You are on page 1of 7

G Model

PHYMED-51597; No. of Pages 7

ARTICLE IN PRESS
Phytomedicine xxx (2014) xxxxxx

Contents lists available at ScienceDirect

Phytomedicine
journal homepage: www.elsevier.de/phymed

Adulteration of Ginkgo biloba products and a simple method to


improve its detection
Hans Wohlmuth a,b, , Kate Savage c , Ashley Dowell c , Peter Mouatt a,c
a
b
c

Medicinal Plant Herbarium, Southern Cross University, PO Box 157, Lismore, NSW 2480, Australia
Integria Healthcare, Gallans Road, Ballina, NSW 2478, Australia
Southern Cross Plant Science, Southern Cross University, PO Box 157, Lismore, NSW 2480, Australia

a r t i c l e

i n f o

Article history:
Received 14 November 2013
Received in revised form
27 November 2013
Accepted 26 January 2014
Keywords:
Ginkgo biloba
Ginkgo extract
Adulteration
Quality control
Ginkgo avonol glycosides
Genistein

a b s t r a c t
Extracts of ginkgo (Ginkgo biloba) leaf are widely available worldwide in herbal medicinal products,
dietary supplements, botanicals and complementary medicines, and several pharmacopoeias contain
monographs for ginkgo leaf, leaf extract and nished products. Being a high-value botanical commodity,
ginkgo extracts may be the subject of economically motivated adulteration. We analysed eight ginkgo
leaf retail products purchased in Australia and Denmark and found compelling evidence of adulteration
with avonol aglycones in three of these. The same three products also contained genistein, an isoavone
that does not occur in ginkgo leaf.
Although the United States Pharmacopeia National Formulary (USP-NF) and the British and European
Pharmacopoeias stipulate a required range for avonol glycosides in ginkgo extract, the prescribed assays
quantify avonol aglycones. This means that these pharmacopoeial methods are not capable of detecting
adulteration of ginkgo extract with free avonol aglycones.
We propose a simple modication of the USP-NF method that addresses this problem: by assaying for
avonol aglycones pre and post hydrolysis the content of avonol glycosides can be accurately estimated
via a simple calculation. We also recommend a maximum limit be set for free avonol aglycones in ginkgo
extract.
2014 Elsevier GmbH. All rights reserved.

Introduction
Extracts of ginkgo (Ginkgo biloba L.) leaf are sold worldwide
as the active ingredient of numerous dietary supplements, botanicals, herbal medicinal products and complementary medicines.
Indeed, ginkgo is currently one of the most widely sold medicinal plants, and the global market for ginkgo has been estimated at
more than US$700 million (Euromonitor International Ltd., 2009).
In the United States, the most recent (2012) data show the retail
market for ginkgo products to be worth US$30 million (Lindstrom
et al., 2013).
Ginkgo is also one of the most intensely studied medicinal
plants, with more than 3000 scientic papers published on the
topic between 2001 and 2009 alone (van Beek and Montoro, 2009).
Ginkgo leaf extracts are recommended for a range of conditions,
including cerebral insufciency, vertigo and tinnitus of vascular

Corresponding author at: Integria Healthcare, Gallans Road, Ballina, NSW 2478,
Australia. Tel.: +61 2 6620 5180; fax: +61 2 6622 3459.
E-mail addresses: hans.wohlmuth@scu.edu.au, hans.wohlmuth@integria.com
(H. Wohlmuth).

origin, and peripheral arterial disease (Blumenthal, 2003; Bone and


Mills, 2013).
The pharmacologically active compounds in ginkgo leaf are
considered to be avonol glycosides (quercetin, kaempferol and
isorhamnetin being the principal aglycones) and terpene lactones
(bilobalide and ginkgolides). Most ginkgo leaf extracts on the market are produced by selective, multi-step extraction processes
involving organic solvents and carry quantitative claims concerning
their content of avonol glycosides and terpene lactones. Accordingly, most ginkgo leaf extracts are more high-tech and high-cost
than typical botanical extracts.
Botanical raw materials including extracts present special challenges in terms of quality control and assurance due to their
chemical complexity and inherent natural variability. The most
fundamental aspects of quality assurance for such materials are
to ensure the correct morphological part(s) from the right botanical taxon is used, and that the material is not adulterated with
other botanical or extraneous material. Adulteration, either accidental or intentional and economically motivated, is a well-known
issue for botanicals, and one that potentially can jeopardise not only
the quality but also the safety of the nished product (Khan, 2006;
Walker and Applequist, 2012). The potential safety issues associated with adulterated or sub-standard ginkgo extracts have been

0944-7113/$ see front matter 2014 Elsevier GmbH. All rights reserved.
http://dx.doi.org/10.1016/j.phymed.2014.01.010

Please cite this article in press as: Wohlmuth, H., et al., Adulteration of Ginkgo biloba products and a simple method to improve its
detection. Phytomedicine (2014), http://dx.doi.org/10.1016/j.phymed.2014.01.010

G Model
PHYMED-51597; No. of Pages 7

ARTICLE IN PRESS
H. Wohlmuth et al. / Phytomedicine xxx (2014) xxxxxx

highlighted by recent toxicology and carcinogenicity studies using


such inferior materials, while other studies using pharmaceutical
quality ginkgo extract have found it to be safe (Koch et al., 2013;
Krenn et al., 2013).
Pharmacopoeial monographs play an important role in the
quality assurance of botanicals and herbal medicinal products
(Vlietinck et al., 2009). Monographs for ginkgo raw materials (leaf and extract) can be found in various pharmacopoeias,
including the United States Pharmacopoeia-National Formulary
(USP-NF) (United States Pharmacopeial Convention, 2013), the
European Pharmacopoeia (EP) and the British Pharmacopoeia (BP)
(British Pharmacopoeia Commission, 2012). The USP-NF monograph for Powdered Ginkgo Extract and the BP/EP monograph
for Rened and Quantied Ginkgo Dry Extract specify a avonoid
content of 2227%, calculated as avonol/avone glycosides and
a maximum content of 5 ppm for ginkgolic acids (putative allergens). In addition, specications are provided for the terpene
lactones, bilobalide and ginkgolide A, B and C, but the ranges
for these differ between the USP-NF and the BP/EP. The USP-NF
also provides monographs for Ginkgo Tablets and Ginkgo Capsules.
Here we report on ginkgo retail products found to be adulterated with free avonol aglycones and also containing the
isoavone genistein, which is not native to ginkgo. We demonstrate that current pharmacopoeial methods are inadequate for the
detection of this type of adulteration, and we propose a simple
modication of the USP-NF method that addresses this problem.

Materials and methods


Plant materials and botanicals
Five samples of dried Ginkgo biloba leaf were obtained from
commercial suppliers. These leaf samples came from ginkgo cultivated in China (3), New Zealand (1) and Australia (1). They were
authenticated by an experienced pharmacognosist (H.W.), and
voucher materials were deposited in the Medicinal Plant Herbarium (PHARM) at Southern Cross University.
Eight retail products containing Ginkgo biloba as the sole active
ingredient were purchased in Australia (6) and Denmark (2). Four
of these were tablets and four were capsules.

Sample preparation
Ginkgo leaf samples were ground to a ne powder in a Retsch
MM301 Mixer Mill (Haan, Germany). Approximately 5.5 g of powdered leaf material was placed in a 250-ml round bottom ask with
50 ml ethanol and 20 ml Milli-Q water and sonicated for 15 min
(Soniclean Ultrasonic bath, Thebarton, SA, Australia). To achieve
hydrolysis of avonol glycosides, 8 ml of 37% hydrochloric acid was
added (with boiling chips) and the mixture reuxed at moderate
temperature in a fume-hood for 2 h 15 min. Once cooled, the solution was transferred quantitatively to a 100-ml volumetric ask and
diluted to volume with Milli-Q water. Samples not hydrolysed were
treated identically, except for the addition of hydrochloric acid.
Tablets were extracted by combining a quantity of 20 and grinding them to a ne powder in a Retsch MM301 Mixer Mill. Depending
on the label claim for avonol glycoside content, between 300 and
1000 mg of the powder was extracted by sonication, as detailed for
the leaf samples. Capsules were opened and their content extracted
in the same way as the tablets. All samples were extracted in triplicate.
HPLC
Reverse-phase HPLC analysis was performed on an Agilent (Palo
Alto, CA, USA) 1100 HPLC system tted with a Phenomenex (Torrance, CA, USA) Synergi C18 4 M (250 mm 4.6 mm i.d.) column,
using an in-house validated method based on the USP34-NF29.
Mobile phase A consisted of 0.5% aqueous phosphoric acid; mobile
phase B consisted of methanol. The gradient eluting mobile phase
was A/B (60:40, v/v) to A/B (50:50, v/v) over 40 min. This was
followed by a 5 min column wash with A/B (5:95) and a 5 min
equilibration period with A/B (60:40) prior to the next injection.
Mobile phase was pumped at 1.2 ml/min, the column temperature 40 C, and the injection volume was 10 l. Data were collected
using a UV/visible light diode array detector collecting absorption spectra from 200 to 400 nm with quantication performed at
270 nm. System control and data evaluation were achieved using
ChemStation for HPLC software. Limits of detection for quercetin,
kaempferol and isorhamnetin were 0.625 g/ml, 0.25 g/ml and
0.07 g/ml, respectively. The range for precision, accuracy and linearity was 0.00250.5 mg/ml for quercetin, 0.0006250.5 mg/ml
for kaempferol and 0.000280.14 mg/ml for isorhamnetin.
LCMS

Chemicals and reagents


Water was obtained from an in-house Milli-Q system (Millipore, Billerica, MA, USA). Methanol (HPLC grade) was obtained
from Merck (Kilsyth, VIC, Australia), ethanol (AR grade) from
Chem-Supply (Gillman, SA, Australia), acetonitrile from Scharlau
(Sentmenat, Spain), phosphoric acid from Ajax Finechem (Sydney, Australia), and hydrochloric acid (AR grade), triuoroacetic
acid, dimethyl sulfoxide (DMSO), kaempferol (97.8%) and genistein
(>98%) from Sigma-Aldrich (Sydney, Australia). Quercetin dihydrate (98.5%) was obtained from Chromadex (Irvine, CA, USA)
and isorhamnetin (>98%) from Chengdu Biopurify Phytochemicals
(Chengdu, Sichuan, China).

Standard preparation
Reference standards (quercetin, kaempferol and isorhamnetin)
were dissolved in DMSO at a concentration of 2.0 mg/mL, then serially diluted in methanol using a Hamilton Microlab 500 Diluter
(Reno, NV, USA). Genistein was dissolved in methanol.

LCMS analysis was performed on an Agilent (Palo Alto, CA,


USA) 1100 LC/MSD system equipped with an atmospheric pressure
chemical ionisation (APCI) source and tted with a Phenomenex
(Torrance, CA, USA) Luna C18 3 M (100 mm 4.6 mm i.d.) column. Mobile phase A consisted of water and mobile phase B of
acetonitrile, both with 0.005% triuoroacetic acid added. The gradient eluting mobile phase was A/B (90:10, v/v) to A/B (5:95, v/v)
over 18 min followed by a 3 min column wash with A/B (5:95), to
A/B (90:10) over 3 min and a 5 min equilibration period with A/B
(90:10). Rate of mobile phase ow was 0.750 ml/min, the column
temperature was 40 C, and the injection volume was 5 l.
MS parameters in the positive ionisation mode were: Vcap
3000 V, nebuliser 60 psig, drying gas ow rate 5.0 l/min, gas
temperature 350 C, corona 4.0 A, vaporiser 350 C, scan range
1001500 m/z, step size 0.15 m/z, peak width 0.1 min, time lter enabled, fragmentor 150 V. System control and data evaluation
were performed with ChemStation for LC/MS software.
Calculation of avonol glycoside content
Flavonol glycoside content was calculated according to the
method provided in the Ginkgo Tablet and Ginkgo Capsule

Please cite this article in press as: Wohlmuth, H., et al., Adulteration of Ginkgo biloba products and a simple method to improve its
detection. Phytomedicine (2014), http://dx.doi.org/10.1016/j.phymed.2014.01.010

G Model
PHYMED-51597; No. of Pages 7

ARTICLE IN PRESS
H. Wohlmuth et al. / Phytomedicine xxx (2014) xxxxxx

Table 1
Content ( w/w) of avonol aglycones quercetin, kaempferol and isorhamnetin in unhydrolysed and hydrolysed samples of ginkgo leaf and retail products (calculated as
quercetin, as per USP36-NF31).
Samplea , b

Quercetin
Unhydrolysed

L1-AU
L2-NZ
L3-CH
L4-CH
L5-CH
T1-AU
T2-AU
T3-AU
T4-DK
C1-AU
C2-AU
C3-AU
C4-DK
a
b

0.000
0.000
0.000
0.000
0.000
0.000
0.036
2.477
0.000
0.013
7.272
7.421
0.027

0.000
0.000
0.000
0.000
0.000
0.000
0.006
0.006
0.000
0.000
0.034
0.124
0.000

Kaempferol
Hydrolysed
4.292
3.715
0.961
0.909
0.869
13.643
3.718
5.593
13.062
12.635
15.416
14.425
17.317

0.046
0.026
0.010
0.013
0.031
0.099
0.014
0.019
0.132
0.064
0.060
0.226
0.102

Isorhamnetin

Unhydrolysed
0.000
0.000
0.000
0.000
0.000
0.000
0.000
1.835
0.000
0.000
7.500
2.230
0.000

0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.015
0.000
0.000
0.030
0.111
0.000

Hydrolysed
4.291
2.833
1.383
1.470
1.349
15.311
4.624
5.991
13.835
15.361
18.120
17.094
26.522

0.061
0.022
0.020
0.032
0.048
0.122
0.005
0.001
0.116
0.060
0.083
0.368
0.140

Unhydrolysed
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.010
0.000
0.000
0.113
0.014
0.000

0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.002
0.000
0.000
0.008
0.002
0.000

Hydrolysed
0.592
0.422
0.416
0.362
0.375
3.449
1.287
1.112
2.832
4.229
2.825
1.870
4.765

0.008
0.025
0.006
0.004
0.011
0.036
0.006
0.007
0.022
0.051
0.011
0.052
0.069

L, leaf; T, tablet; C, capsule.


Country of origin (leaf) or purchase (retail products): AU, Australia; NZ, New Zealand; CH, China; DK, Denmark.

monographs in USP36-NF31. Briey, this method involves the acid


hydrolysis of an extract of the sample, quantication of quercetin,
kaempferol and isorhamnetin by HPLC against reference standards,
and calculation of the quantity (in mg) of each avonol glycoside
in the sample using the formula:
Glycosidequantity =

r 
u

rs

Cs F 50

where ru is the peak area of the relevant aglycone in the sample


solution, rs is the peak area of the aglycone in the corresponding
standard solution, Cs is the concentration (mg/mL) of the aglycone
in the standard solution, and F is the factor used to convert each
aglycone into a avonol glycoside with a mean molecular mass
of 756.7 (2.504 for quercetin, 2.588 for kaempferol, and 2.437
for isorhamnetin). The total quantity of avonol glycosides in
the sample is calculated by summing the values for quercetin,
kaempferol and isorhamnetin glycosides.
Results
The content of quercetin, kaempferol and isorhamnetin in unhydrolysed and hydrolysed extracts of ve ginkgo leaf samples, four
tablet samples and four capsule samples is shown in Table 1. None
of the unhydrolysed leaf samples contained detectable levels of any
of these avonol aglycones. The quercetin content in the hydrolysed
leaf samples varied almost ve-fold (range 0.874.30 w/w), the
kaempferol content more than three-fold (range 1.354.29) and
the isorhamnetin content 1.6-fold (range 0.360.59). The hydrolysed extract of the Australian grown leaf sample L1-AU had the
highest content of all three avonols (total 9.17), followed by the
New Zealand sample (6.97) and the three leaf samples from China
(2.592.76).
Three of the tablets and two of the capsules contained no or
very low levels of free avonols when not subjected to acid hydrolysis. However, unhydrolysed samples of one tablet (T3-AU) and
two capsules (C2-AU and C3-AU) contained high levels of quercetin
(range 2.487.42) and kaempferol (range 1.847.50), whereas
the same samples contained only very low concentrations (0.01)
of isorhamnetin. As shown in Table 2, the concentration of free
avonol aglycones in these three samples comprised up to 51%
(quercetin) and 41% (kaempferol) of the concentration measured
in the same products post hydrolysis. Whether or not high levels of
quercetin and kaempferol were present pre hydrolysis was obvious
by visual inspection of HPLC chromatograms (Fig. 1).
The avonol glycoside content of the retail products, calculated
both pre and post hydrolysis using the formula provided for

Content of Flavonol Glycosides in the USP-NF monographs, is


shown in Table 3. As is evident from this table, the presence of
signicant amounts of avonol aglycones in three of the products
resulted in the avonol glycoside content being overestimated
between 29.0 and 40.9% when applying the pharmacopoeial
method.
As is also shown in Table 3, 75% of the products analysed contained within 10% of their label claim for avonol glycosides,
when the pharmacopoeial method was used. One product contained 14% less and another 34% more than the label claim by this
method.
Fig. 2 shows the ratios between kaempferol and quercetin and
between isorhamnetin and quercetin in the hydrolysed samples of
retail product. The minimum ratios stipulated in the monographs
are shown as horizontal bars (kaempferol to quercetin NLT 0.7;
isorhamnetin to quercetin NLT 0.1). All products conformed to the
relevant USP-NF monographs with respect to avonol ratios after
acid hydrolysis.
All samples were examined for the presence of genistein by
LCDAD and LCMS using positive selective ion monitoring (Fig. 3).
Genistein was identied only in the three samples (T3-AU, C2-AU
and C3-AU) shown to contain high levels of free quercetin and
kaempferol (Table 2). The presence of genistein was conrmed by
re-analysis by LCDAD and LCMS after spiking with a genistein reference standard. Genistein was also readily observable in the HPLC
chromatogram as a peak located between the peaks for quercetin
and kaempferol (Fig. 1).
Discussion
Ginkgo extract is used in numerous dietary supplements, botanicals, herbal medicinal products and complementary medicines
around the world. Being a high-value botanical commodity, ginkgo
extract is a target for economically motivated adulteration, and suspected adulteration with the avonols quercetin and kaempferol,
the avonol glycoside rutin (quercetin 3-rutinoside) or with other
plant extracts containing avonols has been reported (Chandra
et al., 2011; Harnly et al., 2012; He and Roller, 2011; Liu et al., 2005;
Sloley et al., 2003; Tawab, 2010; van Beek and Montoro, 2009).
The pharmacopoeial methods for determining the content of
avonol glycosides in ginkgo extract and products differ somewhat
between the USP-NF and BP/EP monographs, but in both cases the
glycoside content is calculated from the content of free avonol
aglycones, quantied by HPLC following acid hydrolysis. The fact
that avonol glycosides are not quantied directly, but instead calculated based on the aglycone content after hydrolysis, means that

Please cite this article in press as: Wohlmuth, H., et al., Adulteration of Ginkgo biloba products and a simple method to improve its
detection. Phytomedicine (2014), http://dx.doi.org/10.1016/j.phymed.2014.01.010

G Model

ARTICLE IN PRESS

PHYMED-51597; No. of Pages 7

H. Wohlmuth et al. / Phytomedicine xxx (2014) xxxxxx

4
mA U

A-pre

50

40

30

20

10

10

15

mA U

20

25

35

m in

A-post

50

30

40

30

20

10

10

15

20

25

30

35

m in

10

15

20

25

30

35

m in

mA U

B-pre

50

40

30

20

10

0
5
mA U

50

B-post

40

30

20

10

10

15

20

25

30

35

m in

Fig. 1. HPLC chromatograms of adulterated (A) and unadulterated (B) ginkgo products pre and post hydrolysis. (Q) quercetin, (G) genistein, (K) kaempferol, (I) isorhamnetin.

Please cite this article in press as: Wohlmuth, H., et al., Adulteration of Ginkgo biloba products and a simple method to improve its
detection. Phytomedicine (2014), http://dx.doi.org/10.1016/j.phymed.2014.01.010

G Model

ARTICLE IN PRESS

PHYMED-51597; No. of Pages 7

H. Wohlmuth et al. / Phytomedicine xxx (2014) xxxxxx

Table 2
Free avonol aglycones and genistein in unhydrolysed ginkgo products. Concentration of free avonols in unhydrolysed samples is shown as percentage of the concentration
in hydrolysed samples. Presence (+) or absence () of genistein in unhydrolysed samples is shown.
Sample

Type

Quercetin

Kaempferol

Isorhamnetin

Genistein

L1-AU
L2-NZ
L3-CH
L4-CH
L5-CH
T1-AU
T2-AU
T3-AU
T4-DK
C1-AU
C2-AU
C3-AU
C4-DK

Leaf
Leaf
Leaf
Leaf
Leaf
Tablet
Tablet
Tablet
Tablet
Capsule
Capsule
Capsule
Capsule

0.0
0.0
0.0
0.0
0.0
0.0
1.0
44.3
0.0
0.1
47.2
51.4
0.2

0.0
0.0
0.0
0.0
0.0
0.0
0.0
30.6
0.0
0.0
41.4
13.0
0.0

0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.9
0.0
0.0
4.0
0.8
0.0

+
+

Table 3
Flavonol glycoside content and percentage of label claims for ginkgo retail products, based on USP-NF formula for Content of Flavonol Glycosides, calculated pre and post
acid hydrolysis.
Product

T1-AU
T2-AU
T3-AU
T4-DK
C1-AU
C2-AU
C3-AU
C4-DK

Calculated avonol glycoside content


using USP-NF formula (mg/dose unit)

Calculated avonol glycoside content


as percentage of label claim (%)

Pre hydrolysis

Post hydrolysis

USP-NF method (post hydrolysis)

Due to free
aglycones (pre
hydrolysis)

0.00
0.04
9.47
0.00
0.02
15.85
9.09
0.01

29.29
10.91
27.80
26.24
37.95
38.71
31.40
20.65

101.7
102.0
96.5
109.3
94.6
134.4
109.0
86.0

0.0
0.4
32.9
0.0
0.0
55.0
31.6
0.1

adulteration with avonol aglycones may not be detectable using


the current pharmacopoeial methods. In fact, the inadequacy of
calculating the avonol glycoside content in ginkgo solely on the
basis of the aglycone content determined post hydrolysis has previously been identied by other workers (Liu et al., 2005; Sloley
et al., 2003).
As an aid to detecting adulteration of ginkgo extracts, the
USP-NF monographs for Powdered Ginkgo Extract (but not the
equivalent EP/BP monograph), Ginkgo Tablets and Ginkgo Capsules include specications for the relative abundance of quercetin,
kaempferol and isorhamnetin after acid hydrolysis. These criteria
are provided in Identication test B in the monograph for Powdered

Fig. 2. Ratios between kaempferol and quercetin (solid bars) and between isorhamnetin and quercetin (patterned bars) peak areas in hydrolysed samples of tablets and
capsules (see text for details).

Overestimation of avonol
glycoside content using
USP-NF post hydrolysis
method (%)

0.0
0.4
34.0
0.0
0.0
40.9
29.0
0.1

Ginkgo Extract but form part of Identication test A in the monographs for Ginkgo Tablets and Ginkgo Capsules. Since November
2011, the USP-NF monographs have stipulated a kaempferol to
quercetin HPLC peak ratio of not less than 0.7 and an isorhamnetin to quercetin peak ratio of not less than 0.1 (United States
Pharmacopeial Convention, 2013). However, while this test may
assist in the detection of some types of adulteration, it cannot effectively detect adulteration with free avonol aglycones.
Adulteration of ginkgo products
In the present study, we have shown that ve ginkgo leaf samples, sourced from three different countries, contained levels of
quercetin-, kaempferol- and isorhamnetin glycosides that varied
up to ve-fold between samples. However, none of these leaf samples contained detectable levels of free quercetin, kaempferol or
isorhamnetin aglycones. Consistent with the virtual absence of
free avonols in the crude raw material, ve retail products with
ginkgo extract as the sole active ingredient contained no or negligible amounts of free avonols. One of these products (T1-AU) was
Tebonin (Schwabe Pharmaceuticals), which as its active ingredient contains EGb761 , arguably the classic gold standard ginkgo
leaf extract (for which Schwabe holds a worldwide patent). These
results indicate that neither crude ginkgo leaf nor quality ginkgo
leaf extracts would be expected to contain free avonols, except in
very low concentrations in some cases.
In contrast, three other retail products (T3-AU, C2-AU and C3AU) contained signicant levels of free quercetin (2.57.4 w/w)
and kaempferol (1.87.5). The levels of isorhamnetin in these
products were considerably lower, ranging from 0.01 to 0.11.
The same three products were shown to contain the isoavone
genistein (4 ,5,7-trihydroxyisoavone). Isoavones occur predominantly in leguminous plants (family Fabaceae), but have been

Please cite this article in press as: Wohlmuth, H., et al., Adulteration of Ginkgo biloba products and a simple method to improve its
detection. Phytomedicine (2014), http://dx.doi.org/10.1016/j.phymed.2014.01.010

G Model
PHYMED-51597; No. of Pages 7

ARTICLE IN PRESS
H. Wohlmuth et al. / Phytomedicine xxx (2014) xxxxxx

reported from 60 other, mostly angiosperm families (Lapcik, 2007).


In addition to the Fabaceae, genistein has been reported from the
monocotyledonous family Iridaceae and the dicotyledonous families Moraceae, Rosaceae, Myristicaceae and Rutaceae (Reynaud
et al., 2005). Isoavones including genistein have never been
reported from Ginkgo biloba plant material, but there is one published report of genistein isolated from a Chinese ginkgo leaf extract
(Wang et al., 2007). The chemistry of ginkgo leaf and leaf extracts is
exceedingly well known (van Beek, 2002; van Beek and Montoro,
2009), and we believe there can be little doubt that this report
concerns an adulterated ginkgo extract.
The presence of genistein combined with signicant levels of
free quercetin and kaempferol in three of the retail products
analysed provides compelling evidence of adulteration of these
products (all of which listed Ginkgo biloba leaf extract as the sole
active ingredient). While we have not been able to identify the
adulterant, an extract containing both quercetin and kaempferol
(but not isorhamnetin) as well as genistein (or their glycosides)
would appear most likely. One such candidate is Styphnolobium
japonicum (L.) Schott. (syn. Sophora japonica L.), a well-known Chinese medicinal plant known as Japanese pagoda or Chinese scholar
tree. Two botanical drugs derived from this plant are listed in the
Pharmacopoeia of the Peoples Republic of China: huaihua, comprising the dried ower and ower bud, and huaijiao, which consists
of the dried fruit (Chinese Pharmacopoeia Commission, 2005). The
pericarp of this species contains a number of genistein, quercetin
and kaempferol glycosides (Qi et al., 2007), and it has previously
been identied as a likely adulterant of ginkgo extracts (Chandra
et al., 2011; He and Roller, 2011; Tawab, 2010). The adulterated
ginkgo products analysed by us contained genistein, quercetin
and kaempferol in free aglycone form, so if Styphnolobium japonicum were indeed the adulterant, the extract would likely have
undergone processing leading to the hydrolysis of the avonoid
glycosides.
Thus, we have identied three adulterated ginkgo retail products, all of which conformed to the relevant USP-NF monographs
(Ginkgo Tablets or Ginkgo Capsules) in terms of ratios of HLPC peak
areas between kaempferol and quercetin and between isorhamnetin and quercetin in hydrolysed samples (Fig. 2).
Improved quality control for ginkgo
According to the respective USP-NF monographs, ginkgo tablets
and capsules must be prepared from Powdered Ginkgo Extract that
contains between 22.0 and 27.0% avonol glycosides. The adulterated products analysed by us contained signicant levels of free
quercetin and kaempferol. Because the pharmacopoeial method
for calculating avonol glycoside content is based on the quantity of avonol aglycones measured after acid hydrolysis, the actual
avonol glycoside content in these products was overestimated by
between 29% and 41% (Table 3). In other words, these products
did not meet the pharmacopoeial standard for avonol glycoside
content, but this could not be detected by the prescribed pharmacopoeial method. Whether avonoids are present in aglycone or
glycoside form can have profound impact on their pharmacokinetic
and pharmacodynamic characteristics (Murato and Terao, 2003;
Walle, 2004), with obvious potential implications for the safety and
efcacy of the product.
In order to address this problem, we propose a simple modication of the pharmacopoeial test for avonol glycosides. This
involves assaying for avonol aglycone content (as per the pharmacopoeial monograph) before and after acid hydrolysis and basing
the calculation of avonol glycoside content on the difference
between pre and post hydrolysis aglycone content:
A(glycosides) = A(post-hydrolysis) A(pre-hydrolysis)

Fig. 3. Genistein in adulterated ginkgo product. (A) UV spectrum showing max 261;
(B) mass spectrum showing M + 1 = 271.

where A(glycosides) is the amount of avonol aglycones derived from


glycosides in the extract or product (and the value the pharmacopoeial test aims to determine), A(post-hydrolysis) is the total amount
of avonol aglycones after acid hydrolysis (representing the sum of
free aglycones plus aglycones from glycosides) and A(pre-hydrolysis) is
the amount of free avonol aglycones in the unhydrolysed extract
or product.
Applying this modied test ensures that only the amount of
avonol aglycones that exist in glycoside form in the extract or
product is used in the calculation of avonol glycoside content. Any
free aglycones present in the extract or product are not included
in this calculation. This approach prevents the overestimation of
avonol glycoside content that will result by applying the current pharmacopoeial method to gingko extracts or products that
have been adulterated with free avonol aglycones (illustrated by
three of the products we analysed). The proposed change to the
pharmacopoeial method, while adding an additional step, is both
cheap and simple and provides for an accurate determination of
the avonol glycoside content. Accordingly, we believe this change
to the method would represent a signicant improvement to the
USP-NF monographs for Powdered Ginkgo Extract, Ginkgo Tablets
and Ginkgo Tablets, and to equivalent monographs in other pharmacopoeias, including the EP and the BP.
Our analysis of ve commercial samples of ginkgo leaf grown
in three different countries found no evidence (within the limits
of detection) of quercetin, kaempferol or isorhamnetin in aglycone form. Free aglycones have been reported from unhydrolysed
ginkgo leaf, but only in low concentrations (Song et al., 2010; van
Beek and Montoro, 2009). These aglycones were absent from the
ve unadulterated products tested with the exception of quercetin,
which was detected in three of the products at very low levels
(0.0130.036 w/w). In unadulterated material, an increase in
aglycone levels (with accompanying decrease in avonol glycoside
levels) is indicative of degradation during extraction, formulation
or storage and thus undesirable (Sticher et al., 2000). On this basis
we also suggest that a suitable maximum level for free avonol
aglycones be set for ginkgo leaf extract, e.g. 0.5% w/w. Such a
limit would further contribute to the quality assurance of ginkgo
extracts, including limiting adulteration with avonol aglycones
from extraneous sources.
The presence of the isoavone genistein, a compound not occurring naturally in Ginkgo biloba, in the three products that were
adulterated with avonol aglycones suggests that the adulterant
was a plant extract that contains avonols as well as genistein. We
therefore also recommend that manufacturers further improve the
rigour of their quality control processes for ginkgo raw materials

Please cite this article in press as: Wohlmuth, H., et al., Adulteration of Ginkgo biloba products and a simple method to improve its
detection. Phytomedicine (2014), http://dx.doi.org/10.1016/j.phymed.2014.01.010

G Model
PHYMED-51597; No. of Pages 7

ARTICLE IN PRESS
H. Wohlmuth et al. / Phytomedicine xxx (2014) xxxxxx

(especially extracts) by assaying for the presence of genistein and


rejecting as adulterated ginkgo materials found to contain this
extraneous compound.
Conclusions
The need for ongoing vigilance in relation to the quality of herbal
medicinal products and botanicals has been highlighted by our
nding that three out of eight ginkgo retail products tested showed
clear evidence of adulteration. Pharmacopoeial monographs for
botanical raw materials and dose forms, such as those in the USPNF, EP and BP, play an essential role in assuring the quality of such
products. Here we have demonstrated that the tests currently prescribed in these monographs do not provide for the detection of
adulteration of ginkgo extracts and products with free avonol
aglycones.
As a solution to this specic problem, we have proposed a simple
modication of the USP-NF test for Content of Flavonol Glycosides
that allows for the accurate estimation of these compounds, thus
avoiding the overestimation that will result from the presence of
free avonol aglycones (likely of extraneous origin) when applying
the current test.
We believe that the proposed test, along with the setting of a
maximum acceptable level for free avonol aglycones and testing for the presence of genistein (an extraneous compound) would
make a signicant contribution to the quality assurance of gingko
leaf extracts and products.
References
Blumenthal, M., 2003. The ABC Clinical Guide to Herbs. American Botanical Council,
Austin, TX.
Bone, K., Mills, S., 2013. Principles and Practice of Phytotherapy: Modern Herbal
Medicine, second ed. Churchill Livingstone, Edinburgh.
British Pharmacopoeia Commission, 2012. British Pharmacopoeia. The Stationary
Ofce, London.
Chandra, A., Li, Y., Rana, J., Persons, K., Hyun, C., Shen, S., Mulder, T., 2011. Qualitative
categorization of supplement grade Ginkgo biloba leaf extracts for authenticity.
Journal of Functional Foods 3, 107114.
Chinese Pharmacopoeia Commission, 2005. Pharmacopoeia of the Peoples Republic
of China, English ed. Peoples Medical Publishing House, Beijing.
Euromonitor International Ltd., 2009. Country Sector Brieng, Vitamins and Dietary
Supplements. Euromonitor International Ltd., London.
Harnly, J.M., Luthria, D., Chen, P., 2012. Detection of adulterated Ginkgo biloba
supplements using chromatographic and spectral ngerprints. J AOAC Int 95,
15791587.

He, K., Roller, M., 2011. Identication of adulterated compounds in G. biloba extract.
In: Dayan, N., Kromidas, L. (Eds.), Formulating, Packaging and Marketing of Natural Cosmetic Products. Wiley, Hoboken, NJ, pp. 378385.
Khan, I.A., 2006. Issues related to botanicals. Life Sciences 78, 20332038.
Koch, E., Noldner, M., Leuschner, J., 2013. Reproductive and developmental toxicity
of the Ginkgo biloba special extract EGb 761 in mice. Phytomedicine 21, 9097.
Krenn, L., Bilia, A.R., do Ceu Costa, M., Hook, I., Steinhoff, B., Wegener, T., 2013. Now
Ginkgo 10 years after Cimicifuga? Phytomedicine 21, 9899.
Lapcik, O., 2007. Isoavonoids in non-leguminous taxa: a rarity or a rule? Phytochemistry 68, 29092916.
Lindstrom, A., Ooyen, C., Lynch, M.E., Blumenthal, M., 2013. Herb supplement sales
increase 5.5% in 2012: herbal supplement sales rise for 9th consecutive year;
turmeric sales jump 40% in natural channel. Herbalgram 99.
Liu, C., Mandal, R., Li, X.F., 2005. Detection of fortication of ginkgo products using nanoelectrospray ionization mass spectrometry. Analyst 130,
325329.
Murato, K., Terao, J., 2003. Antioxidative avonoid quercet implication of its
intestinal absorption metabolism. Archives of Biochemistry Biophysics 417,
1217.
Qi, Y., Sun, A., Liu, R., Meng, Z., Xie, H., 2007. Isolation and purication of avonoid and
isoavonoid compounds from the pericarp of Sophora japonica L. by adsorption
chromatography on 12% cross-linked agarose gel media. Journal of Chromatography A 1140, 219224.
Reynaud, J., Guilet, D., Terreux, R., Lussignol, M., Walchshofer, N., 2005. Isoavonoids
in non-leguminous families: an update. Natural Products Reports 22,
504515.
Sloley, B.D., Tawk, S.R., Scherban, K.A., Tam, Y.K., 2003. Quality control analyses
for ginkgo extracts require analysis of intact avonol glycosides. Journal of Food
and Drug Analysis 11, 102107.
Song, J., Fang, G., Zhang, Y., Deng, Q., Wang, S., 2010. Fingerprint analysis of Ginkgo
biloba leaves and related health foods by high-performance liquid chromatography/electrospray ionization-mass spectrometry. Journal of AOAC International
93, 17981805.
Sticher, O., Meier, B., van Beek, T.A., 2000. The analysis of ginkgo avonoids. In: van
Beek, T.A. (Ed.), Ginkgo biloba. Harwood Academic Publishers, Amsterdam, pp.
179202.
Tawab, M., 2010. Nahrungsergnzungsmittel mit Ginkgo unter der Lupe, Pharmazeutische Zeitung Online. Govi-Verlag.
United States Pharmacopeial Convention, 2013. United States Pharmacopoeia and
National Formulary (USP36-NF31). United States Pharmacopeial Convention,
Rockville, MD.
van Beek, T.A., 2002. Chemical analysis of Ginkgo biloba leaves and extracts. Journal
of Chromatography A 967, 2155.
van Beek, T.A., Montoro, P., 2009. Chemical analysis and quality control of Ginkgo
biloba leaves, extracts, and phytopharmaceuticals. Journal of Chromatography
A 1216, 20022032.
Vlietinck, A., Pieters, L., Apers, S., 2009. Legal requirements for the quality of herbal
substances and herbal preparations for the manufacturing of herbal medicinal
products in the European Union. Planta Medica 75, 683688.
Walker, K.M., Applequist, W.L., 2012. Adulteration of selected unprocessed botanicals in the U.S. retail herbal trade. Economic Botany 66, 321327.
Walle, T., 2004. Absorption and metabolism of avonoids. Free Radical Biology and
Medicine 36, 829837.
Wang, F., Jiang, K., Li, Z., 2007. [Purication and identication of genistein in Ginkgo
biloba leaf extract]. Chinese Journal of Chromatography 25, 509513.

Please cite this article in press as: Wohlmuth, H., et al., Adulteration of Ginkgo biloba products and a simple method to improve its
detection. Phytomedicine (2014), http://dx.doi.org/10.1016/j.phymed.2014.01.010