Key Odor-Active Compounds in Raw Green and Red Toona sinensis (A. Juss.) Roem. And Their Changes during Blanching

Article abstract of DOI:10.1021/acs.jafc.0c02012

Application of aroma extract dilution analysis and headspace aroma dilution analysis revealed 51 odorants in raw green Toona sinensis and 54 odorants in raw red T. sinensis in the flavor dilution factor range of 8-4096. (E,E)-2,4-Decadienal, nonanal, 2,3,5-trimethylpyrazine, (E,Z)- and (Z,Z)-di-1-propenyl trisulfide, 2-methoxyphenol, and 4-ethylphenol were first identified as key odorants of T. sinensis. Clear differences between green and red T. sinensis in aroma profiles, flavor dilution factors, quantitative data, and odor activity values verified that (E,E)-, (E,Z)-, and (Z,Z)-di-1-propenyl disulfide, (E,E)-, (E,Z)- and (Z,Z)-di-1-propenyl trisulfide, cis- and trans-2-mercapto-3,4-dimethyl-2,3-dihydrothiophene, and dimethyl sulfide caused the distinct sulfury odor note of each variety. Further, hexanal, (E)-2-hexenal, (E)-2-hexen-1-ol, and (E,Z)-2,6-nonadienal led to the green odor note in green T. sinensis, while 2-methoxyphenol and 4-ethylphenol contributed to the intense phenolic aroma note in red T. sinensis. Quantitation experiments and triangle tests in blanched T. sinensis verified that the quick loss of the abovementioned sulfur-containing compounds, aldehydes, the alcohol (E)-2-hexen-1-ol, and phenols was responsible for the changes in the overall aroma profile during blanching.

Full text of DOI:10.1021/acs.jafc.0c02012

Subscriber access provided by Uppsala universitetsbibliotek
Chemistry and Biology of Aroma and Taste
Key Odor-Active Compounds in Raw Green and Red Toona
sinensis (A. Juss.) Roem. and Their Changes during Blanching
Xiaoting Zhai, and Michael Granvogl
J. Agric. Food Chem., Just Accepted Manuscript • Publication Date (Web): 05 Jun 2020
Downloaded from pubs.acs.org on June 5, 2020
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 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 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.
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 64
Journal of Agricultural and Food Chemistry
1
2
3
Key Odor-Active Compounds in Raw Green
and Red Toona sinensis (A. Juss.) Roem.
and Their Changes during Blanching
4
5
§
#,*
Xiaoting Zhai and Michael Granvogl
6
7
8
§ Lebensmittelchemie, Fakultät Chemie, Technische Universität München,
Lise-Meitner-Straße 34, D-85354 Freising, Germany
9
# Institut für Lebensmittelchemie, Fachgebiet für Lebensmittelchemie und Analytische
1
1
0
1
Chemie (170a), Fakultät Naturwissenschaften, Universität Hohenheim,
Garbenstrasse 28, D-70599 Stuttgart, Germany
1
1
2
3
14
15
16
17
*Corresponding Author
Phone: +49 711 459 23979
Fax:
+49 711 459 24096
E-mail:
michael.granvogl@uni-hohenheim.de
1
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 2 of 64
1
8
ABSTRACT: Application of aroma extract dilution analysis and headspace aroma
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
dilution analysis revealed 52 odorants in raw green Toona sinensis and 54 odorants in
raw red T. sinensis in the flavor dilution factor range of 8-4096. (E,E)-2,4-Decadienal,
nonanal, 2,3,5-trimethylpyrazine, (E,Z)- and (Z,Z)-di-1-propenyl trisulfide,
2-methoxyphenol, and 4-ethylphenol were firstly identified as key odorants of T.
sinensis. Clear differences between green and red T. sinensis in aroma profiles, flavor
dilution factors, quantitative data, and odor activity values verified that (E,E)-, (E,Z)-,
and (Z,Z)-di-1-propenyl disulfide, (E,E)-, (E,Z)-, and (Z,Z)-di-1-propenyl trisulfide, cis-
and trans-2-mercapto-3,4-dimethyl-2,3-dihydrothiophene, and dimethyl sulfide
caused the distinct sulfury odor note of each variety. Further, hexanal, (E)-2-hexenal,
(E)-2-hexen-1-ol, and (E,Z)-2,6-nonadienal led to the green odor note in green T.
sinensis, while 2-methoxyphenol and 4-ethylphenol contributed to the intense
phenolic aroma note in red T. sinensis. Quantitation experiments and triangle tests in
blanched T. sinensis verified that the quick loss of the abovementioned
sulfur-containing compounds, aldehydes, the alcohol (E)-2-hexen-1-ol, and phenols
were responsible for the changes in the overall aroma profile during blanching.
3
5
KEYWORDS: T. sinensis, sensomics concept, aroma extract dilution analysis,
3
3
6
7
stable isotope dilution analysis, odor activity values, aroma recombination,
heat-processing, sulfides, phenols.
2
ACS Paragon Plus Environment

Page 3 of 64
Journal of Agricultural and Food Chemistry
3
9
INTRODUCTION
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
Toona sinensis (A. Juss.) Roem. is a popular cultivated tree due to its usage in
pharmacology based on anti-cancer and antioxidative properties as well as in
traditional Chinese diet because of beneficial nutrient contents.1,2 Nowadays, the
planting area of T. sinensis is > one billion square meters, which is used for the
3
cultivation of > 800 billion kilograms of fresh T. sinensis buds every year. Studies on
the volatiles of T. sinensis started about two decades ago, and until now, > 100
volatiles have been identified. Among them, Liu et al. reported on cis- and
trans-2-mercapto-3,4-dimethyl-2,3-dihydrothiophene (both with a T. sinensis-like
smell) as major contributors to the characteristic aroma of fresh T. sinensis (Shanxi,
China) via gas chromatography-mass spectrometry (GC-MS) and gas
chromatography-olfactometry (GC-O).6 Yang et al. verified (E,E)-di-1-propenyl
disulfide, (E,Z)-di-1-propenyl disulfide, hydrogen sulfide, methyl thiirane, hexanal,
(Z)-3-hexenal, (E)-2-hexenal, and (Z)-3-hexen-1-ol responsible for the unique and
pleasant flavor of fresh T. sinensis (Beijing, China) based on the application of
headspace solid phase microextraction combined with gas chromatography-mass
7
spectrometry (HS-SPME-GC-MS), GC-MS, and GC-O. Very recently, Zhai and
Granvogl
characterized
dimethyl
sulfide,
eugenol,
hexanal,
β-ionone,
2-isopropyl-3-methoxypyrazine,
2-methylbutanal,
3-methylbutanal,
and
3-methylnonane-2,4-dione as key aroma-active compounds for the first time in two
commercially dried T. sinensis products (Hubei and Anhui, China) by the application
of the molecular sensory science concept.8
According to the color of the young buds, T. sinensis is classified into red T.
sinensis and green T. sinensis. Red T. sinensis, which is matured earlier, has more
fuchsia leaves, less fiber, and more grease. Young red leaves are considered to have
a better flavor compared to young green leaves, indicating clear differences in key
3
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 4 of 64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
odorants between red and green T. sinensis. However, to the best of our knowledge,
molecular differences of the overall aromas and key odorants between red and green
T. sinensis have not yet been clarified using the comprehensive approach of the
molecular sensory science concept.
Although young T. sinensis buds, picked in early spring (April and May) in China,
are very delicious vegetables, raw buds are rarely used directly as food ingredients.
Thus, blanching of raw T. sinensis sprouts in boiling water is an essential process
prior to further cooking steps, due to the fact that raw sprouts contain some toxic
ingredients (e.g., nitrites). After blanching, no matter if starting with red or green T.
sinensis, the buds are green colored because of the loss of anthocyanines at the
applied temperature (~100 °C), and the aroma is less intense. To date, only one
report identified the composition of odorants in blanched T. sinensis buds using static
headspace aroma dilution analysis (SH-ADA) and aroma extract dilution analysis
7
(AEDA) based on GC-O and GC-MS. Thereby, (E,E)-di-1-propenyl disulfide and
(E,Z)-di-1-propenyl disulfide were found to be potent key odorants. However, no
systematic sensory analysis has been applied to elucidate the changes in the overall
aroma of blanched T. sinensis at a molecular level.
Thus, the aim of the present study was first to elucidate the differences in key
aroma-active compounds between green and red T. sinensis buds and secondly to
reveal the molecular background of aroma changes occurring during blanching by
means of the molecular sensory science concept. Thereby, the key odorants were i)
identified by comparative AEDA (cAEDA) based on GC-O in combination with GC-MS
and ii) quantitated by stable isotope dilution assays (SIDAs) as well as
semiquantitated by internal standard method. Next, iii) odor thresholds were
determined to calculate odor activity values (OAVs; ratio of concentration to
respective odor threshold) and iv) the overall aromas of the respective samples were
4
ACS Paragon Plus Environment

Page 5 of 64
Journal of Agricultural and Food Chemistry
9
9
9
1
2
3
simulated by recombination experiments. Finally, triangle tests were applied to
confirm the analyzed data and get deeper insights into the influence of the changes
induced by blanching of the raw samples on the overall aroma.
9
4
MATERIALS AND METHODS
9
9
9
9
9
5
6
7
8
9
T. sinensis Samples. Raw green and red T. sinensis buds were purchased in a
local vegetable market (Anhui, China) in April 2018. Fresh T. sinensis buds were
frozen by liquid nitrogen, crashed into small pieces, and then powdered by SPEX
SamplePrep 6870 Freezer/Mill (Metuchen, NJ). Finally, the powder was filled into
brown glass bottles and stored at -24 °C prior to analysis. To obtain the respective
blanched samples, T. sinensis buds were heat-treated in boiling water (100 °C) for 1
min, cooled to room temperature, and then prepared as described for the raw sample.
Reference Odorants. The following reference odorants were commercially
available: acetic acid, 2-acetylpyrrole, benzyl alcohol, caryophyllene oxide, decanoic
acid, (E,E)-2,4-decadienal, dimethyl sulfide, dipropyl disulfide, dipropyl trisulfide,
2-ethyl-3,5-dimethylpyrazine, 4-ethylphenol, eugenol, hexanal, hexanoic acid,
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
00
01
02
03
04
05
06
07
08
09
10
11
12
13
14
15
16
(E)-2-hexenal,
(E)-3-hexenoic
acid,
(E)-2-hexen-1-ol,
β-ionone,
α-humulene,
3-hydroxy-4,5-dimethylfuran-2(5H)-one,
isocaryophyllene,
2-isopropyl-3-methoxypyrazine, limonene, linalool, menthol, 2-methoxyphenol,
2-methylbutanoic acid, 3-methylbutanoic acid, methyl 2-methylbutanoate,
methylpyrazine,
2-phenylethanol,
(Sigma-Aldrich
γ-nonalactone,
1-octen-3-ol,
propanoic acid,
Taufkirchen,
phenol,
and
phenylacetic
acid,
α-pinene,
2,3,5-trimethylpyrazine
cyclopentadecanone,
Chemie,
Germany);
2-methylbutanal, 3-methylbutanal, 5-methyl-2-methoxyphenol, and 1-octen-3-one
(Alfa Aesar, Karlsruhe, Germany); aromadendrene, butyrolactone, β-caryophyllene,
(E)-2-decenal, heptanoic acid, methyl hexanoate, nonanoic acid, and valencene
(Fluka, Neu-Ulm, Germany); 2-ethyl-3-methylpyrazine and nonanal (Acros Organics;
5
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 6 of 64
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
Thermo Fisher Scientific, Schwerte, Germany); isoeugenol and (E,Z)-2,6-nonadienal
(Lancaster, Mühlheim/Main, Germany); 2-ethyl-6-methylpyrazine (Pyrazine
Specialties,
Ellenwood,
GA);
vanillin
(Merck,
Darmstadt,
Germany);
3-methylnonane-2,4-dione (Chemos, Regenstauf, Germany); and 2-pyrrolidone (TCI,
Eschborn, Germany). 2-Mercapto-3,4-dimethyl-2,3-dihydrothiophene was a gift from
Firmenich (Geneva, Switzerland).
2
Chemicals. Methyl octanoate, potassium hydroxide, [ H ]-propyl bromide, propyl
3
iodide, sulfur, and tetrahydrofuran were obtained from Sigma-Aldrich Chemie. Ethanol,
hydrochloric acid, sodium carbonate, sodium chloride, and anhydrous sodium sulfate
were from Merck. Deionized water used for HPLC was prepared using a MiliQ
Advantage A10 Water Purification System (Milipore S.A.S., Molsheim, France).
Acetonitrile used for HPLC analysis was of HPLC grade (Merck). Dichloromethane,
diethyl ether, and n-pentane (Merck) were freshly distilled prior to use. All chemicals
were at least of analytical grade.
Stable Isotopically Labeled Internal Standards. The following stable
13
isotopically labeled internal standards were commercially obtained: [ C ]-acetic acid,
2
2
2
2
2
[ H ]-benzyl alcohol, [ H ]-dimethyl sulfide, [ H ]-hexanoic acid, [ H ]-1-octen-3-one,
5
6
3
3-5
13
13
2
[ C ]-phenylacetic acid, [ C ]-2-phenylethanol, and [ H ]-propanoic acid
2
2
2
2
2
(Sigma-Aldrich Chemie); [ H ]-decanoic acid, [ H ]-nonanoic acid, and
2-3
2
2
2
[ H ]-2-pyrrolidone (C/D/N Isotopes, Quebec, Canada); [ H ]-cis-isoeugenol
6
4
2
(aromaLAB, Planegg, Germany); and [ H ]-2-methylbutanoic acid (EQ Laboratories,
9
Augsburg, Germany).
The following standards were prepared as previously described:
2
9
2
10
13
14
[ H ]-(E,E)-2,4-decadienal,
[ H ]-2-ethyl-3,5-dimethylpyrazine,
5
3-5
2
11
2
12
2
[ H ]-4-ethylphenol,
[ H ]-hexanal,
[ H ]-(E)-2-hexenal,
2
3-4
4-6
2
9
13
[ H ]-(E)-2-hexen-1-ol,
[ C ]-3-hydroxy-4,5-dimethylfuran-2(5H)-one,
2
2
6
ACS Paragon Plus Environment

Page 7 of 64
Journal of Agricultural and Food Chemistry
2
15
2
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
[ H ]-β-ionone,
[ H ]-2-isopropyl-3-methoxypyrazine
analogue
to
3
3
2
16
2
17
2
18
[ H ]-2-isobutyl-3-methoxypyrazine,
[ H ]-linalool,
[ H ]-2-methylbutanal,
3
2-3
2
2
5-ethyl-2-methoxyphenol,19
[ H ]-5-methyl-2-methoxyphenol
analogue
to
6-8
2
2-methylbutanoate,12
2
13
[ H ]-methyl
[ H ]-3-methylnonane-2,4-dione,
3
3
2
20
2
13
2
21
24
[ H ]-methylpyrazine,
[ H ]-(E,Z)-2,6-nonadienal,
[ H ]-γ-nonalactone,
2-3
2
2-3
2
22
2
23
2
[ H ]-nonanal,
[ H ]-γ-octalactone,
[ H ]-1-octen-3-ol,
4
2-3
3-6
2
17
2
19
[ H ]-2,3,5-trimethylpyrazine, and [ H ]-vanillin.
3-4
3
The concentrations of the stable isotopically labeled compounds were determined
as recently described.25
Syntheses. Dipropyl Disulfide and Dipropyl Trisulfide.26 Finely powdered
potassium hydroxide (1 g) was added to tetrahydrofuran (THF; 14 mL, containing
0.2% of water), and a white suspension was obtained. After adding powdered sulfur
(0.256 g, 1 mmol) to this vigorously stirred suspension, the reaction mixture was
stirred for another 5 min. A brown coloration was observed, which disappeared upon
addition of a solution of propyl iodide (1.36 g, 8 mmol) in THF (4 mL, containing 0.2%
of water). The mixture was stirred at room temperature for another 2 h, and then
filtered. After evaporation of the solvent via a rotary evaporator (50 °C, 380 mbar), the
residue was dissolved in n-pentane (~2 mL), and purification was performed via
column chromatography using purified silica gel 60 (20 g, 0.040-0.063 mm; Merck) in
a water-cooled glass column (12 °C; 25 cm × 1 cm id) using n-pentane (200 mL). The
reaction yield was 50%, with a dipropyl disulfide/dipropyl trisulfide ratio of 29/71,
based on area counts obtained via gas chromatography-flame ionization detection
(GC-FID). The purity of the mixture was 97%, and the odorants were finally
characterized via GC-MS.
Dipropyl disulfide: MS (EI): m/z (%): 150 (100), 43 (96), 108 (50), 73 (30), 41 (24),
66 (20), 39 (19), 40 (18), 45 (18), 74 (10), 48 (8), 79 (8), 151 (8), 152 (8), 110 (5).
7
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 8 of 64
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
MS (CI): m/z (%): 151 (M + 1, 100).
Dipropyl trisulfide: MS (EI): m/z (%): 182 (100), 75 (98), 41 (40), 43 (40), 73 (30),
45 (13), 47 (13), 184 (12), 44 (11), 117 (11), 39 (10), 98 (10), 140 (10), 183 (10), 64
(8), 105 (8).
MS (CI): m/z (%): 183 (M + 1, 100).
2
2
27
[ H ]-Dipropyl Disulfide and [ H ]-Dipropyl Trisulfide. The same procedure as for
6
6
the synthesis of dipropyl disulfide and dipropyl trisulfide was applied, except that
2
[ H ]-propyl bromide (1.01 g, 8 mmol) in THF (4 mL, containing 0.2% of water) was
3
used as alkylating reagent. The products obtained were characterized by GC-MS.
2
[ H ]-Dipropyl disulfide: MS (EI): m/z (%): 156 (100), 46 (78), 40 (45), 44 (43), 43
6
(40), 111 (28), 112 (25), 45 (20), 73 (15), 47 (13), 75 (13), 76 (13), 41 (12), 67 (9), 66
(8), 158 (8), 42 (7), 157 (7).
MS (CI): m/z (%): 157 (M + 1, 100).
2
[ H ]-Dipropyl trisulfide: MS (EI): m/z (%): 188 (100), 78 (78), 46 (75), 79 (52), 40
6
(45), 44 (40), 43 (30), 73 (25), 45 (22), 190 (16), 189 (13), 64 (10), 41 (8), 48 (8), 123
(8), 75 (6), 108 (6), 99 (5).
MS (CI): m/z (%): 189 (M + 1, 100).
The concentrations of the isotopically labeled standards were determined by a
Trace 2000 gas chromatograph (Thermo, Egelsbach, Germany) equipped with an FID
using methyl octanoate as the internal standard. First, the FID response factor was
determined for each unlabeled reference compound (dipropyl disulfide and dipropyl
trisulfide were commercially bought) and methyl octanoate. Then, the concentration of
the labeled standard was calculated via the peak areas of the labeled compound and
methyl octanoate using the FID response factor determined for the unlabeled
compound.25
Isolation of the Volatiles. T. sinensis powder (20 g) was extracted with
8
ACS Paragon Plus Environment

Page 9 of 64
Journal of Agricultural and Food Chemistry
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
dichloromethane (3 × 200 mL) by vigorously stirring (3 × 0.5 h) at room temperature.
The organic extracts were combined, filtered, and dried over anhydrous sodium
sulfate (10 g). The volatiles were separated from the nonvolatile fraction by high
vacuum distillation using the solvent assisted flavor evaporation (SAFE) technique.28
Fractionation of the Volatiles. The SAFE distillate obtained was separated into
the acidic fraction (AF) and the neutral/basic fraction (NBF) by liquid-liquid extraction
with an aqueous Na CO solution (0.5 mol/L; 3 × 50 mL). The organic phase
2
3
containing the neutral/basic fraction (NBF) was washed with a saturated sodium
chloride solution (3 × 50 mL), dried over anhydrous Na SO , filtered, and
2
4
concentrated at 42 °C to a volume of ∼3 mL using a Vigreux column (50 cm × 1 cm id)
29
and finally to ∼200 µL by microdistillation. The aqueous phase containing the acidic
fraction (AF) was adjusted to a pH value of 2−3 using hydrochloric acid. Afterward, the
odorants were extracted with diethyl ether (1+1 by vol., 3 × 50 mL). The combined
organic phases were washed again with a saturated sodium chloride solution and
dried over anhydrous Na SO . After filtration, the fraction was concentrated as
2
4
described above for NBF.
For an unequivocal compound identification, NBF was further fractionated by
column chromatography using purified silica gel 60 (20 g, 0.040-0.063 mm; Merck) in
a water-cooled glass column (12 °C; 30 cm × 1 cm id) with the following
n-pentane/diethyl ether mixtures: 100:0, 80:20, 50:50, and 0:100 (v:v; 50 mL each).
Each portion obtained (50 mL) was dried over anhydrous Na SO , and then
2
4
concentrated to a final volume of ~200 μL (silica gel fractions, SGF 1-4) as described
above.30
Isolation of Di-1-Propenyl Disulfides and Di-1-Propenyl Trisulfides. The
isolation of di-1-propenyl disulfides and di-1-propenyl trisulfides from SGF 1 obtained
from 4 kg of raw green T. sinensis was performed via high performance liquid
9
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 10 of 64
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
chromatography (HPLC) using a PU-2089 Plus Quaternary pump and an UV-2075
Plus HPLC UV-VIS detector; both Jasco, Pfungstadt, Germany). Aliquots (100 µL) of
SGF 1 (dissolved in acetonitrile (ACN)) were injected onto a Nucleosil 100-5 C18 AB
column (250 mm × 4.6 mm; Macherey-Nagel, Dϋren, Germany) and eluted with the
following gradient: 50/50 ACN/H O for 20 min, then 100/0 ACN/H O for further 20 min.
2
2
Two sulfury smelling fractions obtained after 21 and 29 min were collected, dried with
anhydrous sodium sulfate, and the purity of these two fractions were checked by
GC-O/FID and GC-MS (95% and 92%, respectively). Subsequently, these two
fractions were characterized as di-1-propenyl disulfide (3 isomers) and di-1-propenyl
trisulfide (3 isomers) by GC-MS.
(E,E)-, (E,Z)-, and (Z,Z)-Di-1-propenyl disulfide: MS (EI): m/z (%): 146 (100), 45
(60), 113 (38), 73 (35), 74 (33), 82 (31), 67 (21), 39 (20), 41 (20), 59 (19), 71 (19), 72
(19), 69 (18), 47 (11), 101 (8), 147 (8), 148 (8), 85 (7).
MS (CI): m/z (%): 147 (M + 1, 100).
(E,E)-, (E,Z)-, and (Z,Z)-Di-1-propenyl trisulfide: MS (EI): m/z (%): 114 (100), 178
(62), 45 (60), 105 (45), 73 (33), 41 (31), 61 (31), 39 (24), 71 (24), 99 (23), 106 (20),
112 (20), 118 (18), 47 (12), 180 (12), 64 (10), 116 (9).
MS (CI): m/z (%): 179 (M + 1, 100).
High-Resolution Gas Chromatography-Olfactometry/Flame Ionization
Detection (HRGC-O/FID). HRGC-O/FID was performed using a TRACE GC 2000
(ThermoQuest) equipped with either a DB-FFAP capillary column (30 m × 0.25 mm id,
0.25 µm film thickness) or a DB-5 capillary column (30 m × 0.32 mm id, 0.25 µm film
thickness) (both J&W Scientific; Agilent Technologies, Waldbronn, Germany). Helium
was used as the carrier gas (flow rate = 1.9 mL/min). Two minutes after manual cold
on-column injection of an aliquot of the sample (2 μL) at 40 °C, the oven temperature
was raised with 6 °C/min to 240 °C and then held for 10 min. A Y-type quick-seal
10
ACS Paragon Plus Environment

Page 11 of 64
Journal of Agricultural and Food Chemistry
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
glass splitter (Chrompack, Frankfurt, Germany) was used at the end of the column to
separate the effluent into two equal parts to a sniffing port held at 230 °C and an FID
held at 250 °C, which enabled a simultaneous detection of the respective odor
qualities and the FID chromatogram. Linear retention indices for each compound were
calculated using the retention times of a series of n-alkanes (C6-C26 (DB-FFAP) and
C6-C18 (DB-5), respectively).
Comprehensive High-Resolution Gas Chromatography-Time-of-Flight Mass
Spectrometry (HRGC×HRGC-TOF-MS) for Identification. The instrument consisted
of a gas chromatograph 6890N (Agilent, Böblingen, Germany) equipped with a
DB-FFAP capillary column (30 m × 0.25 mm id, 0.25 µm film thickness) in the first
dimension and a DB-5 capillary column (2 m × 0.15 mm id, 0.30 µm film thickness) in
the second dimension (both J&W Scientific). The front part of the DB-FFAP capillary
column passed a liquid nitrogen-cooled dual stage quad-jet thermal modulator (Leco,
St. Joseph, MI), and the end part was connected via a heated (250 °C) transfer line to
the inlet of a Pegasus III TOF-MS (Leco). Sample injections were performed by a
GC-PAL autosampler (CTC Analytics, Zwingen, Switzerland). Helium was applied as
the carrier gas (flow rate = 2 mL/min). The following temperature program was used:
for the first oven, 40 °C held for 2 min, then raised with 6 °C/min to 230 °C, held for 5
min. The modulation time was set to 4 s. For the second oven, the temperature
started at 70 °C (2 min), raised with 6 °C/min to 250 °C, held for 10 min. Mass spectra
were recorded in electron ionization (EI) mode at 70 eV at a rate of 100 spectra/s. The
scan range was set at m/z 35-350. Data was analyzed by means of GC Image
(Lincoln, NE).
Comparative Aroma Extract Dilution Analysis (cAEDA). The flavor dilution
(FD) factors of the odorants were determined via cAEDA as previously reported.8
Comparative Static Headspace Aroma Dilution Analysis (cSH-ADA) Based
11
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 12 of 64
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
on Static Headspace High-Resolution Gas Chromatography−Olfactometry/Mass
Spectrometry (SH-HRGC-O/MS). To detect very volatile components and the
compounds which were co-eluting with the solvent during AEDA, SH-HRGC-O/MS
8
was performed as previously reported. FD factors of the odorants were determined
via cSH-ADA as previously reported.8
Stable Isotope Dilution Assays (SIDAs). The stable isotopically labeled internal
standards (0.1-400 µg, dissolved in dichloromethane; amount depending on the
concentration of the respective analyte determined in preliminary experiments) and
dichloromethane (30-300 mL) were added to the powdered materials (1-10 g). After
equilibration, the samples were worked-up as described for isolation of the volatiles.
The SAFE distillate obtained was concentrated to ~ 200 µL as described above and
was used for high-resolution gas chromatography-mass spectrometry (HRGC-MS) or
two-dimensional heart-cut high-resolution gas chromatography-mass spectrometry
(HRGC/HRGC-MS) (Table 1).
High-Resolution Gas Chromatography-Mass Spectrometry (HRGC-MS).
HRGC-MS was performed by a gas chromatograph 431 (Varian, Darmstadt)
equipped with a DB-FFAP capillary column (30 m × 0.25 mm id, 0.25 µm film
thickness; J&W Scientific). The initial oven temperature (40 °C) was held for 2 min,
then raised with 6 °C/min to 230 °C, and held for 10 min. The respective response
factors (R ) were determined via analyzing known mixtures of the respective
f
unlabeled analyte and the corresponding stable isotopically labeled internal standard
in five different mass ratios (5:1, 3:1, 1:1, 1:3, 1:5) (Table 1).
Two-Dimensional Heart-Cut High-Resolution Gas Chromatography-Mass
Spectrometry (HRGC/HRGC-MS). The instrument consisted of a TRACE GC 2000
(ThermoQuest) equipped with a cold on-column injector and a DB-FFAP capillary
column (30 m × 0.32 mm id, 0.25 µm film thickness; J&W Scientific) in the first
12
ACS Paragon Plus Environment

Page 13 of 64
Journal of Agricultural and Food Chemistry
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
dimension, and a gas chromatograph CP-3800 (Varian) equipped with a DB-1701
capillary column (30 m × 0.25 mm id, 0.25 µm film thickness; J&W Scientific) in the
second dimension. Sample injections were performed by a CombiPal autosampler
(CTC Analytics). The target compound was transferred via a moving column stream
switching (MCSS) system (ThermoQuest) onto the second column via a cold trap (-
100 °C; cooled by liquid nitrogen), and was finally analyzed by a Saturn 2000 ion trap
mass spectrometer (Varian). The individual temperature programs for each analyte in
both dimensions were optimized according to its heart-cut time in the first dimension
and the retention time in the second dimension. Mass spectra were recorded in CI
mode at 105 eV with methanol as the reactant gas.
Headspace Solid Phase Microextraction Combined with High-Resolution
Gas Chromatography-Mass Spectrometry (HS-SPME-HRGC-MS). Due to their low
boiling points, dimethyl sulfide, 2-methylbutanal, and 3-methylbutanal were
quantitated by SIDAs using HS-SPME-HRGC-MS as previously reported.8
High-Resolution Gas Chromatography-Flame Ionization Detection
(HRGC-FID) for Semiquantitation. A Trace GC Ultra (ThermoQuest) equipped with
a DB-FFAP (30 m × 0.32 mm id, 0.25 µm film thickness; J&W Scientific) was used for
semiquantitation of aromadendrene, β-caryophyllene, caryophyllene oxide,
α-humulene, isocaryophyllene, and valencene via internal standard method as
previously reported (internal standard: cyclopentadecanone).8
Determination of Orthonasal Odor Thresholds (OTs) in Water. Orthonasal
OTs in water of di-1-propenyl disulfide (3 isomers), di-1-propenyl trisulfide (3 isomers),
and 2-mercapto-3,4-dimethyl-2,3-dihydrothiophene (2 isomers) were newly
determined by means of triangle tests according to a previously published protocol.31
Determination of OTs in Air. OTs in air of (E,E)-di-1-propenyl disulfide,
(E,Z)-di-1-propenyl disulfide, (Z,Z)-di-1-propenyl disulfide, (E,E)-di-1-propenyl
13
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 14 of 64
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
trisulfide,
(E,Z)-di-1-propenyl
trisulfide,
(Z,Z)-di-1-propenyl
trisulfide,
and
cis-2-mercapto-3,4-dimethyl-2,3-dihydrothiophene,
trans-2-mercapto-3,4-dimethyl-2,3-dihydrothiophene were determined by the
32
following procedure: isomers of di-1-propenyl disulfide, di-1-propenyl trisulfide, and
2-mercapto-3,4-dimethyl-2,3-dihydrothiophene were dissolved in diethyl ether. After
adding (E)-2-decenal as reference compound, stepwise with diethyl ether diluted
solutions (1+1, v+v) were analyzed by HRGC-O (DB-FFAP capillary column) until no
odorant was detectable at the sniffing port. The odor thresholds in air were then
calculated with the previously published odor threshold of (E)-2-decenal (2.7 ng/L).33
Aroma Profile Analysis (APA). Aqueous solutions of each of the nine reference
compounds in concentrations 50-fold above their respective odor thresholds were
prepared to define the odor descriptors: acetic acid (vinegar-like), eugenol (clove-like),
hexanal (green), β-ionone (flowery), 2-isopropyl-3-methoxypyrazine (earthy),
2-mercapto-3,4-dimethyl-2,3-dihydrothiophene
(cooked
onion-like,
TS-like),
2-methoxyphenol (phenolic), 2-methylbutanal (malty), and 3-methylnonane-2,4-dione
(hay-like, aniseed-like, fishy). Aliquots of the samples (1 g) were presented in covered
®
odorless Teflon vessels (id = 40 mm). Sensory analysis was carried out in a sensory
room at 21 ± 1 °C equipped with individual booths. The sensory panel, consisting of
15-20 weekly trained members able to describe and recognize odor qualities, and
thus, to perform a comparative APA, evaluated the odor intensities of the aroma
attributes of each T. sinensis sample from 0 (not perceivable) to 3 (strongly
perceivable) on a seven-point linear scale by steps of 0.5.
Aroma Recombination. Raw green and red T. sinensis (10 g each) were
extracted with dichloromethane (100 mL each) several times until the residues were
odorless. These odorless residues were used as matrices for the recombinates of raw
green and red T. sinensis as well as the respective blanched samples. Aqueous
14
ACS Paragon Plus Environment

Page 15 of 64
Journal of Agricultural and Food Chemistry
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
solutions of all odorants with an OAV ≥ 1 were prepared from ethanolic stock solutions
and were added to the respective matrices in their original concentrations determined
in the samples. After vigorous shaking for 30 min, the recombinates were evaluated in
the same way as described above for APA.
Triangle Tests. To confirm the data obtained before and after blanching, and to
get deeper insights into the blanching-induced changes of the key odorants, and thus,
of the overall aroma, triangle tests were designed and interpreted according to the
method ISO 4120:2004. Therefore, spiking experiments to green and red blanched T.
sinensis samples were performed by adding reference aroma compounds to obtain
their initial concentrations in the respective raw samples. Then, these spiked samples
were compared to the original raw T. sinensis samples. First, aqueous solutions of all
analyzed aroma compounds with OAVs ≥ 1 were added to blanched T. sinensis
samples in the way that their naturally occurring concentrations determined in raw T.
sinensis were obtained. Subsequently, different compounds eliciting a certain odor
quality were added to blanched T. sinensis sample, again in the amounts to obtain
their naturally occurring concentrations in raw T. sinensis. For “cooked
onion-like/TS-like”, the following odorants were chosen: dimethyl sulfide, (E,E)-, (E,Z)-,
and (Z,Z)-di-1-propenyl disulfide, (E,E)-, (E,Z)-, and (Z,Z)-di-1-propenyl trisulfide, and
cis- and trans-2-mercapto-3,4-dimethyl-2,3-dihydrothiophene. The “green” group
included hexanal, (E)-2-hexenal, (E)-2-hexen-1-ol, and (E,Z)-2,6-nonadienal, the
“phenolic” group 4-ethylphenol and 2-methoxyphenol, the “earthy” group the three
pyrazines
2-ethyl-3,5-dimethylpyrazine,
2-isopropyl-3-methoxypyrazine,
and
2,3,5-trimethylpyrazine, the “malty” group 2-methylbutanal and 3-methylbutanal, and
“vinegar-like” was represented by acetic acid. Finally, aqueous solutions of the
remaining analyzed aroma compounds with OAVs ≥ 1 were added to blanched T.
sinensis in the same way.
15
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 16 of 64
3
77
RESULTS AND DISCUSSION
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
Aroma Profiles of Raw Green and Red T. sinensis. First of all, aroma profile
analysis was performed to evaluate the overall aroma of green and red T. sinensis
using nine odor descriptors. Thereby, the cooked onion-like/TS-like (2.3) odor
impression was the most intense attribute in green T. sinensis, followed by green (1.7),
hay-like/aniseed-like/fishy (1.7), earthy (1.6), malty (1.3), vinegar-like (1.0), phenolic
(0.7), clove-like (0.6), and flowery (0.5). The aroma profile of red T. sinensis revealed
clearly lower intense cooked onion-like/TS-like (1.8) and green (1.1) odor notes,
whereas the phenolic attribute was ranked with a higher intensity (1.3) (Figure 1).
Odorant Screening in Raw Green and Red T. sinensis. Application of cAEDA
to the SAFE distillate obtained from buds of green and red T. sinensis revealed 57
aroma-active areas present in the FD factor range between 8 and 4096 in at least one
of the two T. sinensis aroma extracts (Table 2, Figure 2). For structure identification,
the retention indices on two columns of different polarities (DB-FFAP and DB-5) as
well as the odor qualities and intensities detected at the sniffing port during AEDA
were compared to data available in an in-house database containing > 1000 odorants.
Subsequently, authentic reference compounds were analyzed by GC-O with two
different columns to match the retention indices and odor descriptors. The final step of
identification was based on mass spectrometry in EI and CI mode in comparison to
data obtained from the respective reference compounds. For an unequivocal
identification (avoiding a possible overlap of minor odorants by (aroma-active)
compounds present at higher concentrations), the SAFE distillates were fractionated
by liquid-liquid extraction into the acidic fraction (AF) and the neutral/basic fraction
(NBF). The NBF was further fractionated into four subfractions by silica gel
chromatography (SGF 1-4). All the fractions were analyzed by GC-O and GC-MS to
identify the odorants.
16
ACS Paragon Plus Environment

Page 17 of 64
Journal of Agricultural and Food Chemistry
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
Following the abovementioned procedure, the highest FD factors were obtained
for (E,E)-, (E,Z)-, and (Z,Z)-di-1-propenyl disulfide (21-23, FD factors of 4096 (green)
and
2048
(red),
roasted
onion-like),
cis-
and
trans-
2-mercapto-3,4-dimethyl-2,3-dihydrothiophene (30 and 31, 2048 and 1024, cooked
onion-like/TS-like), and eugenol (57, 1024 and 2048, clove-like) in both raw green and
red T. sinensis samples (Table 2).
Differences in FD factors between green and red T. sinensis were e.g. found for
dimethyl sulfide (1, FD factors of 32 (green) and 16 (red), cabbage-like), hexanal (6,
128 and 32, green/grassy), (E)-2-hexenal (9, 128 and 32, green apple-like),
(E)-2-hexen-1-ol (15, 64 and 16, green/fruity), acetic acid (19, 16 and 64, vinegar-like),
linalool (25, 16 and 64, citrus-like/flowery), (E,Z)-2,6-nonadienal (27, 128 and 32,
green/cucumber-like), 2-methylbutanoic acid (35, 8 and 32, fruity/sweaty),
3-methylbutanoic acid (36, 8 and 32, sweaty), (E,E)-, (E,Z)- and (Z,Z)-di-1-propenyl
trisulfide (39-41, 512 and 128, cooked onion-like), 2-methoxyphenol (44, 8 and 256,
smoky/phenolic), 2-phenylethanol (46, 4 and 32, flowery/honey-like), and
4-ethylphenol (56, < 4 and 64, fecal-like/phenolic) (Table 2).
The presence of the very volatile dimethyl sulfide (1), 3-methylbutanal (2, malty),
and 2-methylbutanal (3, malty) in both raw samples was confirmed via cSH-ADA.
Based on FD factors, nine sulfur-containing compounds including dimethyl sulfide
(1), (E,E)-, (E,Z)-, and (Z,Z)-di-1-propenyl disulfide (21-23), cis- and trans-
2-mercapto-3,4-dimethyl-2,3-dihydrothiophene (30 and 31), and (E,E)-, (E,Z)-, and
(Z,Z)-di-1-propenyl trisulfide (39-41) should contribute to the intense cooked
onion-like/TS-like aroma note of green T. sinensis. Their FD factors in green T.
sinensis were all higher than the respective FD factors in red T. sinensis. In addition,
the green smelling aldehydes hexanal (6), (E)-2-hexenal (9), and
(E,Z)-2,6-nonadienal (27) as well as the green smelling (E)-2-hexen-1-ol (15) showed
17
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 18 of 64
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
higher FD factors in green T. sinensis compared to red T. sinensis, probably
responsible for the intense green aroma note of the green variety. 2-Methoxyphenol
(44) and 4-ethylphenol (56), both present at higher FD factors in red TS (256 and 64)
compared to green T. sinensis (8 and < 4), should lead to the specific phenolic aroma
note of red T. sinensis (Table 2).
Differences in Odorant Concentrations in Raw Green and Red T. sinensis.
To get knowledge about the role of the individual T. sinensis odorants discussed
above and to clarify the aroma differences between the two T. sinensis varieties,
quantitative analysis of aroma compounds revealing FD factors ≥ 32 in at least one
sample as well as of compounds characterized as key odorants in dried T. sinensis in
our previous study8 was performed by SIDAs (Table 1). Aromadendrene,
β-caryophyllene, caryophyllene oxide, α-humulene, isocaryophyllene, and valencene
were semiquantitated by GC-FID via internal standard method using
cyclopentadecanone as standard. Concentrations of the nine sulfur-containing
compounds dimethyl sulfide (14000 µg/kg (green) vs 11000 µg/kg (red)), sum of
di-1-propenyl disulfide isomers (1360 µg/kg vs 595 µg/kg), sum of
2-mercapto-3,4-dimethyl-2,3-dihydrothiophene isomers (230 µg/kg vs 51.1 µg/kg),
and sum of di-1-propenyl trisulfide isomers (1300 µg/kg vs 230 µg/kg)) in green T.
sinensis were much higher than those in red T. sinensis (Table 3). For hexanal,
(E)-2-hexenal, (E)-2-hexen-1-ol, and (E,Z)-2,6-nonadienal, quantitative results
revealed higher concentrations in green T. sinensis compared to those in red T.
sinensis. Quantitation experiments also showed much higher concentrations of
2-methoxyphenol and 4-ethylphenol in red T. sinensis, both present at amounts ≥
factor of 60 compared to raw green T. sinensis (Table 3), confirming the data obtained
during cAEDA.
Also differences in other aroma-active components of green and red T. sinensis
18
ACS Paragon Plus Environment

Page 19 of 64
Journal of Agricultural and Food Chemistry
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
samples were found. For example, green T. sinensis revealed clearly higher
concentrations for 2-methylbutanal (2470 µg/kg vs 1770 µg/kg), nonanoic acid (1760
µg/kg vs 1140 µg/kg), decanoic acid (1570 µg/kg vs 742 µg/kg), 1-octen-3-ol (14.5
µg/kg vs 7.78 µg/kg), and 1-octen-3-one (3.47 µg/kg vs 1.16 µg/kg). All other
compounds, such as acetic acid (772 mg/kg vs 2750 mg/kg), caryophyllene (4.51
mg/kg vs 15.4 mg/kg), propanoic acid (2.26 mg/kg vs 16.7 mg/kg), 2-methylbutanoic
acid (1200 µg/kg vs 7860 µg/kg), 3-methylbutanoic acid (1120 µg/kg vs 10600 µg/kg),
aromadendrene (178 µg/kg vs 8230 µg/kg), phenylacetic acid (90.8 µg/kg vs 1140
µg/kg), linalool (55.4 µg/kg vs 301 µg/kg), 2-phenylethanol (50.6 µg/kg vs 2150 µg/kg),
and 2-isopropyl-3-methoxypyrazine (29.1 µg/kg vs 136 µg/kg) were present at clearly
higher concentrations in red T. sinensis (Table 3).
Differences of Odor Activity Values (OAVs) in Raw Green and Red T.
sinensis. To get information about the contribution of a respective odorant to the
overall aroma of the two T. sinensis samples, OAVs were calculated for each odorant.
Due to the lack of pure isomers of di-1-propenyl disulfide, di-1-propenyl trisulfide, and
2-mercapto-3,4-dimethyl-2,3-dihydrothiophene, their OAVs were calculated with their
respective OTs in water based on their isomer mixtures. In green T. sinensis, 36
odorants were present in concentrations higher than the respective odor thresholds.
Thereby, the highest OAV was calculated for the three isomers of di-1-propenyl
disulfide
(400000),
followed
by
dimethyl
sulfide
(47000),
2-isopropyl-3-methoxypyrazine (7500), β-ionone (6500), (E,Z)-2,6-nonadienal (6300),
the two isomers of 2-mercapto-3,4-dimethyl-2,3-dihydrothiophene (5900), the three
isomers of di-1-propenyl trisulfide (5000), eugenol (2800), 2-methylbutanal (1600),
3-methylbutanal (1400), and hexanal (780) (Table 4).
For red T. sinensis, 41 odorants showed OAVs ≥ 1. The highest OAV was
obtained, again, for the three isomers of di-1-propenyl disulfide (170000), dimethyl
19
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 20 of 64
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
506
sulfide (37000), 2-isopropyl-3-methoxypyrazine (35000), and β-ionone (7000),
followed by eugenol (3800), 3-methylbutanal (1400), the two isomers of
2-mercapto-3,4-dimethyl-2,3-dihydrothiophene (1300), 2-methylbutanal (1200), the
three isomers of di-1-propenyl disulfide (890), and (E,Z)-2,6-nonadienal (770) (Table
4).
A comparison between both raw T. sinensis revealed the sulfur-containing
compounds
di-1-propenyl
disulfide,
dimethyl
sulfide,
2-mercapto-3,4-dimethyl-2,3-dihydrothiophene, and di-1-propenyl trisulfide with
clearly lower OAVs in red T. sinensis. Moreover, (E,Z)-2,6-nonadienal, hexanal,
(E)-2-hexenal, and (E)-2-hexen-1-ol showed clearly lower OAVs in red T. sinensis. In
contrast, 2-methoxyphenol and 4-ethylphenol resulted in clearly higher OAVs in red T.
sinensis (440 and 32) compared to those in green T. sinensis (7 and <1) (Table 4).
Obvious differences in other compounds between green and red T. sinensis were also
found for 2-isopropyl-3-methoxypyrazine (7500 (green) and 35000 (red)),
1-octen-3-one (220 and 73), acetic acid (140 vs 490), linalool (96 and 520), valencene
(8 and 27), caryophyllene (4 and 13), 3-methylbutanoic acid (2 and 22),
aromadendrene (1 and 24), phenylacetic acid (1 and 17), and 2-phenylethanol (<1
and 15) (Table 4).
OTs in Air of Sulfur-Containing Isomers. Due to the good separation of each
isomer
of
di-1-propenyl
disulfide,
di-1-propenyl
trisulfide,
and
2-mercapto-3,4-dimethyl-2,3-dihydrothiophene during GC-O using a DB-FFAP
capillary column, the odor qualities and odor thresholds in air of the respective
isomers were determined. Thereby, (E,E)-di-1-propenyl disulfide (roasted onion-like,
0.015
ng/L),
trans-2-mercapto-3,4-dimethyl-2,3-dihydrothiophene
(cooked
onion-like/TS-like, 0.089 ng/L), (E,Z)-di-1-propenyl disulfide (roasted onion-like, 0.092
ng/L), and (Z,Z)-di-1-propenyl disulfide (roasted onion-like, 0.26 ng/L) showed
20
ACS Paragon Plus Environment

Page 21 of 64
Journal of Agricultural and Food Chemistry
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
extremely low odor thresholds in air. In contrast, the odor thresholds in air of cis-
2-mercapto-3,4-dimethyl-2,3-dihydrothiophene (cooked onion-like/TS-like), (E,E)-,
(E,Z)-, and (Z,Z)-di-1-propenyl trisulfide (all cooked onion-like) were ≥ 1.7 ng/L (Table
5). Interestingly, the odor qualities did not differ between the isomers.
Aroma Recombination Experiments of Raw Green and Red T. sinensis. To
validate the data obtained after identification and quantitation, aroma recombinates of
both samples were prepared in respective odorless residues obtained after solvent
extraction of the initial sample material. To those matrices, all odorants showing OAVs
≥ 1 (Table 4) were added in their natural occurring concentrations. APA showed very
good similarities between both raw T. sinensis samples and the respective
recombinate, proving the successful characterization of the key aroma compounds for
raw green and red T. sinensis (Figure 3).
Influence of the Blanching Process on Key Odorants in Blanched Green and
Red T. sinensis. To characterize the aroma-active compounds in blanched green
and red T. sinensis, cAEDA and cSH-ADA were applied to T. sinensis blanched at
100 °C for 1 min. The odorant screening demonstrated that all the odorants identified
in raw green T. sinensis in the first part of the actual study could be identified again.
Lowered FD factors (30 compounds within the FD factor range of 8-1024) determined
in green T. sinensis, for example, for most of the sulfury and green smelling odorants,
gave first hints, that the loss of these odorants might be responsible for the aroma
profile change after blanching (Table 2 and Figure 4A). Lower FD factors were also
found for, e.g., 3-methylbutanal (32 vs 4), 2-methylbutanal (32 vs 4),
2-isopropyl-3-methoxypyrazine (512 vs 128), eugenol (1024 vs 256), and vanillin (32
vs 4) (Table 2).
A comparison of FD factors in raw and blanched red T. sinensis showed the same
decreasing trends as for green T. sinensis. Additionally, the FD factors of
21
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 22 of 64
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
2-methoxyphenol (256 vs 64) and 4-ethylphenol (64 vs < 4) were lower, which might
also indicate the lower phenolic odor note after blanching (Table 2 and Figure 4B).
Quantitation of Key Odorants in Blanched Green and Red T. sinensis. To get
a deeper insight into the changes of the overall aroma in blanched T. sinensis,
concentrations of selected odorants with high FD factors or compounds showing
obvious differences in their FD factors after blanching were determined via SIDAs or
internal standard method and their OAVs were calculated. Compared to raw green T.
sinensis, the concentrations of dimethyl sulfide (14000 µg/kg (raw) vs 1710 µg/kg
(blanched)), (E)-2-hexenal (3290 µg/kg vs 433 µg/kg), 2-methylbutanal (2470 µg/kg
vs 213 µg/kg), hexanal (1880 µg/kg vs 329 µg/kg), 3-methylbutanal (723 µg/kg vs
18.3 µg/kg), (E)-2-hexen-1-ol (340 µg/kg vs 31.9 µg/kg), and 4-ethylphenol (3.92
µg/kg vs 0.53 µg/kg) decreased by 83 to 97% (Table 3) in blanched green T. sinensis.
The concentrations of cis- and trans-2-mercapto-3,4-dimethyl-2,3-dihydrothiophene
and (E,Z)-2,6-nonadienal decreased between 37 and 50%, of the three isomers of
di-1-propenyl disulfide between 4 and 28%, and of di-1-propenyl trisulfide between 5
and 70%. In general, nearly all odorants were present at decreased concentrations
after blanching, e.g., acetic acid, eugenol, decanoic acid, aromadendrene, β-ionone,
2-phenylethanol, and methyl 2-methylbutanoate. In contrast, only a few compounds
increased: nonanal (+16%), 3-methyl-2,4-nonadione (+18%), (E,E)-2,4-decadienal
(+112%), and 1-octen-3-one (+118%) (Table 3).
Calculation of OAVs in blanched green T. sinensis showed the three isomers of
di-1-propenyl disulfide with the highest OAV of 320000, followed by
2-isopropyl-3-methoxypyrazine (6400), dimethyl sulfide and β-ionone (both 5700), the
three isomers of di-1-propenyl trisulfide (4300), (E,Z)-2,6-nonadienal (3600), the two
isomers of 2-mercapto-3,4-dimethyl-2,3-dihydrothiophene (3100), and eugenol (1600)
(Table 4). A comparison of the OAVs in raw and blanched green T. sinensis confirmed
22
ACS Paragon Plus Environment

Page 23 of 64
Journal of Agricultural and Food Chemistry
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
that the processing step resulted in clearly lower OAVs for di-1-propenyl disulfide (3
isomers), dimethyl sulfide, 2-mercapto-3,4-dimethyl-2,3-dihydrothiophene (2 isomers),
di-1-propenyl trisulfide (3 isomers), (E,Z)-2,6-nonadienal, hexanal, (E)-2-hexenal, and
(E)-2-hexen-1-ol (Table 4).
To indicate whether similar aroma losses also occur in blanched red T. sinensis,
quantitation of the same aroma compounds as analyzed in raw red T. sinensis was
performed (Table 3). Thereby, the same trends as already seen for green T. sinensis
were found for the abovementioned odorants. Monitoring the changes in the
concentrations of 2-methoxyphenol (-60%) and 4-ethylphenol (-70%) indicated that
their clear decreases caused the loss in intensity of the phenolic odor attribute after
blanching.
Further,
(E)-2-hexen-1-ol,
(E)-2-hexenal,
hexanal,
trans-2-mercapto-3,4-dimethyl-2,3-dihydrothiophene, and dimethyl sulfide showed
high losses of 99, 83, 59, 63, and 46%, respectively. Also the amounts of
di-1-propenyl disulfide (3 isomers), di-1-propenyl trisulfide (3 isomers), and
(E,Z)-2,6-nonadienal decreased. Additionally, most other analyzed compounds, such
as 2-methylbutanoic acid, 3-methylbutanoic acid, isocaryophyllene, 3-methylbutanal,
butyrolactone,
benzyl
alcohol,
aromadendrene,
2-phenylethanol,
2-isopropyl-3-methoxypyrazine, 2-ethyl-3,5-dimethylpyrazine, methylpyrazine, and
methyl 2-methylbutanoate decreased by > 65%. On the other hand, identical to the
green
variety,
nonanal
(+14%),
3-methylnonane-2,4-dione
(+4%),
(E,E)-2,4-decadienal (+62%), and 1-octen-3-one (+141%) were again the only four
compounds present at higher concentrations after blanching.
Aroma Profiles and Recombinates of Blanched Green and Red T. sinensis.
To clarify the changes in the quantitative pattern of the key odorants and in the overall
aroma occurring during the blanching process of T. sinensis, APA of blanched green
and red T. sinensis was performed (Figure 4). For green T. sinensis, the sensory
23
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 24 of 64
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
panel analyzed clearly lower scores of the cooked onion-like/TS-like (1.9) and green
(1.3) aroma attributes compared to those of raw T. sinensis (Figure 4A). For red T.
sinensis, obviously lower cooked onion-like/TS-like (1.4), green (1.0), and especially
phenolic (0.7) odor notes were perceived after blanching (Figure 4B).
Aroma recombination experiments revealed very good similarities with the original
blanched T. sinensis samples, both for blanched green and red T. sinensis (Figures
5A and 5B). Thus, all key odorants in both blanched samples were successfully
characterized. In addition, the change to less intensive cooked onion-like/TS-like (for
both T. sinensis), green (especially for green T. sinensis), and phenolic (for red T.
sinensis) aroma notes after blanching was successfully confirmed on a molecular
level for the first time in this study.
Sensory Experiments to Elucidate the Contribution of Single Odorants to
the Overall Aroma. During the blanching process of both green and red T. sinensis,
clear reductions of nearly all aroma-active compounds were found with an exemption
of only four compounds. Therefore, spiking blanched T. sinensis with the amounts of
the odorants lost during blanching should enable a “reconstitution” of the aroma
profile of raw T. sinensis. To verify this assumption, the following triangle experiments
were performed (Table 6). First of all, the respective amounts (equivalent to the losses
during blanching) of odorants with OAVs ≥ 1 were added to the blanched T. sinensis
samples (Table 6, tests S1 and S8) to obtain their concentrations determined in the
original raw T. sinensis samples. While the blanched T. sinensis and the raw T.
sinensis samples were clearly distinguishable by their overall aroma profiles (Figure
4), the spiked blanched T. sinensis could not significantly be distinguished anymore
from the raw T. sinensis (p = 0.5; Table 6), proving the abovementioned assumption.
In a second part of experiments, aroma compounds were grouped according to
six aroma notes, namely “cooked onion-like/TS-like” (S2 and S9), “green” (S3 and
24
ACS Paragon Plus Environment

Page 25 of 64
Journal of Agricultural and Food Chemistry
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
S11), “phenolic” (S10), “earthy” (S4 and S13), “malty” (S5 and S12), and “vinegar-like”
(S6 and S14), and added to the blanched T. sinensis samples. In a last set of
experiments, the odorants with OAVs ≥ 1, but not included in the abovementioned
groups, were added to the blanched T. sinensis samples (remaining compounds; S7
and S15).
The blanched T. sinensis samples spiked with “earthy”, “malty”, “vinegar-like”, and
remaining compounds could significantly be differentiated from the raw T. sinensis
samples (all p < 0.001; only for test S12, p < 0.008; Table 6), indicating that changes
in different odorants are (additionally) responsible for the overall aroma shift.
Interestingly, already after administering only the “cooked onion-like/TS-like” group,
the panelists could not discriminate the spiked blanched samples from the raw
samples (p = 0.3; significantly different not before p < 0.05; Table 6), proving that the
losses of these compounds during blanching is the crucial factor for the change of the
overall aroma of both blanched T. sinensis varieties and that the groups “malty”,
“earthy”, “vinegar-like”, and “remaining compounds” did not have very important
effects. In addition, also spiking of hexanal, (E)-2-hexenal, (E)-2-hexen-1-ol, and
(E,Z)-2,6-nonadienal (“green” group) to the blanched samples ended up with no
discrimination (p = 0.3 and 0.2; Table 6). Due to the low OAVs of 2-methoxyphenol
and 4-ethylphenol (7 and <1) in raw green T. sinensis and high OAVs in raw red T.
sinensis (440 and 32), the “phenolic” group was only spiked to blanched red T.
sinensis and this partial recombinate could also not be differentiated from raw red T.
sinensis (p = 0.3; Table 6).
These results clearly corroborated the importance of the sulfur-containing
compounds
di-1-propenyl
disulfide,
dimethyl
sulfide,
2-mercapto-3,4-dimethyl-2,3-dihydrothiophene, and di-1-propenyl trisulfide as well as
the aldehydes (E,Z)-2,6-nonadienal, hexanal, and (E)-2-hexenal and the alcohol
25
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 26 of 64
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
(E)-2-hexen-1-ol for the overall aroma of both T. sinensis. In addition, the two phenolic
compounds 2-methoxyphenol and 4-ethylphenol were proven to be key odorants of
red T. sinensis.
Possible Sources of Key Odorants in T. sinensis. Sulfur-Containing
Compounds. (E,Z)- and (Z,Z)-Di-1-propenyl trisulfide were identified for the first time
as aroma-active compounds in T. sinensis. In addition, eight sulfur-containing
compounds including (E,E)-, (E,Z)-, and (Z,Z)-di-1-propenyl disulfide, cis- and
trans-2-mercapto-3,4-dimethyl-2,3-dihydrothiopnene, and (E,E)-, (E,Z)-, and
(Z,Z)-di-1-propenyl trisulfide were quantitated for the first time in T. sinensis via SIDA
and showed extremely high OAVs in both T. sinensis, causing the cooked
onion-like/TS-like odor note. Hydrogen sulfide and methyl thiirane, recently reported
7
in T. sinensis by Yang et al., were not found as key odorants in this study, neither in
the raw nor in the blanched samples, which was additionally confirmed by the very
good similarities of both recombinates to the respective original samples.
Di-1-propenyl disulfides and di-1-propenyl trisulfides, also contributing to the
characteristic aroma of onion, garlic, and other Alliaceae, are usually formed during
crushing by a very fast enzymatic degradation of S-(1-propenyl)-L-cysteine sulfoxide
(precursor) cleaved by allinase, followed by a cascade of chemical reactions.34,35
However, Li et al. reported that nonvolatile precursors of di-1-propenyl disulfides and
36
di-1-propenyl trisulfides in T. sinensis are different from those in Alliaceae. Based on
their findings, T. sinensis contains metabolites with S-alkylnorcysteine moieties such
as
(S,S)-γ-glutamyl-(cis-S-1-propenyl)thioglycine
and
(S,S)-γ-glutamyl-(trans-S-1-propenyl)thioglycine, which may release thiols via
cleavage of the amide bond by proteases of T. sinensis, followed by a spontaneous
decomposition of the resulting unstable alk(en)yl norcysteine moiety. Moreover,
2-mercapto-3,4-dimethyl-2,3-dihydrothiophene is a thermal degradation product of
26
ACS Paragon Plus Environment

Page 27 of 64
Journal of Agricultural and Food Chemistry
7
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
di-1-propenyl disulfide. Dimethyl sulfide can be formed from the amino acid
37
S-methylmethionine, which was recently proven as a possible precursor. However,
disruption of the plant cells occurs unavoidably during blanching at 100 °C, resulting in
premature release and subsequent loss of these sulfur-containing compounds35
which are essential for the overall aroma of both raw (more pronounced) and
blanched (less pronounced) T. sinensis.
Green Smelling Aldehydes and (E)-2-Hexen-1-ol. Many lipid-derived odorants
with green odor notes were identified and quantitated via SIDAs in all four T. sinensis
samples. While these compounds have already been identified7,8 and partially
8
quantitated either in raw, cooked, and/or dried T. sinensis, (E,Z)-2,6-nonadienal was
quantitated for the first time in T. sinensis via SIDA in the present study. Hexanal is a
common secondary product arising from peroxidation of linoleic acid by a
hydroperoxide cleavage enzyme system.38 (E)-2-Hexenal is produced by
isomerization of (Z)-3-hexenal that can enzymatically be formed from linolenic acid in
the presence of oxygen.39 Subsequently, (E)-2-hexenal can be converted into
(E)-2-hexen-1-ol via alcohol dehydrogenase in plant leaves.39 In addition,
(E,Z)-2,6-nonadienal was characterized as key odorant in T. sinensis and was present
at OAVs of 6300 in raw green T. sinensis and 770 in raw red T. sinensis. It can be
enzymatically formed from linolenic acid by a sequence of enzyme reactions
containing lipoxygenase, hydroperoxide lyase, and isomerase.40
Phenols. The amounts of 2-methoxyphenol and 4-ethylphenol have obvious
impact on the different overall aroma of green and red T. sinensis. The smoky and
phenolic smelling 2-methoxyphenol was found in raw red T. sinensis with an OAV of
440, 63 times higher compared to raw green T. sinensis (OAV only 7). Rahouti et al.
showed a metabolism starting from ferulic acid via 4-vinylguiacol to vanillin by the low
phenol oxidase producer Paecilomyces variotii. Subsequent oxidation leads to vanillic
27
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 28 of 64
4
1
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
acid and, again, in a final decarboxylation step to 2-methoxyphenol. 4-Ethylphenol
with a fecal-like and phenolic note was characterized not to be a key odorant in green
T. sinensis due to its OAV < 1, whereas an OAV of 32 was obtained for raw red T.
sinensis. The origin of 4-ethylphenol might be related to a sequential activity of two
enzymes which can reduce and decarboxylate hydroxycinnamic acids (e.g., ferulic
acid, p-coumaric acid, and caffeic acid) into hydroxystyrenes (e.g., vinylphenols),
followed by a reduction to the corresponding ethyl derivatives.42
In conclusion, this study characterized 36 key odorants in raw green T. sinensis
and 41 key odorants in raw red T. sinensis, among which (E,E)-2,4-decadienal, (E,Z)-
and (Z,Z)-di-1-propenyl trisulfide, 4-ethylphenol, 2-methoxyphenol, nonanal, and
2,3,5-trimethylpyrazine were identified for the first time. Furthermore, the intense
cooked onion-like/TS-like odorants characterized the green T. sinensis aroma,
namely (E,E)-, (E,Z)-, and (Z,Z)-di-1-propenyl disulfide, dimethyl disulfide, cis- and
trans-2-mercapto-3,4-dimethyl-2,3-dihydrothiophene, and (E,E)-, (E,Z)-, and
(Z,Z)-di-1-propenyl trisulfide. The obvious green note in green T. sinensis was caused
by
(E,Z)-2,6-nonadienal,
hexanal,
(E)-2-hexenal,
and
(E)-2-hexen-1-ol.
2-Methoxyphenol and 4-ethylphenol evoked the specific phenolic odor note in red T.
sinensis. Recombination experiments of the overall aroma of four samples including
raw green and red T. sinensis as well as the respective blanched buds proved the
successful identification and quantitation of their respective key odorants. As
confirmed by sensory experiments, the clear losses of abovementioned
sulfur-containing compounds, aldehydes, (E)-2-hexen-1-ol, and phenols are
responsible for the overall aroma changes during blanching. Thus, these new findings
provide knowledge of molecular differences not only between green and red T.
sinensis, but also between raw and blanched T. sinensis (for both varieties), which
can be used in the future to improve the final overall aroma of T. sinensis after the
28
ACS Paragon Plus Environment

Page 29 of 64
Journal of Agricultural and Food Chemistry
7
7
15
16
blanching process.
7
17
ACKNOWLEDGMENTS
718
719
720
This work was supported by grants of China Scholarship Council (201607060023).
We thank Jörg Stein for his technical assistance in isolating di-1-propenyl disulfide
isomers and di-1-propenyl trisulfide isomers from T. sinensis via preparative HPLC.
29
ACS Paragon Plus Environment