Rapid and Efficient Desulfonation Method for the Analysis of Glucosinolates by High-Resolution Liquid Chromatography Coupled with Quadrupole Time-of-Flight Mass Spectrometry

Article abstract of DOI:10.1021/acs.jafc.7b04662

The goal of our present research was to develop a simple and rapid method for the quantitation of desulfoglucosinolates (desulfoGLS) without using column chromatography. The proposed method involves extraction, concentration, incubation of glucosinolates with a sulfatase enzyme, and HPLC analysis. Identification of desulfoGLS in green kohlrabi was performed by LC-HR-ESI-QTOF-MS in positive-ionization mode. A total of 11 desulfoGLS were identified with neoglucobrassicin (3.32 ± 0.05 μmol/g DW) as the predominant indolyl, whereas progoitrin and sinigrin were the major aliphatic desulfoGLS. The levels of the aliphatic desulfoGLS glucoiberin, progoitrin, and glucoerucin at 7 h were found to be 3.6-, 1.9-, and 1.6-fold higher, respectively, than those produced through the conventional method. This technique was successfully applied in the identification of desulfoGLS from cabbage. The developed method has fewer unit operations, has maximum recovery, and is reproducible in the determination of desulfoGLS in a large number of Brassicaceae samples in a short time.

Full text of DOI:10.1021/acs.jafc.7b04662

Subscriber access provided by READING UNIV
Article
A Rapid and Efficient Desulfonation Method for the Analysis
of Glucosinolates by High Resolution Liquid Chromatography
Coupled with Quadrupole Time-of-Flight Mass Spectrometry
Jashbir Singh, Guddadarangavvanahal Jayaprakasha, and Bhimanagouda S. Patil
J. Agric. Food Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jafc.7b04662 • Publication Date (Web): 22 Nov 2017
Downloaded from http://pubs.acs.org on November 22, 2017
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of Agricultural and Food Chemistry is published by the American Chemical
Society. 1155 Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 31
Journal of Agricultural and Food Chemistry
1
2
3
A Rapid and Efficient Desulfonation Method for the Analysis of
Glucosinolates by High Resolution Liquid Chromatography Coupled with
Quadrupole Time-of-Flight Mass Spectrometry
4
Jashbir Singh, Guddadarangavvanahally K. Jayaprakasha,* Bhimanagouda S. Patil*
5
6
7
Vegetable and Fruit Improvement Center, Department of Horticultural Sciences,
Texas A&M University, 1500 Research Parkway, Suite A120, College Station, TX 77845
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
*Corresponding authors
Tel.: +1 979 458 8090; fax: +1 979 862 4522
eꢀmail: bꢀpatil@tamu.edu
Tel.: +1 979 845 3864; fax: +1 979 862 4522
eꢀmail: gkjp@tamu.edu
1
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 2 of 31
33
______________________________________________________________________________
34
35
36
37
38
39
40
41
42
43
44
45
ABSTRACT: The goal of our present research was to develop a simple and rapid method for the
quantitation of desulfoglucosinolates (desulfoGLS) without using column chromatography. The
proposed method involves extraction, concentration, incubation of glucosinolates with sulfatase
enzyme, and HPLC analysis. Identification of desulfoGLS in green kohlrabi was performed by
LCꢀHRꢀESIꢀQTOFꢀMS in positive ionization mode. A total 11 desulfoGLS were identified with
neoglucobrassicin (3.32±0.05 µmol/g DW) as the predominant indolyl, whereas progoitrin and
sinigrin were the major aliphatic desulfoGLS. The levels of aliphatic desulfoGLS such as
glucoiberin, progoitrin, and glucoerucin were found to be 3.6, 1.9 and 1.6 fold higher than the
conventional method, respectively, at 7 h. This technique was successfully applied to identify
desulfoGLS from cabbage. The developed method has fewer unit operations, with maximum
recovery and is reproducible to determine desulfoGLS in a large number of Brassicaceae
samples in a short time.
46
47
48
KEYWORDS: Kohlrabi, cabbage, desulfoGLS, LCꢀHRꢀESIꢀQTOFꢀMS
______________________________________________________________________________
2
ACS Paragon Plus Environment

Page 3 of 31
Journal of Agricultural and Food Chemistry
49
50
ꢀ INTRODUCTION
Glucosinolates (GLS) are biosynthetically derived from amino acids1 and they occur in all plants
of the Capparales order.2, 3 GLS have a wellꢀdefined structure with a side chain (Rꢀgroup) and
Dꢀ
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
glucopyranose as βꢀthioglucoside attached to (Z)ꢀNꢀhydroximine sulfate esters.4, 5 GLS are
generally precursors of isothiocyanates, which are responsible for the pungent taste as well as
many health beneficial properties6, 7 of the Brassica species, such as prevention of cardiovascular
diseases, cancers, neurodegenerative, and chronic diseases.8 For example, Michaud et al.9
reported a significant correlation between cruciferous vegetable consumption and reduced
incidence of bladder cancer. Moreover GLS breakdown products such as sulphoraphane and
indoleꢀ3ꢀcarbinol are released by breakdown of glucoraphanin, 4 and glucobrassicin, 8, and are
inducers of phase I and II enzymes.10, 11
Due to their high polarity of thioglucosides and relatively lower retention in the stationary
phase, the HPLC separation of intact glucosinolates is a challenge.12 HPLC analysis of GLS in
desulfonated form is well accepted and considered an effective method of quantification due to
improved resolution.13 The desulfonation involves the binding of intact GLS on a weak anion
ꢀ
exchange resin and the removal of the SO3 group from the parent moiety by the addition of aryl
sulfatase enzyme by column chromatography.14, 15 After 18ꢀ24 h of incubation, the desulfoGLS
are eluted with milliꢀQ water12, 16 and analyzed by HPLC. Open columns are not reproducible
and give low yields due their differences in partition coefficients, adsorption and desorption
properties.17, 18 Lee et al.19 reported the desulfonation of glucosinolates from turnip after 24 h
incubation with aryl sulfatase on a Sephadex column and identified by LCꢀQTOFꢀMS with a 55
min run time. Recently, the effect of storage, temperature and radiation treatments on GLS levels
of broccoli was reported by a study using a Sephadex column followed by a 45 min LC
method.20 Similarly, Park et al.21 reported the glucosinolate content in kohlrabi using a mini
3
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 4 of 31
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
column containing DEAEꢀSephadex Aꢀ25. The desulfonation was achieved by overnight
incubation with aryl sulfatase, followed by elution and identification by LCꢀMS. Kohlrabi
desulfoGLS identification has been performed using HPLCꢀPDA, typical UV spectra and LCꢀ
ESIꢀMS system.16, 21, 22
A Scifinder search identified more than 190 papers that used column chromatography for the
removal of the sulphate group for the analysis of desulfonated glucosinolates. This method takes
>18ꢀ24 h incubation, followed by elution, concentration, and quantitation/identification. To
facilitate rapid analysis of GLS in a large number of samples, it is critical to improve the sample
preparation technique with fewer unit operations to minimize the recovery loss. The present
study focused on developing a rapid method to analyze GLS from green kohlrabi in their
desulfonated form without using column chromatography. In addition, LCꢀHRꢀESIꢀQTOFꢀMS
was used to characterize desulfoGLS by accurate mass and the possible fragmentation pattern.
The developed method was successfully applied to validate the desulfonation process in GLS
from cabbage. This is the first report on the identification of desulfoGLS using high resolution
accurate mass spectrometry from green kohlrabi.
ꢀ MATERIALS AND METHODS
Plant Materials. Green kohlrabi (Brassica oleracea var. gongylodes) and green cabbage
(Brassica oleracea var. capitata) were obtained from the local supermarket (College station,
TX), chopped, frozen at ꢀ80 °C for 24 h, lyophilized (Labconco FreeZone, Kansas City, MO),
ground to obtain 60 to 80 mesh size powder and stored in ꢀ80 °C until further analysis.
94
95
Chemicals. Aryl sulfatase from Helix pomatia (Type Hꢀ1, 23050 U/g solid), DEAE
Sephadex Aꢀ25, HPLC grade acetonitrile and analytical grade phosphoric acid were purchased
4
ACS Paragon Plus Environment

Page 5 of 31
Journal of Agricultural and Food Chemistry
96
97
from SigmaꢀAldrich (St. Louis, MO). Progoitrin, 2, sinigrin, 3, glucoraphanin, 4, gluconaphin, 5,
glucoerucin, 7, and gluconasturtiin, 10, were purchased from Chromadex (Irvine, CA). HPLC
grade water (resistivity 18.2 mꢀcm) was obtained from Nanopure water purification system
(Barnstead, Dubuque, IA). Sodium acetate purchased from EMD chemicals (Gibbstown, NJ) and
acetic acid was purchased from J. T. Baker Chemical Co. (Phillipsburg, NJ).
98
99
100
101
102
103
104
105
106
107
108
109
110
Sample Preparation. Stock sample was prepared by extracting 7 g of lyophilized green
kohlrabi with 15 mL of boiling methanol/water (70:30, v/v) in water bath at 80 °C for 10 min to
deactivate the endogenous myrosinase enzyme. The extract was cooled in an ice bath and
vortexed for 2 min then sonicated for 30 min, centrifuged at 4480 g for 15 min and passed
through Whatman filter paper No. 1. The residue was reꢀextracted twice with boiling
methanol/water (70:30) (2 × 5 mL) to ensure the complete extraction of glucosinolates. All three
extracts were pooled and concentrated to dryness under vacuum at 40 °C. The residue was
dissolved in 25 mL of 0.1 M sodium acetate buffer (pH 5), passed through 0.45 micron filters
and distributed into different tubes before being stored at ꢀ20 °C prior to optimization of the
desulfonation method.
111
112
113
114
115
116
Optimization of Rapid Desulfonation Method. Effect of Incubation Time. Five
hundred microliters of sample was transferred into individual screwed capped vials and treated
with 80 µL (1.84 units) of aryl sulfatase enzyme solution (1 mg/mL in buffer). The sample
containing vials were vortexed for 30 s and incubated for 1, 4, 7, 10, 13, 16, 19 and 22 h for the
conversion of intact GLS to desulfoGLS. After each incubation period, samples were passed
through 0.45 µm Teflon PTFE syringe filters and subjected to HPLC analysis.
5
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 6 of 31
117
118
119
120
121
122
Effect of Enzyme Volume. Briefly, 500 µL of extracted samples were treated with 10,
20, 40, 80 and 160 µL (equivalent to 0.23, 0.46, 0.92, 1.84 and 3.68 units) of sulfatase enzyme
and the total volume was adjusted to 1 mL with buffer. The resultant mixtures were vortexed for
30 s then incubated for 1 h and 7 h at 25 °C with constant shaking. After the incubation period,
filtered samples were injected to HPLC. The conversion of the intact GLS to desulfoGLS was
analyzed by HPLC.
123
124
125
126
127
128
129
130
131
132
133
134
135
136
Separation and Quantification of DesulfoGLS by HPLC. The above optimized
conditions were subsequently used to analyze desulfoGLS in green kohlrabi. Briefly, 500 µL of
samples were treated with 80 µL sulfatase enzyme and the total volume adjusted to 1 mL with
buffer, vortexed for 30 s followed by incubation at 25 °C for 7 h with constant shaking. The
resultant sample was filtered and subjected to HPLC analysis. Analysis of desulfoGLS was
performed on a 1525 HPLC system (Waters, Milford, MA) equipped with a 2996 photodiode
array detector and 717plus autosampler. Chromatographic separation was performed on a Zorbax
Eclipse Plus C18 ODS (250 mm × 4.6 mm i.d., 5 µm) column (Agilent, Santa Clara, CA). A
gradient mobile phase consisting of (A) 0.03 M phosphoric acid in water and (B)
water/acetonitrile (70:30, v/v) at a flow rate of 0.8 mL/min. The separation was performed as
isocratic 2% B (8 min), 2%−45% B (9 min), 45%−60% B (8 min), 60%−90% B (6 min), 90%
−2% B (3 min), and 6 min isocratic at 2% B. The column was equilibrated 2 min before the next
injection. A 20 µL sample was injected and the desulfoGLS were monitored at 227 nm.
Empowerꢀ2 software was employed for processing the data.
137
138
139
Standard DesulfoGLS Preparation and Calibration. Intact glucosinolates progoitrin,
2, sinigrin, 3, glucoraphanin, 4, gluconaphin, 5, glucoerucin, 7, and gluconasturtiin, 10, were
prepared at 1 mg/mL concentrations in 0.1 M acetate buffer (pH 5). All these individual
6
ACS Paragon Plus Environment

Page 7 of 31
Journal of Agricultural and Food Chemistry
140
141
142
standards were desulfonated by sulfatase enzyme (80 µL) as described above and confirmed by
LCꢀMS. Further, all these compounds were subjected to serial dilutions (100, 50, 25, 12.5, 6.25,
3.12, 1.56, and 0.78 µg/mL) and peak areas were used for preparing the calibration curves.
143
144
145
146
147
148
149
150
Desulfonation by Conventional Method. To validate the levels of desulfoGLS obtained
from the rapid method, conventional desulfonation was carried out according to Park et al.21.
Briefly, the crude extract (500 µL) was loaded into a column (10 × 6 mm) containing 50 mg
DEAEꢀSephadex anion exchange resin, which was previously soaked in 0.1 M sodium acetate
overnight. The column was washed twice with 1 mL of acetate buffer (pH 5). The desulfonation
was carried out by adding sulfatase enzyme (80 µL) on a column and incubated for 18 h at 25
°C. The column was eluted with nanoꢀpure water to obtain desulfoGLS. The eluent was passed
through 0.45 µm PTFE syringe filter and quantified by HPLC as mentioned above.
151
152
153
154
155
156
157
158
159
160
161
162
Identification of Glucosinolates by LC-ESI-HR-QTOF-MS. The LCꢀMS identification
was performed using HRꢀQTOFꢀMS on a maXis impact mass spectrometer (Bruker Daltonics,
Billerica, MA) coupled to a 1290 HPLC (Agilent, Santa Clara, CA). The desulfoGLS were
separated on a Zorbax Eclipse Plus C18 column (100 × 2.0 mm i.d., 3 µm) (Agilent, Santa Clara)
at 65 °C with a flow rate of 100 µL/min using gradient mobile phase consisting of (A) 0.1%
formic acid in water and (B) 0.1% formic acid in water/ acetonitrile (70:30, v/v). Elution of GLS
was performed as follows, 0−2% B (10 min), 2%−45% B (12 min), 45%−60% B (5 min),
60%−90% B (4 min), 90%−0% B (3 min) and 5 min isocratic A (100%). The column was
equilibrated for 2 min before the next injection. MS and broadband collision induced dissociation
(bbCID) data were acquired in the range of m/z 50–2000. Nitrogen was used for both nebuliser
(gas pressure 2.1 bar) and drying gas (flow rate 8.0 L/min). The capillary ion voltage was 4,500
V and temperature of drying gas was 200 °C. The transfer time of the source was 120.8 ꢁs and
7
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 8 of 31
163
164
165
166
167
168
the prepulse storage time 5 ꢁs. The quadrupole MS and bbCID collision energies were set at 8
and 30 eV, respectively. The spectrum acquisition rate was 1.4 Hz. External instrument
calibration was performed with a sodium formate solution containing 1 mM sodium hydroxide in
isopropanol/0.2% formic acid (1:1, v/v) according to the method used in our published work.23
To identify the intact GLS, samples were run in electrospray negative ionization mode using
same chromatographic conditions.
169
170
171
172
173
174
175
176
177
178
179
180
Confirmation of DesulfoGLS. Certain desulfoGLS were not commercially available.
To confirm the peak purity of identified desulfoGLS, HPLC peak fractions were collected
manually and confirmed by high resolution mass spectrometry (HRꢀMS). For instance, six
injections of 50 µL were injected in HPLC and chromatographic separation was carried out as
described above. Each peak fraction was collected from the PDA detector outlet according to the
expected retention time (tR) of desulfoGLS. Total 11 fractions were collected and immediately
analyzed by the LCꢀMS. The desulfoGLS were separated on Zorbax Eclipse Plus C18 column
(100 × 2.0 mm i.d., 3 µm) at 65 °C with a flow rate of 200 µL/min using a gradient mobile phase
consists of (A) 0.1% formic acid in water and (B) 0.1% formic acid in water/acetonitrile (70:30,
v/v). Elution of GLS was performed as follows, solvent A, 100−0% for 0−8 min, 0−100% A for
8ꢀ11 min, then 2 min equilibrated in between the runs. The MS conditions were maintained as
described above.
181
182
183
184
185
Validation of Rapid Method. To validate the developed method for the desulfonation
technique, lyophilized cabbage sample (500 mg) was extracted with 4 mL of 30% aqueous
methanol at 80 °C for 10 min according to the protocol mentioned above. The extract (500 µL)
was treated with sulfatase enzyme (80 µL), vortexed for 30 sec followed by incubation for 7 h at
25 °C with constant shaking. The resultant sample was filtered and subjected to HPLC analysis.
8
ACS Paragon Plus Environment

Page 9 of 31
Journal of Agricultural and Food Chemistry
186
187
188
Statistical Analysis. Data was analyzed by oneꢀway analysis of variance (ANOVA)
using JMP Pro12 software (SAS, NC). Significance difference between means was observed by
a Student’s t test at 5% probability level (P< 0.05).
189
190
ꢀ RESULTS AND DISCUSSION
191
192
193
194
195
The present study reports the development of a simple, rapid and accurate method for the
quantitation of desulfoGLS in kohlrabi and cabbage. The sample preparation and identification
of desulfoGLS was performed without the tedious steps using DEAEꢀSephadex column
chromatography. The developed method involves fewer unit operations and requires 1ꢀ7 h for the
desulfonation.
196
197
198
199
200
201
202
203
204
205
206
207
208
HPLC Separation of Glucosinolates at Different Time Points. Figure 1 shows the
chromatographic separation of eleven intact glucosinolates (GLS) and their desulfoGLS from
kohlrabi at different time points. Intact glucosinolates glucoiberin, 1, progoitrin, 2, sinigrin, 3,
glucoraphanin, 4, gluconaphin, 5, glucoibervirin, 6, glucoerucin, 7, glucobrassicin, 8, 4ꢀmethoxy
glucobrassicin, 9, gluconasturtiin, 10 and neoglucobrassicin, 11, are denoted as GLS 1ꢀ11
(Figure 1A), whereas desulfoGLS 1ꢀ11 represents their desulfonated forms at 7 h (Figure 1B).
Peaks 8, 9, and 11 eluted at retention times (tR) of 31.9, 33.8, and 37.9 min, respectively, were
tentatively identified as indolyl desulfoGLS. The peaks eluted at tR 10.6, 16.2, 18.2, 18.5, 22.9,
24.4 and 27.8 min were identified as aliphatic desulfoGLS 1, 2, 3, 4, 5, 6 and 7, respectively
(Figure 2). It was observed that for aromatic and indolyl peaks, 8ꢀ11, 70ꢀ90% were converted
into their desulfonated form within 1 h. After 4 h, complete conversion had occurred. Aliphatic
GLS peaks, 1ꢀ7, conversion was completed in 7 h. The desulfonation reaction was monitored for
22 h to compare the results with the conventional method.
9
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 10 of 31
209
210
211
212
213
214
215
216
217
218
219
220
221
222
Effect of Enzyme Volume on Desulfonation. For enzymatic assays, incubation time and
enzyme volume are critical parameters to obtain end products.13, 24 A low volume may lead to
incomplete desulfonation and high enzyme volume may cause a loss of desulfoGLS, which
results in inaccurate quantitation.13 The effect of enzyme volume on desulfonation of GLS is
shown in Figure 3. The results demonstrated that with an increase in enzyme volume, the rate of
desulfonation increased. For instance, indolyl neoglucobrassicin, 11 and aliphatic progoitrin, 2,
desulfoGLS levels increased more than two fold at 40 µL enzyme concentration as compared to
10 µL. The rest of the compounds gradually increased with enzyme volume from 10ꢀ80 µL,
while at higher enzyme volume (80ꢀ160 µL) did not affect yields significantly. Hennig et al.25
reported that at higher enzymatic concentration, the peak intensities of indolyl GLS were
reduced, while at lower concentration incomplete desulfonation of aliphatic GLS (glucoiberin, 1
and glucoraphanin, 4 occurred. Similarly, our results showed lower content of indolyl derivatives
with 160 µL enzyme than with 80 µL. On that basis, we concluded that 80 µL enzyme volume
was sufficient for complete desulfonation of glucosinolates.
223
224
225
226
227
228
229
230
231
Comparison of Conventional and Rapid Desulfonation Methods. The levels of
individual desulfoGLS in kohlrabi were quantified by existing DEAE Sephadex column to
compare the results with the rapid method. Figure 4 shows the levels of individual desulfoGLS
obtained from the rapid optimized method at 7 h and conventional method at 18 h. Seven
desulfoGLS, 1ꢀ7, showed a significantly higher value in the rapid method. The aliphatic
glucosinolates, 3, 4 and 5 had slightly higher 11%, 4% and 21% respectively than the
conventional method. For progoitrin, 2, and glucoerucin, 7, showed 1.9 and 1.6 fold higher
levels, respectively. Similarly, glucoiberin, 1, was 3.8 fold higher than conventional method. The
aromatic and indolyl desulfoGLS, 8ꢀ10, showed no significant differences in their levels.
10
ACS Paragon Plus Environment

Page 11 of 31
Journal of Agricultural and Food Chemistry
232
233
234
235
236
However, neoglucobrassicin, 11, had a significantly higher value 3.65±0.22 µmol/g DW in the
conventional method as compare to the rapid method (3.32±0.06 µmol/g DW). The lower levels
of desulfoGLS using the column may be due to incomplete elution from the resin or interaction
between the anionic resin and the compounds. Thus, a robust and reproducible method is needed
for rapid desulfonation and quantification of GLS.
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
Characterization of Glucosinolates by LC-HR-ESI-QTOF-MS. LCꢀMS/MS analysis
using quadrupole time of flight is considered a reliable and powerful analytical tool for the
identification of GLS, especially when there is a lack of reference compounds.26 In LCꢀHRꢀESIꢀ
QTOFꢀMS, target ions can be detected with sufficient sensitivity within a narrow retention time.
DesulfoGLS were characterized by accurate highꢀresolution mass spectra obtained by
electrospray positive ionization mode from green kohlrabi. Identification of eleven desulfoGLS
was established based on the UV spectra and characteristic mass fragmentation / isotopic ratio
patterns due to sulphur and nitrogen molecules. Table 1 presents the retention times, theoretical
protonated accurate mass, and the key fragments of identified desulfoGLS (aliphatic, aromatic,
indolyl derivatives) in green kohlrabi. Figure 5 represents the extracted ion chromatograms
(EICs) of desulfoGLS identified in green kohlrabi. The tandem mass and +bbCID spectra of
desulfoGLS of aliphatic, aromatic and indolyl derivatives obtained by positive ionization mode
are shown in Figure 6. All mass spectra for the identified desulfoGLS were comprised of the
precursor ion [M+H]+, the sodium adduct [M+Na]+ and the characteristic product ion due to the
loss of the glucose moiety [M+Hꢀ162]+.
252
253
254
Identification of Aliphatic Glucosinolates. A total of 7 aliphatic desulfoGLS were
identified. The first peak was eluted at tR 5.2 min and exhibited an accurate mass at m/z 344.0823
[M+H]+ (mass error +2.67 ppm) and at m/z 366.0650 [M+Na]+ which corresponded to the
11
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 12 of 31
255
256
257
258
259
260
261
262
263
264
265
protonated form of desulfoGLS and its sodium adduct, respectively. It also showed a
characteristic product ion at m/z 182.0301 [M+Hꢀ162]+ which was a result of the loss of a
glucose moiety. The +bbCID spectra displayed a product ion at m/z 118.0318 [M+Hꢀ162ꢀ64]+
which presumably originated by elimination of methyl sulfoxide. On the basis of mass data and
UV spectra, the compound was identified as glucoiberin, 1. Similarly another peak eluted at tR
6.8 min and exhibited an accurate mass value at m/z 310.0952 [M+H]+ (mass error +1.0 ppm)
and its sodium adduct at m/z 332.0775 [M+Na]+. The +bbCID spectra also displayed a product
+
ion at m/z 148.0427 [M+Hꢀ162
]
which appeared due to the loss of a glucose moiety. It also
showed a less abundant product ion at m/z 130.0319 [M+Hꢀ162ꢀ18]+ due to loss of water from a
hydroxylꢀsubstituted side chain. It was thus inferred that the compound at tR 6.8 min was
progoitrin, 2.
266
267
268
269
270
271
272
273
274
275
276
277
Similar fragmentation patterns were observed for other aliphatic glucosinolates (tR 8.5,
9.0, 18.6, 21.4, and 26.0 min) having an accurate mass [M+H]+ at m/z 280.0849 (mass error +0.1
ppm), m/z 358.1005 (mass error ꢀ4.6 ppm), m/z 294.0963 (mass error +14.6 ppm), m/z 328.0850
(mass error +10.1 ppm), and m/z 342.1033 (mass error +1.9 ppm) and their diagnostic ion
fragment due to the loss of a glucose moiety [M+Hꢀ162]+ at m/z 118.0325, m/z 196.0471, m/z
132.0462, m/z 166.0337 and m/z 180.0510 respectively. On the basis of the above fragmentation
patterns the aliphatic glucosinolates were identified as sinigrin, 3, glucoraphanin, 4, gluconaphin,
5, glucoibervirin, 6, and glucoerucin, 7. Interestingly, the +bbCID spectrum of 4 also displayed a
minor product ion at m/z 132.0482 [M+Hꢀ162ꢀ64]+ due to the loss of methyl sulfoxide (CH4OS).
Similarly the +bbCID spectra of 1 and 7 also exhibited a low abundant product ion at m/z
118.0309 [M+Hꢀ162ꢀCH3SH]+ and at m/z 132.0471 [M+Hꢀ162ꢀCH3SH]+ respectively, which
arose due to the loss of a methanethiol moiety.
12
ACS Paragon Plus Environment

Page 13 of 31
Journal of Agricultural and Food Chemistry
278
279
280
281
282
283
284
285
286
287
Identification of Aromatic and Indolyl Glucosinolates Derivatives. Aromatic
desulfoGLS showed characteristic prominent product ion corresponding to tropylium ion.
Gluconasturtiin, 10, at tR 31.6 min exhibited an accurate mass value at m/z 344.1155 [M+H]+
with a mass error +2.14 ppm and its sodium adduct at m/z 366.0978 [M+Na]+. It also exhibited a
prominent product ion at m/z 182.0633 [M+Hꢀ162]+ which corresponds to loss of a glucose
moiety. DesulfoGLS 10 and 1 behaved as isobaric compounds due to same accurate mass in the
positive ion MS, but the +bbCID spectra of both GLS displayed different prominent product ions
due to their distinctive side chain (Rꢀsubstituent). Gluconasturtiin, 10, showed a distinct product
ion at m/z 91.0541 which corresponded to the abundance of tropylium ion while glucoiberin, 1,
displayed a product ion at m/z 118.0318 [M+Hꢀ162ꢀ64]+.
288
289
290
291
292
293
294
295
296
297
298
299
300
The three prominent peaks eluted at tR 27.8 min, 31.5 min and 36.4 min corresponded to
the presence of indolic compounds and were identified as glucobrassicin, 8, 4ꢀmethoxy
glucobrassicin, 9 and neoglucobrassicin, 11, respectively. Indolic glucosinolates were
characterized by odd masses of aglycone fragments which indicated the presence of an even
number of nitrogen atoms in their chemical structure.27 Desulfoglucobrassicin, 8, was putatively
identified using the accurate mass value at m/z 369.1095 [M+H]+ (mass error +5.4 ppm) and its
sodium adduct at m/z 391.0913 [M+Na]+. It also showed a dominant product ion at m/z 207.0580
[M+Hꢀ162]+ due to loss of glucose moiety. Burke et al.28 reported the most important
characteristic product ion of indolyl desulfoglucosinolates was abundance of [Rˊ]+. Similarly, the
+bbCID spectrum of 8 showed a prominent characteristic product ion at m/z 130.0646 [Rˊ]+
which corresponded to indoleꢀ3ꢀmethylene ion (C9H8N+). It also displayed a minor product ion at
m/z 117.0565 [Rˊꢀ13]+. A peak eluted at tR 31.5 min showed an accurate mass value at m/z
399.1205 [M+H]+ (mass error +3.9 ppm), its sodium adduct at m/z 421.1028 [M+Na]+ and
13
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 14 of 31
301
302
303
304
305
306
307
308
309
310
311
312
diagnostic product ion at m/z 237.0692 [M+Hꢀ162]+ The +bbCID spectra also showed a
.
prominent characteristic product ion at m/z 160.0750 [Rˊ]+ which corresponded to 4ꢀmethoxy
indoleꢀ3ꢀmethylene (C10H10NO+). It showed further loss of a methoxy group followed by
protonation to give minor fragment with small abundance at m/z 130.0650 [RˊꢀCH3O+H]+. On
the basis of characteristic fragmentation pattern, the peak was identified as 4ꢀ
methoxyglucobrassin, 9. Similarly, another prominent peak at tR 36.4 min was identified as
neoglucobrassicin, 11, having an accurate mass value at m/z 399.1219 [M+H]+ (mass error
+0.37), its sodium adduct at m/z 421.1035 [M+Na]+, and a product ion peak at m/z 237.0689
[M+Hꢀ162]+. The fragmentation pattern of 11 was similar to 9 but the +bbCID spectra of 11
represented a characteristic intense product ion at m/z 130.0650 [Rˊ]+. Thus, methoxy containing
isomers 9 and 11 were characterized by the difference in the intensities of their product ions
obtained in +bbCID spectra.
313
314
315
316
317
318
319
The extracted ion chromatograms (EICs) of intact GLS of green kohlrabi in negative
ionization mode at 0 h and the EICs at 7 h after incubation with sulfatase enzyme demonstrated
that before the addition of enzyme (0 h), spectra showed an intact GLS peaks but after incubation
with enzyme (7 h), no GLS precursor ions were observed in negative ionization mode. The EICs
in positive ionization mode showed the presence of desulfoGLS at 7 h. The EICs obtained in
positive and negative ionization modes provided strong evidence that the desulfonation process
was successfully working in the rapid method.
320
321
322
323
Confirmation of HPLC DesulfoGLS Fractions by HR-MS. Two different Eclipse Plus
columns were used for analytical HPLC (250 mm × 4.6 mm i.d., 5 µm) and LCꢀMS (100 × 2.0
mm i.d., 3 µm) separation. To confirm the peak identity from analytical HPLC, each peak
fraction was collected manually and analyzed by HRꢀMS. The mass spectra of all the fractions
14
ACS Paragon Plus Environment

Page 15 of 31
Journal of Agricultural and Food Chemistry
324
325
326
327
328
were matched to their desulfoGLS accurate mass and their fragments (Table 1 and Figure 6).
Thus, high resolution instruments are usually operated in an untargeted manner, which allows
accurate mass information on every ion that reaches the detector.29 Moreover, the reliability,
efficiency, sensitivity, and structural information will be more valid for identification of an
unknown molecules using HRꢀQTOF.30, 31
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
Quantification of Glucosinolates. After peak confirmation of desulfoGLS in HPLC and
HRꢀMS, the same methodology was applied to analytical HPLC for the quantification of
desulfoGLS. The quantification of 2, 3, 4, 5, 7 and 10, were achieved using the calibration curves
of their respective desulfoGLS. The remaining aliphatic and indolyl glucosinolates were
expressed relative to 4 and 10, respectively. All calibration curves showed excellent linearity
with high correlation coefficient (R2) > 0.99. The limit of detection (LOD) and limit of
quantification (LOQ) were determined by injecting serial diluted standards solutions until the
signalꢀtoꢀnoise ratio (S/N) for each standard was 3 to 10, respectively.32, 33 In desulfonation
reaction, enzyme volume and incubation time are critical parameters.13, 24, 25 Figure 7A displays
the formation of desulfoGLS at different time points using 80 µL of sulfatase enzyme from
kohlrabi. A total 11 glucosinolates were quantified in green kohlrabi and ranged from
0.16±0.007 to 3.32±0.02 µmol/g DW. DesulfoGLS 8, 9 and 10, showed significantly (P<0.005)
higher content within 4 h. No significant differences were observed in their level after 4 h.
Similarly, the desulfonation of aliphatic GLS, 1, 2, 3, 4, 5, 6 and 7, occurred within 7 h, which
had a significantly (P<0.005) higher level than other time points. There was no significant
variation in their levels after 7 h. In rapid method, aromatic and indolyl derivatives, 8, 10, and
11, were converted to 70ꢀ90% of their desulfonated form after 1 h and reached to their maximum
level in 4 h.
15
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 16 of 31
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
The aliphatic glucoerucin, 7, showed a 70% conversion to desulfonated form within 1 h
and the level increased to 1.6ꢀfold higher at 7 h in rapid method. Similarly, other aliphatic
glucosinolates also exhibited a higher level at 7 h. The glucoiberin, 1 and progoitrin, 2 were 3.6
and 1.9 fold higher, respectively, at 7 h as compared to the conventional method. Similarly,
aliphatic GLS, 3, 4, 5 and 6, exhibited 50ꢀ100% conversion to desulfonated form depending
upon their chemical structure at 4 h, but their levels were 11%, 4%, 21% and 68% higher than
the conventional method at 7 h. The progoitrin, 3, (2.53±0.05 µmol/g DW), sinigrin, 1,
(1.88±0.04 µmol/g DW) and neoglucobrassicin, 11, (3.32±0.06 µmol/g DW) were found to be
major the constituents in green kohlrabi. Previous literature reported the absence of glucoerucin
in pale green kohlrabi (cultivar
'Winner
') whereas it was predominant (8.08 µmol/g DW) in the
purple cultivar (
'Early Purple Vienna'
).21 Progoitrin, 2, (0.52 µmol/g DW) and sinigrin, 3, (2.24
µmol/g DW) were the predominant and neoglucobrassicin, 11, (0.04µmol/g DW) was the lowest
level glucosinolate reported in kohlrabi sprouts.16 Similarly, the seed of kohlrabi cultivars
('White Vienna
', 'Grand Duke' and 'Purple Vienna') showed a significant amount of glucoiberin,
1, (22.84ꢀ26.25 µmol/g FW).22
362
363
364
365
366
367
368
369
To understand the effect of food matrix on desulfonation time, the developed
methodology was applied to green cabbage for the analysis of desulfoGLS. A total of 8
desulfoGLS, 1, 3, 4, 5a, 6, 8, 9 and 11, were identified by LCꢀHRꢀESIꢀQTOFꢀMS and their
enzymatic kinetic reactions at different time points are shown in Figure 7B. Similar to kohlrabi
glucosinolates, the indolyl derivatives of desulfoGLS showed significantly (P<0.005) higher
content within 4 h. At 7 h aliphatic desulfoGLS showed significantly (P<0.005) higher levels
than other time points. The aliphatic desulfoGLS sinigrin (5.62±0.02 µmol/g DW) and indolyl
glucobrassicin, 8, (6.07±0.03 µmol/g DW) were found to be the major glucosinolates in green
16
ACS Paragon Plus Environment

Page 17 of 31
Journal of Agricultural and Food Chemistry
370
371
372
373
cabbage. A similar trend of desulfoGLS levels were report by Park et al.34 in which sinigrin, 3,
ranging from 3.3±0.5
DW) were found to be predominant in various lines of green cabbage. The significant variation
of 4ꢀmethoxy glucobrassicin, 9, (1.77±0.2 2.69±0.4 µmol/g of DW) has been also reported
–13.3±1.3 µmol/g DW and glucobrassicin, 8, (2.8±0.8–7.0±0.8 µmol/g
–
374
375
among the different cabbage cultivars.35 In our study, we also observed a similar level
(1.80±0.03 µmol/g) of 4ꢀmethoxy glucobrassicin, 9.
376
377
378
379
380
The present rapid method showed high recovery of certain GLS and reproducibility due
to minimal unit operations as compared to the chromatographic technique. The use of HRꢀESIꢀ
QTOFꢀMS provides a more reliable approach to identify desulfoGLS. The developed
methodology is simple and can be easily applied to investigate the glucosinolates in other
Brassicaceae species.
381
382
ꢀ ASSOCIATED CONTENT
Supportive information
383
384
385
386
387
388
389
390
391
Possible fragmentation patterns of desulfoGLS, 1-11, in positive ionization mode identified in
green kohlrabi (Figure S1), extracted ion chromatograms (EICs) in negative and positive
ionization modes represent the conversion of green kohlrabi intact glucosinolates to their
desulfonated forms after aryl sulfatase enzyme (80 µL) treatment (Figure S2), extracted ion
chromatograms (EICs) of desulfoGLS fractions, 1ꢀ11 collected from the analytical HPLCꢀPDA
(column: Eclipse Plus C18 ODS) analysis and confirmed by LCꢀHRꢀQTOFꢀMS (column: Eclipse
Plus C18 ODS) analysis (Figure S3), calibration curves of desulfoGLS of progoitrin, 2, sinigrin,
3, glucoraphanin, 4 gluconaphin, 5, glucoerucin, 7, and gluconasturtiin, 10 (Figure S4), HPLCꢀ
PDA overlay chromatograms of green cabbage glucosinolates at 0 h and 7 h (Figure S5), typical
17
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 18 of 31
392
393
394
sample preparation steps involved for the analysis of desulfoGLS from green kohlrabi at
different volume of aryl sulfatase enzyme (Table S1) and linearity, coefficient of variation (R2),
limit of quantification (LOQ) and limit of detection (LOD) of different desulfoGLS (Table S2).
395
396
397
398
399
AUTHOR INFORMATION
Corresponding Authors
*Telephone: +1ꢀ979ꢀ845ꢀ3864. Fax: +1ꢀ979ꢀ862ꢀ4522. Eꢀmail: gkjp@tamu.edu
*Telephone: +1ꢀ979ꢀ862ꢀ4521. Fax: +1ꢀ979ꢀ862ꢀ4522. Eꢀmail: bꢀpatil@tamu.edu
ꢀ ACKNOWLEDGEMENT
400
401
This study was supported by the Agriculture and Food Research Initiative competitive grant
2015ꢀ67013ꢀ22813 of the USDA National Institute of Food and Agriculture.
402
403
404
Notes
The authors declare no competing financial interest.
ꢀ REFERENCES
405
406
407
408
409
410
411
412
413
414
415
416
(1) Hill, J.; Lethenborg, P.; Li, P.; Rahman, M.; Sørensen, H.; Sørensen, J. Inheritance of
progoitrin and total aliphatic glucosinolates in oilseed rape (Brassica napus L). Euphytica
2003, 134, 179–187.
(2) Bjerg, B.; Sørensen, H. Quantitative analysis of glucosinolates in oilseed rape based on
HPLC of desulfoglucosinolates and HPLC of intact glucosinolates. In Glucosinolates in
rapeseeds: Analytical aspects, Springer: 1987; pp 125–150.
(3) Rodman, J. Glucosinolates: methods of analysis and some chemosystematic problems.
Phytochem. Bull 1978, 11, 6–31.
(4) Sørensen, H. Glucosinolates: structureꢀpropertiesꢀfunction. In Canola and rapeseed,
Springer: 1990; pp 149–172.
(5) Olsen, O.; Sørensen, H. Recent advances in the analysis of glucosinolates. J. Am. Oil
Chem. Soc. 1981, 58, 857–865.
18
ACS Paragon Plus Environment

Page 19 of 31
Journal of Agricultural and Food Chemistry
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
(6) Gupta, P.; Kim, B.; Kim, S. H.; Srivastava, S. K. Molecular targets of isothiocyanates in
cancer: recent advances. Mol. Nutr. Food Res. 2014, 58, 1685–1707.
(7) Chung, F. L.; Conaway, C. C.; Rao, C.; Reddy, B. S. Chemoprevention of colonic aberrant
crypt foci in Fischer rats by sulforaphane and phenethyl isothiocyanate. Carcinogenesis
2000, 21, 2287–2291.
(8) Jin, J.; Koroleva, O. A.; Gibson, T.; Swanston, J.; Magan, J.; Zhang, Y.; Rowland, I. R.;
Wagstaff, C. Analysis of phytochemical composition and chemoprotective capacity of
rocket (Eruca sativa and Diplotaxis tenuifolia) leafy salad following cultivation in
different environments. J. Agric. Food Chem. 2009, 57, 5227–5234.
(9) Michaud, D. S.; Spiegelman, D.; Clinton, S. K.; Rimm, E. B.; Willett, W. C.;
Giovannucci, E. L. Fruit and vegetable intake and incidence of bladder cancer in a male
prospective cohort. J. Natl Cancer Inst. 1999, 91, 605–613.
(10) Higdon, J. V.; Delage, B.; Williams, D. E.; Dashwood, R. H. Cruciferous vegetables and
human cancer risk: epidemiologic evidence and mechanistic basis. Pharmacol. Res. 2007,
55, 224–236.
(11) Murray, M. Altered CYP expression and function in response to dietary factors: potential
roles in disease pathogenesis. Curr. Drug Metab. 2006, 7, 67–81.
(12) Ediage, E. N.; Di Mavungu, J. D.; Scippo, M. L.; Schneider, Y. J.; Larondelle, Y.;
Callebaut, A.; Robbens, J.; Van Peteghem, C.; De Saeger, S. Screening, identification and
quantification of glucosinolates in black radish (Raphanus sativus L. niger) based dietary
supplements using liquid chromatography coupled with a photodiode array and liquid
chromatographyꢀmass spectrometry. J. Chromatogr. A 2011, 1218, 4395–4405.
(13) Wathelet, J. P.; Iori, R.; Leoni, O.; Rollin, P.; Quinsac, A.; Palmieri, S. Guidelines for
glucosinolate analysis in green tissues used for biofumigation. Agroindustria 2004, 3,
257–266.
(14) Truscott, R. J.; Burke, D. G.; Minchinton, I. R. The characterisation of a novel
hydroxindole glucosinolate. Biochem. Biophys. Res. Commun. 1982, 107, 1258–1264.
(15) Kim, S. J.; Kawaharada, C.; Jin, S.; Hashimoto, M.; Ishii, G.; Yamauchi, H. Structural
elucidation of 4ꢀ(cysteinꢀSꢀyl) butyl glucosinolate from the leaves of Eruca sativa. Biosci.
Biotechnol. Biochem. 2007, 71, 114–121.
(16) Kusznierewicz, B.; Iori, R.; Piekarska, A.; Namieśnik, J.; Bartoszek, A. Convenient
identification of desulfoglucosinolates on the basis of mass spectra obtained during liquid
chromatography–diode array–electrospray ionisation mass spectrometry analysis: method
verification for sprouts of different Brassicaceae species extracts. J. Chromatogr. A 2013,
1278, 108–115.
(17) Jayaprakasha, G. K.; Nagana Gowda, G. A.; Marquez, S.; Patil, B. S. Rapid separation and
quantitation of curcuminoids combining pseudo twoꢀdimensional liquid flash
chromatography and NMR spectroscopy. J. Chromatogr. B 2013, 937, 25–32.
(18) Chebrolu, K. K.; Jayaprakasha, G. K.; Jifon, J.; Patil, B. S. Purification of coumarins,
including meranzin and pranferin, from grapefruit by solvent partitioning and a
hyphenated chromatography. Sep. Purif. Technol. 2013, 116, 137–144.
(19) Lee, J. G.; Bonnema, G.; Zhang, N.; Kwak, J. H.; de Vos, R. C.; Beekwilder, J. Evaluation
of glucosinolate variation in a collection of turnip (Brassica rapa) germplasm by the
analysis of intact and desulfo glucosinolates. J. Agric. Food Chem. 2013, 61, 3984–3993.
19
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 20 of 31
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
(20) Rybarczyk Plonska, A.; Hagen, S. F.; Borge, G. I. A.; Bengtsson, G. B.; Hansen, M. K.;
Wold, A. B. Glucosinolates in broccoli (Brassica oleracea L. var. italica) as affected by
postharvest temperature and radiation treatments. Postharvest Biol. Tec. 2016, 116, 16–25.
(21) Park, W. T.; Kim, J. K.; Park, S.; Lee, S. W.; Li, X.; Kim, Y. B.; Uddin, M. R.; Park, N. I.;
Kim, S. J.; Park, S. U. Metabolic profiling of glucosinolates, anthocyanins, carotenoids,
and other secondary metabolites in kohlrabi (Brassica oleracea var. gongylodes). J. Agric.
Food Chem. 2012, 60, 8111–8116.
(22) West, L. G.; Meyer, K. A.; Balch, B. A.; Rossi, F. J.; Schultz, M. R.; Haas, G. W.
Glucoraphanin and 4ꢀhydroxyglucobrassicin contents in seeds of 59 cultivars of broccoli,
raab, kohlrabi, radish, cauliflower, brussels sprouts, kale, and cabbage. J. Agric. Food
Chem. 2004, 52, 916–926.
(23) Singh, J.; Jayaprakasha, G.; Patil, B. S. Rapid ultraꢀhighꢀperformance liquid
chromatography/quadrupole timeꢀofꢀflight tandem mass spectrometry and selected
reaction monitoring strategy for the identification and quantification of minor spinacetin
derivatives in spinach. Rapid Commun. Mass Spectrom. 2017, 31, 1803–1812.
(24) Fiebig, H. Desulphation of glucosinolates–potential pitfalls of the HPLC method for
double zero rapeseed. Fett/Lipid 1991, 93, 264–267.
(25) Hennig, K.; Verkerk, R.; Bonnema, G.; Dekker, M. Pitfalls in the desulphation of
glucosinolates in a highꢀthroughput assay. Food Chem. 2012, 134, 2355–2361.
(26) Bianco, G.; Lelario, F.; Battista, F. G.; Bufo, S. A.; Cataldi, T. R. Identification of
glucosinolates in capers by LC‐ESI‐hybrid linear ion trap with Fourier transform ion
cyclotron resonance mass spectrometry (LC‐ESI‐LTQ‐FTICRꢀMS) and infrared
multiphoton dissociation. J. Mass Spectrom. 2012, 47, 1160–1169.
(27) Fabre, N.; Poinsot, V.; Debrauwer, L.; Vigor, C.; Tulliez, J.; Fouraste, I.; Moulis, C.
Characterisation of glucosinolates using electrospray ion trap and electrospray quadrupole
time‐of‐flight mass spectrometry. Phytochem. Anal. 2007, 18, 306–319.
(28) Burke, D. G.; Cominos, X. Identification of desulfoglucosinolates using positiveꢀion fast
atom bombardment mass spectrometry. J. Agric. Food Chem. 1988, 36, 1184–1187.
(29) De Vos, R. C.; Moco, S.; Lommen, A.; Keurentjes, J. J.; Bino, R. J.; Hall, R. D.
Untargeted largeꢀscale plant metabolomics using liquid chromatography coupled to mass
spectrometry. Nat. Protoc. 2007, 2, 778–791.
(30) Ku, K. M.; Kim, M. J.; Jeffery, E. H.; Kang, Y. H.; Juvik, J. A. Profiles of glucosinolates,
their hydrolysis products, and quinone reductase inducing activity from 39 arugula (Eruca
Sativa Mill.) accessions. J. Agric. Food Chem. 2016, 64, 6524–6532.
(31) Garcia, C. J.; GarcíaꢀVillalba, R.; Garrido, Y.; Gil, M. I.; TomásꢀBarberán, F. A.
Untargeted metabolomics approach using UPLCꢀESIꢀQTOFꢀMS to explore the
metabolome of freshꢀcut iceberg lettuce. Metabolomics 2016, 12, 138.
(32) del Álamo, M.; Casado, L.; Hernández, V.; Jiménez, J. J. Determination of free molecular
phenolics and catechins in wine by solid phase extraction on polymeric cartridges and
liquid chromatography with diode array detection. J. Chromatogr. A 2004, 1049, 97–105.
(33) Vial, J.; Jardy, A. Experimental comparison of the different approaches to estimate LOD
and LOQ of an HPLC method. Anal. Chem. 1999, 71, 2672–2677.
(34) Park, S.; Arasu, M. V.; Lee, M.ꢀK.; Chun, J.ꢀH.; Seo, J. M.; Lee, S.ꢀW.; AlꢀDhabi, N. A.;
Kim, S.ꢀJ. Quantification of glucosinolates, anthocyanins, free amino acids, and vitamin C
in inbred lines of cabbage (Brassica oleracea L.). Food Chem. 2014, 145, 77ꢀ85.
20
ACS Paragon Plus Environment

Page 21 of 31
Journal of Agricultural and Food Chemistry
506
507
508
(35) Baek, S. A.; Jung, Y. H.; Lim, S. H.; Park, S. U.; Kim, J. K. Metabolic profiling in chinese
cabbage (Brassica rapa L. subsp. pekinensis) cultivars reveals that glucosinolate content is
correlated with carotenoid content. J. Agric. Food Chem. 2016, 64, 4426–4434.
509
21
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 22 of 31
510
FIGURE CAPTIONS
511
512
513
514
515
516
Figure 1. HPLCꢀUV overlay chromatograms of green kohlrabi glucosinolates at 0 h (intact) and
7 h (desulfoGLS). (A). intact glucosinolates (GLS), (B). desulfoGLS after treating the aryl
sulfatase enzyme (80 µL) without using DEAE Sephadex column at 7 h. Peaks 1-11 represents
glucoiberin, 1; progoitrin, 2; sinigrin, 3; glucoraphanin 4; gluconaphin, 5; glucoibervirin, 6;
glucoerucin, 7; glucobrassicin, 8; 4ꢀmethoxy glucobrassicin, 9; gluconasturtiin, 10; and
neoglucobrassicin, 11.
517
518
Figure 2. Structures of identified aliphatic, aromatic and indolyl desulfoglucosinolates from
kohlrabi and cabbage.
519
520
521
522
523
Figure 3. Effect of aryl sulfatase enzyme volume (10, 20, 40, 80 and 160 µl) for desulfonation of
green kohlrabi glucosinolates after 7 h. The numbers refer to the following glucosinolates:
glucoiberin, 1; progoitrin, 2; sinigrin, 3; glucoraphanin 4; gluconaphin, 5; glucoibervirin, 6;
glucoerucin, 7; glucobrassicin, 8; 4ꢀmethoxy glucobrassicin, 9; gluconasturtiin, 10; and
neoglucobrassicin, 11.
524
525
526
527
528
529
530
Figure 4. Comparison of desulfoglucosinolates from green kohlrabi achieved by rapid method (7
h) and the conventional method (18 h) by DEAE Sephadex Aꢀ25 resin using 80 µL of sulfatase
enzyme. The numbers refer to the following glucosinolates: glucoiberin, 1; progoitrin, 2;
sinigrin, 3; glucoraphanin 4; gluconaphin, 5; glucoibervirin, 6; glucoerucin, 7; glucobrassicin, 8;
4ꢀmethoxy glucobrassicin, 9; gluconasturtiin, 10; and neoglucobrassicin, 11. Data was expressed
as mean of three replicates. Means with the same letter indicate no significant difference
(P<0.05).
531
532
Figure 5. Extracted ion chromatograms of green kohlrabi desulfoglucosinolates, 1ꢀ11, after 7 h
incubation with enzyme obtained from HRꢀESIꢀQTOFMS analysis in positive ionization mode.
533
534
Figure 6. Tandem mass spectra of desulfoglucosinolates identified from green kohlrabi and
cabbage after 7 h by LCꢀHRꢀESIꢀQTOFMS analysis in positive ionization mode.
535
536
537
538
539
540
Figure 7. Levels of desulfoglucosinolates (µmol/g DW) in (A) green kohlrabi and (B) green
cabbage after treatment with 80 µL of sulfatase enzyme at different incubation times (1, 4, 8, 12,
16, 18 and 22 h). The numbers refer to the following glucosinolates: glucoiberin, 1; progoitrin, 2;
sinigrin, 3; glucoraphanin 4; gluconaphin, 5; 4ꢀhydroxy glucobrassicin, 5a; glucoibervirin, 6;
glucoerucin, 7; glucobrassicin, 8; 4ꢀmethoxy glucobrassicin, 9; gluconasturtiin, 10; and
neoglucobrassicin, 11. Data are expressed as mean of three replicates (± standard error).
22
ACS Paragon Plus Environment

Page 23 of 31
Journal of Agricultural and Food Chemistry
541
542
Table 1. Identified Desulfoglucosinolates (desulfoGLS) in Green Kohlrabi by LCꢀHRꢀESIꢀQTOFꢀMS Using Positive Ionization
Mode.
Compound
groups
[MDS+H]+
[MDS+Hꢀ
162]+ m/z
Mass error
(ppm)c
a
DesulphoGLS
Side chain (Rꢀ)
tR
m/zb
5.2
6.8
3ꢀmethylsulfinylpropyl
(2R)ꢀ2ꢀhydroxyꢀ3ꢀbutenyl
2ꢀpropenyl
Aliphatic
Aliphatic
Aliphatic
Aliphatic
Aliphatic
Aliphatic
Aliphatic
344.0823
310.0952
280.0849
358.1005
294.0963
328.085
182.0301
148.0427
118.0325
196.0471
132.0462
166.0337
180.0510
2.67
0.96
1
2
3
4
5
6
7
8.5
0.12
9.0
4ꢀmethylsulfinylbutyl
3ꢀbutenyl
ꢀ4.55
14.57
10.08
1.92
18.6
21.4
26.0
3ꢀmethylthiopropyl
4ꢀmethylthiobutyl
342.1033
27.8
31.5
31.6
36.4
3ꢀindolylmethyl
Indolyl
Indolyl
Aromatic
Indolyl
369.1095
399.1205
344.1155
399.1219
207.0580
237.0692
182.0633
237.0689
5.38
3.88
2.14
0.37
8
9
4ꢀmethoxyꢀ3ꢀindolylmethyl
2ꢀphenethyl
10
11
Nꢀmethoxyꢀ3ꢀindolylꢀmethyl
a tR: retention time (min) of desulfoGLS, b Accurate mass value of desulfoGLS, c Mass error in parts per million (ppm=106 )
543
23
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 24 of 31
544
A) Intact GLS, 1-11
1.0
0.5
11
7
8
1 2
3
4
5
6
9
10
0.0
0.8
0.6
0.4
0.2
0.0
5.0
10.0
20.0
30.0
40.0
B) Desulfo GLS at 7h, 1-11
11
7
4
3
8
1
2
5
6
9
10
5.0
10.0
20.0
30.0
40.0
Minutes
545
546
547
548
Figure 1
24
ACS Paragon Plus Environment

Page 25 of 31
Journal of Agricultural and Food Chemistry
Glucoiberin, 1
Progoitrin, 2
Sinigrin, 3
Glucoraphanin, 4
4ꢀhydroxy glucobrassicin, 5a
Glucoibervirin, 6
Glucoerucin, 7
Gluconaphin, 5
Gluconasturtiin, 10
Glucobrassicin, 8
4ꢀmethoxy glucobrassicin, 9
Neoglucobrassicin, 11
549
550
Figure 2
551
552
553
554
555
25
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 26 of 31
556
1
7
2
8
3
9
4
5
6
10
11
4.5
4
3.5
3
2.5
2
1.5
1
0.5
0
10
20
40
80
160
Enzyme volume (µL)
557
558
Figure 3
559
560
561
562
563
564
26
ACS Paragon Plus Environment

Page 27 of 31
Journal of Agricultural and Food Chemistry
4.5
4
Rapid method
Conventional method
a
b
3.5
3
a
2.5
2
a
b
a
a
a
1.5
1
a
b
a
b
a
b
b
a
a
b
a
0.5
0
b
a
a
1
2
3
4
5
6
7
8
9
10
11
Desulfoglucosinolates
565
566
Figure 4
567
568
27
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 28 of 31
Intens.
Desulfoglucoiberin, 1
5
x10
0
Intens.
Desulfoprogoitrin, 2
5
x10
0
Intens.
Desulfosinigrin, 3
5
x10
2
0
Intens.
Desulfoglucoraphanin, 4
5
x10
0
Intens.
Desulfogluconapin, 5
4
x10
2
0
Intens.
Desulfoglucoibervirin, 6
5
x10
0
Intens.
Desulfoglucoerucin, 7
6
x10
0
Intens.
Desulfoglucobrassicin, 8
4
x10
2
0
Intens.
Desulfoꢀ4ꢀmethoxyglucobrassicin, 9
5
x10
0
Intens.
Desulfogluconasturtiin, 10
5
x10
2
0
Intens.
Desulfoneoglucobassicin, 11
6
x10
0
0
10
20
30
Time [min]
569
570
Figure 5
571
28
ACS Paragon Plus Environment

Page 29 of 31
Journal of Agricultural and Food Chemistry
572
Glucoiberin, 1 (tR 5.2)
+MS
Progoitrin, 2 (tR 6.8)
Sinigrin, 3 (tR 8.5)
+MS
Intens.
5
x10
Intens.
x105
Intens.
x105
[M+Na]+
332.0775
[M+Na]+
302.0671
[M+Na]+
366.0650
+MS
[M+Hꢀ162]+
118.0325
[M+Hꢀ162]+
148.0427
[M+H]+
310.0952
[M+Hꢀ162]+
182.0301
[M+H]+
344.0823
0.5
[M+H]+
280.0849
2
0.5
196.0460
130.0319
145.0489
+bbCID MS
118.0321
282.0301
0
x104
0
x105
0
x104
+bbCID MS
[M+Na]+
332.0769
[M+Na]+
366.0643
+bbCID MS
[M+Na]+
302.0664
1
4
4
0
[M+Hꢀ162ꢀ64]+
[M+Hꢀ162ꢀ18]+
130.0319
118.0318
218.0275
204.0119
302.0652
100 150 200 250 300 350 m/z
132.0476
185.0415
0
0
50 100 150 200 250 300 350
m/z
100
150
200
250
300 m/z
Glucoraphanin, 4 (tR 9.0)
+MS
Gluconaphin, 5 (tR 18.6)
[M+Hꢀ162]+
132.0462
4ꢀhydroxyglucobrassicin, 5a (tR 18.8)
Intens.
6
x10
Intens.
5
x10
[M+Na]+
380.0824
358.1005
[M+H]+
+MS
[M+Na]+
316.0781
[M+H]+
385.2605
Intens.
x104
+MS
[M+Hꢀ162]+
196.0471
0.6
2
0
[M+Hꢀ162]+
223.1431
[M+H]+
294.0963
1
118.0324
261.1444
222.0153
322.2211
0
x106
380.0827
[M+Na]+
5
x10
0
x104
+bbCID MS
[M+Na]+
316.0787
+bbCID MS
[M+Na]+
407.2512
+bbCID MS
[R’]+
146.1185
2
0
0.5
0
[M+Hꢀ162ꢀ64]+
1
0
132.0482
218.0285
245.1335
219.0259
100 150 200 250 300 350 m/z
100 150 200 250 300 350 m/z
50
150
250
350
450 m/z
Glucoibervirin, 6 (tR 21.4)
+MS
Glucoerucin, 7 (tR 26.0)
+MS
Glucobrassicin, 8 (tR 27.8)
+MS
Intens.
x105
Intens.
x105
Intens.
x106
[M+H]+
328.0850
[M+H]+
342.1033
[M+Hꢀ162]+
207.0580
[M+H]+
369.1095
[M+Na]+
350.0666
[M+Hꢀ162]+
180.0510
[M+Hꢀ162]+
166.0337
1
1.0
364.0841
391.0910
0.5
229.1527
281.1102
132.0467
0
x105
0
x105
0
x106
[M+Na]+
350.0666
+bbCID MS
[M+Na]+
364.0849
[M+Na]+
391.0913
+bbCID MS
[Rˊ]+
130.0646
+bbCID MS
1.5
0
0.6
0
1.0
0
[M+Hꢀ162ꢀCH3SH]+
132.0471
[M+Hꢀ162ꢀCH3SH]+
118.0309
[Rˊꢀ13]+
[Rˊˊ]+
117.0565
185.0408
263.9032
173.0283
100 150 200 250 300 350 m/z
100 150 200 250 300 350
m/z
100 150 200 250 300 350 400 m/z
4ꢀmethoxyglucobrassicin, 9 (tR 31.5)
Gluconasturtiin, 10 (tR 31.6)
Neoglucobrassicin, 11 (tR 36.4)
+MS
Intens.
x105
Intens.
x105
Intens.
x104
[M+Hꢀ162]+
182.0633
+MS
+MS
[M+H]+
399.1205
[M+H]+
399.1219
[M+Hꢀ162]+
237.0692
[M+Na]+
366.0978
[M+Hꢀ162]+
237.0689
[M+H]+
1.0
2
344.1155
421.1028
182.0629
0.4
421.1035
237.0683
0
x104
0
0
x104
+bbCID MS
130.0650
+bbCID MS
+bbCID MS
x105
[M+Na]+
421.1033
[M+Na]+
366.0974
[Rˊ]+
[Rˊ]+
160.0750
[R]+
91.0541
1.0
3
0
[M+Na]+
421.1034
1
130.0650
130.0648
204.0448
390.0848
206.0503
100 150 200 250 300 350 400 m/z
259.0503
0
0
100 150 200 250 300 350m/z
50 100 150 200 250 300 350 400 m/z
573
574
Figure 6
29
ACS Paragon Plus Environment