Anal Bioanal Chem (2009) 394:1661–1669
DOI 10.1007/s00216-009-2823-8
ORIGINAL PAPER
Identification and quantification of glucosinolates
in rapeseed using liquid chromatography–ion trap
mass spectrometry
Silvia Millán & M. Carmen Sampedro & Patricia Gallejones & Ander Castellón &
Maria L. Ibargoitia & M. Aranzazu Goicolea & Ramón J. Barrio
Received: 27 February 2009 / Revised: 20 April 2009 / Accepted: 22 April 2009 / Published online: 12 May 2009
# Springer-Verlag 2009
Abstract A rapid and sensitive method for the speciation and
quantification of glucosinolates in rapeseed is described. The
method combines liquid chromatography (LC) with ion trap
mass spectrometry (ITMS) detection. Electrospray ionization
(ESI) has been chosen as the ionization technique for the on-
line coupling of LC with ITMS. Glucosinolates are extracted
from different rapeseeds with MeOH and the extracts are
cleaned-up by solid phase extraction with Florisil cartridges.
Aqueous extracts are injected into LC system coupled to an
ITMS, leading to accurately quantify eight of the most
important glucosinolates in rapeseed, by MS2 mode and
confirming their structure by MS3 acquisition. All the
glucosinolates found in rapeseeds provide good signals
corresponding to the deprotonated precursor ion [M-H]−.
The method is reliable and reproducible, and detection limits
range from 0.5 nmol g−1 to 3.7 nmol g−1 when 200 mg of
dried seeds of certified reference material are analyzed.
Within-day and between-day RSD percentages range
between 2.4–14.1% and 3.9–16.9%, respectively. The
LC-ESI-ITMS-MS method described here allows for a rapid
assessment of these metabolites in rapeseed without a
desulfatation step. The overall process has been successfully
applied to identify and quantify glucosinolates in rapeseed
samples.
S. Millán : M. C. Sampedro : M. A. Goicolea : R. J. Barrio (*)
Department of Analytical Chemistry, Faculty of Pharmacy,
University of the Basque Country,
01006 Vitoria, Spain
e-mail: r.barrio@ehu.es
P. Gallejones : A. Castellón : M. L. Ibargoitia
Neiker Tecnalia,
c/ Berreaga, 1, 48160 Derio,
Biscay, Spain
Keywords Oily seeds . Glucosinolates .
Liquid chromatography . Ion trap mass spectrometry
Introduction
Rape is extensively cultivated as a valuable source of edible
oil, the seed having an oil content of about 40%.
Furthermore, the cake which remains after the oil is
expelled from seed is widely used as feed for farm animals
and poultry. Although it is an important source of protein
and fiber, the presence of a number of antinutritional factors
[1, 2] the most important of which are glucosinolates, limit
their use. Not only do these compounds reduce animal
intake (because they reduce the palatability of the cake), but
they also interfere with the thyroid function, damage vital
organs or interfere with metabolic processes [3]. Therefore
there is a move towards rapeseed that is low in glucosino-
lates and this has prompted the need for modern analytical
methods for the rapid identification and assessment of
glucosinolates in rapeseed.
The general structure of glucosinolates is characterized
by a β-D-thioglucose group, a sulfonated oxime moiety
and a variable side-chain derived from methionine,
tryptophan or phenylalanine [4]: aliphatic glucosinolates
derived from mainly methionine, but also alanine, leucine
or valine. Indolic glucosinolates are derived from trypto-
phan, while aromatic ones come mainly from phenylala-
nine and tyrosine. More than 120 glucosinolates are
known to occur naturally [5] with different side-chains
(aliphatic, aromatic or indolic) which result in a wide
range of biological activity and polarity of these com-
pounds. Glucosinolates are hydrolyzed by endogenous
thioglucosidases, called myrosinases, to produce a wide
range of degradation products (isothiocyanates, nitriles,
1662 S. Millán et al.

20.0±0.2
epithionitriles, oxazolidine-2-thiones, and thiocyanates)

1.20
4.0
with diverse biological activities [6, 7].

422
259
100
100
100–450
19–21

114

10
Several methods have been applied for the determination

NAS
of total glucosinolates including gas chromatography [8, 9],
X-ray fluorescence [10], colorimetric method [11], capillary

17.9±0.1
16.5–19
electrophoresis [12, 13], amperometric flow analyser, [14]

1.20
4.0
and high performance liquid chromatography [15, 16]. In

447
259
100
100
100–470

121

10
1992 the International Organization for Standardization

GBC
(ISO) published an official method [17] for their determi-
nation and quantification involving HPLC of desulfogluco-

15.05–16.5
sinolates obtained after enzymatic desulfatation. However,

15.6±0.1

1.30
the sample preparation required before chromatographic

4.0
110

10
408
259
100
100
90–430
analysis was complex and tedious. Some methods currently

TROP
feature in the published bibliography which determine
intact glucosinolates by HPLC-UV [18, 19] or LC-MS

14.9±0.1
[20–22] with different types of analyzers eliminating the

1.30
14–15.05

4.0
drawbacks mentioned above, but they are basically semi-

386
259
100
100
90–400

104

10
GBN
quantitative methods.
To date, LC/MS-MS analyses of glucosinolates have
been carried out on a quadrupole ion trap [23], quadrupole

12.8±0.1
time-of-flight [24] or on triple-quadrupole mass spectrom-

1.05
Table 2 Data acquisition parameters and MS2 transitions used in LC/MSMS for the quantification of glucosinolates

4.0
eters [25]. Direct analysis of intact glucosinolates is

463
285
100
100
100–500

125
11–14

10
important, because it can reflect the specificity of each
4OH

glucosinolates. Therefore, in this paper, we offer an

alternative for simultaneous identification/confirmation
9.6±0.2
8.4–11
and quantification in a single analysis of eight of the

1.20
4.0
predominant intact glucosinolates in rapeseed: progoitrin
372
259
100
100
90–400

100

10
GNA

(PRO), epiprogoitrin (EPRO), gluconapoleiferin (GNL),

gluconapin (GNA), 4-hydroxyglucobrassicin (4OH), gluco-
8.3±0.2

brassicanapin (GBN), glucobrassicin (GBC), and gluconas-

6.0–8.4

1.30
turtiin (NAS) using an ion trap mass spectrometer which 4.0
402
259
100
100
90–420

109

10
enables MS n experiments. It is difficult and time-
GNL

consuming to identify and accurately quantify several

components such as glucosinolates in complex sample
4.9±0.1
4.4–6.0

matrices. To date, most of the applied methods with this

1.30
4.0

purpose tend to carry out purification and desulfation

388
259
100
100
90–400

105

10
EPRO

processes [26, 27]. Measurement of desulfoglucosinolates

4.0±0.2
3.5–4.4

Table 1 Individual and total GSL contents in mmol kg−1 in the BCR-
1.20
4.0
388
259
100
100
90–400

105

10

190R certified reference material (200 mg)

PRO

BCR-190R mmol kg−1

To waste

PRO Progoitrin 12.56±0.25

0–3.5

EPRO Epiprogoitrin 0.24±0.02

GNL Gluconapoleiferin 0.53±0.04
GNA Gluconapin 3.89±0.08
Compound stability (%)
Retention time (min)

4OH 4-hidroxiglucobrassicin 3.96±0.23

Trap drive level (%)
Time segment (min)
Target mass (m/z)
Product ion (m/z)

Frag. width (m/z)

GBN Glucobrassicanapin 1.37±0.03

Isol. width (m/z)

Frag. ampl. (V)

GBC Glucobrassicin 0.20±0.01

Cut-off (m/z)
Scan (m/z)

NAS Gluconasturtin 0.51±0.08

total GSL 23.25±0.50
Glucosinolates in rapeseed 1663

Fig. 1 Typical LC-ITMS

chromatogram of glucosinolates
in an extract of rapeseed CRM
(ERM-BC190). Peak numbers
correspond to (1) PRO, (2)
EPRO, (3) GNL, (4) GNA, (5)
4OH, (6) GBN, (7) TROP
(internal standard), (8) GBC and
(9) NAS. The ions monitored are
displayed in the right side of each
trace

do not require the use of ion-pair reagents but the time
required for analysis is significantly increased because of
additional sample processing, including extraction, binding
to Sephadex A-25, enzymatic desulfation, and finally
elution of the desulfoglucosinolates.
The aim of our work is therefore to develop an accurate
and sensitive method for the determination of these
compounds in different rapeseed samples based on LC-
MS-MS technique without any desulfation step. The
combination of HPLC and ion trap mass spectrometry has
provided a good solution for the accomplishment of a
secure identification of the target compounds. ESI is a
powerful tool for identifying highly polar, heat-labile
compounds such as glucosinolates, present as ions in
solution. Therefore, due to the polar nature of the target
compounds, ESI was chosen as the ionization technique for
the on-line coupling of LC with MS, and was found to be
particularly sensitive.
Experimental
Chemicals and reagents
Certified reference materials (ERM-BC366, ERM-BC190,
ERM-BC367) from the European Community Bureau of
1664
Table 3 Linearity and detection limits of LC-ESI-ITMS-MS method for rapeseed certified reference material BCR-190R
Compound Regression equation
Progoitrin ye ¼ e0:162x þ 0:066
Epiprogoitrin ye ¼ e0:099x
0:000
Gluconapoleiferin ye ¼ e0:235x
0:000
Gluconapin ye ¼ e0:549x
0:033
4-hidroxiglucobrassicin ye ¼ e0:232x
0:027
Glucobrassicanapin ye ¼ e0:930x
0:029
Glucobrassicin ye ¼ e0:068x þ 0:002
Gluconasturtin ye ¼ e0:133x
0:000
y: ratio of the peak area of glucosinolates to the peak area of the surrogate standard; x: glucosinolates to surrogate concentration ratio.
Reference (BCR, Brussels, Belgium) were purchased from
LGC Standards (Wesel, Germany) in an aluminum plastic-
laminated sachet sealed under nitrogen. For the analysis of
glucosinolates, certified reference material (CRM) with a
medium content of total glucosinolates (ERM–BC190),
which is 23 mmol kg−1, with certified individual glucosi-
nolate content was used to construct the calibration curves.
Commercially available glucotropaeolin (TROP) as potas-
sium salt from ChromaDex Inc. (Santa Ana, CA, USA) was
used as an internal standard.
The gradient HPLC grade organic solvents methanol,
dichloromethane, n-hexane, and ethyl acetate were
purchased from Scharlau Chemie SA (Barcelona, Spain).
Ammonium acetate was obtained from Panreac (Barcelona,
Spain) and formic acid 99% was purchased from Across
Organics (New Jersey, USA). The ultra-high-purity water
(UHP) was prepared from tap water pre-treated by reverse
osmosis (Elix, Millipore, Bedford, MA, USA) prior to
filtration by a Millipore Milli-Q system.
For the clean-up process Bond elute Florisil (1 g, 6 ml)
cartridges were purchased from Varian, Inc. (Palo Alto, CA,
USA) and all the aqueous extracts were filtered prior to
injection into LC-ITMS system with Nylon syringe filters
13 mm, 0.45 μm from Scharlau Chemie S.A. (Barcelona,
Spain).
Preparation of standards and samples
Prior to use, seeds were dried overnight in an oven at 60°C.
The dried seeds (1 g) were crushed with a mortar to a fine
powder and an aliquot of 200 mg transferred to screw-
capped centrifuge tubes. Immediately after crushing, 2 ml
of methanol at 70°C was added into each tube and then
20 μl of 10 mg ml−1 (22.3 mmol l−1) TROP solution in
methanol was spiked to the matrix and samples, while
being kept in magnetic agitation for 15 min and then
centrifuged at 4,400 rpm for 5 min. The extraction process
was repeated twice and the supernatants were finally
S. Millán et al.
Regression coefficient LOQs (µ mol g−1) LODs (µ mol g−1)
MS/MS mode MS/MS mode
0.9950 0.0044 0.0013
0.9979 0.0018 0.0005
0.9983 0.0038 0.0011
0.9949 0.0123 0.0037
0.9953 0.0025 0.0008
0.9945 0.0043 0.0013
0.9916 0.0020 0.0006
0.9992 0.0048 0.0014
collected and were evaporated to dryness under a stream
of nitrogen. It was reconstituted with 500 μl of methanol.
A Florisil cartridge was activated before use with 5 ml
30% (v/v) dichloromethane in hexane. The supernatant
(500 μl) was mixed with 5 ml 30% (v/v) dichloromethane
in hexane and applied to the cartridge. Interferences were
washed with 5 ml 30% (v/v) dichloromethane in hexane and
the cartridge was aspirated to dryness. The glucosinolates
were then eluted with 5 ml 30% (v/v) ethyl acetate in
methanol. The extract was evaporated again to dryness and
reconstituted with 500 μl of Milli-Q water. The final extract
was filtered through Nylon 0.45 μm filters prior to its
injection in the LC-ITMS system.
Samples were prepared to developed calibrations from
seeds with certified individual glucosinolate content (certified
reference material ERM-BC190, Table 1). Seven aliquots of
200 mg of the CRM were extracted as described above
(corresponding to seven levels of calibration), and the
volumes of the final extracts were reconstituted in the range
of 0.1–10 ml. The values were converted into mmol l−1 of the
sample extracts and the calibration curves were constructed.
Chromatographic conditions
A liquid chromatograph model Agilent 1100 series was
equipped with a binary pump, vacuum degasser, an
autosampler, and a thermostatted column compartment.
The analytes were separated by reversed-phase LC
injecting 20 μl aliquot of glucosinolates extract into a
Tracer Extrasil ODS2 column (25 cm×4.6 mm, 5 μm)
heated at 25°C. The mobile phase was prepared from
30 mmol l−1 ammonium acetate adjusted with formic acid
at pH 5.0 (component A) and methanol (component B).
All solutions were filtered through 0.22-µm filters and
sonicated before their use. The gradient program was:
100%A–0%B for 5 min; increased to 80%A–20%B from 5
to 17 min; and achieving initial conditions at 20 min. The
flow rate was held constant at 0.9 ml min−1, and after each
Glucosinolates in rapeseed
Fig. 2 MS/MS product ion
spectra of 1. PRO, 2. EPRO, 3.
GNL, 4. GNA, 5. 4OH, 6. GBN,
7. TROP (internal standard), 8.
GBC, 9. NAS
run the system allowed to equilibrate. Under these
chromatographic conditions progoitrin (PRO), epiprogoitrin
(EPRO), gluconapoleiferin (GNL), gluconapin (GNA),
4-hydroxyglucobrassicin (4OH), glucobrassicanapin (GBN),
glucotropaeolin (TROP), glucobrassicin (GBC), and
gluconasturtiin (NAS) were eluted at retention times of
4.0±0.2, 4.9±0.1, 8.3±0.2, 9.6±0.2, 12.8±0.1, 14.9±0.1,
15.6±0.1, 17.9±0.1, and 20.0±0.2 min, respectively.
Ion trap MS analysis
The MS analyses of glucosinolates were carried out on an
MS n system consisting of a MSD Trap XCT Plus
1665
spectrometer (Agilent Technologies, Palo Alto, CA,
USA), equipped with a G1948A ESI source. System
control and data analysis was provided by the Agilent LC
Chemstation and by Bruker Daltonics Trap Control and
QuantAnalysis. The ESI source was used and operated in
negative ionization mode. Typical operating conditions
were as follows: drying gas (N2) temperature of 365°C,
12 L min−1 drying gas flow, 50 psi nebulizer gas (N2)
pressure, and 3,000 V of capillary voltage. Data were
acquired with a smart target of 70,000 and a max
accumulation time of 200 ms. First, full-scan MS spectra
were obtained by scanning from 50–500m/z. Then, the
chromatogram was divided into ten time segments as
1666
Fig. 3 MS3 spectra of 1. PRO,
2. EPRO, 3. GNL, 4. GNA, 5.
4OH, 6. GBN, 7. TROP (internal
standard), 8. GBC, 9. NAS
shown in Table 2, and MS2 acquisition of the most
abundant ions in the full-scan MS mode was carried out.
Finally, MS3 acquisition was used to confirm the identity of
the analytes in the rapeseeds.
Results and discussion
Initially, the composition of the mobile phase was the
primary target for a correct separation and ionization of the
analytes. The use of a volatile buffer, compatible with an
LC-MS system, increased the retention of the glucosino-
lates on the column and provided a good separation and
S. Millán et al.
resolution of the compounds in less than 25 min. A gradient
method using aqueous and methanol channels, the first one
containing 30 mM ammonium acetate adjusted with formic
acid at pH5, was found to give the best separation, a good
chromatography peak shape and sensitivity for this analysis
(Fig. 1). The glucosinolates were successfully separated and
quantified by LC-ESI-ITMS-MS using internal standard
calibration. Glucotropaeolin (TROP) formulated as potassi-
um salt was chosen as a suitable internal standard because
of its structural similarity to other analytes and due to the
lack of deuterated standards. Otherwise, it has a retention
time that does not correspond to the other eluted compo-
nents and it does not occur naturally in oily seeds. On the
Glucosinolates in rapeseed 1667

Table 4 Quantification of glucosinolates in rapeseed samples (n=3)

Sample Glucosinolates (µ mol g−1)

PRO EPRO GNL GNA 4OH GBN GBC NAS Total

1 5.08±0.97 0.07±0.03 0.46±0.18 2.14±0.61 5.72±0.56 1.18±0.40 0.40±0.15 0.61±0.10 15.66±0.79

2 2.96±0.09 0.05±0.02 0.29±0.01 1.13±0.07 3.02±0.57 0.59±0.02 0.21±0.02 0.49±0.07 8.74±0.34
3 4.90±0.77 0.11±0.05 0.47±0.22 2.59±0.60 5.01±0.32 0.98±0.32 0.30±0.10 0.54±0.12 14.90±0.64
4 5.74±0.92 0.06±0.02 0.36±0.02 1.91±0.15 5.85±0.45 0.87±0.02 0.18±0.02 0.37±0.17 15.34±0.61
5 3.59±0.19 0.05±0.02 0.32±0.07 0.71±0.02 6.16±0.40 0.52±0.17 0.15±0.07 0.29±0.01 11.80±0.28
6 1.38±0.20 0.02±0.01 0.11±0.01 0.25±0.02 4.23±0.04 0.24±0.02 0.09±0.01 0.19±0.04 6.49±0.12
7 1.90±0.01 0.03±0.01 0.19±0.02 0.65±0.04 3.08±0.10 0.59±0.02 0.11±0.01 0.38±0.02 6.93±0.07
8 1.28±0.15 0.03±0.02 0.14±0.04 0.68±0.04 0.65±0.02 0.45±0.02 0.14±0.01 0.23±0.02 3.59±0.10
9 0.31±0.02 0.02±0.01 0.14±0.02 0.25±0.02 0.30±0.04 0.46±0.12 0.01±0.01 0.34±0.01 1.83±0.08
10 2.24±0.04 0.03±0.01 0.27±0.07 0.84±0.10 0.83±0.10 0.60±0.02 0.10±0.04 0.30±0.01 5.22±0.10

other hand, the coelution with other components is not
important since none of the studied glucosinolates display
the same molecular mass, i.e. quasimolecular ion, as
glucotropaeolin.
The chromatograms were segmented into nine windows
to select the optimum parameters for each compound and to
enhance sensitivity. Flow was diverted to waste initially and
data acquisition triggered 3.50 minutes after the run begun,
with the aim of reducing the contamination of the system.
Identification of the major glucosinolates was estab-
lished through the use of negative ion electrospray MS2
mode with secondary confirmation by MS3 acquisition. The
presence of a sulfonate moiety categorizes glucosinolates as
hydrophilic compounds, which occur in nature in the
anionic form. Consecutively, they are readily ionized in
ESI mode, forming deprotonated molecular ions, [M-H]−,
which were selected to generate MSn spectra.
As mentioned above, once the [M-H]− ion has been
identified by negative ion ESI-ITMS, the strategy to
confirm the identity of unknown glucosinolates involves
examination of their fragmentation behavior by multistage
MS. The MS/MS fragmentation enables structure elucida-
tion of the glucosinolates. The optimum fragmentation
amplitude for each analyte was determined infusing the
extract of the certified reference material isolating each
compound and increasing the fragmentation amplitude until
the precursor ion intensity was reduced to 5–15% of its
major product ion response. The cut-off values (the
minimal value of m/z ratio, so that the ions with smaller
values than these quantities are not trapped by the IT) were
set to the default value (27%) from the precursor ions m/z
ratio. The observed fragment ion at m/z 259 was formed
through rearrangements and loss of R-CNS [28] from the
glucosinolates molecule [M-H-R-CNS]− for the majority of
the analytes except for 4OH, the main product ions of
which resulted in m/z 267 and 285m/z which were due to
the loss of (C6H10O5–H2S) and (C6H10O5–OH) from the
precursor [M-H]− ion, respectively.
The most abundant product ions (259m/z for PRO,
EPRO, GNL, GNA, GBN, TROP, GBC, NAS, and 285m/z
for 4OH) were chosen for the quantitative analyses,
although other characteristic ions can be observed in the
MS/MS spectra (Table 3). The ions with m/z 195
[C6H11O5S]−, m/z 259 [M-H-R-CNS]−, and m/z 275
[M-H-R-CNO]−, which correspond to the fragment ions
from the glycone side chain, were found in all the examined
glucosinolates as they share a common structure (Fig. 2).
The MS3 spectrum from m/z 259 and 285 ions, selected
as precursors, displayed a common fragment ion m/z 97
corresponding to the sulfate group HSO4−. Additional
typical ions are m/z 139, 169, 199, and 241 for all
compounds except for 4OH. The ions m/z 139, m/z 169,
and m/z 199 obtained from the m/z 259 fragment ion
correspond to a loss of C4H8O4, C3H6O3, and C2H4O2 from
the sugar, respectively. The latter ion m/z 241 was identified
as the fragment C6H9O8S− (Fig. 3). Therefore, the fragment
product ions mentioned above were considered as diagnostic
product ions of the common moiety of glucosinolates.
Evaluation of the analytical method
Evaluation of the analytical method was carried out with
aqueous extracts at different dilutions of rapeseed certified
reference material (BCR-190R), resulting in good chro-
matographic profiles. Glucotropaeolin (TROP) was selected
as surrogate standard and calibration curves were generated
by plotting the ratio of the peak area of glucosinolates to
that of the surrogate standard as a function of their
concentration ratios. Seven-point calibration plots were
applied over the following ranges: from 0.15–7.7 mmol
1668
l−1 for PRO, 0.003–0.15 mmol l−1 for EPRO, 0.006–
0.33 mmol l−1 for GNL, 0.05–2.4 mmol l−1 for GNA and
4OH, 0.02–0.84 mmol l−1 for GBN, 0.002–0.12 mmol l−1
for GBC, and 0.006–0.31 mmol l−1 for NAS. The ion trap
mass analyzer revealed a linear response in the selected
ranges with good correlation coefficients (Table 3).
The LC-ESI-ITMS-MS quantification and detection
limits were estimated with extracts at different dilutions of
rapeseed certified reference material ERM-BC366, ERM-
BC190, and ERM-BC367, with an increasing content of
individual glucosinolates. The selected quantification method
for real samples was MS/MS because of its high selectivity,
which allows unequivocal identification of target compounds.
Detection limits were calculated for a signal-to-noise ratio as 3
and ranged from 0.5 nmol g−1 to 3.7 nmol g−1 for dried seeds.
Limits of quantification, calculated on the basis of a signal-
to-noise ratio of 10, ranged from 1.8 to 12.3 nmol g−1
(Table 4).
To determine the precision of the LC-ESI-ITMS-MS
method relative standard deviations were calculated on ten
extracts of rapeseed CRM at two different levels of
concentration under the selected conditions. The same
samples were also analyzed over a period of five successive
days to determine the inter-day RSD. In most cases, the
RSD was often less than 10%: for low concentration
(calibration level 1) varying from 2.4 to 14.1% for intra-
day and from 3.9 to 16.9% for inter-day precision and for
high concentration (calibration level 6) the RSD varied
from 1.6 to 10.9% for intra-day and from 6.3 to 10.7% for
inter-day precision.
Real samples analysis
The developed method can provide comprehensive gluco-
sinolates profiles and concentration from a single sample
and simultaneously collect confirmatory spectra of each
analyte identified.
Concentrations of the glucosinolates in ten different
samples of rapeseeds were quantified on the basis of
internal standard calibration plots. The samples come from
a field experiment where different N and S fertilizer doses
were applied. Results for the quantification of glucosino-
lates in oily seeds show that eight glucosinolates of interest
were found in complex matrix samples at different
concentration levels, the most abundant being progoitrin
(PRO), gluconapin (GNA), 4-hydroxiglucobrassicin (4OH),
and glucobrassicanapin (GBN). Table 5 displays the results
of quantitative determination of the most important glucosi-
nolates in oily seeds in these representative samples. It is
noted that the total glucosinolates content varied considerably
for different samples. The highest concentration of total
compounds was found to be 16.08 μmol g−1 and the lowest
corresponded to 1.83 μmol g−1.
S. Millán et al.
Conclusions
The results demonstrate that the proposed method can
determine the total glucosinolate content in different rapeseed
samples. Eight intact glucosinolates, progoitrin, epiprogoitrin,
gluconapoleiferin, gluconapin, 4-hydroyglucobrassicin,
glucobrassicanapin, glucobrassicin, and gluconasturtiin were
detected and quantified by direct analysis LC-ESI-IT MS-MS
in negative ion mode on the basis of a reproducible
fragmentation of glucosinolates to a sulfated glucose anion,
except for 4OH the characteristic fragment ion of which is
probably formed by cleavage of thio-glucose after OH-
rearrangement from the sugar moiety [24]. The m/z 259
fragment ion is also formed for 4OH, but this pathway is not
favored for this compound. Indeed, the fragmentation process
described for 4OH occurs for the other compounds but the
resulting fragment ion intensities are weak. Sensitivity of the
method is sufficient for the quantification of the analytes in
the real samples.
Acknowledgements The authors would like to thank the Central
Service of Analysis—Araba Campus of the University of the Basque
Country (SGiker) for its excellent technical assistance.
References
1. Fenwick GR, Heaney RK, Mullin WJ (1983) Glucosinolates and
their breakdown products in food and food plants. Crit Rev Food
Sci Nutr 18:123–201
2. Brand TS, Smith N, Hoffman LC (2007) Anti-nutritional factors
in canola produced in the Western and Southern Cape areas of
South Africa. S Afr J Anim Sci 37:45–50
3. Tripathi MK, Mishra AS (2006) Glucosinolates in animal
nutrition: a review. Anim Feed Sci Tech 132:1–27
4. Graser G, Schneider B, Oldham NJ, Gershenzon J (2000) The
methionine chain elongation pathway in the biosynthesis of
glucosinolates in Eruca sativa (Brassicaceae). Arch Biochem
Biophys 378:411–419
5. Chen S, Andreasson E (2001) Update on glucosinolate metabolism
and transport. Plant Physiol Biochem 39:743–758
6. Kliebenstein DJ, Kroymann J, Brown P, Figuth A, Pedersen D,
Gershenzon J, Mitchell-Olds T (2001) Genetic control of natural
variation in Arabidopsis glucosinolate accumulation. Plant Physiol
126:811–825
7. Halkier BA, Gershenzon J (2006) Biology and biochemistry of
glucosinolates. Annu Rev Plant Biol 57:303–333
8. Slominski BA, Campbell LD (1987) Gas chromatographic
determination of indole glucosinolates—a reexamination. J Sci
Food Agr 40:131–143
9. Daxenbichler ME, Spencer GF, Carlson DG, Rose GB, Brinker AM,
Powell RG (1991) Glucosinolate composition of seeds from 297
species of wild plants. Phytochemistry 30:2623–2638
10. Schnug E, Haneklaus S, Zhao F, Evans EJ (1992) Relations
between total sulfur and glucosinolate content in rapeseed -
calibration of the X-ray fluorescence (X-RF) method. Fett Wiss
Technol 94:246–249
11. Nasirullah, Krishnamurthy MN (1996) A method for estimating
glucosinolates in mustard/rapeseeds and cake. J Food Sci Tech
33:498–500
Glucosinolates in rapeseed
12. Karcher A, El Rassi Z (1999) Capillary electrophoresis of
glucosinolates and their degradation products. Electrophoresis
20:3181–3189
13. Bringmann G, Kajahn I, Neusuess C, Pelzing M, Laug S, Unger M,
Holzgrabe U (2005) Analysis of the glucosinolate pattern of
Arabidopsis thaliana seeds by capillary zone electrophoresis coupled
to electrospray ionization-mass spectrometry. Electrophoresis
26:1513–1522
14. Tsiafoulis CG, Prodromidis MI, Karayannis MI (2003) Development
of a flow amperometric enzymatic method for the determination of
total glucosinolates in real samples. Anal Chem 75:927–934
15. Kaushik N, Agnihotri A (1999) High-performance liquid
chromatographic method for separation and quantification of
intact glucosinolates. Chromatographia 49:281–284
16. Wade KL, Garrard IJ, Fahey JW (2007) Improved hydrophilic
interaction chromatography method for the identification and
quantification of glucosinolates. J Chromatogr A 1154:469–472
17. ISO 9167-1. Rapeseed-Determination of glucosinolates content.
Part 1., Ed. International Organization for Standardization (ISO),
Geneva (1992)
18. Zrybko CL, Fukuda EK, Rosen RT (1997) Determination of
glucosinolates in domestic and wild mustard by high-performance
liquid chromatography with confirmation by electrospray mass
spectrometry and photodiode-array detection. J Chromatogr A
767:43–52
19. Szmigielska AM, Schoenau JJ, Levers V (2000) Determination of
glucosinolates in canola seeds using anion exchange membrane
extraction combined with the high-pressure liquid chromatography
detection. J Agric Food Chem 48:4487–4491
20. Mellon FA, Bennett RN, Holst B, Williamson G (2002) Intact
glucosinolate analysis in plant extracts by programmed cone
voltage electrospray LC/MS: performance and comparison with
LC/MS/MS methods. Anal Biochem 306:83–91
21. Lee K-C, Cheuk M-W, Wan C, Lee AW-M, Zhao Z-Z, Jiang Z-H,
Cai Z (2006) Determination of glucosinolates in traditional
1669
Chinese herbs by high-performance liquid chromatography and
electrospray ionization mass spectrometry. Anal Bioanal Chem
386:2225–2232
22. Lee K-C, Chan W, Liang Z, Liu N, Zhao Z, Lee AW-M, Cai Z
(2008) Rapid screening method for intact glucosinolates in
Chinese medicinal herbs by using liquid chromatography coupled
with electrospray ionization ion trap mass spectrometry in negative
ion mode. Rapid Commun Mass Spectrom 22:2825–2834
23. Cataldi TRI, Rubino A, Lelario F, Bufo SA (2007) Naturally
occurring glucosinolates in plant extracts of rocket salad
(Eruca sativa L.) identified by liquid chromatography coupled
with negative ion electrospray ionization and quadrupole ion-
trap mass spectrometry. Rapid Commun Mass Spectrom
21:2374–2388
24. Fabre N, Poinsot V, Debrauwer L, Vigor C, Tulliez J, Fouraste I,
Moulis C (2007) Characterisation of glucosinolates using electrospray
ion trap and electrospray quadrupole time-of-flight mass spectrometry.
Phytochem Anal 18:306–319
25. Bennett RN, Carvalho R, Mellon FA, Eagles J, Rosa EAS (2007)
Identification and quantification of glucosinolates in sprouts
derived from seeds of wild Eruca sativa L. (salad rocket) and
Diplotaxis tenuifolia L. (wild rocket) from diverse geographical
locations. J Agric Food Chem 55:67–74
26. Tolra RP, Alonso R, Poschenrieder C, Barcelo D, Barcelo J (2000)
Determination of glucosinolates in rapeseed and Thlaspi caerulescens
plants by liquid chromatography-atmospheric pressure chemical
ionization mass spectrometry. J Chromatogr A 889:75–81
27. Fahey JW, Wade KL, Stephenson KK, Chou FE (2003) Separation
and purification of glucosinolates from crude plant homogenates
by high-speed counter-current chromatography. J Chromatogr A
996:85–93
28. Rochfort SJ, Trenerry VC, Imsic M, Panozzo J, Jones R (2008)
Class targeted metabolomics: ESI ion trap screening methods for
glucosinolates based on MSn fragmentation. Phytochemistry
69:1671–1679