker regions. Phytoextracts
obtained through hydroalcoholic treatment, with the lower percentage of ethanol (<40%), and aqueous processing, at the lowest
temperature, had major rate of sequencing and identification success.
This study proves that DNA barcoding is a useful tool for Medicinal
and Aromatic Plants (MAPs) traceability, which would provide consumers with safe and high-quality herbal products.

Full text

Journal of Applied Botany and Food Quality 92, 33 - 38 (2019), DOI:10.5073/JABFQ.2019.092.005
1Department of Biotechnology and Biosciences, University of Milano-Bicocca, Milano, Italy
2FEM2 Ambiente Srl, Milano, Italy
DNA barcoding to trace Medicinal and Aromatic Plants from the eld to the food supplement
Jessica Frigerio1, 2, Tommaso Gorini1, Andrea Galimberti1, Ilaria Bruni1,
Nicola Tommasi1, Valerio Mezzasalma2, Massimo Labra1*
(Submitted: October 8, 2018; Accepted: January 3, 2019)
* Corresponding author
Summary
The global market of food supplements is growing, along with con-
sumers demand for high-quality herbal products. Nevertheless,
substitution fraud, and adulteration cases remain a common safety
problem of global concern. In the last years, the DNA barcoding ap-
proach has been proposed as a valid identication method and it is
now commonly used in the authentication of herbal and food pro-
ducts. The objective of this study was to evaluate whether DNA
barcoding can be applied to trace the plant species from the start-
ing raw material to the nished commercial products. We selected a
panel of 28 phytoextracts obtained through three different extraction
methods (i.e., maceration, percolation and sonication) with different
solvents (i.e., ethanol, deionized water and glycerol). Furthermore,
we chose six plant species for which we collected and analysed all
the intermediates of the industrial production. We sequenced and
analyzed the sequence variability at DNA barcoding (psbA-trnH,
ITS) and minibarcoding (rbcL 1-B) marker regions. Phytoextracts
obtained through hydroalcoholic treatment, with the lower per-
centage of ethanol (<40%), and aqueous processing, at the lowest
temperature, had major rate of sequencing and identication success.
This study proves that DNA barcoding is a useful tool for Medicinal
and Aromatic Plants (MAPs) traceability, which would provide con-
sumers with safe and high-quality herbal products.
Key Word s: Herbal products, ITS, MAPs, Minibarcode, psbA-trnH,
Phytoextracts, rbcL
Introduction
Medicinal and Aromatic Plants (MAPs) and their preparations are
products used in medicine, cosmetics and food industr y, belonging
to plants, fungi, algae or lichens (Efsa, 2009). Such products are
prepared using plants or their parts to exploit their therapeutic and
healthy properties (e.g., antioxidant, anti-inammatory), as well as
their avor or scent (Who, 1999). According to a report published
by the Persistence Market Research, the global market of herbal
supplements had a value of USD 40 billion in 2017 and is expected
to reach a market valuation in excess of USD 65 billion by 2025
(PErsistEncE Ma rkEt rEsEarch, 2017). In the last years, the in-
creasing consumption of natural food supplements and the growing
awareness of consumers concerning the healthy benets of these
products have been progressively enhancing the market of MAPs
(Efsa, 2009). Although most of the herbal products used as food in-
gredients have been available to consumers since decades, the regu-
lation of these products differs greatly among jurisdictions. While
some countries consider MAPs as reliable ingredients for food pro-
duction, others regulate them as healthy products or medicines. For
example, in the European Union (EU), most products containing
Medicinal and Aromatic Plants are sold as food supplements and re-
gulated under the food law (silano et al., 2011); in Australia dietary
supplements are considered medicinal products and in Canada they
are subject to the complex regulation of the Natural Health Pro-
ducts Directorate (NNHPD) of Health Canada (hEalth can., 2015)
as medical products (loW et al., 2017). The lack of a clear and shared
global regulation and the large market demand of high-quality plant-
based items led to safety problems with the increase of substitution,
fraud and adulteration cases. Anyway, more attention is required to
guarantee a high quality level of MAPs which necessitates a stable
raw material and its assurance. As a matter of fact, frequ e n t l y, va lu -
able plants are substituted with cheaper raw materials, such as the
case of saffron substituted with safower (BosMali et al., 2017).
However, adulteration is not necessarily intentional, and herbal pro-
ducts may be altered due to inadvertent substitution, misidentication
or confusion resulting from the use of different vernacular names in
the countries of production.
According to the World Health Organization (WHO), the adultera-
tion of herbal products is a potential threat to consumers’ safety.
This condition opens two key issues which refer to the denition of
suitable toxicological evaluations to estimate the risks for human
health and the setup of an efcient identication system to trace the
herbal products from the eld to the traded MAPs items. Usually,
macroscopic and microscopic examinations are the classic strategies
adopted to verify the identity of fresh plants or the origin of plant
portions. These tools can be used to trace the herbal products when
the plants are processed immediately after being harvested. How-
ever, when the herbs undergo drying, fragmentation and pulveri-
zation processes the morphological traits cannot be longer used to
reliably assess the botanic source. Moreover, many herbal ingredi-
ents are obtained by infusion, maceration, distillation or pressing. In
these cases, only dedicated chemical analyses of the complex mix-
tures could permit to achieve a reliable plant identication.
In the last years the DNA barcoding approach was proposed as a
valid molecular identication method to provide species-level reso-
lution and it is now more and more used in the authentication of
taxonomic provenance of herbal and food products (nEWsMastEr
et al., 2013; GaliMBErti et al., 2013; MohaMMEd et al., 2017). How-
ever, the most important limit of this molecular tool is that it can
preferentially be adopted on unprocessed material (e.g., dry, frag-
mented and shredded plant portions) and several difculties are
encountered when dealing with extracts or with any other process
that results in the degradation of the DNA. Recently, some manu-
scripts described the efcacy of minibarcode regions (i.e., the analy-
sis of smaller genome portions − 100-150 bp − usually associated
to the largest DNA barcodes) for the identication of processed
plant extracts (raclariu et al., 2017; lit tlE, 2014). To date, any
study addressed the efcacy of a DNA barcoding-based approach
to trace herbal products along the entire production chain. In this
work, we selected Medicinal and Aromatic Plants in the form of
phytoextracts obtained by several industrial companies and sub-
jected to different kind of industrial processes and phytoextraction
strategies. The objective of this survey was to evaluate whether
or not the DNA barcoding approach (using standard barcodes or
minibarcode regions) could be applied to trace the plant species from

34 J. Frigerio, T. Gorini, A. Galimberti, Il. Bruni, N. Tommasi, V. Mezzasalma, M. Labra
the eld to the nished commercial product in the case of food sup-
plements. Therefore, we evaluated which are the industrial processes
mostly affecting the efcacy of DNA analysis such as sample pre-
treatment methods, solvents used for extraction and which are the
most suitable DNA markers to achieve a reliable MAPs traceability.
Material and methods
Study design
To test the efcacy of DNA traceability at different steps of the
industrial production chain of MAPs, we selected a panel of 28 com-
mercial phytoextracts (Tab. 1) sold by three main European com-
panies. The selected items were obtained starting from 17 plant
species (in the initial form of dried raw material) and were processed
by the same companies adopting three main extraction procedures,
namely maceration, sonication and percolation. Maceration consists
in the solubilization of the plant material in different solvents like
water and alcohol (e.g., ethanol), while percolation involves the slow
descent of a solvent through the plant raw material until it absorbs
the molecules of interest. Both methods rely on liquid ltration and
concentration. Differently, the sonication provokes cellular cavita-
tion and the release of the phytocomplexes in the solvent used. Three
different extraction solvents were considered in this study and spe-
cically, ethanol (i.e., alcoholic), deionized water (i.e., aqueous) and
glycerol. During maceration the temperature was maintained under
the threshold of 55 °C while in percolation higher temperatures (i.e.,
> 80 °C) were maintained, and sonication was mainly performed
at 30-40 °C. After the percolation process, some phytoextracts are
dried at very high temperatures (about 200 °C). Processing details
for each tested phytoextract are shown in Tab. 1.
To evaluate the efcacy of DNA barcoding to trace the intermediates
of industrial production after different steps of phytoextraction, we
selected a panel of six commercial products obtained only by alco-
holic and aqueous extraction procedures (Tab. 2). For these samples,
we collected and molecularly analysed through DNA barcoding, any
intermediate of production (Fig. 1).
DNA Extraction and DNA barcoding analysis
All commercial products from Tab. 1 were tested for authenticity
by sequencing three candidate markers, namely the standard DNA
barcoding plastidial intergenic spacer psbA-trnH (stEvEn and suB-
raManyaM, 2009), the nuclear ITS region (primers ITS p5-u4,
chEnG et al., 2016) and the minibarcode region rbcL 1-B (littlE,
2014). Primer details and size of amplied fragments are provided in
Tab. A .1.
A total of 50 mg of dried plant raw material, 150 μL of phytoex-
tract (and intermediate products of phytoextraction, see Tab. 2) were
treated for DNA extraction by using the EuroGOLD Plant DNA Mini
Kit (Euroclone, Pero, Italy). Each commercial phytoextract product
was subjected to DNA extraction in three replicates. Puried DNA
concentration of each sample was estimated uorometrically by us-
ing NanoDrop™ One/OneC Microvolume UV-Vis Spectrophoto-
meter (Thermo Scientic™).
A PCR amplication for each candidate marker was performed us-
ing puReTaq Ready-To-Go PCR beads (GE Healthcare Life Sciences,
Italy) in a 25 μL reaction volume according to the manufacturer’s
instructions containing 1 μL 10mM of each primer and up to 3 μL of
DNA template. PCR cycles consisted of an initial denatu ration step for
7 min at 94 °C, followed by 35 cycles of denaturation (45 s at 94 °C),
annealing (30 s at different temperatures; see Tab. A.1) and extension
(1 min at 72 °C), and, hence, a nal extension at 72 °C for 7 min.
In the case of the intermediates of production listed in Tab. 2, we
amplied and sequenced only the minibarcode locus rbcL 1B.
Amplicons occur rence was assessed by electrophoresis on agarose
gel using 1.5% agarose TAE gel stained with ethidium bromide and
amplicon length was measured by comparison against 100 bp ladder.
When a sample did not produce any band or showed multiple or non-
specic amplicons, the reaction was repeated increasing the amount
of template DNA up to 10 μL.
Puried amplicons were bidirectionally sequenced using an ABI
3730XL automated sequencing machine at Eurons Genomics (Ebers-
berg, Germany). The 3' and 5' terminal portions of each sequence
were clipped to generate consensus sequences for each sample.
After manual editing, primer removal and pairwise alignment, the
obtained sequences for dried raw material were submitted to the
international GenBank through the EMBL platform (see Tab. A.2
for accession numbers).
For all the tested samples (Tab. 1 and Tab. 2), the reliability of DNA
barcoding identication was assessed by adopting a standard com-
parison approach against a GenBank database with BLASTn. Each
barcode sequence was taxonomically assigned to the plant species
with the nearest matches (maximum identity > 99% and query co-
verage of 100%) according to Bruni et al. (2015). We performed the
identication separately for the three markers.
Results and discussion
Good DNA quality (i.e., A260/A230 and A260/A280 absorbance
ratios within the range 1.8 - 2.2) and extraction yield (20-40 ng/μl)
were obtained from all the 17 raw material samples. The three can-
Fig. 1: The industrial owchart of MAPs production. The numbers indicate
the intermediate steps of the industrial process for which the DNA
barcoding efcacy has been veried (Tab. 2).

DNA barcoding to trace MAPs from the eld to the food supplement 35
Sonication 12.94 1.58 × × ×
EtOH < 40%
EtOH > 40%
WATER
GLYCEROL
Tab. 1: List of the analysed MAPs samples with details concerning their industrial processing to obtain the nal phytoextracts. Average yield of DNA extraction
(with standard deviation) and assessment of positive sequencing of DNA barcoding markers (×) are also reported.
SAMPLES Industrial processing DNA Extraction yield DNA BARCODING MARKERS
Phytoextraction Solvent Value Standard psbA - trnH ITS rbcL 1 - B
Process ng/μl Deviation
Achillea millefolium L. Sonication 12.26 1.55 ×
Echinacea pallida (Nutt.) Nutt. Sonication 16.29 0.97 ×
Harpagophytum procumbens (Burch.) DC.
ex Meisn.
Melissa ofcinalis L. Percolation 44.63 1.32 ×
Mentha × piperita L. Sonication 12.3 0.83 × × ×
Tilia platyphyllos Scop. Sonication 9.78 1.28 × × ×
Zingiber ofcinale Roscoe Sonication 13.16 2.13 × × ×
Arctium lappa L. Maceration 1.23 0.2
Echinacea angustifolia DC. Maceration 2.83 0.5
Melissa ofcinalis L. Percolation 1.4 0.44
Passiora incarnata L. Maceration 1.74 0.17
Taraxacum ofcinale Weber ex F.H. Wigg. Maceration 2.47 0.54
Thymus vulgaris L. Maceration 1.78 0.38
Arctostaphylos uva-ursi (L.) Spreng. Percolation 3.1 0.33
Cetraria islandica (L.) Ach. Percolation 2.27 0.94
Echinacea purpurea (L.) Moench Percolation 3.47 0.71 ×
Epilobium angustifolium L. Percolation 1.88 0.63
Malva sylvestris L. Percolation 2.34 0.69
Arctostaphylos uva-ursi (L.) Spreng. Sonication 13.36 1.05 ×
Echinacea purpurea (L.) Moench Sonication 46.41 0.81 × × ×
Epilobium angustifolium L. Sonication 14.78 0.97 ×
Melissa ofcinalis L. Sonication 12,73 0,76 × × ×
Arctium lappa L. Maceration 2.69 0.64
Echinacea angustifolia DC. Maceration 4.73 0.85 ×
Melissa ofcinalis L. Maceration 2.55 0.76
Passiora incarnata L. Maceration 2.09 0.6
Taraxacum ofcinale Weber ex F.H. Wigg. Maceration 3.12 0.3
Thymus vulgaris L. Maceration 2.71 0.8
Tab. 2: List of six commercial MAPs (phytoextracts) traced along their entire production chain. Each sample (intermediates of industrial production) was
treated for DNA extraction and DNA barcoding analysis using the minibarcode region rbcL 1-B. Numbers indicate the industrial processing step as
described in Fig. 1. Y= correct plant identication by DNA barcoding at rbcL 1-B locus, N= DNA extraction or amplication failure, - = Sample was
not collected and analysed.
Plant species Solvent Steps of the industrial production process
1 2 3 4 5 6
Achillea millefolium L. 20% Ethanol Y Y N Y Y Y
Zingiber ofcinale Roscoe 30% Ethanol Y Y N Y Y Y
Thymus vulgaris L. 60% Ethanol Y N Y N N -
Melissa ofcinalis L. 70% Ethanol Y N Y N N -
Echinacea purpurea (L.) Moench Water Y Y N Y Y -
Melissa ofcinalis L. Water Y Y N Y Y Y

36 J. Frigerio, T. Gorini, A. Galimberti, Il. Bruni, N. Tommasi, V. Mezzasalma, M. Labra
didate genetic markers exhibited high PCR success and the obtained
PCR products were successfully sequenced with high-quality bi-
directional sequences. The BLASTn analysis suggested that all the
obtained sequences corresponded with 100% maximum identity to
the species declared by each company.
We are aware that multiple cases of 100% maximum identity within
the same plant genus could occur, especially concerning the DNA
minibarcoding rbcL 1B region. For example, this plastid region failed
in discriminating Echinacea purpurea (L.) Moench from conge-
nerics like Echinacea angustifolia DC. or Echinacea pallida (Nut t.).
Such events suggest that in some conditions, the main limit of DNA
minibarcoding relies on the reduced discrimination power among
congenerics but it allows to detect plant contaminations when the
adulterant/s belong to genera different from the target one. Never-
theless, it should be considered that when DNA content is expected
to be low (or of low quality), the use of shorter DNA barcoding re-
gions offers the best compromise between amplication universality,
sequence quality and taxonomic discrimination (littlE, 2014).
Concerning the 17 dry raw material samples, our results agree with
the assumptions of nEWMastEr and co-workers (2013) who sug-
gested that a DNA barcoding approach could be successfully applied
to verify the identity of commercial herbal products and to reveal
cases of contamination or substitution. Therefore, when herbal pro-
ducts are directly used as ingredients of complex food, medicine or
cosmetics items, they are subjected to “soft” processing actions such
as cleaning, drying and cutting and the DNA barcoding (achieved
using long barcode fragments > 300 bp) represents a useful tool to
trace plant species during the processing (dE Mattia et al., 2011).
As expected, the efcacy of DNA extraction and amplication
decreased when we analyzed the 28 commercial phytoextracts
and their intermediates of industrial processing. Overall, the DNA
amount obtained after extraction processes ranged from 1.5 to more
than 40 ng/μL (Tab. 1). The extracts obtained through hydroalcoholic
treatment, with the lower percentage of ethanol (< 40%), and aqueous
processing, at the lowest temperature, contained more DNA than the
other samples (Tab. 1 and Tab. 2).
In the samples where the DNA barcoding analysis worked well, no
contamination or adulteration (i.e., the occurrence of DNA bacodes
of other plant species) were observed. Unfortunately, in some groups
of extracts, the molecular analysis did not provide reliable DNA ex-
traction or high-quality sequences. At technical level, we hypothe-
size that in general, the high concentration of ethanol used in the
industrial processing steps lead to DNA precipitation. This was con-
rmed by the data reported in Tab. 2, where samples of Thymus vul-
garis L. and Melissa ofcinalis L. processed with ethanol at high
concentration, showed residual DNA in the extraction waste rather
than in the phytoextract. For this reason, both the DNA extraction
and DNA barcoding authentication failed when applied to the suc-
cessive intermediate products of industrial processing and DNA was
no longer available in the nal herbal supplements. In this case, we
conclude that for this kind of industrial production, a DNA-based
approach is not suitable to achieve a reliable traceability of the ini-
tial plant raw material. Similarly, high temperatures of water du-
ring aqueous extraction, followed by a drying step (about 200 °C)
probably lead to DNA fragmentation and degradation (karni et al.,
2013) as observed in ve of the samples processed with a percolation
procedure (Tab. 1). Conversely, the use of more lukewarm water (i.e.,
< 55° C) allows to achieve a successful DNA extraction, ampli-
cation and sequencing of DNA barcoding markers (Tab. 1). More-
over, such conditions also allow the traceability of the intermediate
pro-ducts of industrial processing as observed for Echinacea pur-
purea (L.) Moench and Melissa ofcinalis L. extracted using water
as solvent.
Concerning glycerol extracts, although this solvent does not act di-
rectly on DNA molecules, it usually contains ethylhexylglycerin and
phenoxyethanol, which are typically used as additives. According to
Langsrud and co-workers (2016) these antibacterial agents could be
responsible for DNA loss. For this reason, also the analysis of the
DNA minibarcode region did not produce amplicons in glycerine
extracts (Tab. 1).
Concerning the industrial treatments, the sonication seems to keep
the DNA of raw materials more intact than the other processes (i.e.,
maceration and percolation). Our results also show that sonicated
samples contained higher amounts of DNA (i.e. from 9.73 to 44 ng/
μl, Tab. 1) compared to the other categories, thus allowing a success-
ful amplication and sequencing of the DNA minibarcode marker.
Concerning the quality of extracted genetic material, the purity of
DNA is more important than the extraction yield to achieve a good
amplication and then a reliable identication (sonG et al., 2017). It
should also be considered that secondary metabolites, like polyphe-
nols and polysaccharides, which are normally extracted along with
DNA, may interfere with PCR amplication (sahu, thanGara j,
and kathirEsan, 2012). These molecules could bind DNA cova-
lently and make the extraction products impure, with several pro-
blems for the successive molecular analysis. For example, tannic
acids could bind and inactivate Taq polymerase (oPEl, chunG, and
Mccord, 2010). However, in our analysis we hypothesize that the
main amplication problem for the phytoextracts is the fragmen-
tation of DNA. In all the tested cases, the DNA minibarcode locus
rbcL 1-B was most easily amplied and sequenced (Tab. 1) than the
other two DNA barcoding markers. This suggests that the DNA ob-
tained from phytoextracts are richer in small DNA fragments (80-
200 bp). Such condition is in line with the data reported in recent
review articles (MohaMMEd et al., 2017) suggesting that DNA bar-
coding is a reliable and suitable technique only for the herbal product
that preserve a good quality DNA and with poor fragmentation. In
the other cases DNA minibarcoding is the most efcient and reliable
tool for traceability purposes (sonG et al., 2017).
Nowadays, analytical chemistry methods (TLC, HPLC) represent
the most used tools to verify the quality of MAPs, however, these
approaches are usually directed to dene the concentration of spe-
cic bioactive molecules or to estimate chemical contaminants (e.g.,
heavy metals) rather than to identify the occurrence of plant con-
taminants (sGaMMa et al., 2017). Conversely, the DNA barcoding
approach is globally recognized as one of the most reliable DNA-
based approaches to identify species if a well populated reference
dataset of DNA barcode sequences for the target taxa is available
(GaliMBErti et al., 2013). Moreover, in the case of contamination
(or substitution), DNA analyses also allow to simultaneously iden-
tify any species (i.e., DNA metabarcoding) using High Throughput
Sequencing (HTS) sequencing systems (GaliMBErti et al., 2015;
MEzzasalM a et al., 2017). For these reasons, the Pharmacopoeia
guidelines of some countries such as that of UK (British PharMa-
coPoEia coMMission, 2017) indicate the DNA barcoding as one
of the ofcial traceability systems in the sector of herbal products.
Our data suppor t this proposal and the ability of DNA minibarcode
makers to provide a reliable tracing of the intermediate products
of industrial production. However, it is important to underline that
some industrial processes demanding high temperatures and the use
of solvents, such as a high concentration of ethanol, can induce DNA
degradation and make this molecular tool less effective.
In conclusion, this study leads to two main considerations about
the future application of DNA barcoding as a quality control tool
in the sectors where the Medicinal and Aromatic Plants constitute
relevant ingredients (e.g., food, cosmetics and pharmacology). First
of all, the current industrial trends promote the adoption of extrac-
tion processes from plant raw material, which rely on the reduction of
energy consumption (i.e., low temperatures), and on the use of more
‘green’ solvents (e.g., water) to obtain exhausted waste products that
can be used in other supply chains (e.g., fertilizers). The adoption and

DNA barcoding to trace MAPs from the eld to the food supplement 37
spread of this trend should lead to an increased integrity and quality
of DNA in MAPs (and related intermediate products) and therefore
enhance the success of DNA barcoding as a universal traceability
system.
Secondly, the continuous advances in High Throughput Sequencing
and the resulting possibility of exploring multiple short genetic
regions simultaneously (i.e. 150-200 bp), could increase the sensi-
tivity of a DNA-based identication.An HTS-DNA metabarcoding
approach would allow to check the presence of several plant con-
taminants in the same sample, even if occurring at low concentra-
tions (nEW MastEr et al., 2013). sGaMMa and co-workers (2017)
proposed the introduction of DNA metabarcoding to evaluate the
quality and authenticate herbal drug material in the industrial con-
text. The authors proposed a dedicated DNA barcoding owchart for
industrial traceability purposes. Our results could be taken into ac-
count to improve this owchart and to also adapt it to the traceability
of intermediates of industrial production. Interestingly, valEntini
and co-workers (2017) recently proposed an innovative nanoparticle-
DNA barcoding hybrid system called NanoTracer that could poten-
tially revolutionize the world of traceability as it allows for rapid and
naked-eye molecular traceability of any food and requires limited
instrumentation and cost-effective reagents.
This and other similar applications (aartsE et al., 2017) open the op-
portunity to really boost the issue of herbal supplements traceability,
not only with the industrial actors as the main stakeholders, but also
involving a wider cicle of specialists.
Acknowledgments
This work was supported by Regione Lombardia in the frame-
work of the Program ‘Accordi per la ricerca e l’innovazione’, pro-
ject name: Food Social Sensor Network Food NET, grant number:
E47F17000020009. The funder had no role in conducting the re-
search and/or during the preparation of the article. FEM2-Ambiente
s.r.l. provided support in the form of a salary for authors Jessica
Frigerio and Valerio Mezzasalma but did not play a role in the study
design, data collection and analysis, decision to publish, or prepa-
ration of the manuscript and only provided nancial support in the
form of research materials. The authors are indebted with Dr. Paola
Re for support in preparing the graphical part of the manuscript.
References
aartsE, A., scholtEns, I.M. J., van dEr a, H.J.G., BoErsMa-GrEvE, M.M.,
Prins, T.W., va n Gink El, L.A., kok, E.j., BovEE, T.F. H ., 2017: Ev a-
luation of a loop-mediated isothermal amplication (LAMP) method for
rapid on-site detection of horse meat. Food Control, 81, 9-15.
D OI : 10.1016/j. fo od co nt.2 017.05. 02 5
BosMali, i., ordoudi, s.a., tsiMidou, M.z., MadEsis, P., 2017: Greek PDO
saffron authentication studies using species specic molecular markers.
Food Res. Int. 100(Pt 1), 899-907. DOI: 10.1016/j.foodres.2017.08
British Phar macopoeia Commission, 2015: Working Party (DNA): Identi-
cation Techniques SUMMA RY MINUTES. https://www.pharmaco-
poeia.com/le/DNA---January-2016.pdf
Bruni, i., GaliMBErti, a., car idi, l., scaccaBarozzi, d., dE Mattia, f.,
casiraGhi, M., laBra, M., 2015: A DNA barcoding approach to iden-
tify plant species in multiower honey. Food Chem. 170, 308-315.
D OI : 10.1016/j. fo od ch em.2 014.0 8. 06 0
chEnG, t., Xu, c., lEi, l., li, c., zhanG, y., zhou, s., 2016: Barcoding
the kingdom Plantae: new PCR primers for ITS regions of plants with
improved universality and specicity. Mol. Ecol. Resour. 16(1), 138-149.
D OI: 10.1111/1755- 099 8.1 2438
dE Mattia, f., Bruni, i., GaliMBErt i, a., cattanEo, f., casiraGhi, M.,
laBra, M., 2011: A comparative study of different DNA barcoding
markers for the identication of some members of Lamiacaea. Food Res.
Int. 44(3), 693-702. DOI: 10.1016/j.foodr es. 2010.12.032
GaliMBErti, a., Bruno, a., MEzzasalMa, v., dE Matt ia, f., Bruni, i.,
laBra, M., 2015: Emerging DNA-based technologies to characterize
food ecosystems. Food Res. Int. 69, 424-433.
D OI : 10.1016/J. FOODRE S. 2015.01. 017
GaliMBErti, a., dE Mattia, f., losa, a., Bruni, i., fEdEr ici, s.,
casiraGhi, M. MartEllos, s., la Bra, M., 2013: DNA barcoding as a
new tool for food traceability. Food Res. Int. 50(1), 55-63.
DOI: 10.1016/J.F OO DR ES.2 012.09.03 6
EFSA, 2009: Guidance on Safety assessment of botanicals and bota nical
preparations intended for use as ingredients in food supplements. EFSA
Journal, 7(9). DOI: 10.2903/j.efsa.2009.1249
hEalth can., 2015: Natural and Non-prescription Health Products Di rec-
torate (N NHPD). Updated on May 1, Health Can., Ottawa.
http://www.hc-sc.gc.ca/dhp-mps/prodnatur/index-eng.php
karni, M., zidon, d., Polak, P., zalEvsky, z., shEfi, o., 2013: Thermal
Degradat ion of DNA. DNA Cell Biol. 32(6). DOI: 10.1089/dna.2013.2056
lanGsrud, s., stEinh auEr, k., lüthj E, s., WEBEr, k., Goroncy-BErMEs,
P., holc k, a.l., 2016: Ethylhexylglycerin impai rs membrane integrity
and enhances the lethal effect of Phenoxyethanol. PLOS ONE, 11(10),
e016522 8. DOI: 10.1371/jour nal.po ne.0165228
littlE, d. P., 2014: A DNA mini-ba rcode for land plants. Mol. Ecol. Resour.
14(3), 437-4 46. DO I: 10. 1111/1755 -09 98.12194
loW, t.y., WonG, k.o., yaP, a.l.l., dE haan, l.h.j., riEtjEns, i.M.c.M.,
2017: T he regulator y framework across international jurisdictions for
risks associated with consumption of botan ical food supplements. Com-
pr. Rev. Food Sci. F. 16(5), 821-834. DOI: 10.1111/1541-4337.12289
MEzzasal Ma, v., GanoP oulos, i., Gal iMBErti, a., cornara, l., fErri,
E., laBra, M., 2017: Poisonous or non-poisonous plants? DNA-based
tools and applications for accurate identication. Int. J. Legal Med.
131(1), 1-19. D OI : 10.10 07/s0 0414- 016-1460 -y
MohaMMEd aBuBak ar, B., Mohd sallEh, f., shaMsir oM ar, M.s.,
WaGiran, a., 2017: Review: DNA barcoding and chromatography n-
gerprints for the authentication of botanicals in herbal medicinal pro-
ducts. Evid-Based Compl. Alt. 2017, 1-28. DOI: 10.1155/2017/1352948
nEWMastE r, s.G., GrGuric, M., shanMuGhanandhan, d., raMal inGaM,
s., raGuPathy, s., 2013: DNA barcoding detects contamination and
substitution in Nor th American herbal products. BMC Med. 11(1), 222.
DO I: 10. 1186 /1741-7015-11-22 2
oPEl, k.l., chu nG, d., Mccord, B.r., 2010: A Study of PCR Inhibition
mechanisms using real time PCR. J. Forensic Sci. 55(1), 25-33.
D OI: 10.1111/j. 155 6 -40 29.2 0 09.01 245. x
PErsistEncE MarkEt rEsEarch, 2017: Global market study on bot anical
supplements: Drugs application segment to hold max imum value share
dur in g 2017-2025. https://www.prnewswire.com/news-releases/global-
market-study-on-botanical-supplements-drugs-application-segment-to-
hold-maximum-value-share-during-2017---2025-300484161.html
raclariu, a.c., PaltinEan, r., vlasE, l., laBarr E, a., Manzan illa, v.,
ichiM, M.c., crisan, G., BrystinG, a.k., dE BoEr, h., 2017: C om pa ra-
tive authentication of Hyper icum perforatum herbal products using DNA
metabarcoding, TLC and HPLC-MS. Sci. Rep. 7(1), 1291.
D OI: 10.1038/s 41598-017-01389-w
sahu, s.k., thanGa raj, M., kathirEsan, k., 2012: DNA Extraction pro-
tocol for plants with high levels of secondary metabolites and poly-
saccharides without using liquid nitrogen and phenol. ISR N Mol. Biol.
2012, 1-6. DOI: 10.5402/2012/205049
sGaMMa, t., lockiE-WilliaMs, c., krEuzEr, M., WilliaMs, s., schEyhinG,
u., koch, E., slatEr, a., hoWard, c., 2017: DNA barcoding for indus-
trial quality assurance. Planta Med. 83(14/15), 1117-1129.
D OI: 10.1055/s- 0043-1134 48
silano, v., coPPEns, P., lar rañaGa-GuEtaria, a., Mi nGhEtti, P., roth-
EhranG, r., 2011: Regulations applicable to plant food supplements and
related products in the European Union. Food Funct. 2(12), 710.
DOI:10.1039/c1fo10105f
sonG, M., donG, G.-Q., zhanG, y.-Q., liu, X., sun, W., 2017: Identication

38 J. Frigerio, T. Gorini, A. Galimberti, Il. Bruni, N. Tommasi, V. Mezzasalma, M. Labra
of processed Chinese medicinal materials using DNA mini-barcoding.
Chin. J. Nat. Med. 15(7), 481-486. DOI: 10.1016/S1875-5364(17)30073- 0
stEvEn G.n., suBraManyaM, r., 20 09: Testing plant barcoding in a sister
species complex of pantropical Acacia (Mimosoideae, Fabaceae). Mol.
Eco. Resour. 9, 172-180. DOI: 10.1111/j .1755- 099 8. 2009. 026 42 .x
valEntini, P., GaliMBErti, a., MEzzasalMa, v., dE Mattia, f., casiraGhi,
M., laBr a, M., PoMPa, P.P., 2017: DNA barcoding meets nanotechno-
logy: Development of a Universal Colorimetric Test for Food Authen-
tication. Angew. Chem. Int. Ed. 56(28), 8094-8098.
D OI: 10.1002/anie. 201702120
Who, 1999: WHO Monographs on selected medicinal plants. Vol. 1,
Medicinal Plants, Geneva.
http://apps.who.int/medicinedocs/pdf/s2200e/s2200e.pdf
Address of the cor responding author:
Massimo Labra, Department of Biotechnology and Biosciences, University
of Milano-Bicocca, Piazza della Scienza 2, 20126 Milan, Italy
E-mail: massimo.labra@unimib.it
© The Author(s) 2019.
This is an Open Access article distributed under the terms of
the Creative Commons Attribution 4.0 International License (https://creative-
commons.org/licenses/by/4.0/deed.en).

Supplementary material I
From Appendix
Table A.1: List of primer pairs used for DNA barcoding analysis.
Locus
name
Name
5' - 3' Sequence
Tm °C
Amplified samples
size
Reference
rbcL 1
TTGGCAGCATTYCGAGTAACTCC
50
80-200 bp
PALMIERI (2009)
rbcL
rbcL B
AACCYTCTTCAAAAAGGTC
50
psbA
GTTATGCATGAACGTAATGCTC
53
300-600 bp
STEVEN & SUBRAMANYAM (2009)
psbA-trnH
trnH
CGCGCATGGTGGATTCACAATCC
53
ITS p5
CCTTATCAYTTAGAGGAAGGAG
55
300-750 bp
CHENG (2016)
ITS
ITS u4
RGTTTCTTTTCCTCCGCTTA
55
Table A.2: List of all the analysed plant species. The table include the voucher specimens and GenBank accession numbers of the
raw material samples used to authenticate the phytoextracts and their intermediates of industrial production treated in this study.
Species Type of sample Specimen voucher Company GenBank Accession Number (rbcL 1-B; psbA-trnH; ITS)
Achillea millefolium L. Dried raw material FEM_DBE_001 Company 1 LS999840; LS999856; LS999873
Arctium lappa L. Dried raw material FEM_DBE_007 Company 3 LS999841; LS999857; LS999874
Arctostaphylos uva-ursi (L.) Spreng. Dried raw material FEM_DBE_010 Company 2 LS999842; LS999858; LS999875
Cetraria islandica (L.) Ach. Dried raw material FEM_DBE_013 Company 2 ND; ND; LS999876
Echinacea angustifolia DC. Dried raw material FEM_DBE_015 Company 3 LS999843; LS999859; LS999877
Echinacea pallida (Nutt.) Nutt. Dried raw material FEM_DBE_018 Company 1 LS999844; LS999860; LS999878
Echinacea purpurea (L.) Moench Dried raw material FEM_DBE_020 Company 2 LS999845; LS999861; LS999879
Epilobium angustifolium L. Dried raw material FEM_DBE_026 Company 2 LS999846; LS999862; LS999880
Harpagophytum procumbens (Burch.) Dried raw material FEM_DBE_029 Company 1 LS999847; LS999863; LS999881
Malva sylvestris L. Dried raw material FEM_DBE_031 Company 2 LS999848; LS999864, LS999882
Melissa officinalis L. Dried raw material FEM_DBE_033 Company 2 LS999849; LS999865; LS999883
Mentha x piperita L. Dried raw material FEM_DBE_045 Company 1 LS999850; LS999866; LS99984
Passiflora incarnata L. Dried raw material FEM_DBE_047 Company 3 LS999851; LS999867; LS999885
Taraxacum officinale Weber ex F.H. Wigg. Dried raw material FEM_DBE_050 Company 3 LS999852; LS999868; LS999886
Thymus vulgaris L. Dried raw material FEM_DBE_053 Company 3 LS999853; LS999869; LS999887
Tilia platyphyllos Scop. Dried raw material FEM_DBE_059 Company 1 LS999854; LS999870; LS999888
Zingiber officinale Roscoe Dried raw material FEM_DBE_061 Company 1 LS999855; LS999871; LS999889

Supplementary material
Tab. A.1: List of primer pairs used for DNA barcoding analysis.
Tab. A.2: List of all the analysed plant species. The table include the voucher specimens and GenBank accession numbers of the raw material samples used to
authenticate the phytoextracts and their intermediates of industrial production treated in this study.
From Appendix
Table A.1: List of primer pairs used for DNA barcoding analysis.
Locus
name Name 5' - 3' Sequence Tm °C Amplified samples
size Reference
rbcL 1 TTGGCAGCATTYCGAGTAACTCC 50 80-200 bp PALMIERI (2009)
rbcL rbcL B AACCYTCTTCAAAAAGGTC 50
psbA GTTATGCATGAACGTAATGCTC 53 300-600 bp STEVEN & SUBRAMANYAM (2009)
psbA-trnH
trnH CGCGCATGGTGGATTCACAATCC 53
ITS p5 CCTTATCAYTTAGAGGAAGGAG 55 300-750 bp CHENG (2016)
ITS
ITS u4 RGTTTCTTTTCCTCCGCTTA 55
Table A.2: List of all the analysed plant species. The table include the voucher specimens and GenBank accession numbers of the
raw material samples used to authenticate the phytoextracts and their intermediates of industrial production treated in this study.
Species
Type of sample
Specimen voucher
Company
GenBank Accession Number (rbcL 1-B; psbA-trnH; ITS)
Achillea millefolium L.
Dried raw material
FEM_DBE_001
Company 1
LS999840; LS999856; LS999873
Arctium lappa L.
Dried raw material
FEM_DBE_007
Company 3
LS999841; LS999857; LS999874
Arctostaphylos uva-ursi (L.) Spreng.
Dried raw material
FEM_DBE_010
Company 2
LS999842; LS999858; LS999875
Cetraria islandica (L.) Ach.
Dried raw material
FEM_DBE_013
Company 2
ND; ND; LS999876
Echinacea angustifolia DC.
Dried raw material
FEM_DBE_015
Company 3
LS999843; LS999859; LS999877
Echinacea pallida (Nutt.) Nutt.
Dried raw material
FEM_DBE_018
Company 1
LS999844; LS999860; LS999878
Echinacea purpurea (L.) Moench
Dried raw material
FEM_DBE_020
Company 2
LS999845; LS999861; LS999879
Epilobium angustifolium L.
Dried raw material
FEM_DBE_026
Company 2
LS999846; LS999862; LS999880
Harpagophytum procumbens (Burch.)
Dried raw material
FEM_DBE_029
Company 1
LS999847; LS999863; LS999881
Malva sylvestris L.
Dried raw material
FEM_DBE_031
Company 2
LS999848; LS999864, LS999882
Melissa officinalis L.
Dried raw material
FEM_DBE_033
Company 2
LS999849; LS999865; LS999883
Mentha x piperita L.
Dried raw material
FEM_DBE_045
Company 1
LS999850; LS999866; LS99984
Passiflora incarnata L.
Dried raw material
FEM_DBE_047
Company 3
LS999851; LS999867; LS999885
Taraxacum officinale Weber ex F.H. Wigg.
Dried raw material
FEM_DBE_050
Company 3
LS999852; LS999868; LS999886
Thymus vulgaris L.
Dried raw material
FEM_DBE_053
Company 3
LS999853; LS999869; LS999887
Tilia platyphyllos Scop.
Dried raw material
FEM_DBE_059
Company 1
LS999854; LS999870; LS999888
Zingiber officinale Roscoe
Dried raw material
FEM_DBE_061
Company 1
LS999855; LS999871; LS999889
