Structure and Function Relationship Study of Allium Organosulfur Compounds on Upregulating the Pi Class of Glutathione S-Transferase Expression

1

Structure and Function Relationship Study of Allium Organosulfur

2

Compounds on Upregulating the Pi Class of Glutathione S-Transferase

3

Expression

4

Chia-Wen Tsai,

^†

Kai-Li Liu,

^‡

Chia-Yuan Lin,

^†

Haw-Wen Chen,

^†

and Chong-Kuei Lii*

^,†

5 ^†

Department of Nutrition, China Medical University, No. 91, Hsueh-Shih Road, Taichung 404, Taiwan

6 ^‡

Department of Nutrition, Chung Shan Medical University, No. 110, Sec. 1, Chien-Kuo N. Road, Taichung 402, Taiwan

7

ABSTRACT: Allium organosulfides are potential chemopreventive compounds due to their effectiveness on the induction of phase

8

II detoxification enzyme expression. In this study, we examined the structure and function relationship among various alk(en)yl

9

sulfides on the expression of the pi class of glutathione S-transferase (GSTP) in rat Clone 9 cells, and what mechanism is involved.

10

Cells were treated with 300 μM dipropyl sulfide (DPS), dipropyl disulfide (DPDS), propyl methyl sulfide (PMS), and propyl methyl

11

disulfide (PMDS) for 48 h. DPDS and PMDS displayed more potency on GSTP protein and mRNA induction than that of DPS and

12

PMS. Next, we compared the effectiveness of DPDS, PMDS, and diallyl disulfide (DADS), which have the same number of sulfur

13

atoms but differ in the side alk(en)yl groups. The maximum increases on protein expression, mRNA level, and enzyme activity were

14

noted in cells treated with DADS, followed by DPDS and PMDS. A reporter assay showed that three disulfides increased GSTP

15

enhancer I (GPE I) activity (P < 0.05) in the order DADS > DPDS g PMDS. Electromobility gel shift assays showed that the DNA

16

binding of GPE I to nuclear proteins reached a maximum at 1 to 3 h after alk(en)yl disulfide treatment. Supershift assay revealed that

17

c-jun bound to GPE I. Silencing of extracellular signal-regulated kinase (ERK) 2 expression inhibited c-jun activation and GSTP

18

induction. Results suggest that both the type of alk(en)yl groups and number of sulfur atoms are determining factors of allium

19

organosulfides on inducing GSTP expression, and it is likely related to the ERK-c-Jun-GPE I pathway.

20

KEYWORDS: allium organosulfides, pi class of glutathione S-transferase, extracellular signal-regulated kinase, c-Jun, Clone 9 cells

21 22

23

’ INTRODUCTION

24

The genus Allium vegetables garlic and onion have garnered

25

significant interest owing to their reported health benefits, which

26

include antithrombotic, antiatherosclerotic, antidiabetic, and

27

anticancer properties.

^1-3

Epidemiologic evidence suggests that

28

increased dietary consumption of garlic reduces the risk of

29

colorectal, laryngeal, and endometrial cancer.

^4,5

This anticarci-

30

nogenic activity has been attributed to the rich content of

31

organosulfur compounds in garlic and onion. The type and

32

content of different garlic and onion products differ dramatically,

33

which is dependent on the means of plant tissue storage and

34

processing.

⁶

For instance, by immersing fresh garlic into a vinegar

35

or wine, S-acetylcysteine and S-acetylmercaptocysteine are two

36

major organosulfur compounds in the aged garlic. By steam-

37

distillation, garlic oil is composed of volatile alk(en)yl sulfides

38

including diallyl disulfide (DADS), diallyl trisulfide (DATS),

39

diallyl sulfide (DAS), and allyl methyl trisulfide, and a trace of

40

propyl methyl disulfide (PMDS).

⁷

In onion oil, the organosulfur

41

compounds include dipropyl disulfide (DPDS), dipropyl sulfide

42

(DPS), propyl methyl sulfide (PMS), and PMDS.

^8,9

43

Cancer chemoprevention of garlic and onion organosulfur

44

compounds has been proposed to be mainly due to their modula-

45

tion on carcinogen metabolism, including the effects on both phase I

46

and II detoxification enzymes

^10,11

and cell cycle.

¹²

It has been

47

demonstrated that G2/M arrest resulted by DADS is related to an

48

increase of cyclin B1 protein levels in human gastric cancer BGC823

49

cells.

¹³

The antitumorigenic effect of DADS and DPDS can be

50

attributed to the transcriptional upregulation of phase II detoxifica-

51

tion enzymes, including glutathione S-transferases (GST),

UDP-glucuronyl transferases, and NAD(P)H-dependent quinone

52

oxidoreductase, which accelerate carcinogen excretion.

^10,11

Higher

53

tissue levels of phase II detoxification enzymes lower susceptibility

54

to chemical carcinogenesis.

¹⁴ 55

Among phase II detoxification enzymes, GST represents a

56

major group that catalyzes the conjugation of glutathione with a

57

variety of electrophilic xenobiotics and facilitates their excretion.

¹⁵ 58

GST is divided into cytosolic, mitochondrial, and microsomal

59

families. Seven distinct classes of cytosolic GST have been

60

identified: alpha, mu, pi, sigma, theta, delta, and zeta.

¹⁶

Compared

61

with other isozymes, the pi class of GST (GSTP) is more effective

62

in the detoxification of the electrophilic R,β-unsaturated carbonyl

63

compounds that are generated by radical reactions of lipids.

¹⁷ 64

There has been considerable interest in the properties of GSTP,

65

particularly in relation to its role in cell transformation and

66

carcinogenesis.

¹⁸

GSTP activity has been used to evaluate the

67

potency of chemoprevention agents in benzo[a]pyrene-induced

68

cancer.

¹⁹

Moreover, in transgenic male Wistar rats, overexpression

69

of GSTP inhibits the early phase of liver carcinogenesis.

²⁰

The

70

importance of GSTP in cancer prevention is further supported by

71

the fact that benzo[a]pyrene-induced lung cancer is significantly

72

elevated in GSTP-null mice.

²¹

Therefore, the expression of GSTP

73

is regarded as an important determinant of protection against

74

various chemical insults.

75

Received: November 2, 2010 Accepted: February 22, 2011 Revised: January 26, 2011

ARTICLE pubs.acs.org/JAFC

r XXXX American Chemical Society A dx.doi.org/10.1021/jf104254r|J. Agric. Food Chem. XXXX, XXX, 000–000

76

The inducibility of GSTP is generally attributed to the

77

existence of a strong enhancer named GSTP enhancer I (GPE

78

I), which has two 12-O- tetradecanoylphorbol 13-acetate (TPA)

79

responsive element (TRE)-like sequences in the 5

⁰

upstream

80

region.

²²

This enhancer on GSTP expression is regulated by

81

multiple factors, mainly the activator protein-1 (AP-1), which is

82

known to be a heterodimer or homodimer composed of c-Jun

83

and c-Fos.

²³

Several cellular stresses and cytotoxic chemicals

84

engage the activation of mitogen-activated protein kinases,

85

including c-Jun NH

2

-terminal kinase, extracellular signal-regu-

86

lated kinase (ERK), and p38 kinase, which in turn activate AP-1.

87

Recently, we reported that DADS on GSTP expression is likely

88

related to the activation of ERK and AP-1 in Clone 9 cells.

²⁴

89

Activation of ERK signaling regulates the binding of c-Jun to the

90

TRE in human lung cancer cells.

²⁵

Therefore, ERK-c-Jun-GPE I

91

signaling pathway may play an important role in GSTP expres-

92

sion by garlic and onion organosulfur compounds.

93

We previously reported that alkenyl sulfides DADS and DATS

94

upregulate GSTP mRNA and protein expression.

²⁴

However,

95

much less is known about alkyl sulfides. In this study, we

96

investigated the effect of DPS, DPDS, PMS, and PMDS on

97

GSTP expression in rat liver Clone 9 cells and the induction

98

potency was compared to that of DADS. In addition, the possible

99

mechanism involved on GSTP transcription was examined.

100

’ MATERIALS AND METHODS

101

Materials. DPS, DPDS, PMS, PMDS, ethacrynic acid, HEPES,

102

bovine serum albumin, deoxynucleotide triphosphate, poly(dI-dC), and

103

β-mercaptoethanol were obtained from Sigma-Aldrich (St. Louis, MO).

104

DADS was purchased from Tokyo Kasei Chemical Co. (Tokyo, Japan).

105

RPMI-1640 medium and penicillin-streptomycin solution were ob-

106

tained from Gibco Laboratory (Grand Island, NY). Trizol and lipofec-

107

tamine were ordered from Invitrogen (Carlsbad, CA). Fetal bovine

108

serum was purchased from Hyclone (Logan, UT). RNase inhibitor,

109

oligo dT, and moloney murine leukemia virus reverse transcriptase were

110

purchased from Promega Company (Madison, WI).

111

Cell Culture. Clone 9 cells, which were derived from normal rat

112

livers, were obtained from Bioresources Collection and Research Center

113

(BCRC, Taiwan). They were grown in RPMI-1640 medium supple-

114

mented with 10 mM HEPES, 1  10

⁵

units/L penicillin, 100 mg/L

115

streptomycin, and 10% fetal bovine serum at 37 °C in a humidified

116

atmosphere of 5% CO

2

and 95% air. For all studies, cells between

117

passages 4 and 10 were used. The cells were plated on 35 mm plastic

118

tissue culture dishes (Falcon, Lakes, NJ) at a density of 2.5  10

⁵

cells

119

per dish and were allowed to grow for 24 h. Fresh culture medium

120

containing 300 μM DPS, DPDS, PMS, PMDS, or DADS (Figure 1

F1

) was

121

then added, and the cells were incubated for 48 h. Cells treated with 0.1%

122

DMSO were used as controls.

123

Western Blot. Cells were washed twice with cold phosphate-

124

buffered saline and were then lysed with potassium phosphate buffer

125

(pH 7.0). The homogenates were then centrifuged at 10500g for 30 min

126

at 4 °C. Protein concentrations were measured by using Coomassie Plus

Protein Assay Reagent Kit (Pierce Chemical Company, Rockford, IL).

127

Five micrograms of cellular protein was separated by 10% SDS-

128

polyacrylamide gel electrophoresis. Separated proteins were transferred

129

to polyvinylidene fluoride membranes (Millipore, Bedford, MA). Pro-

130

tein immunoblot analysis was carried out by use of the following: anti-

131

GSTP (Transduction Laboratories, Lexington, KY), GSTA, and GSTM

132

(Oxford Biomedical Research, Oxford, MI); β-actin (Sigma Chemical,

133

St. Louis, MO); ERK1/2, c-Jun, phospho-ERK1/2, and phospho-c-Jun

134

(all from Santa Cruz Biotechnology Inc., Santa Cruz, CA) as primary

135

antibody, and horseradish peroxidase-conjugated goat anti-rabbit IgG,

136

goat anti-mouse IgG (all from Perkin-Elmer Life Sciences, Boston, MA),

137

or rabbit anti-goat IgG (R&D Systems Inc., Minneapolis, MN) as

138

secondary antibody. The blots were visualized by using an enhanced

139

chemiluminescence kit (Perkin-Elmer Life Science, Boston, MA).

140

RT-PCR. Total RNA was extracted by using Trizol reagent. A total of

141

0.1 μg of RNA was used for the synthesis of first-stand cDNA. Reverse

142

transcription was carried out in a programmable thermal cycler and was

143

performed in 20 μL containing 25 mM Tris-HCl, 50 mM (NH

4

)

2

SO

4

,

144

0.3% β-mercaptoethanol, 0.1 g/L bovine serum albumin, 5 mM MgCl

2

,

145

1 mM of each deoxynucleotide triphosphate, 2.5 units of RNase

146

inhibitor, and 0.5 mM oligo dT and moloney murine leukemia virus

147

reverse transcriptase. The reaction mixture was incubated for 1 cycle at

148

42 °C for 15 min, 99 °C for 5 min, and 4 °C for 10 min. The sequences

149

for the RT-PCR primers were as follows: for GSTP (forward, 5

⁰

-

150

TTCAAGGCTCGCTCAAGTCCAC-3

⁰

; reverse, 5

⁰

-CTTGATCTT

151

GGGGCGGGCACTG-3

⁰

); for glyceraldehyde-3-phosphate dehydro-

152

genase (GAPDH) (forward, 5

⁰

-GACGTGCCGCCTGGAGAAA-3

⁰

;

153

reverse, 5

⁰

-GGGGGCCGAGTTGGGATAG-3

⁰

). The PCR reactions

154

were performed as follows: 5 min at 94°C; 35 cycles of 40 s at 94 °C, 40 s

155

at 60 °C, and 120 s at 72 °C; and a final extension of 5 min at 68 °C. The

156

PCR amplicons were then electrophoresed in 1%-agarose gels contain-

157

ing 1X TAE buffer (40 mM Tris, 20 mM glacial acetic acid, and 2 mM

158

EDTA).

159

Transfection and Small Interfering RNA (siRNA). Cells were

160

plated in 35 mm plastic tissue culture dishes at 70-80% confluence and

161

then transfected with four synthesized ERK2 siRNAs (100 nM) or

162

nontargeting control siRNA (si-control) by using DharmaFECT siRNA

163

transfection reagent (all from Thermo Fisher Scientific, Lafayette, CO)

164

for 24 h. The sense sequences of ERK2 siRNAs were as follows: (1) 5

⁰

-

165

ACACUAAUCUCUCGUACAU-3

⁰

; (2) 5

⁰

-AAAAUAAGGUGCC-

166

GUGGAA-3

⁰

; (3) 5

⁰

-UAUACCAAGUCCAUUGAUA-3

⁰

; and (4) 5

⁰

-

167

UCGAGUUGCUAUCAAGAAA-3

⁰

. Cells were treated with allium

168

organosulfides for indicated time and then lysed, and cell lysates were

169

subjected to immunoblotting.

170

Enzyme Activity Assays. GST activity was measured by using

171

ethacrynic acid as the substrate because of its better selectivity of the pi

172

class isozyme.

²⁶

Briefly, the reaction mixture in a final volume of 1 mL

173

contained 100 mM potassium phosphate buffer (pH 6.5), 0.5 mM

174

glutathione, 0.2 mM ethacrynic acid, and an appropriate amount of the

175

total proteins. The ethacrynate-glutathione conjugate formed was

176

measured at 270 nm.

177

Transient Transfection and Luciferase Activity Assay. The

178

Luc-GPE reporter with -2713 to -2605 (GPE I) bp of the GSTP gene

179

promoter region was constructed according to our previous study.

²⁷ 180

Clone 9 cells were plated at a density of 2  10

⁵

cells on 35 mm plastic

181

tissue culture dishes, and the dishes were incubated until 70% confluence

182

was reached. Cells were transiently transfected for 5 h with 1 μg of the

183

Luc-GPE vector by lipofectamine reagent and were then exposed to

184

allium disulfides for an additional 20 h. Cells were then washed twice

185

with cold phosphate-buffered saline and were lysed in 100 μL of lysis

186

buffer. Luciferase activity was measured by using Luciferase Assay

187

Reagent (Clontech, Palo Alto, CA) according to the manufacturer’s

188

instructions. The luciferase activity of each sample was corrected on the

189

basis of β-galactosidase activity, which was measured at 420 nm with

190

Figure 1. Structures of allium organosulfur compounds.

B dx.doi.org/10.1021/jf104254r |J. Agric. Food Chem. XXXX, XXX,000–000

191

O-nitrophenyl β-

D

-galactopyranoside as a substrate. The value for cells

192

treated with DMSO vehicle alone was regarded as 1.

193

Electromobility Gel Shift Assay. Crude nuclear extracts were

194

prepared according to the method described previously.

²⁴

The Light-

195

Shift Chemiluminescent electromobility gel shift assay (EMSA) Kit

196

(Pierce Chemical Company, Rockford, IL) and synthetic biotin-labeled

197

double-stranded GPE I consensus oligonucleotide (forward, 5

⁰

-AG-

198

TAGTCAG TCACTATGATTCAGCAAC-3

⁰

; reverse, 5

⁰

-GTTGCTG

199

AATCATAGTGACTGACTACT-3

⁰

) were used to measure whether

200

allium sulfides changed GPE I binding activity with nuclear proteins.

201

Unlabeled double-stranded GPE I (200 ng) and a mutant double-

202

stranded oligonucleotide were also used to confirm specific binding.

203

Two micrograms of nuclear protein, poly(dI-dC), and biotin-labeled

204

double-stranded GPE I oligonucleotide were mixed with the binding

205

buffer to a final volume of 20 μL and were incubated at room

206

temperature for 30 min. The nuclear protein-DNA complex was

207

separated by electrophoresis on a 6% Tris-boric acid-EDTA-poly-

208

acrylamide gel and was then electrotransferred to a Hybond-N

^þ

nylon

209

membrane (GE Healthcare, Buckinghamshire, U.K.). The membrane

210

was treated with streptavidin-horseradish peroxidase, and the nuclear

211

protein-DNA bands were developed by using an enhanced chemilu-

212

minescence kit. In the supershift assay, nuclear protein was incubated

213

with monoclonal anti-c-Jun antibody for 30 min after the binding

214

reactions and was subjected to electrophoresis as described above.

215

Statistical Analysis. Statistical analysis was performed with com-

216

mercially available software (SAS Institute Inc., Cary, NC). Data were

217

analyzed by means of one-way ANOVA, and the significant difference

218

among treatment means was assessed by use of Ducan’s test. A value of

219

P < 0.05 was considered to be significant.

220

’ RESULTS

221

Allium Alk(en)yl Sulfides on the Expression of GST Iso-

222

zymes. To ensure that no cytotoxicity resulted from the treat-

223

ment with these organosulfur compounds, we first performed a

224

cell viability assay. The 3-(4,5-dimethylthiazol-2-yl)-2,5-diphe-

225

nyltetrazolium bromide method showed that each of the garlic

226

organosulfides tested at the concentration up to 300 μM for 48 h

227

resulted in cell viability greater than 85% (data not shown).

228

We first examine four alkyl sulfides DPS, DPDS, PMS, and

229

PMDS on GSTP expression in the Clone 9 cells. The immuno-

230

blot assay showed that two alkyl disulfides, i.e. DPDS and PMDS,

231

induced GSTP protein expression (Figure 2

F2

A). In addition, an

232

increase of the other GST isozyme GSTM also resulted from

233

DPDS and PMDS, although the extent of induction was less than

234

that noted for GSTP. GSTA was not affected by both DPDS and

235

PMDS. The levels of GSTP, GSTM, and GSTA, however, had a

236

minor change resulting from sulfides with a single sulfur atom, i.e.

237

DPS and PMS. RT-PCR revealed that changes of GSTP mRNA

238

levels were consistent with those noted for protein expression

239

(Figure 2B). Moreover, DPDS and PMDS dose-dependently

240

increased GSTP protein levels in Clone 9 cells (Figure 2C).

241

These results suggested that the induction of allium alkyl sulfides

242

on GSTP expression was positively related to the number of

243

sulfur atoms. This is similar to the findings reported in our

244

previous work,

²⁷

the sulfur atom numbers of three garlic diallyl

245

sulfides are positively related to the induction efficiency on GSTP

246

transcription.

247

Next, the differential induction on GSTP expression by alkyl

248

disulfides (DPDS and PMDS) and alkenyl disulfide (DADS) was

249

determined. As shown, all three disulfides increased GSTP

250

protein (Figure 3

F3

A) and mRNA (Figure 3B) levels and DADS

251

displayed the greatest induction compared with that of DPDS

and PMDS (P < 0.05). We additionally used ethacrynic acid as a

252

substrate to measure GSTP enzyme activity. As noted, DPDS,

253

PMDS, and DADS resulted in an increase of enzyme activity by

254

132%, 125%, and 298%, respectively, as compared with that of

255

the control cells (Figure 3C).

256

GSTP Promoter Activity. To demonstrate the importance of

257

GPE I in the GSTP expression in response to allium disulfides,

258

we created a reporter construct (Luc-GPE) by ligating the

259

genomic 109-bp GPE I segment (-2713 to -2605 bp) to the

260

luciferase coding region. Results clearly indicated that the

261

reporter activity was increased by DPDS, PMDS, and DADS.

262

Luciferase activity in cells treated with DPDS, PMDS, and DADS

263

was 109%, 78%, and 260%, respectively, higher than control cells

264

(P < 0.05) (Figure 4 ). Again, the disulfides with allyl groups had

265 F4

the greatest increase among three compounds tested (P < 0.05).

266

Nuclear Protein Binding Activity to GPE I. EMSA indicated

267

that, in the presence of DPDS, PMDS, and DADS, the binding of

268

nuclear proteins to DNA reached a maximum at 1-3 h

269

(Figure 5 ) in Clone 9 cells. The specificity of the DNA-protein

270 F5

interaction for GPE I was demonstrated by a competitive assay

271

with 100-fold excess of unlabeled double-stranded oligonucleo-

272

tide (cold) and also with a mutant double-stranded oligonucleo-

273

tide (mut). A similar increase in the DNA binding activity of GPE

274

I was also noted by TPA, a GSTP inducer.

275

Figure 2. Changes of the pi class of glutathione S-transferase (GSTP) protein and mRNA levels by allium alkyl sulfides. Clone 9 liver cells were cultured with 0.1% DMSO alone (control, C) or with 100, 200, or 300 μM dipropyl sulfide (DPS), dipropyl disulfide (DPDS), propyl methyl sulfide (PMS), or propyl methyl disulfide (PMDS) for 48 h. (A) Expression of GSTP, GSTA, and GSTM protein was determined by immunoblotting. (B) Changes in GSTP mRNA levels induced by treatment with alkyl sulfides. (C) Changes in GSTP protein level induced by treatment with alkyl sulfides. The protein and mRNA levels were quantified by densitometry, and the level in the control cells was set at 1. Values are expressed as means (SD), n = 3. Means not sharing a common letter differ significantly, P < 0.05.

C dx.doi.org/10.1021/jf104254r |J. Agric. Food Chem. XXXX, XXX,000–000

276

To further identify the transcription factor that is activated by

277

DADS and bind to GPE I, a supershift assay with highly specific

278

antibodies directed against c-Jun was performed (Figure 5). The

279

increase of nuclear protein-DNA interaction by DADS

decreased in the presence of c-Jun antibody, and the supershift

280

appeared in a dose-dependent manner.

281

ERK on GSTP Expression. The phosphorylation of ERK1/2

282

was increased by treating Clone 9 cells with DPDS, PMDS, or

283

DADS. With accompanying ERK activation, phosphorylation of

284

c-Jun resulted (Figure 6 A). It has been reported that ERK2 but

285 F6

Figure 4. GSTP enhancer I (GPE I) is required for the upregulation of the pi class of glutathione S-transferase (GSTP) by dipropyl disulfide (DPDS), propyl methyl disulfide (PMDS), or diallyl disulfide (DADS).

The GPE I-linked (-2713 to -2604 bp) Luc-reporter was transfected into the Clone 9 cells, and then the cells were treated with 300 μM DPDS, PMDS, or DADS for 20 h. Values are means ( SD, n = 3. Groups not sharing a common letter differ significantly, p < 0.05.

Figure 3. The expression of the pi class of glutathione S-transferase (GSTP) by various alk(en)yl disulfides in Clone 9 cells. Cells were treated with DMSO alone (control, C) or with 300 μM dipropyl disulfide (DPDS), propyl methyl disulfide (PMDS), or diallyl disulfide (DADS) for 48 h. GSTP protein (A), mRNA (B), and enzyme activity (C) were determined. The protein and mRNA levels were quantified by densitometry, and the level in the control cells was set at 1. Values are expressed as means (SD), n = 3. Means not sharing a common letter differ significantly, P < 0.05. Ethacrynic acid was used as a substrate for measuring GSTP activity because of its better specificity. Values are means ( SD, n = 3-4. Groups not sharing a common letter differ significantly, P < 0.05.

Figure 5. Activation of GPE I binding activity by various alk(en)yl disulfides in Clone 9 cells. Cells were treated with 300 μM dipropyl disulfide (DPDS), propyl methyl disulfide (PMDS), or diallyl disulfide (DADS) for the indicated times, and nuclear extracts were prepared to measure GPE I binding activity by electromobility gel shift assay (EMSA). Unlabeled double-stranded GPE I (200 ng) and a mutant double-stranded oligonucleotide were also used to confirm specific binding. For supershift assay, nuclear proteins isolated from the cells treated with DADS for 1 h were first reacted with GPE I oligonucleotides for 30 min and were then incubated with 1 μg (1) or 2 μg (2) of antibodies to c-Jun for an additional 30 min at room temperature. The subsequent supershift complexes were separated by 6% acrylamide gel electrophoresis. One representative immunoblot out of four indepen- dent experiments is shown.

Figure 6. ERK2 knockdown suppressed alk(en)yl disulfide-induced GSTP protein expression. (A) Cells were treated with DMSO alone (C) or with 300 μM diallyl disulfide (DADS) and dipropyl disulfide (DPDS) for 0.5 h, and the phosphorylation of ERK and c-Jun was determined.

(B) Cells were transfected with ERK2 siRNA (si-ERK2) or nontargeting control siRNA (si-control) for 24 h. The activation of c-Jun and ERK in the Clone 9 cells treated with DADS or DPDS for 0.5 h is shown. For GSTP protein determination, the transfected cells were treated with 300 μM DADS or DPDS for 48 h. One representative immunoblot out of three independent experiments is shown.

D dx.doi.org/10.1021/jf104254r |J. Agric. Food Chem. XXXX, XXX,000–000

dipropyl disulfide (DPDS), propyl methyl disulfide (PMDS), and diallyl disulfides (DADS) P

change figure 6 in the attachment

286

not ERK1 plays a key role in rat hepatocyte. ERK2 is more

287

important than ERK1 in replication of hepatocytes,

²⁸

and ERK2

288

might regulate the expression of GSTP in this study. Therefore,

289

knockdown of ERK2 by siRNA transfection was tested to

290

determine the critical role of ERK2 in c-Jun activation and, thus,

291

GSTP induction by allium disulfides. Immunoblots revealed that

292

the activation of ERK1/2 and c-Jun was stimulated in the

293

presence of DADS and DPDS (Figure 6B). In control groups,

294

the activation of ERK2 and c-Jun and expression of GSTP

295

protein were slightly inhibited in ERK2 siRNA-transfected cells.

296

With ERK2 siRNA, cellular ERK2 level was dramatically de-

297

creased (vs si-control), which resulted in the phosphorylation of

298

ERK2 by DADS and DPDS being alleviated. Activation of c-Jun

299

and induction of GSTP expression by both disulfides was then

300

suppressed.

301

’ DISCUSSION

302

Several lines of evidence have suggested that GSTP may play

303

an important role in chemoprevention. GSTP is involved in the

304

protection against alkylation caused by 4-nitroquinoline

305

1-oxide.

²⁹

A point mutation in the GSTP gene that leads to a

306

decrease in enzyme activity was also reported to be associated

307

with increased cancer risk of the oral cavity, bladder, lung,

308

testicles, larynx, and breast.

³⁰

The importance of GSTP in cancer

309

prevention is also supported by the fact that the 7,12-dimethyl-

310

benz anthracene-induced skin cancer was significantly elevated in

311

the GSTP null mice.

³¹

Therefore, higher GSTP activity allows

312

cells to be better in protecting against chemical insult.

313

The Clone 9 cells, a permanently growing and nontrans-

314

formed rat liver cell line, were derived from normal rat liver and

315

retain an epithelial morphology. They have been used extensively

316

as a model for hepatocyte functions, including mediating the

317

expression of a number of phase II detoxification enzymes. In this

318

study, DPDS, PMDS, and DADS are effective inducers of GSTP

319

gene transcription, and DADS shows the greatest potency. In

320

in vivo and in vitro studies, we reported that the induction of

321

GSTP protein and mRNA levels in rat liver by DAS, DADS, and

322

DATS was in the order of DATS g DADS > DAS.

^27,32

However,

323

there was a study that reported that these compounds were not

324

good inducers of GST in Hepa 1c1c7 cells.

³³

It is not clear at

325

present what causes such a differential structure-function

326

relationship in modulating the GSTP, and this requires further

327

study. Different experimental models and varied binding affinity

328

of organosulfur compounds and their metabolic products may

329

explain in part this discrepancy. In vivo study indicated that DPS

330

in rat was metabolized into sulfone.

³⁴

DADS was found to be

331

reduced to allyl mercaptan in an isolated perfused rat liver and

332

oxidized to diallyl thiosulfinate in rat liver microsomes.

^35,36

333

DPDS is oxidized to dipropyl thiosulfinate in rat liver micro-

334

somes, whereas it is transformed to propylglutathione sulfide and

335

propyl mercaptan by liver cytosol.

³⁷

336

In the present study, results clearly indicated that DPDS and

337

PMDS displayed higher induction on GSTP mRNA and protein

338

expression (Figure 2) than those of DPS and PMS, suggesting

339

that the number of sulfur atoms plays a role in the upregulation of

340

this phase II detoxification enzyme. Similar to this finding, the

341

positive relationship between the number of sulfur atoms and

342

GSTP expression has also been noted on DAS, DADS, and

343

DATS in primary hepatocytes.

²⁷

The reason that the compounds

344

containing more sulfur atoms exhibit better inductive effect on

345

the GSTP expression is unclear. Bose et al. suggested that the

disulfide chain might provide an appropriate spacing of the allyl

346

groups in the GSTP-inducing activity of DADS.

³⁸

In addition, it

347

is also possible that the induction of phase II enzymes is often

348

associated with oxidative stress,

^24,39

and disulfides have been

349

shown to cause oxidative damage though their ability to generate

350

“active oxyen” species via redox cycling.

⁴⁰ 351

In addition to the sulfur atom number, the levels in GSTP

352

mRNA and protein were higher in cells treated with DADS than

353

those exposed to DPDS and PMDS (Figure 3). It indicated that

354

allium disulfides with allyl group exert stronger inducibility on

355

GSTP expression than that with saturated propyl and methyl

356

groups. The presence of allyl groups as well as the disulfide chain

357

is required for maximum induction of GSTP in vivo by garlic

358

organosulfur compounds.

³⁸

Allium organosulfur compounds

359

(such as DADS) that contain disulfur atoms and diallyl groups

360

are more potent in inhibiting benzo[a]pyrene-induced forest-

361

omach cancer than are those containing monosulfur (such as

362

DAS) and propyl groups (such as DPDS).

^11,41

The chemopre-

363

ventive efficacy of these organosulfur compounds correlated with

364

their ability to increase the expression of GSTP.

⁴²

Taken

365

together, both the number of sulfur atoms and the type of

366

alk(en)yl groups of allium organosulfur compounds are deter-

367

mining factors on upregulating GSTP expression. Among those

368

sulfides examined, DADS, which is composed of two sulfur atoms

369

and two allyl groups, showed the greatest induction, followed by

370

DPDS and PMDS, and DPS and PMS had only minor effects.

371

Structure-function relationship study has been widely used to

372

examine the relative biological activity among phytochemicals

373

sharing similar structure.

^12,41

In the case of garlic organosulfur

374

compounds, DATS revealed better growth inhibition of A375

375

skin cancer cells than did DADS and DAS.

¹²

In flavonoids, the

376

order of potency at suppressing human liver HepG2 cancer cells

377

is chalcones > flavones > isoflavones >flavanones.

⁴³ 378

GPE I, which consists of two TRE-like sequences, acts as an

379

enhancer and is required for the basal and inducible expression of

380

GSTP in rat livers by a number of stimuli, such as lipoic acid and

381

sulforaphane.

^22,44

In this study, a 109 bp GPE I-Luc reporter was

382

constructed and the change of luciferase activity was determined

383

in the presence of alk(en)yl disulfides. Consistent with the

384

changes of GST mRNA and protein levels, three disulfides tested

385

significantly increased luciferase activity, and the greatest in-

386

crease was noted in cells treated with DADS (Figure 4). It is

387

interesting to explore how these allium organosulfur compounds

388

work differentially on GSTP transcription. These results indi-

389

cated that the differential induction potency among allium alk-

390

(en)yl disulfides is likely to work through the modulation of GPE

391

I activity, and the activation of intracellular signal transduction

392

and transcription factors is the most likely explanation.

393

AP-1, which is composed mainly of c-Jun and c-Fos protein

394

dimers, is the main transcription factor that binds to the TRE-like

395

element in GPE I.

²³

c-Jun is a member of a multiprotein family that

396

has been implicated in several signal transduction pathways

397

associated with cellular growth, neuronal regeneration, and cellular

398

stress.

^45-47

The results of our supershift assay in the present study

399

clearly indicated that c-Jun was involved in the formation of the

400

nuclear protein-GPE I complexes induced by DADS (Figure 5).

401

Based on the fact that c-Jun is required for cellular defense against

402

chemical agents,

^48,49

it is likely that c-Jun functions as an important

403

component that activates GPE I, followed by increasing GSTP

404

expression, in the liver cells exposed to allium sulfides. In addition

405

to c-Jun, the binding of other transcription factors to the TRE-like

406

element in GPE I cannot be excluded. Nuclear factor erythroid-2

407 E dx.doi.org/10.1021/jf104254r |J. Agric. Food Chem. XXXX, XXX,000–000

408

related factor 2 (Nrf2) is one of the candidates that has attracted a

409

lot of attention, because of the sequence homology between the

410

TRE-like sequences on GPE I (5

⁰

-AGTCAGTCACTATGATT-

411

CAGCA-3

⁰

) and the conserved sequences of the antioxidant

412

response element (ARE, 5

⁰

-GTGACNNNGCA-3

⁰

). The binding

413

of Nrf2 to the ARE is well-known to upregulate the transcription of

414

several antioxidant enzymes and phase II detoxification enzymes,

415

including heme oxygenase 1, glutamate-cysteine ligase, GSTM,

416

and NAD(P)H:quinone oxidoreductase.

^50,51

Although it is not

417

determined in this study, Nrf2 binding to GPE I has been reported

418

to be responsible for upregulating GSTP transcription by DATS in

419

primary rat hepatocytes

⁴⁴

and also in the early carcinogenesis stage

420

of rat H4IIE hepatoma cells.

⁵²

Moreover, the induction of GSTP

421

by 6-methylsulfinylhexyl isothiocyanate of wasabi and oltipraz is

422

completely abrogated in Nrf2-deficient mice.

⁵³

423

Recently, in human pulmonary epithelial cells, ERK signaling,

424

but not JNK1/2 and p38, was reported to be the main MAPK

425

involved in activating the binding of c-Jun to the TRE by TPA.

²⁵

426

The finding is supported in our result that allium sulfides

427

increased the phosphorylation of ERK1/2, and then activation

428

of c-Jun in Clone 9 cells (Figure 6A). Silibinin suppressed human

429

osteosarcoma MG-63 cell invasion is attributed to its inhibition

430

on the ERK-dependent c-Jun induction of matrix metalloprotei-

431

nase-2.

⁵⁴

In this study, ERK2 was chosen to be knockdown

432

because ERK2 is more important than ERK1 in replication of

433

hepatocytes.

²⁸

As noted, when ERK2 expression was silenced by

434

ERK2 siRNA, c-Jun activation by DADS and DPDS was sup-

435

pressed (Figure 6B). In parallel, DADS and DPDS induction on

436

GSTP expression disappeared. These findings strongly suggest

437

that ERK2-c-Jun signaling is likely to play an important role in

438

upregulating the transcription of this phase II detoxification

439

enzyme. The result is consistent with our previous study that

440

DATS on GSTP expression is dependent on the ERK-AP-1

441

signaling pathway. The activation of this pathway is likely related

442

to transient changes in cellular redox states.

²⁴

Moreover, Xu et al.

443

indicated that the activation of the ERK signaling pathway is

444

important for transcriptional activity of AP-1 and is involved in

445

the regulation of cell death elicited by sulforaphane and phe-

446

nethyl isothiocyanate in human prostate cancer PC-3 cells.

⁵⁵

An

447

understanding of the role of the ERK-c-Jun-mediated signal

448

pathway in GSTP transcriptional regulation will help to clarify

449

the possible molecular mechanism of allium organosulfur com-

450

pounds in drug metabolism and cancer prevention.

451

In conclusion, both the number of sulfur atoms and the type of

452

alk(en)yl groups are determining factors in the effectiveness of

453

garlic and onion sulfides on upregulating GSTP expression.

454

Moreover, differences in the potency among allium sulfides can

455

be partly attributed to their differential activation of the ERK-c-

456

Jun-GPE I signaling pathway.

457

’ AUTHOR INFORMATION

458

Corresponding Author

459

*Department of Nutrition, China Medical University, Taichung,

460

Taiwan; email: [email protected]; Fax: þ886-4-22062891.

461

Funding Sources

462

Supported by NSC 96-2320-B-235-002-MY2 and CMU97-175.

463

’ ABBREVIATIONS USED

464

AP-1, activator protein-1; ARE, antioxidant response element;

465

DADS, diallyl disulfide; DAS, diallyl sulfide; DATS, diallyl

trisulfide; DPDS, dipropyl disulfide; DPS, dipropyl sulfide;

466

EMSA, electromobility gel shift assay; ERK, extracellular signal-

467

regulated kinase; GAPDH, glyceraldehyde-3-phosphate dehy-

468

drogenase; GST, glutathione S-transferase; GSTP, pi class of

469

GST; GPE I, GSTP enhancer I; Nrf2, nuclear factor erythroid-

470

2 related factor 2; PMDS, propyl methyl disulfide; PMS, propyl

471

methyl sulfide; siRNA, small interfering RNA; TPA, 12-O-tetra-

472

decanoylphorbol 13-acetate; TRE, TPA responsive element

473

’ REFERENCES

474

(1) Das, I.; Saha, T. Effect of garlic on lipid peroxidation and

475

antioxidation enzymes in DMBA-induced skin carcinoma. Nutrition

476

2009, 25, 459–71.

477

(2) Vidyashankar, S.; Sambaiah, K.; Srinivasan, K. Dietary garlic and

478

onion reduce the incidence of atherogenic diet-induced cholesterol

479

gallstones in experimental mice. Br. J. Nutr. 2009, 101, 1621–9.

480

(3) Mariee, A. D.; Abd-Allah, G. M.; El-Yamany, M. F. Renal

481

oxidative stress and nitric oxide production in streptozotocin-induced

482

diabetic nephropathy in rats: the possible modulatory effects of garlic

483

(Allium sativum L.). Biotechnol. Appl. Biochem. 2009, 52, 227–32.

484

(4) Galeone, C.; Pelucchi, C.; Dal Maso, L.; Negri, E.; Montella, M.;

485

Zucchetto, A.; Talamini, R.; La Vecchia, C. Allium vegetables intake and

486

endometrial cancer risk. Public Health Nutr. 2009, 12, 1576–9.

487

(5) Schulz, M.; Lahmann, P. H.; Boeing, H.; Hoffmann, K.; Allen, N.;

488

Key, T. J.; Bingham, S.; Wirfalt, E.; Berglund, G.; Lundin, E.; Hallmans,

489

G.; Lukanova, A.; Martinez Garcia, C.; Gonzalez, C. A.; et al. Fruit and

490

vegetable consumption and risk of epithelial ovarian cancer: the

491

European Prospective Investigation into Cancer and Nutrition. Cancer

492

Epidemiol., Biomarkers Prev. 2005, 14, 2531–5.

493

(6) Banerjee, S. K.; Mukherjee, P. K.; Maulik, S. K. Garlic as an

494

antioxidant: the good, the bad and the ugly. Phytother. Res. 2003,

495

17, 97–106.

496

(7) Yu, T. H.; Wu, C. M.; Liou, Y. C. Volatile compounds from garlic.

497

J. Agric. Food Chem. 1989, 37, 725–730.

498

(8) Takahashi, M.; Shibamoto, T. Chemical compositions and

499

antioxidant/anti-inflammatory activities of steam distillate from freeze-

500

dried onion (Allium cepa L.) sprout. J. Agric. Food Chem. 2008,

501

56, 10462–7.

502

(9) Jarvenpaa, E. P.; Zhang, Z.; Huopalahti, R.; King, J. W. Determi-

503

nation of fresh onion (Allium cepa L) volatiles by solid phase micro-

504

extraction combined with gas chromatography-mass spectrometry. Z.

505

Lebensm.-Unters. -Forsch. A. 1998, 207, 39–43.

506

(10) Garcia, A.; Morales, P.; Arranz, N.; Delgado, M. E.; Rafter, J.;

507

Haza, A. I. Antiapoptotic effects of dietary antioxidants towards N-ni-

508

trosopiperidine and N-nitrosodibutylamine-induced apoptosis in HL-60

509

and HepG2 cells. J. Appl. Toxicol. 2009, 29, 403–13.

510

(11) Srivastava, S. K.; Hu, X.; Xia, H.; Zaren, H. A.; Chatterjee, M. L.;

511

Agarwal, R.; Singh, S. V. Mechanism of differential efficacy of garlic

512

organosulfides in preventing benzo(a)pyrene-induced cancer in mice.

513

Cancer Lett. 1997, 118, 61–7.

514

(12) Wang, H. C.; Yang, J. H.; Hsieh, S. C.; Sheen, L. Y. Allyl sulfides

515

inhibit cell growth of skin cancer cells through induction of DNA

516

damage mediated G2/M arrest and apoptosis. J. Agric. Food Chem. 2010,

517

58, 7096–103.

518

(13) Ling, H.; Wen, L.; Ji, X. X.; Tang, Y. L.; He, J.; Tan, H.; Xia, H.;

519

Zhou, J. G.; Su, Q. Growth inhibitory effect and Chk1-dependent

520

signaling involved in G2/M arrest on human gastric cancer cells induced

521

by diallyl disulfide. Braz. J. Med. Biol. Res. 2010, 43, 271–8.

522

(14) Clarke, J. D.; Dashwood, R. H.; Ho, E. Multi-targeted preven-

523

tion of cancer by sulforaphane. Cancer Lett. 2008, 269, 291–304.

524

(15) Burg, D.; Mulder, G. J. Glutathione conjugates and their

525

synthetic derivatives as inhibitors of glutathione-dependent enzymes

526

involved in cancer and drug resistance. Drug Metab. Rev. 2002, 34, 821–63.

527

(16) Strange, R. C.; Spiteri, M. A.; Ramachandran, S.; Fryer, A. A.

528

Glutathione-S-transferase family of enzymes. Mutat. Res. 2001,

529

482, 21–6.

530

F dx.doi.org/10.1021/jf104254r |J. Agric. Food Chem. XXXX, XXX,000–000

531

(17) Berhane, K.; Widersten, M.; Engstrom, A.; Kozarich, J. W.;

532

Mannervik, B. Detoxication of base propenals and other alpha, beta-

533

unsaturated aldehyde products of radical reactions and lipid peroxida-

534

tion by human glutathione transferases. Proc. Natl. Acad. Sci. U.S.A. 1994,

535

91, 1480–4.

536

(18) Satoh, K.; Kitahara, A.; Soma, Y.; Inaba, Y.; Hatayama, I.; Sato,

537

K. Purification, induction, and distribution of placental glutathione

538

transferase: a new marker enzyme for preneoplastic cells in the rat

539

chemical hepatocarcinogenesis. Proc. Natl. Acad. Sci. U.S.A. 1985,

540

82, 3964–8.

541

(19) Hu, X.; Benson, P. J.; Srivastava, S. K.; Xia, H.; Bleicher, R. J.;

542

Zaren, H. A.; Awasthi, S.; Awasthi, Y. C.; Singh, S. V. Induction of

543

glutathione S-transferase pi as a bioassay for the evaluation of potency of

544

inhibitors of benzo(a)pyrene-induced cancer in a murine model. Int. J.

545

Cancer 1997, 73, 897–902.

546

(20) Nakae, D.; Denda, A.; Kobayashi, Y.; Akai, H.; Kishida, H.;

547

Tsujiuchi, T.; Konishi, Y.; Suzuki, T.; Muramatsu, M. Inhibition of early-

548

phase exogenous and endogenous liver carcinogenesis in transgenic rats

549

harboring a rat glutathione S-transferase placental form gene. Jpn. J.

550

Cancer Res. 1998, 89, 1118–25.

551

(21) Ritchie, K. J.; Henderson, C. J.; Wang, X. J.; Vassieva, O.; Carrie,

552

D.; Farmer, P. B.; Gaskell, M.; Park, K.; Wolf, C. R. Glutathione

553

transferase pi plays a critical role in the development of lung carcinogen-

554

esis following exposure to tobacco-related carcinogens and urethane.

555

Cancer Res. 2007, 67, 9248–57.

556

(22) Okuda, A.; Imagawa, M.; Maeda, Y.; Sakai, M.; Muramatsu, M.

557

Structural and functional analysis of an enhancer GPEI having a phorbol

558

12-O-tetradecanoate 13-acetate responsive element-like sequence found

559

in the rat glutathione transferase P gene. J. Biol. Chem. 1989,

560

264, 16919–26.

561

(23) Angel, P.; Imagawa, M.; Chiu, R.; Stein, B.; Imbra, R. J.;

562

Rahmsdorf, H. J.; Jonat, C.; Herrlich, P.; Karin, M. Phorbol ester-

563

inducible genes contain a common cis element recognized by a TPA-

564

modulated trans-acting factor. Cell 1987, 49, 729–39.

565

(24) Tsai, C. W.; Chen, H. W.; Yang, J. J.; Sheen, L. Y.; Lii, C. K.

566

Diallyl disulfide and diallyl trisulfide up-regulate the expression of the pi

567

class of glutathione S-transferase via an AP-1-dependent pathway. J.

568

Agric. Food Chem. 2007, 55, 1019–26.

569

(25) Adiseshaiah, P.; Li, J.; Vaz, M.; Kalvakolanu, D. V.; Reddy, S. P.

570

ERK signaling regulates tumor promoter induced c-Jun recruitment at

571

the Fra-1 promoter. Biochem. Biophys. Res. Commun. 2008, 371, 304–8.

572

(26) Mannervik, B.; Alin, P.; Guthenberg, C.; Jensson, H.; Tahir,

573

M. K.; Warholm, M.; Jornvall, H. Identification of three classes of

574

cytosolic glutathione transferase common to several mammalian species:

575

correlation between structural data and enzymatic properties. Proc. Natl.

576

Acad. Sci. U.S.A. 1985, 82, 7202–6.

577

(27) Tsai, C. W.; Yang, J. J.; Chen, H. W.; Sheen, L. Y.; Lii, C. K.

578

Garlic organosulfur compounds upregulate the expression of the pi class

579

of glutathione S-transferase in rat primary hepatocytes. J. Nutr. 2005,

580

135, 2560–5.

581

(28) Fremin, C.; Ezan, F.; Boisselier, P.; Bessard, A.; Pages, G.;

582

Pouyssegur, J.; Baffet, G. ERK2 but not ERK1 plays a key role in

583

hepatocyte replication: an RNAi-mediated ERK2 knockdown approach

584

in wild-type and ERK1 null hepatocytes. Hepatology 2007, 45, 1035–45.

585

(29) Morrow, C. S.; Smitherman, P. K.; Townsend, A. J. Role of

586

multidrug-resistance protein 2 in glutathione S-transferase P1-1-

587

mediated resistance to 4-nitroquinoline 1-oxide toxicities in HepG2

588

cells. Mol. Carcinog. 2000, 29, 170–8.

589

(30) Taningher, M.; Malacarne, D.; Izzotti, A.; Ugolini, D.; Parodi, S.

590

Drug metabolism polymorphisms as modulators of cancer susceptibility.

591

Mutat. Res. 1999, 436, 227–61.

592

(31) Henderson, C. J.; Smith, A. G.; Ure, J.; Brown, K.; Bacon, E. J.;

593

Wolf, C. R. Increased skin tumorigenesis in mice lacking pi class

594

glutathione S-transferases. Proc. Natl. Acad. Sci. U.S.A. 1998, 95, 5275–80.

595

(32) Wu, C. C.; Sheen, L. Y.; Chen, H. W.; Kuo, W. W.; Tsai, S. J.;

596

Lii, C. K. Differential effects of garlic oil and its three major organosulfur

597

components on the hepatic detoxification system in rats. J. Agric. Food

598

Chem. 2002, 50, 378–83.

(33) Xiao, H.; Parkin, K. L. Induction of phase II enzyme activity by

599

various selenium compounds. Nutr. Cancer 2006, 55, 210–23.

600

(34) Nickson, R. M.; Mitchell, S. C. Fate of dipropyl sulphide and

601

dipropyl sulphoxide in rat. Xenobiotica 1994, 24, 157–68.

602

(35) Teyssier, C.; Guenot, L.; Suschetet, M.; Siess, M. H. Metabo-

603

lism of diallyl disulfide by human liver microsomal cytochromes P-450

604

and flavin-containing monooxygenases. Drug Metab. Dispos. 1999,

605

27, 835–41.

606

(36) Egen-Schwind, C.; Eckard, R.; Kemper, F. H. Metabolism of

607

garlic constituents in the isolated perfused rat liver. Planta Med. 1992,

608

58, 301–5.

609

(37) Teyssier, C.; Siess, M. H. Metabolism of dipropyl disulfide by

610

rat liver phase I and phase II enzymes and by isolated perfused rat liver.

611

Drug Metab. Dispos. 2000, 28, 648–54.

612

(38) Bose, C.; Guo, J.; Zimniak, L.; Srivastava, S. K.; Singh, S. P.;

613

Zimniak, P.; Singh, S. V. Critical role of allyl groups and disulfide chain in

614

induction of pi class glutathione transferase in mouse tissues in vivo by

615

diallyl disulfide, a naturally occurring chemopreventive agent in garlic.

616

Carcinogenesis 2002, 23, 1661–5.

617

(39) Nakamura, Y.; Ohigashi, H.; Masuda, S.; Murakami, A.; Mor-

618

imitsu, Y.; Kawamoto, Y.; Osawa, T.; Imagawa, M.; Uchida, K. Redox

619

regulation of glutathione S-transferase induction by benzyl isothiocya-

620

nate: correlation of enzyme induction with the formation of reactive

621

oxygen intermediates. Cancer Res. 2000, 60, 219–25.

622

(40) Munday, R. Bioactivation of thiols by one-electron oxidation.

623

Adv. Pharmacol. 1994, 27, 237–70.

624

(41) Guyonnet, D.; Belloir, C.; Suschetet, M.; Siess, M. H.; Le Bon,

625

A. M. Antimutagenic activity of organosulfur compounds from Allium is

626

associated with phase II enzyme induction. Mutat. Res. 2001,

627

495, 135–45.

628

(42) Hu, X.; Benson, P. J.; Srivastava, S. K.; Mack, L. M.; Xia, H.;

629

Gupta, V.; Zaren, H. A.; Singh, S. V. Glutathione S-transferases of female

630

A/J mouse liver and forestomach and their differential induction by anti-

631

carcinogenic organosulfides from garlic. Arch. Biochem. Biophys. 1996,

632

336, 199–214.

633

(43) Loa, J.; Chow, P.; Zhang, K. Studies of structure-activity

634

relationship on plant polyphenol-induced suppression of human liver

635

cancer cells. Cancer Chemother. Pharmacol. 2009, 63, 1007–16.

636

(44) Lii, C. K.; Liu, K. L.; Cheng, Y. P.; Lin, A. H.; Chen, H. W.; Tsai,

637

C. W. Sulforaphane and alpha-lipoic acid upregulate the expression of

638

the pi class of glutathione S-transferase through c-jun and Nrf2 activa-

639

tion. J. Nutr. 2010, 140, 885–92.

640

(45) Raivich, G.; Bohatschek, M.; Da Costa, C.; Iwata, O.; Galiano,

641

M.; Hristova, M.; Nateri, A. S.; Makwana, M.; Riera-Sans, L.; Wolfer,

642

D. P.; Lipp, H. P.; Aguzzi, A.; Wagner, E. F.; Behrens, A. The AP-1

643

transcription factor c-Jun is required for efficient axonal regeneration.

644

Neuron 2004, 43, 57–67.

645

(46) Johnson, R. S.; van Lingen, B.; Papaioannou, V. E.; Spiegelman,

646

B. M. A null mutation at the c-jun locus causes embryonic lethality and

647

retarded cell growth in culture. Genes Dev. 1993, 7, 1309–17.

648

(47) Merienne, K.; Friedman, J.; Akimoto, M.; Abou-Sleymane, G.;

649

Weber, C.; Swaroop, A.; Trottier, Y. Preventing polyglutamine-induced

650

activation of c-Jun delays neuronal dysfunction in a mouse model of

651

SCA7 retinopathy. Neurobiol. Dis. 2007, 25, 571–81.

652

(48) Wainford, R. D.; Weaver, R. J.; Hawksworth, G. M. The

653

immediate early genes, c-fos, c-jun and AP-1, are early markers of

654

platinum analogue toxicity in human proximal tubular cell primary

655

cultures. Toxicol. in Vitro 2009, 23, 780–8.

656

(49) Dhandapani, K. M.; Hadman, M.; De Sevilla, L.; Wade, M. F.;

657

Mahesh, V. B.; Brann, D. W. Astrocyte protection of neurons: role of

658

transforming growth factor-beta signaling via a c-Jun-AP-1 protective

659

pathway. J. Biol. Chem. 2003, 278, 43329–39.

660

(50) Chen, C.; Pung, D.; Leong, V.; Hebbar, V.; Shen, G.; Nair, S.; Li,

661

W.; Kong, A. N. Induction of detoxifying enzymes by garlic organosulfur

662

compounds through transcription factor Nrf2: effect of chemical structure

663

and stress signals. Free Radical Biol. Med. 2004, 37, 1578–90.

664

(51) Chanas, S. A.; Jiang, Q.; McMahon, M.; McWalter, G. K.;

665

McLellan, L. I.; Elcombe, C. R.; Henderson, C. J.; Wolf, C. R.; Moffat,

666 G dx.doi.org/10.1021/jf104254r |J. Agric. Food Chem. XXXX, XXX,000–000

667

G. J.; Itoh, K.; Yamamoto, M.; Hayes, J. D. Loss of the Nrf2 transcription

668

factor causes a marked reduction in constitutive and inducible expres-

669

sion of the glutathione S-transferase GSTA1, GSTA2, GSTM1, GSTM2,

670

GSTM3 and GSTM4 genes in the livers of male and female mice.

671

Biochem. J. 2002, 365, 405–16.

672

(52) Ikeda, H.; Nishi, S.; Sakai, M. Transcription factor Nrf2/MafK

673

regulates rat placental glutathione S-transferase gene during hepatocar-

674

cinogenesis. Biochem. J. 2004, 380, 515–21.

675

(53) Ramos-Gomez, M.; Kwak, M. K.; Dolan, P. M.; Itoh, K.;

676

Yamamoto, M.; Talalay, P.; Kensler, T. W. Sensitivity to carcinogenesis

677

is increased and chemoprotective efficacy of enzyme inducers is lost in

678

Nrf2 transcription factor-deficient mice. Proc. Natl. Acad. Sci. U.S.A.

679

2001, 98, 3410–5.

680

(54) Hsieh, Y. S.; Chu, S. C.; Yang, S. F.; Chen, P. N.; Liu, Y. C.; Lu,

681

K. H. Silibinin suppresses human osteosarcoma MG-63 cell invasion by

682

inhibiting the ERK-dependent c-Jun/AP-1 induction of MMP-2. Carci-

683

nogenesis 2007, 28, 977–87.

684

(55) Xu, C.; Shen, G.; Yuan, X.; Kim, J. H.; Gopalkrishnan, A.;

685

Keum, Y. S.; Nair, S.; Kong, A. N. ERK and JNK signaling pathways are

686

involved in the regulation of activator protein 1 and cell death elicited by

687

three isothiocyanates in human prostate cancer PC-3 cells. Carcinogen-

688

esis 2006, 27, 437–45.

H dx.doi.org/10.1021/jf104254r |J. Agric. Food Chem. XXXX, XXX,000–000