Hydrolysis before Stir-Frying Increases the Isothiocyanate Content of

Jan 22, 2018 - Broccoli is found to be a good source of glucosinolates, which can be hydrolyzed by endogenous myrosinase to obtain chemopreventive ...
2 downloads 3 Views 426KB Size
Subscriber access provided by READING UNIV

Article

Hydrolysis before stir-frying increases the isothiocyanate content of broccoli Yuanfeng Wu, Yuke Shen, Xuping Wu, Ye Zhu, Jothame Mupunga, Wenna Bao, Jun Huang, Jianwei Mao, Shiwang Liu, and Yuru You J. Agric. Food Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jafc.7b05913 • Publication Date (Web): 22 Jan 2018 Downloaded from http://pubs.acs.org on January 23, 2018

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 29

Journal of Agricultural and Food Chemistry

1

Hydrolysis before Stir-frying Increases the Isothiocyanate Content of Broccoli

2

Yuanfeng Wu*,†,§, Yuke Shen†,#, Xuping Wu†,#, Ye Zhu†,#, Jothame Mupunga†,#,

3

Wenna Bao†,#, Jun Huang†,#, Jianwei Mao†,#, Shiwang Liu†,#, and Yuru You†,#

4

† School of Biological and Chemical Engineering, Zhejiang University of Science

5

and Technology, Hangzhou, 310023, Zhejiang, China

6

§ Guangdong Province Key Laboratory for Green Processing of Natural Products

7

and Product Safety, Guangzhou, 510640, Guangdong, China

8

# Zhejiang Provincial Key Lab for Chem & Bio Processing Technology of Farm

9

Produces, Hangzhou, 310023, Zhejiang, China

10

*Corresponding author, Tel: +86-571-85070396, Fax: +86-571-85070378

11

Email: [email protected];

12

E-mail for all co-authors:

13

Wu Xuping: [email protected];

14

Zhu Ye: [email protected];

15

Shen Yuke: [email protected];

16

Jothame Mupunga: [email protected];

17

Bao Wenna: [email protected];

18

Huang Jun: [email protected];

19

Mao Jianwei: [email protected];

20

Liu Shiwang: [email protected];

21

You Yuru: [email protected]

22

ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry

23

Abstract: Broccoli is found to be a good source of glucosinolates, which can be

24

hydrolyzed by endogenous myrosinase to get chemopreventive isothiocyanates (ITCs),

25

among them sulforaphane (SF) is the most concerned agent. Studies have shown that

26

cooking greatly affected the levels of SF and total ITCs in broccoli. However, little is

27

documented about the stability of these compounds during cooking. In this study, we

28

proved that the half-lives of SF and total ITCs during stir frying were 7.7 and 5.9 min,

29

respectively, while the myrosinase activity decreased by 80% after 3 min of stir-frying,

30

SF and total ITCs were more stable than myrosinase. Thus, the contents of SF and

31

total ITCs decreased during stir-frying largely because myrosinase was destroyed.

32

Subsequently, it was confirmed that compared to direct stir-fry, hydrolysis of

33

glucosinolates in broccoli for 90 min and followed by stir-frying increased the SF and

34

total ITCs concentration by 2.8 and 2.6 times, respectively. This method provides high

35

quantities of beneficial ITCs even after cooking.

36

Keywords: Broccoli, stir-frying, sulforaphane, isothiocyanate, myrosinase

37

ACS Paragon Plus Environment

Page 2 of 29

Page 3 of 29

Journal of Agricultural and Food Chemistry

38

Introduction

39

Broccoli has been widely approved for its beneficial effects on human health because

40

it contains high concentrations of vitamins, minerals, polyphenols and in particular a

41

group of phytochemicals named glucosinolates1. Epidemiological studies have shown

42

that broccoli has potential beneficial effects, including the prevention or reversal of

43

cancer2, hepatic steatosis3, and cardiovascular disease4. The health effects of broccoli

44

are related to its high content of glucosinolates5. When fresh broccoli florets or

45

broccoli sprouts are crushed, chopped or chewed, glucosinolates are hydrolyzed to

46

equimolar amounts of glucose and unstable thiono compounds. The latter are

47

spontaneously converted into bioactive isothiocyanates (ITCs) by the action of an

48

endogenous enzyme called myrosinase, or non-bioactive nitriles by the action of

49

epithiospecifier protein (ESP)5,6. The most abundant glucosinolate found in broccoli is

50

glucoraphanin, followed by glucobrassicin, which are hydrolyzed to sulforaphane (SF)

51

and indole-3-carbinol by the action of myrosinase7.

52

Before consumption, broccoli is commonly cooked. Cooking methods such as

53

steaming and microwaving are common in Western society, whereas stir-frying is the

54

most popular method in China8. SF has many bioactivities, and methods to retain or

55

improve its content in cooked broccoli are vital. Some studies have shown that

56

cooking methods greatly alter the content of total glucosinolates and their

57

hydrolysates, as well as the myrosinase activity in brassica vegetables9. For example,

58

steaming broccoli for 2 min has a significant inhibitory effect on SF production10. A

59

study by Rungapamestry et al.11 showed that hydrolysis of glucoraphanin to SF and

ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry

60

absorption of SF in humans after consumption of lightly-cooked broccoli

61

(microwaved for 2 min) were about three times those after consumption of

62

fully-cooked broccoli (microwaved for 5.5 min), indicating that a longer cooking time

63

inhibits SF production. Jones et al.9 reported that SF production decreased from 51.9

64

to less than 1.1 mg/kg dry weight (DW) after microwaving or boiling for 2-5 min, and

65

steaming for 2 min resulted in SF yield loss of about 50%. Therefore, the authors

66

suggested that for optimum SF intake, broccoli should be eaten raw or lightly steamed.

67

Vermeulen et al.12 found that, compared with microwaved broccoli, raw broccoli

68

consumption led to much higher SF levels in urine and blood. SF was not detectable

69

in broccoli florets after boiling for 15 min or steaming for 23 min, whereas

70

pressure/microwave cooking for 2 min only caused 17% loss of SF13. Generally,

71

compared with raw broccoli hydrolyzed by endogenous myrosinase during digestion,

72

intake of cooked broccoli usually provides approximately one tenth of the amount of

73

SF14-16. These results show that only raw or lightly processed broccoli provides high

74

levels of beneficial SF.

75

The stability of SF under different conditions and in different systems has been

76

reported in a number of studies17,18. As an inherently reactive small molecule, SF is

77

reported to be thermally sensitive and readily degradable17,18. Surprisingly, few

78

methods have reported the SF concentrations in stir-fried broccoli, and to the best of

79

our knowledge, no report has focused on SF stability in the stir-frying process. In a

80

previous study, we showed that the half-life (t1/2) of SF at 90°C and pH 6.0 was 1.5

81

h19. Considering that the stir-frying time is usually 4 min, SF and total ITCs may not

ACS Paragon Plus Environment

Page 4 of 29

Page 5 of 29

Journal of Agricultural and Food Chemistry

82

be as labile as we thought. Nevertheless, myrosinase is very labile, which in broccoli

83

tissue and juice were stable up to 45°C and 40°C, respectively20,21. Consequently, we

84

expect that since myrosinase is much more labile than ITCs, myrosinase will be

85

destroyed during cooking, and formation of SF and total ITCs will greatly decrease.

86

Thus, allowing for hydrolysis before stir-frying should increase the SF and total ITCs

87

contents in broccoli.

88

In this study, we evaluated the stabilities of SF, total ITCs, and myrosinase during

89

stir-frying process. A method for modifying the broccoli stir-frying process to increase

90

SF and total ITCs formation is proposed.

91

Materials and Methods

92

Materials

93

Broccoli seeds, namely LS-1, were preserved at Zhejiang Provincial Key Lab for

94

Chem & Bio Processing Technology of Farm Products. Broccoli and carrots were

95

purchased from a local market. One single cultivar of broccoli purchased at one time

96

was used to study myrosinase activity, and another single cultivar of broccoli

97

purchased at another time was used to study ITCs contents before and after stir-frying.

98

Sulforaphane (98%) and sinigrin hydrate (99%) were purchased from Sigma

99

Chemical Co. (St. Louis, MO, USA), 1,2-Benzenedithiol was purchased from TCI Co.

100

(Tokyo, Japan), Glucose GOD/PAP Kit was purchased from Nanjing Jiancheng

101

Biotech Inc. (Jiangsu, China), and Protein Quantitation Kit was purchased from

102

Sangon Biotech (Shanghai, China). Methanol (Merck, USA) was HPLC grade, and all

103

other chemicals used were of analytical grade and purchased either from Aladdin

ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry

104

(Shanghai, China) or Sangon Biotech (Shanghai, China).

105

Total ITCs extraction methods

106

Total ITCs was extracted by the method previously reported19. Briefly, 50 g broccoli

107

seed meal was ground using a Chinese herbal medicine grinder for 15 s. 400 ml of

108

hexane was added into the seed meal and stirred for 3 h. The seed meal was then

109

transferred and dried in a fume hood overnight to obtain de-fatted seed meal. In the

110

hydrolysis and extraction step, 200 ml ethyl acetate and 100 ml potassium phosphate

111

buffer (0.05 M, pH 5.8) were added simultaneously into the seed meal and was

112

agitated for 4 h, then 20g sodium chloride and 10 g anhydrous sodium sulfate were

113

added to the solution and agitated. The ethyl acetate layer was transferred and the

114

residual seed meal was extracted twice by equal volume of ethyl acetate, the resulting

115

extracts were combined and dried in a rotary evaporator (Yarong Instrument,

116

Shanghai, China). The residue was diluted with water and stored in a refrigerator at

117

-20°C.

118

Stability of SF and total ITCs during stir-frying

119

100g of carrots were cut into 0.3 cm dice, and 10 ml broccoli extracts with SF

120

concentration of 4.17 mg/ml was added and mixed. An electromagnetic oven with

121

output power of 1200 W was preheated for 1 min, then carrots and broccoli extracts

122

mixture were stir-fried in a wok, at 0, 1, 2, 3, 4, 6, 8 and 10 min, 5 g sample was taken

123

and was rapidly cooled by placing them in an ice water bath. Each sample was

124

extracted by 15 ml ethyl acetate for three times, and the resulting ethyl acetate layers

125

were combined and evaporated at 30 °C. The residue was dissolved by methanol and

ACS Paragon Plus Environment

Page 6 of 29

Page 7 of 29

Journal of Agricultural and Food Chemistry

126

constant-volumed to 10 ml. The SF and total ITCs concentrations of each sample

127

were measured by HPLC, as described below. Three independent replicates were

128

carried out for each experiment.

129

Determination of myrosinase activity during stir-frying

130

Myrosinase activity was measured by the method of Guo et al.22 with some

131

modifications. Briefly, an electromagnetic oven with the output power of 1200 W was

132

preheated for 1 min, then a total of 100g broccoli were stir-fried in a wok, as

133

described above. At 0, 1, 2, 3, 4, 6 and 8 min, 5 g sample was taken and was cooled

134

down by placing them into an ice water bath. Each sample was divided into two equal

135

parts, one part was heated at 80 °C overnight to obtain dry weight. Another part was

136

homogenized with 15 ml 0.1 M phosphate buffer (pH 6.5) in an ice bath for 3 min.

137

The homogenate was centrifuged at 10000 g and 4°C for 15 min. 500 µl supernatant

138

was mixed with 500 µl distilled water and boiled for 5 min, then the amount of

139

glucose in the mixture (A1) was quantified using glucose GOD/PAP kit (Nanjing

140

Jiancheng Biotech Inc., Jiangsu, China). Another 500 µl supernatant was mixed with

141

500 µl 0.25 mM sinigrin solution, and incubated at 37 °C for 15 min. The reaction

142

was stopped by boiling for 5 min. The amount of glucose in the mixture (A2) was

143

analyzed using the glucose GOD/PAP kit. The glucose formed in the enzymatic

144

hydrolysis was calculated by subtracting A1 from A2. One myrosinase unit was

145

expressed as 1 nmol glucose formed per min. The specific activity was defined as

146

units per gram of broccoli (dry weight). Three independent replicates were carried out

147

for each experiment.

ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry

148

The Hydrolysis rate of myrosinase

149

2.5g broccoli were homogenized with 15 ml 0.1 M phosphate buffer (pH 6.5) in an ice

150

bath for 3 min. The homogenate was centrifuged at 10000 g and 4°C for 15 min to get

151

the supernatant. 0.25 ml of 2.5 mM sinigrin were mixed with 0.25 ml supernatant,

152

well capped and incubated at 37 °C. At 0, 10, 15, 30, 45, 60, 90, 120, 150 and 210

153

min intervals, one tube was randomly taken from the water bath, and boiled for 5 min

154

to quench the reaction. The amount of glucose was measured using the glucose

155

GOD/PAP kit. Three independent replicates were carried out for each experiment.

156

The effect of hydrolysis on ITCs content after stir-frying

157

Fresh broccoli florets were cut into small pieces (size approximately 0.2 × 0.2 × 0.2

158

cm) and were divided into three groups, namely hydrolyzed and stir-fried (HS) group,

159

directly stir-fried (DS) group and raw broccoli (RB) group. For the HS group, 100 g

160

of chopped fresh broccoli were mixed with 3 ml water and incubated in a 37°C water

161

bath for 90 min. An electromagnetic oven with the output power of 1200 W was

162

preheated for 1 min, and the broccoli was stir-fried in a wok for 4 min. After which

163

the broccoli meal was mixed with 200 ml ethyl acetate and 100 ml 0.05 M potassium

164

phosphate buffer (pH 7.0), agitated for 30 min, 25 g NaCl were then added and the

165

mixture was shaken and filtered by filter paper. The ethyl acetate layer was transferred,

166

the water layer was extracted by 200 ml ethyl acetate for another 2 times. The ethyl

167

acetate layers were combined and evaporated. The residue was dissolved using 25 ml

168

methanol. For the DS group, the electromagnetic oven with output power of 1200 W

169

was preheated for 1 min, 100 g of chopped broccoli florets were stir-fried for 4 min,

ACS Paragon Plus Environment

Page 8 of 29

Page 9 of 29

Journal of Agricultural and Food Chemistry

170

then the broccoli florets were rapidly cooled down by placing them in an ice water

171

bath. The florets were mixed with 200 ml ethyl acetate and 100 ml potassium

172

phosphate buffer, and the proceeding steps were as same as those done on the HS

173

group. For the RB group, 100 g of chopped fresh broccoli was mixed with 200 ml

174

ethyl acetate and 100 ml 0.05 M potassium phosphate buffer (pH 7.0), and the

175

proceeding extraction steps were also as the same as those done on the HS group.

176

Three independent replicates were carried out for each condition, Total ITCs

177

concentration in the methanol solution was measured by HPLC, individual ITCs,

178

nitriles and degradation products were determined by GC-MS, as described below.

179

Total ITCs Quantification

180

Total ITCs quantification was performed using the cyclocondensation method23.

181

Briefly, ITCs sample was mixed with 10 mM of 1,2-benzenedithiol and 25 mM of

182

potassium phosphate buffer. After incubated for 2 h at 65°C, the mixture was cooled

183

down to room temperature and then centrifuged at 16000 g for 10 min. The

184

supernatant was collected and analyzed using a Waters e2695 HPLC System (Milford,

185

MA, USA) equipped with a 4.6×250 mm i.d., 5 µm WondaCract ODS-2 column

186

(Shimadzu, Japan). The mobile phase consisted of 80% methanol and 20% water at a

187

flow rate of 1.0 ml/min, and 10 µl sample was injected onto the column. The

188

cyclocondensation product, 1,3-benzodithiole-2-thione, was detected by absorption at

189

wavelength of 365 nm using a Waters 2489 detector (Milford, MA, USA). Standard

190

curve was prepared using varying amounts of pure SF to estimate the ITCs

191

concentration in each sample.

ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry

192

SF quantification

193

SF was quantified using a Waters e2695 HPLC system (Milford, MA, USA) by the

194

method previously reported19. 10 microliters of sample were injected onto a 4.6×250

195

mm i.d., 5 µm WondaCract ODS-2 column (Shimadzu, Japan). The mobile phase

196

consisted of methanol and water, the flow rate was 1 ml/min, and the column oven

197

temperature was 25 °C. Gradient conditions: Methanol was increased from 20% in

198

water to 60% over 10 min period, then increased to 100% in 2 min and maintained for

199

another 2 min. SF was detected by absorbance at 241 nm using a Waters 2489

200

detector (Milford, MA, USA).

201

GC-MS analysis

202

Glucosinolate hydrolysis products were determined by a 7890A GC System (Agilent

203

Technologies Inc, USA) by the method previously reported24. Mass spectra were

204

obtained by an 5975C inert XL/CI MSD with Triple-Axis Detector enabling recording

205

of ions from m/z 35 to m/z 500. The capillary column was an HP-5MS UI capillaries

206

(0.25 μm film thickness, 30 m × 0.25 mm i.d.). The injection volume was 1.0 µL.

207

The injector and detector temperature were set at 270 and 280°C, respectively. The

208

column oven temperature was began at 50°C for 2 min, increased at 10 °C/min to

209

190 °C, then increased at 20 °C/min to 300 °C, and was maintained for 5 min. The

210

carrier gas was ultra-high purity helium at flow rate of 1 ml/min, and the split ratio

211

was 10:1. Cyclohexanone was used as the internal standard.

212

Statistical Analysis

213

All experiments were done in triplicates. Treatment effects were determined by

ACS Paragon Plus Environment

Page 10 of 29

Page 11 of 29

Journal of Agricultural and Food Chemistry

214

analysis of variance (one-way ANOVA) using SPSS 22.0 software (IBM Corporation,

215

NY, USA). Differences were considered significant at p values0.9, showing that the

252

degradation of SF and total ITCs during stir-frying followed first order reaction

253

kinetics. The t1/2 of SF and total ITCs were calculated by linear regression of the

254

observed data in Figure 2, and were 7.7 min (SF) and 5.9 min (total ITCs). In our

255

previous study, the t1/2 of SF in a pH 6.0 solution at 90°C was 1.5 h19, much longer

256

than that observed in this study. This is reasonable because SF was not in solution in

257

the present study and the temperature was higher than 90°C. Considering that the

ACS Paragon Plus Environment

Page 13 of 29

Journal of Agricultural and Food Chemistry

258

stir-frying time is always much shorter than 10 min, our results indicate that SF and

259

total ITCs are somewhat stable during stir-frying.

260

Myrosinase stability and activity during stir-frying

261

The stability of myrosinase during stir-frying was studied. Fresh broccoli was

262

chopped, stir-fried for 8 min, and the myrosinase activity at 0, 1, 2, 4, 6 and 8 min

263

were measured (Figure 3). Based on sinigrin hydrolysis measured with a glucose

264

assay kit, the myrosinase activity of raw broccoli was 128 U/g DW broccoli. Guo et

265

al.22 reported a myrosinase activity of 75.2 U/mg protein. Okunade et al.29 reported

266

that the myrosinase activity of brown mustard was the highest (2.04 un/mg) whilst

267

that of yellow mustard was the lowest (0.48un/mg). The variations in myrosinase

268

activity within and between Brassica species have been attributed to genetic and/or

269

environmental factors (agronomic and climatic conditions)30 as well as different

270

enzyme activity definitions.

271

Myrosinase activity decreased by 80% after stir-frying for 3 min (Figure 3). This

272

result is not unexpected because myrosinase has been reported to be easily inactivated

273

at normal cooking temperatures, regardless of the cooking method used31. Myrosinase

274

in Brassicaceae lost approximately 90% of its activity after heating at 60°C for only 3

275

min32. Microwave cooking of red cabbage for 4.8 min at 900 W caused almost

276

complete loss of myrosinase activity33. Gliszczyńska-Świgło et al.34 reported that the

277

black mustard enzyme activity in cabbage was lost after 2 min of microwave cooking

278

or 7 min of steaming. Compared to other myrosinase sources, broccoli myrosinase has

279

a low thermal stability35, and heating broccoli florets to 70°C or above for 5 min

ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry

280

inactivates the myrosinase36. Generally, myrosinase is temperature sensitive and

281

cooking methods such as steaming or microwaving are known to decrease or

282

completely loss of the hydrolytic activity.

283

Furthermore, the enzyme-hydrolysis rate of glucosinolates was studied. The result

284

showed that the hydrolysis rate was very slow. In the first 60 min, the glucose content

285

increased very slowly. After 90 min, the glucose content increased to a certain extent

286

(0.34 µM) and then level off with further increases (Figure 4). This result was in line

287

with earlier studies. Tian et al.37 studied the effect of the hydrolysis time on SF

288

production from broccoli sprouts, and found that the best hydrolysis time for SF

289

production was 1.5 h. Vastenhout et al.38 evaluated the kinetics of glucosinolate

290

hydrolysis by Sinapis alba myrosinase, found that production of ITCs increased

291

gradually in 80 min. Similar results were reported by Klingaman et al.39 using another

292

glucosinolate. The results for these reports were different, because the hydrolysis rate

293

is dependent on the structure of the ITC25. However, from these results, we could infer

294

that the hydrolysis rate is very slow.

295

In summary, the enzymatic activity study showed that myrosinase was labile, and that

296

the cooking procedure inactivated myrosinase. The hydrolysis rate is very slow,

297

therefore during cutting or mincing of broccoli, only a small quantity of the

298

glucosinolates are hydrolyzed to ITCs.

299

The effect of hydrolysis on the ITCs content after stir-frying

300

Based on the above results, we hypothesized that hydrolysis of glucosinolates in

301

broccoli by endogenous myrosinase for a certain length of time (e.g., 90 min) before

ACS Paragon Plus Environment

Page 14 of 29

Page 15 of 29

Journal of Agricultural and Food Chemistry

302

stir-frying, the nutritional value would then increase because ITCs are more stable

303

than myrosinase. To verify this, the total and individual ITCs contents in the

304

hydrolyzed and stir-fried (HS) group, directly stir-fried (DS) group and raw broccoli

305

(RB) group were analyzed. The total ITCs concentration in broccoli was quantified by

306

the cyclocondensation method, which is both accurate and rapid23, and individual

307

ITCs and nitriles were determined by GC-MS. The cutting style greatly affected the

308

ITCs content in broccoli, and the total ITCs concentration increased by 133% in

309

florets processed into chops compared to other cutting styles40. Thus, we processed

310

broccoli florets into chops, and waited for 90 min before stir-frying to allow for

311

production of as much ITCs as possible. During stir-frying, broccoli is usually cooked

312

for about 4 min, therefore stir-frying of 4 min was used in this study. The results

313

verified our hypothesis, the ITCs formation in the RB, HS and DS groups were

314

37.92±13.35, 30.90±10.88 and 11.89±2.52 mg/(100g DW broccoli), respectively

315

(Figure 5). The ITCs production in RB and HS groups were higher than DS group, as

316

compared to DS group, the ITCs content in HS group increased by 2.6 folds, while

317

there was no significant difference between RB and HS groups.

318

Volatile compounds from the methylene chloride extracts of the RB, HS and DS

319

groups, such as ITCs, nitriles and their degradation products, were identified and

320

quantified by GC-MS (Table 1). SF was the most abundant ITCs, followed by

321

1-butene 4-isothiocyanate and erucin, only trace amounts of PEITC were detected in

322

the RB group. After stir-frying, there were no significant differences of SF and erucin

323

contents between RB and HS groups, while in the DS group, SF content decreased

ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry

324

significantly, and only trace amounts of erucin were found in the DS group.

325

Some degradation products of SF were also found. Among them, dimethyl sulfone,

326

S-methyl methylthiosulfinate, S-methyl methylthiosulfonate have been reported as

327

thermal degradation products of SF, and the mechanisms for the formation of these

328

compounds from SF have been proposed by Jin et al18. Concentration of these

329

degradation products was higher in the RB group but much lower in the HS and DS

330

groups. We supposed that these compounds are volatile and volatilized in the

331

stir-frying procedure.

332

Sulforaphane nitrile and 3-methyl-2-butenenitrile, produced in the action of

333

epithiospecifier proteins (ESP), were also found. ESP is the cofactor of myrosinase,

334

which can direct the hydrolysis reaction towards nitriles in the presence of Fe2+ ions36.

335

Unlike ITCs, nitriles are much lower in the DS and HS groups. For example,

336

3-methyl-2-butenenitrile was not present in the DS and HS groups, but in the RB

337

group, its concentration was 5.43±1.59 mg/100 g DW broccoli. There were no

338

significant differences between the concentrations of sulforaphane nitrile in the HS

339

and DS groups, but in the RB group, the concentration was higher. Nitriles are much

340

stable than ITCs, which can present even during heat treatments at temperature higher

341

than 100°C and hardly degrade any further26. Therefore, their losses during stir-frying

342

predominantly occur because of their volatility.

343

A significant increase (p