Cancer Chemoprevention by Sulforaphane, a Bioactive Compound

Home , Glucotropaeolin, Mustard oil bomb, Sulforaphane

DISSERTATION

Presented in Partial Fulfillment of the Requirements for the Degree Doctor of Philosophy in the Graduate School of The Ohio State University

Yanyan Li

Graduate Program in Food Science and Nutrition

The Ohio State University

2011

Dissertation Committee:

Dr. Steven Schwartz, Advisor

Dr. Duxin Sun, Co-advisor

Dr. Steven Clinton

Dr. Hua Wang

Dr. Earl Harrison

Copyrighted by

Yanyan Li

2011

Abstract

Sulforaphane, a bioactive compound from broccoli and broccoli sprouts, possess potent cancer chemopreventive activity. In the current studies, we have revealed a novel molecular target of sulforaphane in pancreatic cancer, evaluated the effect of sulforaphane on breast cancer stem cells, and compared different broccoli sprout preparations for delivery of sulforaphane for future chemoprevention studies.

We showed that heat shock protein 90 (Hsp90), a molecular chaperone regulating the maturation of a wide range of oncogenic proteins, as a novel target of sulforaphane.

Different from traditional Hsp90 inhibitors that block ATP binding to Hsp90, sulforaphane disrupted Hsp90-p50 Cdc37 interaction, induced Hsp90 client degradation, and inhibited pancreatic cancer in vitro and in vivo . We traced its activity to a novel

interaction site of Hsp90. Proteolytic fingerprinting and LC-MS revealed sulforaphane

interaction with Hsp90 N-terminus and p50 Cdc37 central domain. LC-MS tryptic peptide

mapping and NMR spectra of full-length Hsp90 identified a covalent sulforaphane adduct

in sheet 2 and the adjacent loop in Hsp90 N-terminal domain. Furthermore, we

investigated the combination efficacy of sulforaphane and 17-allylamino 17-

demethoxygeldanamycin (17-AAG) in pancreatic cancer. 17-AAG, an Hsp90 inhibitor

that blocks ATP binding to Hsp90, has been evaluated in clinical trials; however,

hepatotoxicity limits its application as a single agent. Our data indicated that ii

sulforaphane potentiated the efficacy of 17-AAG through enhanced abrogation of Hsp90 function, while lowered the dose-limiting toxicity of 17-AAG. Concomitant use of sulforaphane and 17-AAG synergistically down-regulated Hsp90 client proteins.

New evidence has shown the existence of cancer stem cells (CSCs) in breast cancer. Targeting CSCs may reduce cancer recurrence. Our data showed that sulforaphane inhibited breast CSCs and down-regulated Wnt/ β-catenin self-renewal pathway. Sulforaphane (1-5 µM) decreased aldehyde dehydrogenase-positive cell population by 65%-80% in human breast cancer cells, and reduced the size and number of mammospheres by 8-125-fold and 45%-75%, respectively, as evidenced by Aldefluor and mammosphere formation assays. Daily injection with 50 mg/kg sulforaphane for two weeks eliminated breast CSCs in nonobese diabetic/severe combined immune-deficient

(NOD/SCID) xenograft mice, thereby abrogating tumor growth after re-implantation of primary tumor cells into the secondary mice. Western blotting and reporter assay showed that sulforaphane down-regulated Wnt/ β-catenin pathway.

All these studies support the development of broccoli sprout preparations for

chemoprevention studies. Therefore, we developed three preparations, compared their

ability to deliver sulforaphane in vivo , and evaluated the pharmacokinetics and tissue

distribution after oral administration. The sulforaphane-rich preparation generated by

two-step procedure contained the highest amount of sulforaphane, 11 and 5 times higher

than the freeze-dried sprouts with and without plant enzymes, respectively; and produced

the greatest plasma response among all the three preparations, with the peak plasma

concentration of sulforaphane 6 and 2.3 times higher, and the AUC 7.9 and 2.2 times

iii

higher, compared to the other two preparations. Consumption of 2.5 mg/g body weight of the sulforaphane-rich preparation resulted in rapid absorption and distribution, achieving high levels of sulforaphane and its glutathione conjugate in plasma and tissues. This study provides a broccoli sprout preparation that can serve as a good source of sulforaphane for further evaluation of chemopreventive efficacy.

Dedication

Dedicated to my family

Acknowledgements

I would like to express my deepest gratitude to my advisors, Dr. Steven Schwartz and Dr. Duxin Sun for their invaluable guidance, inspiration, and support throughout my graduate years. Their conscientious attitude and willingness to explore new avenues made this research possible. Their immeasurable support and every effort for my unusual situation are the best thing that could have happened to me and my family.

I wish to thank my dissertation committee members, Dr. Steven Clinton, Dr. Earl

Harrison and Dr. Hua Wang for their valuable inputs and encouragement.

I wish to thank Dr. Yael Vodovotz for her continuous support and help.

I thank Dr. Max Wicha (University of Michigan) for his precious time and resources, and Dr. Hasan Korkaya and Dr. Suling Liu for their help in the project of breast cancer stem cells.

I thank Dr. Stefan Rüdiger (Utrecht University, Netherlands), Dr. Kate Carroll

(The Scripps Research Institute) and Dr. Young Ho Seo for their help in the study of

Hsp90. Their expertise in NMR and LC-MS made this project possible.

I am grateful to everyone in the Sun lab and Schwartz lab for their help, valuable inputs, and friendship.

I owe special thanks to my husband, for his immeasurable love, steadfast support and encouragement through this entire process. vi

Vita

2005...... B.S. Biology, Nanjing Normal University

2005 to 2006 ...... Fellowship, The Ohio State Biochemistry

Program, The Ohio State University

2006 to present ...... Graduate Research Associate, Department

of Food Science and Technology, The Ohio

State University

Publications

1. Yanyan Li , Max S. Wicha, Steven J. Schwartz, and Duxin Sun. Implications of cancer stem cell theory for cancer chemoprevention by natural dietary compounds. J Nutr Biochem, 2010 Feb 3. Epub ahead of print. 2. Yanyan Li , Tao Zhang, Hasan Korkaya, Suling Liu, Hsiu-Fang Lee, Bryan Newman, Yanke Yu, Shawn G. Clouthier, Steven J. Schwartz, Max S. Wicha, and Duxin Sun. Sulforaphane, a dietary component of broccoli/broccoli sprouts, inhibits breast cancer stem cells. Clin Cancer Res, 2010; 16(9); 2580-90. 3. Yanyan Li , Tao Zhang, Yiqun Jiang, Hsiu-Fang Lee, Steven J. Schwartz, and Duxin Sun. (-)-Epigallocatechin-3-gallate inhibits Hsp90 function by impairing Hsp90 association with co-chaperones in pancreatic cancer cell line Mia Paca-2. Mol Pharm, 2009; 6(4):1152-9. 4. Yanyan Li , Tao Zhang, Steven J. Schwartz, and Duxin Sun. New developments in Hsp90 inhibitors as anti-cancer therapeutics: mechanisms, clinical perspective and more potential. Drug Resist Updates, 2009; 12(1): 17-27. 5. Tao Zhang, Yanyan Li , Zhenkun Zhu, Mancang Gu, Bryan Newman, and Duxin vii

6. Sun. MEK inhibition potentiates the activity of Hsp90 inhibitor 17-AAG against pancreatic cancer cells. Mol Pharm, 2010; 7(5): 1576-1584. 7. Yiqun Jiang, Denzil Bernard, Yanke Yu, Tao Zhang, Yanyan Li , Shaomeng Wang, Xueqi Fu, and Duxin Sun. Split renilla luciferase protein-fragment-assisted complementation (SRL-PFAC) to characterize Hsp90/Cdc37 complex and identify critical residues in protein-protein interactions. J Biol Chem, 2010; 285(27): 21023- 36. 8. Yanke Yu, Adel Hamza, Tao Zhang, Mancang Gu, Peng Zou, Bryan Newman, Yanyan Li , A.A. Leslie Gunatilaka, Chang-Guo Zhan, and Duxin Sun. Withaferin A targets heat shock protein 90 in pancreatic cancer cells. Biochem Pharmacol, 2010; 79(4): 542-51. 9. Tao Zhang, Yanyan Li , Yanke Yu, Peng Zou, Yiqun Jiang, and Duxin Sun. Characterization of celastrol to inhibit Hsp90 and Cdc37 interaction. J Biol Chem, 2009; 284(51): 35381-9.

Fields of Study

Major Field: Food Science and Nutrition

viii

Table of Contents

Abstract ...... ii

Dedication ...... v

Acknowledgements ...... vi

Vita ...... vii

List of Tables ...... xvi

List of Figures ...... xvii

Chapter 1: Literature Review ...... 1

1.1 Cruciferous Vegetables, Glucosinolates and Isothiocyanates ...... 1

1.1.1 Conversion of glucosinolates to isothiocyanates ...... 1

1.1.2 Loss of glucosinolates and isothiocyanates during cooking ...... 2

1.2 Chemopreventive Activity of Sulforaphane ...... 2

1.2.1 Inhibition of Phase 1 enzymes and induction of Phase 2 enzymes ...... 4

1.2.2 Induction of apoptosis and cell cycle arrest ...... 5

1.2.3 Inhibition of angiogenesis and metastasis ...... 6

1.2.4 Inactivation of NF κB pathway ...... 7

1.2.5 Inhibition of histone deacetylases (HDACs) ...... 7

1.2.6 Activation of MAPK pathway ...... 7

1.3 Absorption and Metabolism of Sulforaphane ...... 8

1.4 Pancreatic Cancer Therapy...... 9 ix

1.4.1 Conventional treatment approaches...... 10

1.4.2 Targeted therapy ...... 11

1.5 Heat Shock Protein 90 ...... 12

1.5.1 Hsp90 machinery ...... 13

1.5.2 Hsp90 inhibitors targeting N-terminal ATP binding ...... 15

1.5.3 Hsp90 inhibitors targeting Hsp90-cochaperone interactions...... 17

1.6 Chemotherapy of Breast Cancer ...... 18

1.7 Cancer Stem Cells ...... 20

1.7.1 Techniques for studying breast cancer stem cells ...... 20

1.7.2 Self-renewal pathways of breast cancer stem cells ...... 22

1.8 Targeting Self-renewal Pathways by Bioactive Food Compounds ...... 25

1.8.1 Curcumin ...... 25

1.8.2 Epigallocatechin-3-gallate (EGCG) ...... 27

1.8.3 Quercetin ...... 28

1.8.4 Sulforaphane ...... 28

1.9 Research Objectives and Hypothesis ...... 29

Chapter 2: Sulforaphane Inhibits Hsp90 Function by Disrupting Hsp90-p50 Cdc37 Complex in Pancreatic Cancer Cells through Direct Interaction with Specific Residues ...... 35

2.1 Abstract ...... 35

2.2 Introduction ...... 36

2.3 Materials and Methods ...... 38

2.3.1 Cell culture ...... 38

2.3.2 Reagents...... 38

2.3.3 Western blotting analysis...... 38

2.3.4 Triton-soluble and triton-insoluble protein fraction ...... 39

2.3.5 MTS cell proliferation assay ...... 39

2.3.6 Caspase-3 activity assay ...... 39

2.3.7 Pancreatic cancer xenograft ...... 40

2.3.8 Protein cloning, expression, and purification ...... 40

2.3.9 ATP-sepharose binding assay ...... 41

2.3.10 Hsp90 co-immunoprecipitation ...... 41

2.3.11 Proteolytic fingerprinting assay ...... 42

2.3.12 LC-MS ...... 42

2.3.13 NMR spectroscopy ...... 42

2.3.14 Statistical analysis...... 43

2.4 Results ...... 43

2.4.1 Sulforaphane induces proteasomal degradation of Hsp90 client proteins in pancreatic cancer cells ...... 43

2.4.2 Sulforaphane inhibits pancreatic cancer cells in vitro and exhibits anticancer activity in pancreatic cancer xenograft ...... 44

2.4.3 Sulforaphane inhibits Hsp90 by blocking Hsp90-p50 Cdc37 complex in pancreatic cancer cells ...... 45

2.4.4 Proteolytic fingerprinting and LC-MS detect sulforaphane binding with Hsp90 and p50 Cdc37 ...... 46

2.4.5 NMR reveals the sulforaphane binding sites on Hsp90 ...... 48

2.5 Discussion ...... 50

2.6 Conclusions ...... 55

Chapter 3: Sulforaphane Potentiates the Efficacy of 17-Allylamino 17- Demethoxygeldanamycin against Pancreatic Cancer through Enhanced Abrogation of Hsp90 Chaperone ...... 62

3.1 Abstract ...... 62

3.2 Introduction ...... 63

3.3 Materials and Methods ...... 66

3.3.1 Cell culture ...... 66

3.3.2 Reagents...... 67

3.3.3 MTS cell proliferation assay ...... 67

3.3.4 Caspase-3 activity assay ...... 67

3.3.5 Western blotting analysis...... 68

3.3.6 ATP-sepharose binding assay ...... 68

3.3.7 Hsp90 co-immunoprecipitation ...... 69

3.3.8 Pancreatic tumor xenograft ...... 69

3.3.9 Statistical Analysis ...... 70

3.4 Results ...... 70

3.4.1 Sulforaphane sensitizes pancreatic cancer cells to 17-AAG in vitro ...... 70

3.4.2 Sulforaphane blocks Hsp90-p50 Cdc37 interaction while 17-AAG inhibits ATP binding to Hsp90 ...... 71

3.4.3 Combination of sulforaphane and 17-AAG synergistically down-regulates Hsp90 client proteins in pancreatic cancer cells ...... 72

3.4.4 Sulforaphane potentiates the therapeutic efficacy of 17-AAG in pancreatic cancer xenograft model in vivo ...... 73

3.5 Discussion ...... 74

3.6 Conclusions ...... 77 xii

Chapter 4: Sulforaphane Targets Breast Cancer Stem Cells by Down-regulating Wnt/ β- catenin Self-renewal Pathway ...... 84

4.1 Abstract ...... 84

4.2 Introduction ...... 85

4.3 Materials and Methods ...... 87

4.3.1 Cell culture ...... 87

4.3.2 Reagents...... 88

4.3.3 MTS cell proliferation assay ...... 89

4.3.4 Caspase-3 activity assay ...... 89

4.3.5 Mammosphere formation assay ...... 89

4.3.6 Aldefluor assay ...... 90

4.3.7 Primary NOD/SCID mouse model ...... 90

4.3.8 Dissociation of tumors ...... 91

4.3.9 Secondary NOD/SCID mouse model ...... 91

4.3.10 Western blotting analysis...... 92

4.3.11 TOP-dGFP lentiviral β-catenin reporter assay ...... 92

4.3.12 Statistical analysis...... 93

4.4 Results ...... 93

4.4.1 Sulforaphane inhibits proliferation and induces apoptosis of breast cancer cells ...... 93

4.4.2 Sulforaphane inhibits breast cancer stem/progenitor cells in vitro ...... 94

4.4.3 Sulforaphane eliminates breast cancer stem cells in vivo ...... 95

4.4.4 Sulforaphane down-regulates Wnt/ β-catenin pathway in breast cancer cells .. 97

4.5 Discussion ...... 98

xiii

4.6 Conclusions ...... 103

Chapter 5: Development of Broccoli Sprout Preparations to Deliver Optimal Sulforaphane Levels for Cancer Chemoprevention ...... 110

5.1 Abstract ...... 110

5.2 Introduction ...... 111

5.3 Materials and Methods ...... 114

5.3.1 Reagents...... 114

5.3.2 Broccoli sprout preparations ...... 114

5.3.3 Determination of glucoraphanin in broccoli sprout preparations ...... 115

5.3.4 Determination of sulforaphane in broccoli sprout preparations ...... 116

5.3.5 Determination of sulforaphane, sulforaphane-GSH and glucoraphanin in animal samples ...... 117

5.3.6 Pharmacokinetic analysis ...... 118

5.4 Results ...... 119

5.4.1 Content of sulforaphane and glucoraphanin in three different broccoli sprout preparations ...... 119

5.4.2 Pharmacokinetic profile of sulforaphane, sulforaphane-GSH, and glucoraphanin after oral administration of the broccoli sprout preparations in mice ...... 120

5.4.3 Tissue distribution of sulforaphane and sulforaphane-GSH ...... 122

5.5 Discussion ...... 123

5.6 Conclusions ...... 128

Chapter 6: Conclusions and Further Work ...... 142

Appendix: Green Tea Epigallocatechin-3-gallate Inhibits Hsp90 Function by Impairing Hsp90 Association with Cochaperones in Pancreatic Cancer Cells ...... 146

xiv

References ...... 168

List of Tables

Table 5.1. Glucoraphanin and sulforaphane content in three different broccoli sprout preparations………………………………………………………………………….… 130

Table 5.2. Pharmacokinetic parameters of sulforaphane, sulforaphane-GSH, and glucoraphanin after a single oral dose of 2.5 mg/g body weight of three different broccoli sprout preparations in mice…………………………………………..………. 131

Table 5.3. Recovery of sulforaphane and glucoraphanin in urine and feces after a single oral dose of 2.5 mg/g body weight of three different broccoli sprout preparations in mice…………………………………………………..………………. 132

xvi

List of Figures

Figure 1.1. Chemical structures of glucosinolate, glucoraphanin, isothiocyanate, and sulforaphane...... 31

Figure 1.2. Chemical structures of Hsp90 N-terminal ATP binding site inhibitors……. 32

Figure 1.3. Cancer stem cell theory…………………………………………………….. 33

Figure 1.4. Chemical structures of curcumin, EGCG, and quercetin…………………... 34

Figure 2.1. Sulforaphane induces proteasomal degradation of Hsp90 client proteins…. 56

Figure 2.2. Sulforaphane exhibits anticancer activity in vitro and in vivo ……………... 57

Figure 2.3. Influence of sulforaphane on ATP binding of Hsp90 and Hsp90- cochaperone association in Mia Paca-2 cells………………………………..….……… 58

Figure 2.4. Proteolytic fingerprinting assay of Hsp90 and p50 Cdc37 …………………… 59

Figure 2.5. LC-MS analysis of Hsp90 and p50 Cdc37 …………………………………… 60

Figure 2.6. Sulforaphane binding to Hsp90 mapped by NMR………………………… 61

Figure 3.1. Simultaneous treatment with sulforaphane and 17-AAG enhances anti- proliferative effect in pancreatic cancer cells………………………………………….. 79

Figure 3.2. Combination of sulforaphane and 17-AAG induces a more pronounced activation of caspase-3 in pancreatic cancer cells……………………………………... 80

xvii

Figure 3.3. Effect of sulforaphane and 17-AAG on ATP binding to Hsp90 and Hsp90- cochaperone association in pancreatic cancer cells…………………………………...... 81

Figure 3.4. Co-administration of sulforaphane and 17-AAG synergistically down- regulates Hsp90 client proteins in pancreatic cancer cells……………………………... 82

Figure 3.5. Sulforaphane and 17-AAG combination exhibits a potentiated anticancer activity in pancreatic cancer xenograft model………………………………...………... 83

Figure 4.1. Sulforaphane inhibits proliferation and induces apoptosis in breast cancer cells…………………………………………………………………………………..... 104

Figure 4.2. Inhibitory effect of sulforaphane on mammosphere formation………...…. 105

Figure 4.3. Inhibitory effect of sulforaphane on ALDH-positive cell population…….. 106

Figure 4.4. Sulforaphane decreases tumor size and ALDH-positive cell population in primary breast cancer xenografts……………………………………………………... 107

Figure 4.5. Sulforaphane eradicates breast cancer stem cells in vivo as assessed by re- implantation in secondary mice……………………………………………………..... 108

Figure 4.6. Sulforaphane down-regulates Wnt/ β-catenin self-renewal pathway……... 109

Figure 5.1. Development of three broccoli sprout preparations…………………….... 133

Figure 5.2. Plasma concentration profile of sulforaphane (A), sulforaphane-GSH (B), and glucoraphanin (C) after a single dose of 2.5 mg/g body weight of three different broccoli sprout preparations in mice…………………………….……………………. 134

Figure 5.3. Cumulative urinary levels of sulforaphane (A) and glucoraphanin (B) after a single dose of 2.5 mg/g body weight of three different broccoli sprout preparations in mice………………...………………………………………………… 135

xviii

Figure 5.4. Tissue levels of sulforaphane (A) and sulforaphane-GSH (B) in the liver after a single dose of 2.5 mg/g body weight of broccoli sprout preparations in mice……………………………………………………………………………………. 136

Figure 5.5. Tissue levels of sulforaphane (A) and sulforaphane-GSH (B) in the kidney after a single dose of 2.5 mg/g body weight of broccoli sprout preparations in mice… 137

Figure 5.6. Tissue levels of sulforaphane (A) and sulforaphane-GSH (B) in the lung after a single dose of 2.5 mg/g body weight of broccoli sprout preparations in mice……………………………………………………………………………….…… 138

Figure 5.7. Tissue levels of sulforaphane (A) and sulforaphane-GSH (B) in the heart after a single dose of 2.5 mg/g body weight of broccoli sprout preparations in mice……………………………………………………………………………………. 139

Figure 5.8. Tissue levels of sulforaphane (A) and sulforaphane-GSH (B) in the muscle after a single dose of 2.5 mg/g body weight of broccoli sprout preparations in mice… 140

Figure 5.9. Tissue levels of sulforaphane in the mammary fat pad after a single dose of 2.5 mg/g body weight of broccoli sprout preparations in mice……………….……. 141

Figure A.1. (-)-EGCG inhibites pancreatic cancer proliferation and increases caspase-3 activity……………………………………………………… ……………... 163

Figure A.2. Effect of (-)-EGCG on Hsp90 client proteins……. ……………………… 164

Figure A.3. Influence of (-)-EGCG on Hsp90/cochaperones association…………….. 165

Figure A.4. (-)-EGCG binds to the C-terminus but not N-terminus of Hsp90………... 166

Figure A.5. Effect of (-)-EGCG on ATP binding to Hsp90…………………………... 167

xix

Chapter 1: Literature Review

1.1 Cruciferous Vegetables, Glucosinolates and Isothiocyanates

Numerous studies have substantiated the chemopreventive properties of high consumption of cruciferous vegetables, especially broccoli and broccoli sprouts, against various cancers [1]. These effects have been attributed to the activity of isothiocyanates that are converted from their inactive precursors, glucosinolates [1, 2].

1.1.1 Conversion of glucosinolates to isothiocyanates

Cruciferous vegetables are characterized by their high content of glucosinolates,

which are β-thioglucoside N-hydroxysulfates with amino acid side chains (Fig. 1.1) [3,

4]. Glucosinolates are converted to isothiocyanates (Fig. 1.1) by the action of a β-

thioglucosidase, myrosinase. This enzyme is physically separated from glucosinolates in

intact plant cells [3]. Disruption of the plant during harvesting, chopping, blending,

cooking, and chewing leads to loss of cellular compartmentalization and subsequent

mixing of glucosinolates and myrosinase to form isothiocyanates [3, 5]. The hydrolysis of

glucosinolates can also occur in the mammalian gastrointestinal tract, mediated by the

myrosinase produced by gut microflora [6]. Epithiospecifier protein (ESP), another

protein naturally occurring in the cruciferous vegetables, binds to and directs the

intermediate of glucosinolates toward epithionitrile formation [7].

1.1.2 Loss of glucosinolates and isothiocyanates during cooking

The significant loss of total glucosinolates in boiled broccoli is primarily due to the leaching into the cooking water [8]. A small proportion of glucosinolates may also be broken down thermally during cooking when the cooking time is longer than 10 min [9].

Vallejo et al. reported that steam cooking had minimal effects on glucosinolate content of broccoli florets compared to high pressure boiling, conventional boiling, and microwave cooking [8]. They confirmed that leaching was the major factor contributing to the loss of glucosinolates in boiled broccoli florets. They also suggested that the great loss in glucosinolate content in microwaved florets was probably due to the high cooking water evaporation that contained leached compounds [8].

Isothiocyanates, on the other hand, have been demonstrated to be relatively

thermo-labile [10]. Sulforaphane was rapidly degraded when the temperature was 60 °C

or above and was degraded by more than 90% after a 20 min treatment at 90 °C [10].

Sulforaphane was found to decompose to volatile compounds due to high temperature

[11].

1.2 Chemopreventive Activity of Sulforaphane

Sulforaphane was found to be converted from glucoraphanin by myrosinase (Fig.

1.1), the principal glucosinolate in broccoli and broccoli sprouts [12]. Broccoli sprouts contain approximately 20 times more glucoraphanin than broccoli, which represents

typically 74% of all glucosinolates in the sprouts [13]; and thus are commonly used for delivery of glucoraphanin and sulforaphane [4].

Sulforaphane has been shown to be not only effective in preventing chemically induced cancers in animal models, including colon, lung, breast, pancreatic, skin and stomach cancer [12, 14-20], but also inhibit the growth of established tumors [21, 22].

Oral or intraperitoneal administration of sulforaphane inhibited the tumor growth in prostate PC-3 and pancreatic Panc-1 xenografts [21, 23]. Sulforaphane was suggested to be responsible for the inhibition of ultraviolet-induced skin carcinogenesis by broccoli sprout extracts, which were applied topically after UV treatment [24]. Sulforaphane- supplemented diet significantly inhibited the formation of intestinal polyps in ApcMin/+ mice [25]. Tumors in these mice spontaneously occur because of a mutation of the tumor suppressor gene, adenomatous polyposis coli (APC) [25].

Although sulforaphane itself has not been evaluated in human clinical trials, several pilot and Phase I human trials have been conducted using different sources of sulforaphane. The risk of premenopausal breast cancer was shown to be inversely associated with broccoli consumption [26]. A recent pilot study detected an accumulation of sulforaphane in human breast tissue, with 1.45 ± 1.12 pmol/mg for the right breast and

2.00 ± 1.95 pmol/mg for the left, in eight women who consumed broccoli sprout preparation containing 200 µmol sulforaphane about 1 h before the surgery, as well as increased activity of Phase 2 detoxification enzymes [27]. A Phase I trial showed that broccoli sprouts caused no significant toxicity when administered orally at 8-h intervals for 7 days as 25 µmol isothiocyanates (mainly sulforaphane) [13]. In another study, it

was well tolerated in 200 adults who consumed broccoli sprout solution containing 400

µmol glucoraphanin (precursor of sulforaphane) nightly for 2 weeks [28].

Early research focused on induction of Phase 2 enzymes and inhibition of Phase 1

enzymes by sulforaphane, which enhances the detoxification of carcinogens and “blocks”

carcinogenesis at the initiation stage of cancer [1, 29]. Recent studies suggest that

sulforaphane provides protection against tumor development during the “post-initiation”

phase by controlling cell proliferation, apoptosis, cell cycle, angiogenesis and metastasis

[1, 30].

1.2.1 Inhibition of Phase 1 enzymes and induction of Phase 2 enzymes

The reactions catalyzed by Phase 1 enzymes are usually implicated in the

bioactivation of carcinogens (conversion of procarcinogens to carcinogens) [31, 32].

Sulforaphane can modulate the levels and activities of various Phase 1 enzymes, leading

to reduced activation of procarcinogens [1]. Sulforaphane was shown to direct interacts

with cytochrome P450 enzymes (CYPs) or regulate their transcription in the cell [1, 33].

Sulforaphane dose-dependently inhibits the activities of CYP1A1 and CYP2B1/2 in rat

hepatocytes [1, 33]. Similarly, the transcript level of CYP3A4 in human hepatocytes is

reduced by sulforaphane [1, 33].

Phase 2 metabolisms limit further biotransformation of Phase 1 metabolites by

conjugating endogenous substrates (e.g., glutathione, glucuronide, and sulfate) to the

Phase 1 metabolites [34], resulting in enhanced elimination and excretion (detoxification)

[35]. Sulforaphane induces Phase 2 enzymes, including NAD(P)H:quinine reductase

(NQO1), γ-glutamylcysteine synthetase ( γ-GCS), glutathione S-transferase (GST), and

UDP glucuronosyltransferases (UGT) [36-38]. Regulation of these genes is associated with disruption of Nrf2-Keap1 interaction, increased nuclear translocation of Nrf2 and binding of antioxidant response element (ARE) sites [1, 31]. In unstimulated cells, Kelch- like ECH-associated protein 1 (Keap1) sequesters (nuclear factor erythroid-2 related factor 2) Nrf2 and bridges it to ubiquitinase. Sulforaphane was shown to react with specific cysteine residues of Keap1, promoting Nrf2 dissociation from Keap1 and allowing subsequent activation of ARE-driven gene expression [39, 40].

1.2.2 Induction of apoptosis and cell cycle arrest

Numerous studies have supported the ability of sulforaphane to induce apoptosis in vitro and in vivo . Sulforaphane affects classical molecular targets involved in the

apoptosis pathways including down-regulation of anti-apoptotic Bcl-2 and Bcl-XL , up-

regulation of pro-apoptotic Bax expression, proteolytic activation of caspase-3, and the

degradation/cleavage of poly(ADP-ribose) polymerase [41]. In addition, it was shown

that Bax activation, downregulation of IAP family proteins, and apoptotic protease

activating factor-1 induction are also involved in the regulation of sulforaphane induced

cell death [42]. The production of reactive oxygen species (ROS) has also been suggested

to be implicated in sulforaphane-induced apoptosis. Sulforaphane administration caused

ROS generation and disruption of mitochondrial membrane potential, which led to

cytochrome c release and apoptosis of PC-3 prostate cancer cells [43].

There is substantial evidence to support sulforaphane as a potent inducer of cell

cycle arrest. Sulforaphane has been shown to arrest cells in G1 phase [44-46], S phase

[47], and G2/M phase [23, 47-49], depending on the cell line. G2/M arrest is the

predominant stage of cell cycle arrest induced by sulforaphane [48, 50]. In PC-3 prostate cancer cells, sulforaphane induced arrest of G2/M phase was associated with a decrease in protein levels of cyclin B1, cell division cycle 25B (Cdc25B), Cdc25C, leading to accumulation of inactive form of cyclin-dependent kinase 1 (Cdk1) [51]. Chk2-mediated phosphorylation of Cdc25C at Ser-216 also plays a role in sulforaphane induced G2/M arrest [51]. Sulforaphane treatment caused a modest increase in S phase fraction and a marked increase in G2/M fraction in LnCaP cells [52]. Down-regulation of cyclin D1, cyclin E, Cdk4, and Cdk6 protein levels was correlated with sulforaphane induced S phase arrest [52].

1.2.3 Inhibition of angiogenesis and metastasis

More recent studies demonstrate that sulforaphane is also capable of inhibiting angiogenesis and metastasis. Sulforaphane was shown to interfere with essential steps of neovascularization from proangiogenic signaling and basement membrane integrity to endothelial cell proliferation, migration, and tube formation [53]. These effects were associated with transcriptional down-regulation of vascular endothelial growth factor

(VEGF), hypoxia-inducible factor-1α (HIF-1α), c-Myc and matrix metalloproteinase-2

(MMP-2). Sulforaphane also inhibited the proliferation and tubular formation on matrigel of human umbilical vein endothelial cells in vitro [54], and was responsible for suppression of MMP-9 activity and invasiveness of human MDA-MB-231 breast cancer cells by broccoli extracts [55].

1.2.4 Inactivation of NF κB pathway

Constitutive activation of nuclear factor kappa-B (NF κB) is common in various human malignancies [56, 57]. NF κB is normally sequestered by I κB in the cytosol.

Phosphorylation of I κB by IKK leads to the degradation of I κB and subsequent release

and nuclear translocation of NF κB [31]. Sulforaphane was shown to strongly inhibit LPS-

induced NF κB luciferase activity and I κB phosphorylation in HT-29 cells [58].

Sulforaphane inhibited NF κB transcriptional activity, nuclear translocation of NF κB subunit, and NF κB regulated VEGF, cyclin D1, and Bcl-XL expression in PC-3 cells, which was mediated through inhibition of IKK phosphorylation [59].

1.2.5 Inhibition of histone deacetylases (HDACs)

Enhanced HDAC activity and expression is common in many cancer, which results in de-regulation of differentiation, cell cycle and apoptosis [1]. Sulforaphane inhibited HDAC activity and increased acetylated histones bound to the P21 promoter in

HCT116 human colorectal cancer cells [60]. In BPH1, PC3, and LnCaP prostate cancer cells, sulforaphane inhibited HDAC activity accompanied by an increase in global histone acetylation, enhanced interaction of acetylated histone H4 with the promoter region of the P21 gene and the Bax gene, and induction of p21 and Bax protein expression [61].

1.2.6 Activation of MAPK pathway

Mitogen-activated protein kinase (MAPK) pathway, which includes the extracellular signal-regulated kinases (ERK) and c-Jun NH 2-terminal kinases (JNK), has also been reported to be activated by sulforaphane [62]. The activation of ERK and JNK

signaling pathways is important for transcriptional activity of activator protein-1 (AP-1) and is involved in the regulation of cell death elicited by sulforaphane in PC-3 cells [62].

1.3 Absorption and Metabolism of Sulforaphane

The primary factor determining the absorption and bioavailability of sulforaphane involves the conversion of glucoraphanin to sulforaphane. Myrosinase was found in the cruciferous vegetables and the microbial flora of mammalian intestinal tract. Thorough chewing of fresh broccoli sprouts exposes the glucosinolates to plant myrosinase and significantly increase excretion of sulforaphane metabolites [63]. A study reported that cooking procedures can inactivate the enzyme and significantly reduce the bioavailability of sulforaphane up to 3-fold [64].

One of the advantages of sulforaphane is high bioavailability [30]. Sulforaphane from broccoli extracts is efficiently and rapidly absorbed in human small intestine, and quickly distributed throughout the body [1, 65]. Human perfusion experiment showed that 74 ± 29% of sulforaphane from broccoli extracts can be absorbed in the jejunum and that a portion of that returns to the lumen of the jejunum as sulforaphane-glutathione conjugate [65]. Plasma concentrations of isothiocyanate equivalents peaked between 0.94 and 2.27 µM in humans subjects 1 h after a single dose of 200 µmol broccoli sprout isothiocyanates (mainly sulforaphane), with half life times of 1.77 ± 0.13 h [66]. In rats, following a 50 µmol oral gavage, sulforaphane was detectable after 1 h and peaked at 20

µM at 4 h, with a half life of approximately 2.2 h [67].

Sulforaphane is primarily metabolized through the mercapturic acid pathway in vivo [68]. After absorption the electrophilic central carbon of the –N=C=S group in sulforaphane quickly reacts with glutathione (GSH) to form sulforaphane-GSH conjugate, which is catalyzed by glutathione S-transferase (GST). Sulforaphane-GSH is further metabolized to cysteinylglycine (sulforaphane-Cys-Gly), cysteine (sulforaphane-

Cys), and N-acetylcysteine conjugates (sulforaphane-NAC), which are subsequently excreted in the urine [69, 70]. Approximately 72% of a single oral dose of sulforaphane was recovered in the urine as NAC conjugates in rats in 24 h, but only about 1% was detected in the second 24-h urine sample [71], indicating that urinary elimination is almost completed within 24 h after dosing. The urinary excretion of isothiocyanates

(primarily sulforaphane) equivalents amounted to approximately 73.7% in 24 h in rats fed with broccoli sprout extract [72]. In human subjects, 58.3 ± 2.8% and 77.9 ± 6.4% of a single dose of approximately 200 µmol isothiocyanates (mainly sulforaphane) contained in broccoli sprout extracts was recovered in the urine as sulforaphane equivalents in 8 h and 72 h, respectively [66]. About 37% of sulforaphane was recovered in urine as sulforaphane mercapturic acids in 24 h after consumption of fresh broccoli by human subjects [73].

1.4 Pancreatic Cancer Therapy

Pancreatic cancer is the fourth leading cause of cancer death in the United States

[74]. It is an aggressive malignant disease, characterized by invasiveness, rapid

progression and profound resistance to treatment [75]. Although it only accounts for 3% of all cancers, the survival rate is extremely low, with an overall 5-year survival <4% over the past 20 years [74]. Despite the efforts in the last three decades, advances in screening, chemotherapy, and radiotherapy have had minimal impact on this aggressive disease [76].

1.4.1 Conventional treatment approaches

Treatment options for pancreatic cancer vary with the disease stage [77]. In general, pancreatic tumors are classiﬁed as resectable, locally advanced or metastatic disease for different treatments. At the time of diagnosis, about 20% of pancreatic cancers are detected at the resectable stage [77]. These patients are considered eligible for surgery. Nonetheless, complete eradication of the disease is rarely achieved even after resection [78]. After surgery or at times when surgery is not possible, adjuvant chemotherapy or radiotherapy may be offered. The active metabolite of 5-fluorouracil (5-

FU) works at the S phase of cell division and inhibits RNA and DNA synthesis [79].

Similarly, gemcitabine is also a cell-cycle specific agent, inhibiting DNA synthesis and function in the S phase [79]. An improvement in clinical response was noticed in gemcitabine-treated patients compared to 5-FU treated patients [80]. After gemcitabine became the standard chemotherapy drug for pancreatic cancer, various agents were combined with gemcitabine for better impact. Although many Phase II studies demonstrated the efficacy of combining gemcitabine with cytotoxic agents, unfortunately, none of the randomized Phase III trials have showed statistically significant improvement in overall survival [79, 81-87]. Radiotherapy has been given to

improve the unsatisfactory results of surgery; however, it has exhibited limited effect.

Intra-operative radiotherapy reduced the local recurrence rate after surgery, but it did not significantly improve the survival rate [88].

1.4.2 Targeted therapy

The advances in molecular biology have dramatically improved our knowledge of pancreatic cancer pathogenesis, which has provided the foundation for targeted therapy.

The first genetic changes detected in the progression are mutations of K-Ras, a member of the Ras family [75, 89]. Oncogenic K-Ras is involved in the initiation or early phase of pancreatic tumorigenesis, and more than 85% of the pancreatic cancer patients have K-Ras mutation at the early stage of cancer development [90]. Activation of Ras requires farnesyltransferase [89]. Tipifarnib, a farnesyltransferase inhibitor, was combined with gemcitabine in a Phase III trial [91]. Other Ras related inhibitors under investigation include romidepsin, a histone deacetylase inhibitor which inhibits Ras- mediated signal transduction, and fanesylthiosalicylic acid (salirasib) which dissociates

Ras from membrane binding site [89].

Over-expression of EGFR (a transmembrane receptor tyrosine kinase) and its ligands EGF and TGF-α (43, 46 and 54%, respectively) is frequently observed in pancreatic cancer [89, 92]. Two therapeutic approaches, small molecule tyrosine kinase inhibitors (TKIs) and monoclonal antibodies directed against the extracellular ligand binding domain, have been clinically evaluated. Erlotinib is an orally active TKI that binds to the ATP binding site of EGFR. In a Phase III trial, erlotinib in combination with gemcitabine demonstrated a small but statistically significant increase in the survival,

which led to the approval of erlotinib as the first targeted therapy for patients with advanced pancreatic cancer [93]. Cetuximab is a chimeric monoclonal antibody specifically binding to the extracellular domain of EGFR [86]. A Phase II trial combining cetuximab with gemcitabine demonstrated longer survival [86]. However, cetuximab did not show encouraging results in a Phase III trial in patients with locally advanced and metastatic pancreatic cancer [89].

VEGF, an important mediator of angiogenesis, is over-expressed in more than

90% of pancreatic cancer [94]. Bevacizumab is one of the most commonly used VEGF inhibitors [79]. To date, there are two Phase III trials comparing the use of bevacizumab with and without gemcitabine, and a Phase III trial evaluating gemcitabine plus erlotinib with or without bevacizumab [79]. However, none of these have shown any survival benefit yet. Other VEGF inhibitors such as sorafenib, axitinib and sunitinib have been investigated in Phase I and II trials [79, 89].

Many other targets are under investigation in preclinical or early clinical stages, including TGF-β and SMAD4 pathway, insulin-like growth factor pathway, Src pathway, hedgehog pathway, Notch pathway, Wnt pathway, telomerase, microRNAs and cancer stem cells [89].

1.5 Heat Shock Protein 90

Although the targeted therapy has yielded encouraging results in preclinical

settings, the application in clinical trials has not translated into statically significant

improvements. Given that pancreatic carcinogenesis is characterized by complex molecular basis that are involved many oncogenic proteins, targeting a single pathway is unlikely to be effective. Thus, heat shock protein 90 (Hsp90), an essential molecular chaperone that regulates the stability and maturation of a wide range of oncogenic proteins, may provide advantages as a promising target for pancreatic cancer therapy [95,

96].

1.5.1 Hsp90 machinery

The molecular chaperone Hsp90 is a highly abundant protein, constituting about

1-2% of total proteins under non-stress conditions in most tissues [97, 98]. Hsp90 controls the stability, activation, and maturation of many important proteins, including transmembrane tyrosine kinases (Her-2, EGFR), metastable signaling proteins (Akt, Raf-

1 and IKK), mutated signaling proteins (p53, v-Src), chimeric signaling proteins (Bcr-

Abl), cell cycle regulators (Cdk4, Cdk6), and steroid receptors (androgen, estrogen, and progesterone receptors) [96, 99]. Over the past several years, the application of global analysis has extended Hsp90 clientele to more than 200 proteins [100-102]. Many of these client proteins are mutated and/or over-expressed in cancers [103, 104].

Hsp90 exists as a homodimer, and each monomer consists of three highly conserved domains: an N-terminal ATP-binding domain (25 kDa), a middle domain (35 kDa) and a

C-terminal dimerization domain (12 kDa) [105]. The N-terminus of Hsp90 contains a specific ATP binding pocket [106]. The major role of the middle domain is to discriminate various types of client proteins to adjust the molecular chaperone for proper substrate activation [107]. The C-terminal dimerization domain strengthens the weak

association between the two N-terminal domains of the Hsp90 dimer [108]. The C- terminal domain of eukaryotic Hsp90 has a conserved pentapeptide (MEEVD) implicated in binding to the tetratricopeptide repeat (TPR) domain of cochaperones, such as Hop

(Hsp organizing protein) and Sti1 (stress-inducible protein 1, yeast homologue of Hop)

[103, 108].

The “open” state of the Hsp90 dimer, with its two N-termini separated, can capture client proteins [109]. ATP binding triggers the closure of the ATP pocket “lid” and brings the N-termini close to each other, resulting in the formation of a compacted, ring-shaped Hsp90 dimer [109, 110]. These conformational alterations lead to a “closed” state to “clamp” client proteins inside [108]. The ATPase activity of Hsp90 itself drives the chaperone cycle [96].

The chaperone cycle driven by ATPase activity is composed by the formation and dissociation of an array of multi-chaperone complexes between Hsp90 and various cochaperones. One of the cochaperones, p50 Cdc37 is a kinase-specific cochaperone of

Hsp90 [111]. Protein kinases are the largest class of Hsp90 clients. The Hsp70/Hsp40

complex first prepares a newly synthesized or misfolded protein kinase for interaction

with the N-terminal domain of p50 Cdc37 , followed by recruitment of Hsp90 to the

complex with the help of Hop [112, 113]. The C-terminal side chain of p50 Cdc37

associates with the “lid” of Hsp90, which closes the N-terminal ATP binding pocket

[114]. Crystallographic studies reveal that the insertion of p50 Cdc37 C-terminus to the

Hsp90 N-terminal ATP pocket inhibits the ATPase activity of Hsp90 and prevents its N- terminal dimerization [115]. This holds Hsp90 in an “open” conformation in the

intermediate complex for client loading [114]. Although the release of p50 Cdc37 C- terminus from Hsp90 N-terminal clamp is required for the transition from “open” to the

“closed” conformation, p50Cdc37 may stay in the complex by interacting with client protein [114]. Other cochaperones, such as p23 and Aha1, may be required as well [110].

More details of kinase maturation in the complex remain to be understood.

1.5.2 Hsp90 inhibitors targeting N-terminal ATP binding

Since geldanamycin (Fig. 1.2) was demonstrated to possess potent anticancer effects through inhibiting Hsp90 [116, 117], a great deal of efforts have been devoted to this area [116, 118].

Geldanamycin is a competitive inhibitor of ATP binding to Hsp90 [119]. Binding of geldanamycin in the N-terminal ATP pocket restrains Hsp90 in its ADP-bound conformation and prevents the subsequent “clamping” of Hsp90 around a client protein

[120, 121], resulting in ubiquitination and proteasomal degradation of the client [122].

Although geldanamycin exhibited potent anticancer activities in preclinical studies, it had little clinical potential mostly due to the high hepatotoxicity observed in animal models

[123]. As a result, geldanamycin derivatives with similar anticancer activities but better toxicological properties were synthesized, such as 17-allylamino-17- desmethoxygeldanamycin (17-AAG) [124, 125], 17-dimethylaminoethylamino-17- demethoxygeldanamycin (17-DMAG) [124, 126], and 17-allylamino-17- demethoxygeldanamycin hydroquinone hydrochloride (IPI-504) (Fig. 1.2) [124, 127].

17-AAG entered Phase I trials in 1999 [128, 129] and several intravenous formulations have completed Phase I testing [130, 131]. Early signs of therapeutic

activity have been seen in melanoma, breast cancer, prostate cancer, and multiple myeloma [129, 132-135]. Phase II clinical trials for 17-AAG are currently ongoing [130,

131], which mainly focus on tumor types hallmarked by specific Hsp90 chaperoning targets, such as leukemia expressing Bcr-Abl and Her-2 positive breast cancer [129, 136].

17-AAG was recently administered in combination with trastuzumab in patients whose disease progressed following trastuzumab treatment; this trial has demonstrated promising anti-tumor activity and acceptable toxicity [133]. Currently 17-DMAG, a more water-soluble analogue of 17-AAG, has entered Phase I and Phase II clinical testing, and displayed higher oral bioavailability, lower toxicity, and increased stability compared to

17-AAG [130, 137]. Another water soluble hydroquinone hydrochloride analogue of 17-

AAG is IPI-504 [104]. IPI-504 is in Phase I and Phase II clinical trials to evaluate its potential for treating cancer that has become resistant to therapy with tyrosine kinase inhibitors [138].

Like GA, radicicol (Fig. 1.2) [124] was also shown to compete with nucleotide for the N-terminal ATP pocket of Hsp90 [119]. Radicicol displayed anticancer activity in vitro but not in vivo due to its chemical and metabolic instability [103, 139]. Hence,

synthetic efforts have been directed to generate radicicol derivatives with improved

stability and in vivo efficacy [140]. Several oxime derivatives and cycloproparadicicol

have been developed and shown to possess anticancer activity in preclinical animal

models with tolerable toxicity [139, 141, 142].

1.5.3 Hsp90 inhibitors targeting Hsp90-cochaperone interactions

Hsp90 requires a series of cochaperones to assemble a multi-chaperone complex for its function. These cochaperones bind and leave the complex at various stages to regulate the chaperoning process [143]. Arresting the chaperone cycle at these stages by targeting Hsp90-cochaperone interactions is likely to achieve similar consequences with the direct inhibition of Hsp90 [144, 145]. Inhibitors that block Hsp90-cochaperone interaction may offer more specificity. Silencing of Aha1 decreases client protein activation and increases cellular sensitivity to the Hsp90 inhibitor 17-AAG [146].

Simultaneous knockdown both Hsc70 and Hsp72 induces proteasome-dependent degradation of Hsp90 client proteins, G1 cell cycle arrest, and extensive tumor-specific apoptosis [147]. Silencing of p50 Cdc37 promotes the proteasome-mediated degradation of kinase clients via a degradation pathway independent of Hsp90 binding, and enhances apoptosis in combination with 17-AAG [148]. The compounds disrupting Hsp90 and Hop interaction have been identified and shown activity in human breast cancer cells [149,

150].

The p50 Cdc37 , an Hsp90 cochaperone, acts as an adaptor to load protein kinase to

Hsp90 complex [95, 111]. Most of the Hsp90 clients p50 Cdc37 associated with are crucial elements implicated in signal transduction, cell proliferation and survival [95]. The important role of p50 Cdc37 in the Hsp90 chaperoning cycle may indicate its involvement in

malignancy. Indeed, p50 Cdc37 over-expression is found in prostate cancer cells and tissues

[151], and its up-regulation during hepatocarcinogenesis can protect cancer development

from inhibitory compounds and promote tumorigenesis [152]. In addition, p50 Cdc37 is

highly expressed in proliferative tissues [153] and it can function as an oncogene to induce hyperproliferative disorders and tumor formation in a genetically engineered mouse model [151, 154]. Recently, our group has shown that celastrol and withaferin A block Hsp90-p50 Cdc37 interaction, thereby inhibiting Hsp90 chaperone function and

inducing degradation of Hsp90 client proteins, eventually leading to anticancer activity

[144, 155, 156].

1.6 Chemotherapy of Breast Cancer

Breast cancer, one of the most common cancers in Western society, is the second leading cause of cancer-related deaths in the population of American women [157, 158].

Although breast cancer mortality rates have declined in recent years and the five-year survival has been improved to 89% in the last few decades [159], it remains a formidable health problem [158].

Many chemotherapeutic agents, including classes anthracyclines (e.g., doxorubicin and epirubicin), fluoropyrimidines (e.g., 5-deoxyfluorouridine and cyclophosphamide), and taxanes (e.g., paclitaxel and docetaxel), have demonstrated activities in breast cancer [158, 159]. Chemotherapy has been shown to prolong survival in patients with early, primary breast cancer [160]. The early chemotherapy trials mostly used single agents such as methotrexate or cyclophosphamide until the standard combination chemotherapy of cyclophosphamide, methotrexate and 5-fluorouracil (CMF) schedule was developed [161-163]. Combination chemotherapy regimens generally

produce higher response rates than a single agent [164, 165]. The advantages of combination chemotherapy over single agent regimen was demonstrated in a series of clinical trials [166]. Furthermore, anthracycline-containing regimens were shown to produce a moderate improvement over the standard combination of CMF, in terms of five-year disease-free survival rates [160]. A large, multicenter study reported that patients treated with doxorubicin/cyclophosphamide (AC) followed by paclitaxel had a significantly better disease-free survival and overall survival than patients treated with

AC only [160].

For patients with advanced, metastatic breast cancer, chemotherapy is frequently used to palliate symptoms and to prolong patient survival [158, 167, 168]; however, metastatic breast cancer remains an incurable disease with current therapeutic approaches

[166, 169]. Advanced breast cancer can be treated using chemotherapy in combination regimens [158]. The standard chemotherapy options for advanced, estrogen receptor- negative breast cancer include combinations such as CMF, CMF/vincristine/prednisone

(CMFVP), and 5-fluorouracil/doxorubicin/cyclophosphamide (FAC) [160]. In the past decade, taxanes have be introduced and emerged as powerful options in the treatment of metastatic breast cancer [160, 170]. A number of studies have shown that taxane- containing chemotherapy combinations such as paclitaxel/doxorubicin or cyclophosphamide/doxorubicin/docetaxel provide advantages in response rate and time to progression over non-taxane-containing combination chemotherapies [171, 172]. Other agents used for treatment of advanced breast cancer include vinorelbine and gemcitabine

[173]. Gemcitabine has demonstrated improved effect when patients had been exposed to

paclitaxel first [158]. Further evaluations are undergoing to determine the optimal combinations and dosing regimens for treating breast cancer [170].

1.7 Cancer Stem Cells

In recent years, a great deal of research has demonstrated the existence of cancer stem cells (CSCs) or tumor-initiating cells (TICs) in several human cancers, including breast cancer [174-180]. The CSC theory asserts that many types of cancer are initiated from and maintained by a minor population of tumorigenic cells that are capable of continuous self-renewal and differentiation [181, 182] (Fig. 1.3A). This cell population undergoes unlimited proliferation and gives rise to differentiated cells, developing new tumors phenotypically recapitulating the original tumors [183] (Fig. 1.3B). In addition, recent studies indicate that CSCs may be responsible for tumor relapse and resistance to therapy [30, 184]. However, most currently available therapeutic approaches, including chemotherapy and radiotherapy, lack the ability to effectively kill these CSCs [182, 185-

187], although they can considerably shrink tumor sizes [185]. The cancer may eventually develop drug resistance and recurrence [167, 183, 185, 188, 189]. Therefore, this CSC population has become a target for cancer prevention and therapy [183].

1.7.1 Techniques for studying breast cancer stem cells

Evidence supporting the CSC model was initially obtained from acute myeloid leukemia (AML) [190, 191]. Dick et al. isolated a cell subpopulation with surface marker

CD34 +CD38 -, which was able to recapitulate the phenotypes of the original patient

neoplasms along serial passaging through multiple NOD/SCID recipient mice [174, 190,

192]. The first work in isolation and characterization of CSCs in solid tumors was conducted by Al-Hajj et al. [175]. A breast cancer cell population expressing the surface

marker, CD44 +CD24 -/low Lin -, was able to initiate tumors with the same heterogeneity as the primary tumor from 100 cells [175]. Similarly, enzymatic activity of aldehyde dehydrogenase 1 (ALDH) was also demonstrated to be a selective marker to enrich for breast cancer stem/progenitor cells [193]. These two phenotypes, ALDH-positive and

CD44 +CD24 -/low Lin -, were identified as possessing a small overlap that has the highest

tumorigenic capacity, generating tumors from as few as 20 cells [193]. These cell

markers have been widely used to evaluate the ability of drugs to target cancer

stem/progenitor cells [194-196].

Another technique that has been developed to isolate and characterize breast

cancer stem/progenitor cells is mammosphere culture [197-200]. This is based on the

ability of stem/progenitor cells to grow in serum-free, non-adherent suspension as

spherical clusters, while differentiated cells fail to survive under the same condition [197,

198]. Cancer stem/progenitor cells are capable of yielding secondary spheres and

differentiating along multiple lineages [197]. Decreases in mammosphere formation in

primary culture in the presence of drug treatment and in subsequent passages that are

cultured in the absence of drugs indicate an inhibitory effect of the drug on self-renewal

capacity of breast cancer stem/progenitor cells [194, 197].

CSCs are able to generate the diverse cells that comprise the tumor through

continuous self-renewal and differentiation [201]. There is a reliable in vivo model often

used to evaluate the drug efficacy against breast CSCs [175, 201, 202]. The NOD/SCID mice are first implanted with human breast cancer cells or human primary tumors in the mammary fat pad. After treatment, the dissociated tumor cells are analyzed for CSC population based on the specific cell markers, and the same number of living tumor cells from control and treated mice are re-implanted to a second group of mice which do not receive any treatment [182]. The ability of breast cancer cells from the primary

NOD/SCID xenografts to re-generate tumors upon re-implantation in the mammary fat pads of secondary mice reflects the inhibitory effect of the treatment on breast CSCs

[182]. Failure of tumor initiation indicates the effectiveness of the treatment against breast CSCs.

1.7.2 Self-renewal pathways of breast cancer stem cells

CSCs produce the tumor mass through continuous self-renewal and

differentiation, which may be regulated by similar signaling pathways occurring in

normal stem cells [181, 185]. Understanding the mechanisms that underlie the self-

renewal behavior of CSCs is of greatest importance for discovery and development of

anticancer drugs targeting CSCs. So far, several major pathways including Wnt/ β- catenin, Hedgehog, and Notch have been identified to play pivotal roles in self-renewal of breast CSCs [203-205].

Wnt/ β-catenin pathway was demonstrated to modulate cell proliferation, migration, apoptosis, differentiation, and stem cell self-renewal [206-209]. It has been shown that Wnt/ β-catenin signaling is implicated in the maintenance of CSCs of breast

[210, 211], leukemia [212-214], melanoma [215], colon [216], liver [217], lung [218] cancers.

β-Catenin, the essential mediator of canonical Wnt signaling, participates in two distinct functions in the cell, depending on its cellular localization. Membrane-localized

β-catenin is sequestered by the epithelial cell-cell adhesion protein E-cadherin to maintain cell-cell adhesion [219]. On the other hand, cytoplasmic accumulation of β- catenin and its subsequent nuclear translocation, followed by cooperation with the transcription factors T cell factor/lymphoid enhancer factor (TCF/LEF) as a transcription activator, eventually leads to activation of Wnt target genes such as c-Jun , c-Myc , fibronectin , and cyclin D1 [181, 220-225]. Binding of Wnt proteins, a family of secreted

proteins, to Frizzled receptors results in the cytoplasmic accumulation of β-catenin [226].

In the absence of Wnt signaling, β-catenin forms a multi-protein complex with glycogen

synthase kinase 3β (GSK3 β), adenomatous polyposis coli, casein kinase1 α, and axin

[227]. When β-catenin is phosphorylated at Ser33/Ser37/Thr41 by GSK3 β, it is

immediately subject to ubiquitin-proteasome degradation [227, 228].

The link between Wnt/ β-catenin and PI3K/Akt pathway has been established by several studies. Activated Akt (i.e., phospho-Akt Ser473) was shown to be able to phosphorylate Ser9 on GSK3 β, which may decrease the activity of GSK3 β, thereby

stabilizing β-catenin [229-231]. Furthermore, Korkaya et al. demonstrated that PI3K/Akt

pathway is important in regulating the mammary stem/progenitor cells by promoting β-

catenin downstream events through phosphorylation of GSK3 β [182].

Another major pathway that is involved in stem cell self-renewal is hedgehog signaling pathway [198, 203, 232, 233]. Liu et al. have demonstrated that the hedgehog pathway plays a crucial role in regulating self-renewal of normal and malignant human mammary stem cells by utilizing both in vitro and mouse model systems [203].

In the absence of hedgehog ligands (Sonic Hedgehog, Desert Hedgehog, and

Indian Hedgehog), their transmembrane receptor Patched (Ptch) associates with

Smoothened (Smo) and blocks Smo function [181, 232, 234]. When secreted hedgehog ligands bind to Ptch, Smo is released, triggering dissociation of transcription factors,

Gli1, Gli2, and Gli3 from Fused (Fu) and suppressor of Fused (SuFu), leading to transcription of an array of genes, such as cyclin D , cyclin E , Myc , and elements of EGF pathway [181, 232, 234, 235].

Notch signaling is known to control cell proliferation and apoptosis to modulate the development of many organs [236]. A number of recent studies have demonstrated that Notch-activated genes and pathways can drive tumor growth through the expansion of CSCs [198, 236-241]. Notch pathway is believed to be dysregulated in CSCs, ultimately leading to uncontrolled CSC self-renewal [236]. Notch pathway was shown to play an important role in the self-renewal function of malignant breast cancer CSCs [204,

242].

Five Notch proteins, Notch-1 to Notch-4, have been identified to express as transmembrane receptors in a variety of stem/progenitor cells [243]. Binding of surface- bound ligands (Jagged1, Jagged2, Delta-like1, Delta-like3, and Delta-like4) triggers serial cleavage events at the Notch proteins by ADAM protease family and γ-secretase

[243-245]. Subsequently, the intracellular domain of Notch is released and translocates into the nucleus, where it acts as a transcription co-activator of recombination signal sequence-binding protein J κ (RBP-J) to activate downstream target genes, e.g., c-Myc ,

cyclin D1 , p21 , NF-κB [245-251].

1.8 Targeting Self-renewal Pathways by Bioactive Food Compounds

Since a large number of epidemiological studies have demonstrated an association

between consumption of fruits and vegetables and the reduced risk of various cancers,

naturally-occurring dietary compounds have received increasing attention for their

efficacy in cancer chemoprevention [252]. The anticancer activities of many dietary

components have been reported for both in vitro and in vivo studies [253-260]. Naturally-

occurring dietary compounds are advantageous in several aspects as chemoprevention

agents: (1) they are present in commonly consumed food, which is readily available to

most people in daily life; (2) they usually have very low or no toxicity, in contrast to most

chemotherapy drugs; (3) many of these compounds have shown potential as an adjunct to

chemotherapy drugs in some clinical trials. Although the reports were very limited for

dietary compounds to inhibit CSCs, some of them have been shown to be directly or

indirectly involved in modulation of CSC self-renewal pathways.

1.8.1 Curcumin

Curcumin (Fig. 1.4) is a well-known dietary polyphenol present in an Indian

spice, turmeric, which is usually used in preparation of mustard and curry [261].

Curcumin possesses anti-inflammatory and anti-oxidant activities [261, 262], and has been studied as a chemoprevention agent in several cancer models [258, 263].

Jaiswal et al. suggested that curcumin induced caspase-3-mediated cleavage of β- catenin, leading to inactivation of Wnt/ β-catenin signaling in HCT116 intestinal cancer cells [264]. The work of Park et al. strengthened the point that curcumin decreased β- catenin/TCF transcription activity in all tested cancer cell lines, including gastric, colon, and intestinal cancer cells, which was attributed to the reduced amount of nuclear β-

catenin and TCF-4 proteins [261]. Moreover, analysis of gene transcription profile

revealed that the expression of Wnt receptor Frizzled-1 was potently suppressed by

curcumin [265]. Curcumin was also shown to be able to attenuate response of β-catenin to Wnt-3a in colon cancer cells through down-regulation of p300, a positive regulator of

Wnt/ β-catenin signaling [266]. In addition, Wang and his colleagues demonstrated that curcumin down-regulated Notch-1 mRNA level in pancreatic cancer cells, indicating a transcriptional inactivation of Notch-1 by curcumin [267]. Curcumin-induced inactivation of NF-κB DNA-binding activity was potentially mediated by Notch-1 signaling pathway [267].

Very recently, Kakarala et al. demonstrated that curcumin was able to target breast stem/progenitor cells, as evidenced by suppressed mammosphere formation along serial passage and by a decrease in the percent of ALDH-positive cells [268]. On the contrary, curcumin had little impact on differentiated cells [268]. By utilizing a TCF-LEF reporter assay system in MCF7 cells, these authors confirmed that the effect of curcumin

on breast cancer stem/progenitor cells was mediated through its potent inhibitory effect on Wnt/ β-catenin signaling [268].

1.8.2 Epigallocatechin-3-gallate (EGCG)

Green tea is one of the most widely consumed beverages in the world.

Epidemiological studies suggest an association between green tea consumption and cancer prevention effects. The various polyphenolic catechins contained in green tea are thought to largely account for its chemopreventive activity against certain types of cancer. in particular, many studies indicate that EGCG (Fig. 1.4), the most abundant catechin in green tea, is a potent chemoprevention agent.

EGCG was demonstrated to block Wnt signaling by stabilizing mRNA of HBP1, a suppressor of Wnt signaling, thereby reducing breast cancer cell tumorigenic proliferation as well as invasiveness [269, 270]. The nuclear import of β-catenin was

decreased in adenomas isolated from EGCG-treated Apc Min/+ mice, a widely used

transgenic model recapitulating human colon cancer that bears an Adenomatous

Polyposis Coli (APC) gene mutation [271, 272]. In addition, several studies revealed that

EGCG suppressed Akt activation in both colon cancer cell lines and in vivo mouse

models [271-274]. Additionally, EGCG has been found to negatively regulate NF-κB

activity and inhibit the ATP- or IL-1β induced activation of NF-κB [275-279].

Combination of EGCG and doxorubicin was suggested to eradicate prostate cancer stem

cells. Relatively low levels of EGCG plus doxorubicin eradicated established tumors (in

NOD/SCID mice) that were derived from pancreatic tumor-initiating cells [280]. EGCG

was demonstrated to inhibit self-renewal capacity of prostate cancer stem cells [281].

Furthermore, EGCG was found to inhibit EMT by inhibiting the expression of vimentin, slug, snail and nuclear β-catenin, and the activity of LEF-1/TCF responsive reporter, and

also retards migration and invasion of prostate cancer stem cells [281].

1.8.3 Quercetin

Quercetin (Fig. 1.4), a dietary polyphenol commonly detected in apples,

cranberries, blueberries, and onions, was shown to possess antioxidant and

anticarcinogenic activities.

Quercetin was suggested to be a potent inhibitor of β-catenin/TCF signaling in

SW480 colon cancer cells, and the reduced β-catenin/TCF transcriptional activity was

due to the decreased nuclear β-catenin and TCF-4 proteins [282]. Another group reported

that the inhibition of colon cancer cell growth by quercetin is related to the inhibition of

cyclin D1 and surviving expression through Wnt/ β-catenin signaling pathway [283].

Very recently, quercetin was demonstrated to potently eliminate pancreatic cancer stem cell characteristics by affecting clonogenicity, spheroid formation, ALDH1 activity along with signaling involved in apoptosis resistance, proliferation, angiogenesis, NF-κB and

epithelial-mesenchymal transition [284]. Moreover, quercetin was found to synergize

with green tea EGCG in inhibiting the self-renewal properties of prostate cancer stem

cells, inducing apoptosis, and blocking cancer stem cell migration and invasion [281].

1.8.4 Sulforaphane

In a very recent report, Kallifatidis et al. suggested that sulforaphane could

abrogate the resistance of pancreatic TICs to TRAIL (tumor necrosis factor-related

apoptosis-inducing ligand) by interfering with TRAIL-activated NF-κB signaling [285].

Hence, they concluded that combination of sulforaphane with TRAIL would be a promising strategy for targeting pancreatic TICs [285]. Sulforaphane was previously shown to induce down-regulation of β-catenin in human cervical carcinoma HeLa and

hepatocarcinoma HepG2 cells [41]. On the other hand, several studies have reported the

activity of sulforaphane to down-regulate Akt pathway in ovarian, prostate, and

colorectal cancers [286-288]. Very recently, PI3K/Akt pathway was demonstrated to play

an important role in regulating breast stem/progenitor cells by promoting β-catenin downstream events through phosphorylation of GSK3 β [182].

1.9 Research Objectives and Hypothesis

The objective of this research is to explore the chemopreventive mechanisms of sulforaphane against pancreatic cancer and breast cancer. This will be investigated in the following aspects:

1) To reveal a novel molecular target of sulforaphane in pancreatic cancer and to

trace the mechanism of disruption of Hsp90-p50 Cdc37 complex. It is

hypothesized that sulforaphane inhibits Hsp90 function by disrupting Hsp90-

p50 Cdc37 complex in pancreatic cancer cells through direct interaction with

specific residues.

2) To evaluate the effect of sulforaphane on breast cancer stem cells and to study

its impact on self-renewal pathway of cancer stem cells. It is hypothesized that

sulforaphane targets breast cancer stem cells by inhibiting Wnt/ β-catenin self-

renewal pathway.

3) To develop three different broccoli sprout preparations, compared their ability

to deliver sulforaphane in vivo , and evaluated the pharmacokinetics and tissue

distribution in mice. It is hypothesized that the sulforaphane-rich broccoli

sprout preparation will deliver optimal sulforaphane levels to plasma and

tissues.

Glucosinolate Glucoraphanin

Isothiocyanate Sulforaphane

Figure 1.1. Chemical structures of glucosinolate, glucoraphanin, isothiocyanate, and sulforaphane

Geldanamycin R = OCH 3 IPI-504 17-AAG R = NHCH 2CH=CH 2 17-DMAG R = NHCH 2CH 2N(CH 3)2

Radicicol

Figure 1.2. Chemical structures of Hsp90 N-terminal ATP binding site inhibitors

A B

Cancer stem cell Self- Differentiation Isolation of renewal cancer stem cell Self- renewal Cancer stem cell Progenitor cell Inoculation to immunologically- permissive mouse (NOD/SCID mouse)

Mature cell

Recapitulation of parental tumor

Figure 1.3. Cancer stem cell theory

Curcumin EGCG

Quercetin

Figure 1.4. Chemical structures of curcumin, EGCG, and quercetin

Chapter 2: Sulforaphane Inhibits Hsp90 Function by Disrupting Hsp90-p50 Cdc37

Complex in Pancreatic Cancer Cells through Direct Interaction with Specific

Residues

2.1 Abstract

Inhibition of the ATP binding of molecular chaperone Hsp90 is a promising

strategy against cancer. However, disruption of protein-protein interactions in Hsp90

complex is not well understood. Here, we demonstrate that the natural product

sulforaphane blocks Hsp90-p50 Cdc37 interaction, inhibits Hsp90 function, and exhibits

anticancer activity, and we trace its activity to a novel interaction site of Hsp90.

Sulforaphane inhibited pancreatic cancer in vitro and in vivo . Sulforaphane blocked the interaction of Hsp90 with its cochaperone p50 Cdc37 and induced the degradation of Hsp90 clients in pancreatic cancer cells. Proteolytic fingerprinting and LC-MS revealed sulforaphane interaction with the N-terminal domain of Hsp90 and the central domain of p50 Cdc37 . LC-MS and NMR identified a covalent sulforaphane adduct in sheet 2 and the

adjacent loop in Hsp90’s N-terminal domain. These data suggest a novel mechanism of

sulforaphane that disrupts protein-protein interactions in Hsp90 complex for its

chemoprevention activity.

2.2 Introduction

Numerous studies have shown the chemoprevention efficacy of high consumption of broccoli and broccoli sprouts against various cancers [1]. Sulforaphane is a major compound from broccoli/broccoli sprouts to possess chemopreventive activity [12, 18].

Previous studies suggest that sulforaphane modulates multiple targets [1, 29, 194], which regulates many cellular activities including oxidative stress, apoptosis induction, cell cycle arrest, angiogenesis and metastasis suppression, and detoxification of carcinogens

[1, 30].

The regulation of multiple targets by sulforaphane may be through direct inhibition of each individual target, or through a common modulator that regulates multiple targets. A recent study reported that sulforaphane enhanced acetylation of heat shock protein 90 (Hsp90) by inactivating HDAC6 in prostate cancer cells, thereby inhibiting its association with androgen receptor (AR), leading to destabilization of AR protein and disruption of AR signaling [289]. Hsp90 regulates the maturation of a wide range of oncogenic client proteins [96]. Up to date, more than two hundred of Hsp90 client proteins have been identified, which includes mutated signaling proteins (Raf, p53), transmembrane tyrosine kinases (HER-2, EGFR), signaling proteins (Akt, HIF-1α),

and cell cycle regulators (Cdk4), hormone receptors (AR) [96, 290]. Previous studies

have shown that sulforaphane down-regulates a group of signaling proteins (such as Akt,

Cdk4, AR, HIF-1α), while these proteins are also Hsp90 client proteins [287, 289, 291].

It remains unknown if sulforaphane directly modulates Hsp90.

In this study, we intend to investigate if sulforaphane directly modulates Hsp90 and thereby reducing the Hsp90 client protein levels in pancreatic cancer cells. Several oncogenic proteins have been reported to be critical regulator in pancreatic cancers, such as kinases (e.g., K-Ras and Akt ), mutated tumor-suppressor genes (e.g., p53 mutant)

[292], and growth factor receptors (e.g., EGFR) [293]. Modulation of Hsp90 and its client

proteins by sulforaphane may provide rationale for chemoprevention of pancreatic

cancer.

We first studied the effect of sulforaphane to down-regulate Hsp90 client proteins

through proteasomal degradation pathway in pancreatic cancer cells. Given that Hsp90

chaperone function is dependent on an array of cochaperones (Hop, Hsp70, Hsp40,

p50 Cdc37 , p23 Sba1 , etc.) and ATP cycle [124], we are interested in characterizing the

impact of sulforaphane on Hsp90-cochaperone interactions and ATP binding to Hsp90.

Our data showed that sulforaphane inhibits Hsp90 through an ATP-binding independent

manner since sulforaphane did not interfere with ATP binding to Hsp90; instead,

sulforaphane may directly interact with specific amino acid residues of Hsp90 and

p50 Cdc37 to disrupt the complex, which is distinct from other Hsp90 inhibitors targeting

ATP-binding pocket. The p50 Cdc37 acts as a critical molecular adaptor to recruit protein

kinases to the Hsp90 chaperone machinery [95, 111]. Most of the Hsp90 clients that are

associated with p50 Cdc37 are crucial elements implicated in signal transduction, cell proliferation and survival [95]. Further studies using proteolytic fingerprinting assay, LC-

MS, and NMR indicated that sulforaphane directly modifies specific amino acid residues of Hsp90 and p50 Cdc37 , which may lead to disruption of Hsp90-p50 Cdc37 complex.

2.3 Materials and Methods

2.3.1 Cell culture

Human pancreatic cancer cell lines Mia Paca-2, Panc-1, AsPc-1, and BxPc-3, were obtained from American Type Culture Collection. Authentication of these cell lines in their origin sources included morphology analysis, growth curve analysis, isoenzyme analysis, short tandem repeat analysis, and mycoplasma detection. Panc-1, AsPc-1, and

BxPc-3 were maintained in RPMI1640 medium (Invitrogen) supplemented with 10% fetal bovine serum (Fisher Scientific) and 1% penicillin-streptomycin (Invitrogen). Mia

Paca-2 was maintained in DMEM medium (Invitrogen) supplemented with 10% fetal bovine serum and 1% penicillin-streptomycin-glutamine.

2.3.2 Reagents

Sulforaphane was purchased from LKT Laboratories. The following antibodies were used for immunoblotting: Akt (Cell Signaling), p23 Sba1 (Abcam), p53 mutant, Cdk4, p50 Cdc37 , Hsp90, β-actin (Santa Cruz Biotechnology).

2.3.3 Western blotting analysis

After treatment, cells were harvested by washing twice with ice-cold PBS and lysed in RIPA lysis buffer (20 mM Tris-HCl, 150 mM NaCl, 1% NP-40, 5 mM EDTA, 1 mM sodium orthovanadate, pH 7.5) supplemented with protease inhibitors (Pierce) on ice for 20 min. After centrifugation at 14,000 rpm for 15 min at 4 °C, the supernatant was recovered. Protein concentration was determined with BCA Protein Assay Reagents

(Pierce). Equal amounts of protein were subject to SDS-PAGE, and transferred to PVDF

membrane (BioRad). The membrane was then incubated with appropriate primary antibodies at 4 °C overnight, followed by 2 h incubation with secondary antibodies at room temperature.

2.3.4 Triton-soluble and triton-insoluble protein fraction

Sulforaphane (15 µM) was added after pre-incubation cells with 10 µM MG132.

After 24 h treatment, cells were collected and lysed in a lysis buffer (50 mM Tris-HCl,

150 mM NaCl, 1 mM EDTA, 1 mM EGTA, 1% Triton X-100, 2 mM sodium

orthovanadate, 5 mM sodium fluoride, 1 mM phenyl methyl sulfonyl fluoride and 5,000

U/ml aprotinin) on ice for 10 min. The samples were centrifuged at 16,000 g for 15 min

at 4 °C, and the supernatant (triton-soluble fraction) was collected. The pellets (triton-

insoluble fraction) were further lysed in 2% SDS in 50 mM Tris-HCl and boiled for 15

min.

2.3.5 MTS cell proliferation assay

Cells were seeded in 96-well microplates at a density of 3,000-5,000 cells per

well. After 72 h treatment with sulforaphane, cell viability was assessed by MTS assay

(Promega) according to the manufacturer’s instruction. The number of living cells is

proportional to the absorbance at 490 nm.

2.3.6 Caspase-3 activity assay

Cells were treated with sulforaphane and collected after 24 h. Caspase-3 activity

assay was based on the manufacturer’s instruction of Caspase-3/CPP32 Fluorometric

Assay Kit (Biovision Research Products). The cleavage of a caspase-3 substrate was

quantified by using a fluorescence microtiter plate reader with a 400 nm excitation filter and a 505 nm emission filter.

2.3.7 Pancreatic cancer xenograft

The animal study protocol was approved by the University Committee on Use and

Care of Animals (UCUCA) at University of Michigan. Four- to six-week-old athymic

(nu/nu) female mice were obtained from NCI. Mia Paca-2 cells (5×10 6-10×10 6) mixed

with Matrigel were subcutaneously implanted into the right and left flanks of the mice.

Tumor volume was calculated with V=1/2 (width 2 × length). After tumor volume reached

100-150 mm 3, mice were randomized into three groups for treatment. The mice were i.p.

injected with vehicle (saline), 25 mg/kg or 50 mg/kg sulforaphane (five times per week)

for four weeks. Tumor size was monitored twice a week and normalized to the initial

volumes.

2.3.8 Protein cloning, expression, and purification

The plasmids pET28a(+)-hHsp90 β(529-723) and pGEX4T.1-p50 Cdc37 for expression of human Hsp90 βC (Hsp90 β-G529-D723) and GST-p50 Cdc37 protein were

kindly provided by Dr. Thomas Ratajczak (University of Western Australia). The

purified Hsp90 βN protein (Hsp90 β-P1-E245) was kindly provided by Dr. Dan Bolon

(University of Massachusetts Medical School). To clone p50 Cdc37 middle domain

(p50 Cdc37 M, p50 Cdc37 -K147-K276), we used 5’-

GAGGCATATG AAACACAAGACCTTCGTGGAAAAATACG-3’ and 5’-

CCACCTCGAG CTTCATGGCCTTCTCGATGC-3’ as primers, double digested the

PCR product with NdeI and XhoI, and cloned to pET-30a-c(+) vectors. The plasmids

were transformed into Escherichia coli strain Rosetta 2(DE3) (EMD Biosciences Inc.) according to the protocol provided by manufacturer. Protein expression was induced by

0.2 mM IPTG. His-tagged proteins were purified by affinity chomatography through mixing with HisPurTM Cobalt Resin (Pierce), and GST-tagged proteins with glutathione

4B-sepharose (GE Healthcare). The GST tag was removed by thombin cleavage. Purified proteins were dialyzed against PBS, and the purity was assessed by SDS-PAGE. Proteins were stored at -70 °C after adding glycerol to 10%.

2.3.9 ATP-sepharose binding assay

Cells were treated with sulforaphane for 24 h, and then lysed in TNESV buffer

(50 mM Tris, 2 mM EDTA, 100 nM NaCl, 1 mM Na 3VO 4, 25 mM NaF, 1% Triton X-

100, pH 7.5) supplemented with protease inhibitors at 4 °C for 30 min. Protein (200 µg)

was incubated with 25 µl pre-equilibrated γ-phosphate-linked ATP-sepharose (Jena

Bioscience GmbH) in 200 µl incubation buffer (10 mM Tris-HCl, 50 mM KCl, 5 mM

MgCl 2, 20 mM Na 2MoO 4, 0.01% NP-40, pH 7.5) overnight at 4 °C. The protein bound to

sepharose beads were analyzed by Western blotting.

2.3.10 Hsp90 co-immunoprecipitation

Cells were treated with sulforaphane for 24 h, and lysed in 20 mM Tris-HCl (pH

7.4), 25 mM NaCl, 2 mM DTT, 20 mM Na 2MoO 4, 0.1% NP-40, and protease inhibitors.

Protein (500 µg) was incubated with H9010 antibody (Axxora) for 1 h at 4 °C, followed

by overnight incubation with protein A/G agarose at 4 °C. The bound proteins were

analyzed by Western blotting.

2.3.11 Proteolytic fingerprinting assay

Purified protein (0.5 µg) was pre-incubated with DMSO or sulforaphane in assay buffer (10 mM Tris-HCl, 50 mM KCl, 5 mM MgCl 2, 0.1 mM EDTA, pH 7.4) on ice for 1 h, and was digested with trypsin for 6 min. The reaction was terminated by adding SDS buffer and boiling for 3-5 min. The digested products from Hsp90 βN (Hsp90 β-P1-E245) and p50 Cdc37 were analyzed by Western blotting with Hsp90 antibody (N-17, Santa Cruz

Biotechnology) and p50 Cdc37 antibody (C-19, Santa Cruz Biotechnology), respectively.

2.3.12 LC-MS

Hsp90 βN (Hsp90 β-P1-E245, 80 µM) and p50 Cdc37 M (p50 Cdc37 -K147-K276, 80

µM) were incubated with DMSO or sulforaphane in PBS at 37 °C for 30 min. Small

molecules were removed by gel filtration using P-30 micro Bio-Spin columns. Samples

were analyzed on Shimadzu LCMS-2010EV after separation on a PLRP-S polymeric RP-

HPLC column (50×2.1 mm, 1000Å, 5 µm) with a gradient of 5% to 35% B in 30 min (A:

0.1% formic acid in H 2O; B: 0.1% formic acid in CH 3CN) at a flow rate of 0.2 ml/min.

For peptide mapping, Hsp90 βN (80 µM) was incubated with DMSO or sulforaphane (2 mM) for 30 min, and was then treated with DTT (10 mM) for 1 h, followed by

Sequencing Grade Modified Trypsin (Promega) for 24 h at 37 °C. Samples were analyzed on a Vydac Everest reverse-phase C18 monomeric column (250×2.1 mm, 300Å,

5 µm) with a gradient of 5% to 65% B in 60 min at a flow rate of 0.2 ml/min.

2.3.13 NMR spectroscopy

Hsp90 was recombinantly produced with N-terminal hexa-histidine tag and

specifically methyl labeled at Ile side chains as described [294]. The NMR experiments

were performed with 150 µM of Hsp90 in 100% D 2O, 25 mM sodium phosphate, pH 7.2,

300 mM NaCl, 1 mM TCEP, and sulforaphane concentration was varied (dissolved in

deuterated DMSO, final concentration 1% at 1.5 mM sulforaphane) for NMR

measurements. 13 C-1H-methyl TROSY spectra [295] were recorded at 25˚C on a Bruker

AvanceII 900 MHz spectrometer equipped with a TCI cryoprobe. Reported chemical shifts are referenced ag ains t DSS, and processing and peak position plots were done as described [294]. The cross peaks of the methyl-TROSY spectra were 2D-line shape fitted using mixed Lorentzian/Gaussian deconvolution parameters (TopSpin). The rainbow spectra indicate increasing intensity from blue to read over 14 contour levels. All contours above the intensity of the cross peak at 0.72 ppm ( 1H) and 10.8 ppm ( 13 C) are

red. Comparisons with Hsp90 in the absence of sulforaphane were corrected for DMSO

effects.

2.3.14 Statistical analysis

Statistical analysis was performed using student t-test. Data are presented as mean

± SD (n ≥ 3, P < 0.01).

2.4 Results

2.4.1 Sulforaphane induces proteasomal degradation of Hsp90 client proteins in

pancreatic cancer cells

We first examined whether sulforaphane could reduce the levels of Hsp90 client

proteins, Akt, Cdk4, and p53 mutant, in Mia Paca-2 and Panc-1. Both cell lines contain

p53 mutant and activated Ras [296-298]. As shown in Fig. 2.1A and 2.1B, the levels of these proteins declined in a time- and concentration-dependent manner in response to sulforaphane treatment. Sulforaphane (15 µM) down-regulated Akt, Cdk4, and p53 mutant by 70-95% after 48 h, and 25 µM of sulforaphane decreased these client proteins by 80-95% after 24 h (Fig. 2.1A and 2.1B).

We further investigated whether sulforaphane-induced Hsp90 client protein down-regulation was through proteasomal degradation. MG132, a well-known proteasome inhibitor, was pre-incubated with pancreatic cancer cells to block proteasome function. Inhibition of proteasome leads to accumulation of unfolded proteins, which form insoluble “aggresomes” in the triton-insoluble fraction of cell lysates. As shown in

Fig. 2.1C, Akt and Cdk4 were found to accumulate in the triton-insoluble fraction after sulforaphane treatment in the presence of MG132, while they were undetectable in triton- insoluble fraction after sulforaphane treatment alone. These data indicate that sulforaphane induced proteasomal degradation of Hsp90 client proteins in pancreatic cancer cells.

2.4.2 Sulforaphane inhibits pancreatic cancer cells in vitro and exhibits anticancer activity in pancreatic cancer xenograft

We evaluated the anti-proliferative activity of sulforaphane in human pancreatic cancer cell lines, Mia Paca-2, Panc-1, AsPc-1, and BxPc-3, by MTS cell proliferation assay. After treatment with increasing concentrations of sulforaphane for 72 h, the percentage of viable cells relative to untreated cells is plotted in Fig. 2.2A. Cell viability decreased as the concentration of sulforaphane increased in a similar pattern among all

the four cell lines, with the IC 50 ’s around 10-15 µM. Then we used Mia Paca-2 to

illustrate the effect of sulforaphane on caspase-3 activation. The results of caspase-3

assay showed that sulforaphane induced activation of caspase-3 in a concentration-

dependent manner (Fig. 2.2B). In comparison with control cells, 10 µM sulforaphane was

able to enhance the caspase-3 activity to more than 3-fold ( P = 0.003), and more than a 4-

fold increase was observed when cells were exposed to 15 µM sulforaphane ( P = 0.001).

These data suggest that sulforaphane inhibits pancreatic cancer cell growth and induces

apoptosis in vitro .

We next evaluated the efficacy of sulforaphane in Mia Paca-2 xenograft model.

Two weeks after cancer cell implantation, the mice were injected (i.p.) with either vehicle

or sulforaphane at 25 mg/kg or 50 mg/kg for 1 month. At the end of treatment, control

tumors reached an average size of 715 ± 66 mm³, while 25 mg/kg and 50 mg/kg sulforaphane treatment shrank the tumors to 538 ± 88 mm³ and 354 ± 52 mm³, respectively (Fig. 2.2C). These correspond to 25% (P = 0.002) and 50% ( P = 0.0001) inhibition of tumor growth. Meanwhile, sulforaphane showed no apparent toxicity as determined by body weight measurement (Fig. 2.2D).

2.4.3 Sulforaphane inhibits Hsp90 by blocking Hsp90-p50 Cdc37 complex in pancreatic

cancer cells

Most Hsp90 inhibitors, e.g., 17-AAG, bind to Hsp90 N-terminal ATP pocket,

preventing the maturation of Hsp90 client proteins [299]. As shown in Fig. 2.3A,

sulforaphane (5-30 µM) was unable to decrease the amount of ATP-bound Hsp90, which

suggests that sulforaphane did not affect the ATP-binding capacity of Hsp90. As positive

control, 17-AAG (5 µM) blocked the ATP binding to Hsp90 and decreased the amount of

ATP-bound Hsp90 in the ATP-sepharose binding assay.

Since Hsp90 chaperone function depends on multiple cochaperones to form the complex, we next investigated the effect of sulforaphane on Hsp90-cochaperone interaction using co-immunoprecipitation assay. Sulforaphane (15-30 µM) treatment for

24 h in Mia Paca-2 cells reduced the amount of p50 Cdc37 co-precipitated with Hsp90 by approximately 3- to 5-fold (Fig. 2.3B). However, sulforaphane (15-30 µM) did not change the amount of p23 Sba1 in precipitated Hsp90 complex. The total protein levels of

Hsp90, p50 Cdc37 , and p23 Sba1 were unchanged after sulforaphane treatment (Fig. 2.3B).

These data suggest that sulforaphane blocked Hsp90-p50 Cdc37 interaction, but not Hsp90-

p23 Sba1 interaction.

2.4.4 Proteolytic fingerprinting and LC-MS detect sulforaphane binding with Hsp90 and p50 Cdc37

The p50 Cdc37 is a substrate-targeting factor of Hsp90 that binds via its middle

domain (p50 Cdc37 -K147-K276) to the N-terminal domain of Hsp90 [111, 114]. To examine how sulforaphane impairs the association of Hsp90 with p50 Cdc37 , we first

performed proteolytic fingerprinting assay to test if sulforaphane can bind to Hsp90 N-

terminus and p50 Cdc37 . In the absence of sulforaphane, the N-terminal domain of Hsp90

(Hsp90N, Hsp90-P1-E245) was highly sensitive to trypsin digestion. Pre-incubation of

sulforaphane with Hsp90N prevented it from trypsin hydrolysis, with a strong band

representing intact Hsp90N even at a high concentration of trypsin (Fig. 2.4A).

Sulforaphane also showed moderate protection of p50Cdc37 against trypsin digestion (Fig.

2.4B). When sulforaphane was pre-incubated with p50Cdc37 , stronger bands of small molecular weight were observed after trypsin digestion in comparison to control.

We next conducted LC-MS analysis of Hsp90N and p50 Cdc37 middle domain

(p50 Cdc37 M) to test whether sulforaphane can form covalent adducts with these protein regions. After incubation with 2 mM sulforaphane, the mass spectrum of Hsp90N exhibited a mass increase which corresponded to the molecular weight of one sulforaphane molecule (Fig. 2.5A). Moreover, a ten-fold increase in the concentration of sulforaphane (20 mM) enhanced the percentage of Hsp90N tagged with one sulforaphane molecule and gave rise to Hsp90N tagged with two sulforaphane molecules (Fig. 2.5A).

Further peptide mapping study indicated that a short peptide Ile 72-Arg 81

(IDIIPNPQER) was tagged with sulforaphane (Fig. 2.5B). Given that sulforaphane was shown to form adduct with lysine present in other proteins [300], the covalent modification may occur at Arg 81.

The p50 Cdc37 molecule can be structurally dissected into three domains, an N-

terminal kinase-binding domain (p50 Cdc37 N, residues 1-127), an Hsp90 binding middle

domain (p50 Cdc37 M, residues147-276), and a C-terminal domain (p50 Cdc37 C, residues 283-

378) [301]. As shown in Fig. 2.5C, incubation of 2 mM sulforaphane with p50 Cdc37 M produced three and four sulforaphane-tagged p50 Cdc37 M, while 20 mM sulforaphane incubation generated p50 Cdc37 M tagged with several sulforaphane molecules. Higher

concentrations of sulforaphane (200 mM) afforded a complex mass spectrum due to a

large number of sulforaphane labeling of p50 Cdc37 M.

These results suggest that sulforaphane may covalently interact with Hsp90N and p50 Cdc37 M. These interactions may lead directly or indirectly to blocking Hsp90 interaction with p50 Cdc37 .

2.4.5 NMR reveals the sulforaphane binding sites on Hsp90

Experiments with isolated domains, however, may not accurately represent the full length protein. Hsp90 full-length dimer in solution with molecular weight of 170 kDa is not suitable for LC-MS analysis due to its size. Therefore, we used NMR approach to analyze the interaction of Hsp90 with sulforaphane using the full-length Hsp90 protein.

NMR spectroscopy is a non-invasive technique to map binding sites and conformational changes in solution and proved useful for the molecular understanding of chaperone interaction with ligands [294, 295, 302, 303]. We studied sulforaphane binding to the Hsp90 full length protein by NMR using an isoleucine-specific labelling strategy, which allowed us to overcome the notorious NMR size limitation for the 170 kDa Hsp90 dimer. Binding of sulforaphane to Hsp90 resulted in well-resolved NMR spectra (Fig.

2.6A). When comparing those to the spectra to Hsp90 obtained in the absence of sulforaphane, we found that the majority of peaks are identical, while a subset of peaks shifts (Fig. 2.6B).

We previously assigned the N-terminal and middle domains of Hsp90 [294], which allowed us to map the chemical shift changes within those domains (Fig. 2.6C).

Two binding clusters were particularly intriguing. We noticed chemical shift changes in the N-terminal domain in sheet 2 for Ile 74 and Ile 75 (Fig. 2.6D). Both Ile’s are part of the short peptide IDIIPNPQER covering sheet 2 and the adjacent loop, which we

independently identified to be modified with sulforaphane by LC-MS tryptic peptide mapping (Fig. 2.5B). The NMR data are in agreement with the mass spectrometric findings that only one sulforaphane adduct was found on the N-terminal domain of

Hsp90 at a comparable concentration (Fig. 2.5A). The signals of two additional N- terminal Ile side chains, Ile 43 and Ile 125, are shifting. Ile 43 is found in α-helix in the

neighborhood of Ile 74 and the change of environment of this residue might have affected

Ile 43 indirectly. In the middle domain, we also identified chemical shifts in four Ile side

chains. Three of those (Ile 369, Ile 440 and Ile 482) are sufficiently close to Cys side

chains, which in other proteins are known to undergo reaction with sulforaphane. Cys

365 is 6.3 Å away from Ile 369 and Cys 520 is 5.3 Å and 13.6 Å away from Ile 440 and

Ile 482 respectively. Ile 287 is found in the loop in the interface of the N-terminal and

middle domain of Hsp90 and this could be shifting due to change in the domain

orientation.

As controls, we also compared the sulforaphane-dependent shifts with those of

nucleotides and the ATP binding inhibitor geldanamycin [294]. While all of those ligands

show a coherent pattern clustering around the nucleotide binding pocket, sulforaphane

binds in a very different mode. The NMR data further confirms that sulforaphane does

not block the nucleotide binding pocket of Hsp90, in agreement with the results of ATP-

sepharose binding assay (Fig. 2.3A).

In addition, we also analyzed whether sulforaphane binding could have any effect on binding of Hsp90’s co-chaperones p23 Sba1 (Fig. 2.6C) and p50 Cdc37 (Fig. 2.6E). In both cases, we concluded that direct obstruction of the binding interface for those proteins is

unlikely since most of the residues involved in binding of those cochaperones were not affected by sulforaphane. Ile 125, which belongs to the p50 Cdc37 interaction interface,

shifts upon sulforaphane interaction, but the nearby Ile 122 is not affected. This indicates

that Ile 125 shifting is most likely caused by an allosteric effect, which may disturb the

interaction p50 Cdc37 .

2.5 Discussion

Pancreatic cancer is one of the leading causes of cancer death in the United States

[304]. The currently available therapeutics (such as gemcitabine) have shown very limited success on treatment of this aggressive disease [90]. Given that pancreatic carcinogenesis is characterized by complex molecular basis that are involved many oncogenic proteins, Hsp90, an essential molecular chaperone that regulates the stability and maturation of a wide range of oncogenic proteins, may provide advantages as a promising target for pancreatic cancer therapeutics [95, 96]. Inhibition of Hsp90 may result in simultaneous degradation of multiple client proteins, such as kinases, hormone receptors, and transcription factors in cancers [96, 290]. Among these proteins phosphoinositide 3-OH kinase (PI3K)/Akt pathway and p53 mutant are important molecular targets in pancreatic cancer [305, 306]. K-Ras mutation is known to occur in approximately 90% of human pancreatic ductal adenocarcinomas [304, 307]; and

PI3K/Akt is a critical effector pathway of activated K-Ras [304, 308]. Moreover,

PI3K/Akt pathway is constitutively active in the majority of pancreatic cancer [304, 309];

and amplification or activation of Akt2 is encountered in up to 60% of pancreatic cancer

[304, 306, 310]. Over-expression or activation of p53 mutant is a critical event in human pancreatic carcinogenesis as well [305]. Thus, interference with Hsp90 function and subsequent down-regulation of these client proteins provides a rational strategy to abrogate signaling against pancreatic cancer.

In the current study, we show that sulforaphane, a dietary component from broccoli/broccoli sprouts, inhibits Hsp90 chaperone function and promotes proteasomal degradation of Hsp90 client proteins (Fig. 2.1), resulting in pancreatic cancer cell death

(Fig. 2.2). Sulforaphane has been reported as a potent chemoprevention agent, which provides many advantages over traditional Hsp90 inhibitors, including high bioavailability and low toxicity [30]. Sulforaphane from broccoli extracts is efficiently and rapidly absorbed in human small intestine, and quickly distributed throughout the body [1, 65]. Although the accumulation of sulforaphane in human pancreas tissue is not clear yet, plasma concentrations of sulforaphane equivalents reached as micromolar concentrations in the blood after administration of a single dose of 200 µmol broccoli sprout isothiocyanates (mainly sulforaphane) [66].

Hsp90 consists of an N-terminal ATP-binding domain (Hsp90-P1-E245), a

middle domain (Hsp90-K246-D528) and a C-terminal dimerization domain (Hsp90-

G529-D723) [311, 312]. Since Hsp90 chaperone function depends on the conformational

changes driven by its ATPase activity [299], numerous Hsp90 inhibitors, represented by

geldanamycin and 17-AAG, have been developed to inhibit its chaperone function by

interfering with ATP binding [124]. These small molecules competitively bind to the

deep ATP pocket of Hsp90, thereby blocking the ATPase cycle and result in proteasomal degradation of its client proteins [116, 121]. Although 17-AAG has entered clinical trials, its hepatotoxicity, low water solubility, low oral-bioavailability, may limit its application

[313]. Recently, our group has demonstrated that disruption of Hsp90-p50 Cdc37 complex inhibits Hsp90 chaperone function and induces degradation of Hsp90 client proteins, leading to anticancer activity [144, 155, 156].

In contrast to these classical Hsp90 inhibitors, sulforaphane impaired Hsp90 chaperone function by blocking formation of the Hsp90-p50 Cdc37 complex (Fig. 2.1 &

2.3). We also examined detailed mechanism of how sulforaphane blocked Hsp90- p50 Cdc37 complex formation. Proteolytic fingerprinting assay suggested that sulforaphane

may interact directly with Hsp90N and p50 Cdc37 , thereby protecting the proteins from

trypsin digestion (Fig. 2.4). Sulforaphane did not enter the Hsp90’s ATP binding pocket

but was found by both NMR and LC-MS to interact with the sheet 2 of Hsp90 N-terminal

domain and its adjacent loop (Fig. 2.5 & 2.6). Sulforaphane covalently modifies the short

peptide IDIIPNPQER covering sheet 2 and the adjacent loop, most likely at Arg 81.

The sulforaphane-induced shifts in the N-terminal and middle domain of Hsp90

suggest that sulforaphane may not directly block the interface of Hsp90- p50 Cdc37 . In particular, NMR did not show any signal shift of Ile 122. Those data and finding by others suggest that sulforaphane may interact directly with p50 Cdc37 . Sulforaphane was

described previously to mediate chemopreventive effects through its ability to react with

the thiol side chain of cysteine residues in target proteins [314, 315]. When analyzing the

position of the six cysteines in Hsp90, we noted that only Cys 365 is to a minor extent

solvent accessible, all others are entirely buried. In the case of p50 Cdc37 , three out of the four cysteines are buried, however the thiol group of Cys 203 is solvent exposed. This

Cys is close to the Hsp90 binding interface, which makes it a likely candidate for a sulforaphane adduct that could directly block the Hsp90-p50 Cdc37 interaction.

In comparison, our data suggest that sulforaphane did not affect Hsp90-p23 Sba1

interaction (Fig. 2.3B & 2.6C), although p23 Sba1 binds to a similar region of Hsp90 as p50 Cdc37 does. The p50 Cdc37 binds to the Hsp90 client protein via the N-terminus and facilitates the client loading to Hsp90 via its middle domain interaction with the N- terminal domain of Hsp90 [111, 114, 301]. Disruption of Hsp90-p50 Cdc37 complex by sulforaphane may prohibit kinase client loading to Hsp90 and result in client degradation

(Fig. 2.1). The unchanged level of Hsp90-p23 Sba1 complex upon sulforaphane treatment indicates that either the chaperone cycle still proceeds to later stages without client folding or some non-kinase client proteins are loaded onto Hsp90 without the help of p50 Cdc37 but require p23 Sba1 for maturation [111].

The selective disruption of Hsp90-p50 Cdc37 but not Hsp90-p23 Sba1 indicates the

specificity of sulforaphane inhibition of protein-protein interaction in Hsp90 complex.

The crystal structures of Hsp90-p50 Cdc37 and Hsp90-p23 Sba1 complex may provide some clues for the specificity. The p50 Cdc37 binding Hsp90 is in an open confirmation with small conformational changes compared with nucleotide-free structure [114]. The majority of the interaction involves Met 164, Leu 165, Ala 204, and Leu 205 of p50 Cdc37 ,

and Ala 103, Ala 107, Ala 110, Gly 111, Ala 112, Met 116 and Phe 120 of Hsp90 [114].

Results of our recent split Renilla luciferase protein fragment-assisted complementation

(SRL-PFAC) bioluminescence assay have shown that mutation of a single amino acid can reduce the Hsp90-p50 Cdc37 interaction by 70-95% [316]. Thus, it is not surprising that

binding of sulforaphane to Hsp90 and p50 Cdc37 resulted in disruption of the complex. By

contrast, extensive interaction interface is involved in Hsp90-p23 Sba1 complex when p23 Sba1 binds to the ATP-bound, closed confirmation of Hsp90, which may make this

interface less sensitive to sulforaphane.

In addition, we observed that sulforaphane slightly protected Hsp90 C-terminus

(Hsp90 βC, Hsp90 β-G529-D723) against trypsin digestion. LC-MS analysis of Hsp90C

suggested that sulforaphane may covalently interact with Hsp90C. At present, no NMR

assignment of the C-terminal domain is available, but we could identify three residues

that could be allocated to the C-terminal domain. For those three residues we did not

observe sulforaphane-dependent shifts (Figs. 6ab). The NMR data cannot fully exclude,

though, that sulforaphane might interact with flexible areas in the C-terminus. The

dynamic segments of Hsp90 correspond to the large area around the cross peak at 0.85

(1H) and 12.5 ( 13 C), which shows line broadening upon sulforaphane binding. This may

indicate an effect of sulforaphane on the dynamics in Hsp90. The discrepancy between

LC-MS and NMR studies could be explained by the fact that experiments with isolated

domains may expose binding site that might not be available in the context of the full

length protein. Another possible reason is that interaction between sulforaphane and some

residues might not cause Ile shift.

2.6 Conclusions

In summary, our study has revealed a novel molecular mechanism of sulforaphane for its chemoprevention activity. Sulforaphane directly interacts with specific amino acid residues of Hsp90 and p50 Cdc37 , disrupts Hsp90-p50 Cdc37 complex, inhibits Hsp90

chaperone function, induces degradation of Hsp90 client proteins, and eventually

suppresses pancreatic cancer growth. These data suggest a novel mechanism of

sulforaphane that disrupts protein-protein interactions in Hsp90 complex for its

chemoprevention of pancreatic cancer.

A Mia Paca-2 15 µM SF Panc-1 15 µM SF 0 12 24 48 h 0 12 24 48 h Akt

Cdk4 p53 mutant β-actin

B Mia Paca-2 24 h Panc-1 24 h SF (µM) 0 5 15 25 0 5 15 25 Akt Cdk4

p53 mutant β-actin

C Mia Paca-2 Panc-1 10 µM MG132 - - + - - + 15 µM SF - + + - + + Triton-soluble Akt Triton-insoluble

Triton-soluble Cdk4 Triton-insoluble

Triton-soluble β-actin Triton-insoluble

Figure 2.1. Sulforaphane induces proteasomal degradation of Hsp90 client proteins

AB Mia Paca-2

140 5 P=0.001 120 4.5 4 100 P=0.003 3.5 80 3 Mia Paca-2 60 2.5 Panc-1

Cell Viability Viability Cell (%) 40 2 AsPc -1 20 1.5 BxPc -3 1 0 0.5 0.010 0.1 1 10 100 0

Concentration of Sulforaphane (µM) in Activity Caspase-3Increase Fold 0 5 10 15

Concentration of Sulforaphane (µM)

C D 1000

) Control 3 Control 30 25 mg/kg SF 800 25 mg/kg SF ) 50 mg/kg SF g 50 mg/kg SF P=0.002 600 P=0.0001 25 400 20 200 Body Weight ( Weight Body Tumor Volume (mm Volume Tumor 0 15 10 20 30 40 50 10 15 20 25 30 35 40 45 Days Days

Figure 2.2. Sulforaphane exhibits anticancer activity in vitro and in vivo

ATP Pull-down SF (µM) 17-AAG (µM) 0 5 15 30 0 5 Hsp90

B IP: Hsp90 Supernatant SF (µM) 0 5 15 30 0 5 15 30 p50 Cdc37

p23 Sba1 Hsp90

Figure 2.3. Influence of sulforaphane on ATP binding of Hsp90 and Hsp90-cochaperone association in Mia Paca-2 cells

A Ctrl 1 mM SF Trypsin (µg/ml) 0 15 50 0 15 50 Hsp90N→ ← 36 kD

← 22 kD

B Ctrl 1 mM SF Trypsin(µg/ml) 0 15 50 100 0 15 50 100 p50 Cdc37 → ← 50 kD

← 36 kD

← 22 kD

← 16 kD

Figure 2.4. Proteolytic fingerprinting assay of Hsp90 and p50 Cdc37

Hsp90N + 0 mM SF Hsp90N + 2 mM SF Rel. Intensity Rel. Intensity Rel.

m/z m/z

Hsp90N + 20 mM SF Hsp90N + 200 mM SF Rel. Intensity Rel. Intensity Rel.

m/z m/z

B Rel. Intensity Rel.

m/z

C p50 Cdc37 M + 0 mM SF p50 Cdc37 M + 2 mM SF Rel. Intensity Rel. Intensity Rel.

m/z m/z

p50 Cdc37 M + 20 mM SF p50 Cdc37 M + 200 mM SF Rel. Intensity Rel. Intensity Rel.

m/z m/z Figure 2.5. LC-MS analysis of Hsp90 and p50 Cdc37

A C

D E

Figure 2.6. Sulforaphane binding to Hsp90 mapped by NMR

Chapter 3: Sulforaphane Potentiates the Efficacy of 17-Allylamino 17-

Demethoxygeldanamycin against Pancreatic Cancer through Enhanced Abrogation

of Hsp90 Chaperone Function

3.1 Abstract

Heat shock protein 90 (Hsp90), an essential molecular chaperone that regulates the stability of a wide range of oncogenic proteins, is a promising target for cancer therapeutics. We investigated the combination efficacy and potential mechanisms of sulforaphane, a dietary component from broccoli and broccoli sprouts, and 17-allylamino

17-demethoxygeldanamycin (17-AAG), an Hsp90 inhibitor, in pancreatic cancer. MTS assay demonstrated that sulforaphane sensitized pancreatic cancer cells to 17-AAG in vitro . Caspase-3 was activated to 6.4-fold in response to simultaneous treatment with sulforaphane and 17-AAG, whereas 17-AAG alone induced caspase-3 activity to 2-fold compared to control. ATP binding assay and co-immunoprecipitation revealed that sulforaphane disrupted Hsp90-p50 Cdc37 interaction while 17-AAG inhibited ATP binding to Hsp90. Concomitant use of sulforaphane and 17-AAG synergistically down-regulated

Hsp90 client proteins in Mia Paca-2 cells. Co-administration of sulforaphane and 17-

AAG in pancreatic cancer xenograft model led to more than 70% inhibition of the tumor

growth while 17-AAG alone only suppressed the tumor growth by 50%. Our data suggest that sulforaphane potentiates the efficacy of 17-AAG against pancreatic cancer through enhanced abrogation of Hsp90 function. These findings provide a rationale for further evaluation of broccoli/broccoli sprout preparations combined with 17-AAG for better efficacy and lower dose-limiting toxicity in pancreatic cancer.

3.2 Introduction

Pancreatic cancer, an aggressive malignancy, is the fourth leading cause of cancer death in the United States [317], and the overall 5-year survival rate after diagnosis for pancreatic cancer patients is below 5% [318]. Currently available therapeutics such as surgery, chemotherapy, and radiotherapy have shown very limited success on treatment of this aggressive disease [90]. Since a large number of epidemiological studies have demonstrated an association between the reduced risk of various cancers and consumption of fruits and vegetables, naturally-occurring dietary compounds have been tested for cancer chemoprevention. For example, vitamin D was shown to reduce proliferation of pancreatic cancer in vitro by down-regulation of glycogen synthase

kinase 3 [319]. Curcumin was demonstrated to suppress the growth of human pancreatic

cancer in nude mice [320]. A recent study found that curcumin potentiates anticancer

activity of gemcitabine in pancreatic cancer mouse model through inhibition of NF-κB

target genes, cell proliferation, and angiogenesis [321].

Numerous studies have substantiated the protective effect of high consumption of cruciferous vegetables, especially broccoli/broccoli sprouts, against carcinogenesis and cancer progression [1, 2]. These effects have been attributed to the activity of isothiocyanates that are converted from glucosinolates [1, 2]. In particular, sulforaphane was found to be derived from glucoraphanin, a major glucosinolate in broccoli/broccoli sprouts [12]. Sulforaphane has been demonstrated to be not only effective in preventing chemically induced cancers in animal models [12, 14-16] but also inhibited the growth of established tumors [21, 22]. Previous studies suggest that sulforaphane modulates multiple targets, such as NF-κB, Chk2, p21, MAPK, death receptor, histone deacetylase,

Stat3, Nrf2, β-catenin [1, 29, 31, 41, 194], which regulates diverse cellular activities

including oxidative stress, apoptosis induction, cell cycle arrest, angiogenesis and

metastasis suppression [1, 15, 30]. In addition, the chemoprevention efficacy of

sulforaphane is also linked to the induction of phase II metabolism enzymes and

inhibition of phase I metabolism enzymes, which enhances the detoxification of

carcinogens [1, 29, 285].

Given that pancreatic carcinogenesis is characterized by complex molecular basis

that are involved many oncogenic proteins, heat shock protein 90 (Hsp90), an essential

molecular chaperone that regulates the stability and maturation of a wide range of client

oncogenic proteins, has recently emerged as a promising target for pancreatic cancer

therapeutics [95, 96]. Inhibition of Hsp90 may result in simultaneous degradation of

multiple oncogenic client proteins, such as kinases, hormone receptors, and transcription

factors in cancers [96, 290, 322]. Among these proteins, phosphoinositide 3-OH kinase

(PI3K)/Akt pathway and p53 mutant are important molecular targets in pancreatic cancer

[306, 311, 323, 324]. K-Ras mutation is known to occur in approximately 90% of human pancreatic ductal adenocarcinomas [304, 307, 325]; and PI3K/Akt is a critical effector pathways of activated K-Ras [304, 308, 326, 327]. Moreover, PI3K/Akt pathway is constitutively active in the majority of pancreatic cancer [304, 309]; and amplification or activation of Akt2 occurred in 60% of pancreatic cancer [304, 306, 310, 328]. Over- expression or activation of p53 mutant is an important event in human pancreatic carcinogenesis as well [305]. Thus, interference with Hsp90 function and subsequent down-regulation of these client proteins provides a rational strategy to abrogate signaling against pancreatic cancer.

Since Hsp90 chaperone function depends on the conformational changes driven

by its ATPase activity [299], an array of Hsp90 inhibitors have been developed to inhibit

its chaperone function by interacting with the N-terminal ATP binding domain [124]. 17-

Allylamino 17-demethoxygeldanamycin (17-AAG), one of the most well-known Hsp90

inhibitors, competitively binds to N-terminal ATP pocket of Hsp90, and induces a

conformational change in the Hsp90 molecule, leading to proteasomal degradation of its

client proteins [124, 125]. 17-AAG has been evaluated in a great deal of preclinical

studies and clinical trials; however, hepatotoxicity seems to be the most significant dose-

limiting factor for its application as a single agent [313]. Thus, a more appropriate

strategy is to combine 17-AAG with other agent(s) to lower 17-AAG dose and to achieve

better anticancer effect [313]. Indeed, 17-AAG is currently evaluated in clinical trials as

part of combination therapy for solid tumors and hematological malignancies [313].

Our study revealed that sulforaphane directly interacts with specific amino acid residues of Hsp90 and p50 Cdc37 , which may lead to disruption of Hsp90- p50 Cdc37 complex, thereby inhibiting Hsp90 chaperone function and inducing degradation of

Hsp90 client proteins in pancreatic cancer cells [329]. As an Hsp90 cochaperone, p50 Cdc37 acts as a critical molecular adaptor to load protein kinases to the Hsp90 chaperone machinery [95, 111]. Most of the Hsp90 clients that are associated with p50 Cdc37 are crucial components in signal transduction, cell proliferation and survival

[95]. Given that the formation of Hsp90- p50 Cdc37 complex is required for Hsp90 activity

[95, 96], we hypothesized that sulforaphane could enhance 17-AAG-induced inhibition of

Hsp90 chaperone function, thereby potentiating the therapeutic efficacy of 17-AAG against pancreatic cancer.

3.3 Materials and Methods

3.3.1 Cell culture

Human pancreatic cancer cell line, Mia Paca-2, was maintained in DMEM medium (Invitrogen, Carlsbad, CA) supplemented with 10% fetal bovine serum (Fisher

Scientific, Pittsburgh, PA) and 1% penicillin-streptomycin-glutamine (Invitrogen,

Carlsbad, CA). Human pancreatic cancer cell line, Panc-1, was maintained in RPMI1640 medium (Invitrogen, Carlsbad, CA) supplemented with 10% fetal bovine serum and 1% penicillin-streptomycin (Invitrogen, Carlsbad, CA).

3.3.2 Reagents

Sulforaphane was purchased from LKT Laboratories (St. Paul, MN), and 17-AAG was obtained from LC Laboratories (Woburn, MA). Drugs were dissolved in DMSO as a stock solution. The following antibodies were used for immunoblotting: Akt (Cell

Signaling, Beverly, MA), p23 (Abcam, Cambridge, MA), Raf-1, p53 mutant, Cdk4, p50 Cdc37 , Hsp90, β-actin (Santa Cruz Biotechnology, Santa Cruz, CA).

3.3.3 MTS cell proliferation assay

Cells were seeded in 96-well microplates at a density of 3,000 to 5,000 cells per well. Cells were treated with increasing concentrations of 17-AAG with or without 5 µM sulforaphane as indicated. After 48 h incubation cell viability was assessed by MTS assay

(Promega, Madison, WI) according to the manufacturer’s instruction. The number of living cells in the culture is directly proportional to the absorbance at 490 nm of a formazan product reduced from MTS by living cells.

3.3.4 Caspase-3 activity assay

Mia Paca-2 cells were treated with 10 µM sulforaphane, 0.1 µM 17-AAG, or combination of the two drugs, and collected after 24 h. The Caspase-3 activity assay was based on the manufacturer’s instruction of Caspase-3/CPP32 Fluorometric Assay Kit

(Biovision Research Products, Mountain View, CA). Cellular protein was extracted with the supplied lysis buffer, followed by determination of protein concentration using BCA

Protein Assay Reagents (Pierce, Rockford, IL). The cleavage of DEVD-AFC, a substrate of caspase-3, was quantified by using a fluorescence microtiter plate reader with a 400 nm excitation filter and a 505 nm emission filter.

3.3.5 Western blotting analysis

The procedure for Western blotting analysis was briefly described below. After treated with 10 µM sulforaphane, 0.1 µM 17-AAG, or combination of the two drugs for

24 h, cells were harvested by washing twice with ice-cold PBS, collected in RIPA lysis buffer (20 mM Tris-HCl, 150 mM NaCl, 1% NP-40, 5 mM EDTA, 1 mM Na 3VO 4, pH

7.5) supplemented with a protease inhibitor cocktail (Pierce, Rockford, IL), and incubated on ice for 20-30 min. Afterwards cell lysate was centrifuged at 14,000 rpm for

15 min at 4 °C, and the supernatant was recovered. Protein concentration was determined with BCA Protein Assay Reagents (Pierce, Rockford, IL). Equal amounts of protein were subject to SDS-PAGE, and transferred to PVDF membrane (BioRad, Richmond, CA).

The membrane was then incubated with appropriate primary antibodies at 4 °C overnight, followed by 2 h incubation with secondary antibodies at room temperature.

3.3.6 ATP-sepharose binding assay

The ATP-Sepharose binding assay was similar to that in a previous report [330].

Cells were treated with 15 µM sulforaphane or 5 µM 17-AAG for 24 h, and then lysed in

TNESV buffer (50 mM Tris, 2 mM EDTA, 100 nM NaCl, 1 mM Na 3VO 4, 25 mM NaF,

1% Triton X-100, pH 7.5) supplemented with protease inhibitors for 30 min. After centrifugation, supernatant was recovered and protein concentrations were determined with BCA Protein Assay Reagents. Protein (200 µg) was incubated with 25 µl pre- equilibrated γ-phosphate-linked ATP-Sepharose (Jena Bioscience GmbH, Jena,

Germany) in 200 µl incubation buffer (10 mM Tris-HCl, 50 mM KCl, 5 mM MgCl 2, 20

mM Na 2MoO 4, 0.01% NP-40, pH 7.5) overnight at 4 °C. The sepharose beads were

washed four times with incubation buffer, and the bound proteins were resolved by SDS-

PAGE and analyzed by Western blotting.

3.3.7 Hsp90 co-immunoprecipitation

The Hsp90 co-immunoprecipitation was similar to that in a previous report [330].

Cells were treated with 15 µM sulforaphane or 5 µM 17-AAG for 24 h, and then lysed in

20 mM Tris-HCl (pH 7.4), 25 mM NaCl, 2 mM DTT, 20 mM Na 2MoO 4, 0.1% NP-40,

and protease inhibitors. After centrifugation, supernatant was recovered and protein

concentrations were determined with BCA Protein Assay Reagents. Protein (500 µg) was

first incubated with H9010 antibody (Axxora, San Diego, CA) for 1 h at 4 °C, followed

by overnight incubation with protein A/G agarose (Santa Cruz Biotechnology, Santa

Cruz, CA). The bound proteins were resolved by SDS-PAGE and analyzed by Western

blotting.

3.3.8 Pancreatic tumor xenograft

The animal study protocol was approved by the University Committee on Use and

Care of Animals (UCUCA) at the University of Michigan. Four- to six-week-old athymic

(nu/nu) female mice were obtained from NCI (National Cancer Institute at Frederick).

Mia-Paca-2 cells (5×10 6-10×10 6) mixed with Matrigel (BD Biosciences, San Jose, CA)

were implanted subcutaneously into the right and left flanks of the mice. Tumor volumes

were calculated with V=1/2 (width 2×length). After tumor volumes reach 100-150 mm 3,

mice were randomized into four groups for treatment, with six animals each group. The

mice were i.p. injected with vehicle, 25 mg/kg 17-AAG (three times per week), 25 mg/kg

sulforaphane (five times per week), or combination of the two for four weeks.

Sulforaphane was dissolved in saline while 17-AAG in 10% DMSO, 70%

Cremophor/ethanol (3:1), and 20% PBS [144]. Tumor growth was monitored twice a week and normalized to the initial volumes. Mice were humanely euthanized at the end of drug treatment.

3.3.9 Statistical Analysis

All experiments were performed independently at least three times. Statistical analysis was performed using student t-test. Data are presented as mean ± SD (n ≥ 3, P <

0.01).

3.4 Results

3.4.1 Sulforaphane sensitizes pancreatic cancer cells to 17-AAG in vitro

In order to examine the anticancer effect of the combined treatment of sulforaphane and 17-AAG in pancreatic cancer cells, we incubated Mia Paca-2 and Panc-

1 cells with these drugs alone or in combination. As shown in Fig. 3.1A, sulforaphane inhibited the cell proliferation of Mia Paca-2, with an IC 50 approximately 13 µM.

Similarly, this compound suppressed Panc-1 growth with an IC 50 around 14 µM (Fig.

3.1B). Thus, the concentration of sulforaphane chosen for combination treatment was 5

µM at which below 10% of pancreatic cancer cells were eliminated. Sulforaphane significantly potentiated the anti-proliferative effect of 17-AAG in both cell lines. In Mia

Paca-2 cells, the IC 50 of 17-AAG (0.07 µM) when combined with sulforaphane (5 µM)

was more than 4-fold lower than the IC 50 of 17-AAG alone (0.31 µM). Panc-1 cell line

was resistant to 17-AAG with IC 50 of 10 µM. The resistance was attenuated in the

presence of sulforaphane; the IC 50 of 17-AAG (5 µM) when combined with sulforaphane

(5 µM) was 2-fold lower than the IC 50 of 17-AAG alone in Panc-1 cells. These data suggest that sulforaphane significantly enhanced the anti-proliferative effect of 17-AAG.

To further confirm the enhanced effect of combination of sulforaphane and 17-

AAG against pancreatic cancer cells, we measured the apoptosis by caspase-3 activity in

Mia Paca-2 cells. While 0.1 µM of 17-AAG alone and 10 µM of sulforaphane alone induced caspase-3 activity to only 2-fold and 3.4-fold, the combination of the two drugs achieved a dramatic increase of caspase-3 activity to 6.4-fold (Fig. 3.2). These results demonstrated that sulforaphane sensitized pancreatic cancer cells to 17-AAG in vitro .

3.4.2 Sulforaphane blocks Hsp90-p50 Cdc37 interaction while 17-AAG inhibits ATP binding to Hsp90

17-AAG is well known to inhibit Hsp90 activity by blocking N-terminal ATP binding pocket of Hsp90. In our previous study we characterized the impact of sulforaphane on Hsp90-cochaperone interactions and ATP binding to Hsp90. Our data showed that sulforaphane could inhibit Hsp90 through an ATP-binding independent manner in pancreatic cancer cells since sulforaphane did not interfere with ATP binding to Hsp90; instead, sulforaphane directly interacts with specific amino acid residues of

Hsp90 and p50 Cdc37 to disrupt the complex, which is distinct from other Hsp90 inhibitors

targeting ATP-binding pocket [329]. Therefore, we performed ATP-sepharose binding

assay and Hsp90 co-immunoprecipitation to further confirm that sulforaphane and 17-

AAG interfere with Hsp90 chaperone function through different mechanisms.

As shown in Fig. 3.3A, 5 µM of 17-AAG dramatically blocked ATP binding to

Hsp90, which was evidenced by the decreased amount of Hsp90 in ATP-sepharose binding assay. In contrast, 15 µM of sulforaphane did not alter the amount of Hsp90 pulled down by ATP sepharose beads.

On the other hand, sulforaphane significantly abrogated the interaction between

Hsp90 and p50 Cdc37 , whereas 17-AAG had no effect on Hsp90-p50 Cdc37 complex

formation (Fig. 3.3B). In Fig. 3.3B, immunoprecipitation (IP) of Hsp90 by its antibody

also pulled down cochaperones that were associated with Hsp90. Sulforaphane (15 µM)

largely eliminated p50 Cdc37 in the immunoprecipitated complex, while 17-AAG (5 µM) did not change the p50 Cdc37 level in the IP assay. Another cochaperone, p23, has been

demonstrated to associate with ATP-bound conformation of Hsp90 [331]. 17-AAG binds

to the ATP pocket and locks the Hsp90 molecule in the intermediate multi-chaperone

complex [330, 332], hence Hsp90 will no longer be available for p23 binding [155, 156].

As shown in Fig. 3.3B, 17-AAG decreased the interaction between Hsp90 and p23,

whereas sulforaphane did not affect Hsp90-p23 complex formation. These were in

consistent with the results of ATP-sepharose binding assay. These data suggest that

sulforaphane may enhance 17-AAG-induced inhibition of Hsp90 chaperone function

through disruption of Hsp90-p50 Cdc37 interaction.

3.4.3 Combination of sulforaphane and 17-AAG synergistically down-regulates Hsp90

client proteins in pancreatic cancer cells

In order to further investigate the combination effect of sulforaphane and 17-AAG

on Hsp90 chaperone function, we tested low concentrations of sulforaphane and 17-AAG

for their impact on the levels of Hsp90 oncogenic client proteins using Western Blotting.

Our previous study revealed that sulforaphane induced the degradation of Hsp90 client proteins in a concentration-dependent manner [329]. Here, the concentration of sulforaphane we chose was 10 µM, which showed a moderate effect on the destabilization of these proteins compared to control (29%, 43%, 14%, and 14% for Akt, p53, Raf-1, and Cdk4, respectively) (Fig. 3.4). As shown in Fig. 3.4, in comparison with

17-AAG alone, concomitant use of sulforaphane and 17-AAG further down-regulated

Akt, p53 mutant, Raf-1, and Cdk4 in Mia Paca-2 cells by 78%, 87%, 93%, and 44%, respectively. These data indicate that sulforaphane and 17-AAG synergistically increased degradation of Hsp90 client proteins that are critical to pancreatic carcinogenesis, which may contribute to the enhanced effect against pancreatic cancer cells in vitro .

3.4.4 Sulforaphane potentiates the therapeutic efficacy of 17-AAG in pancreatic cancer

xenograft model in vivo

To test the combination anticancer efficacy of sulforaphane and 17-AAG in vivo ,

we evaluated them in a pancreatic cancer xenograft model. It has been reported in the

literatures that 17-AAG (50-100 mg/kg) [333-336] and sulforaphane (50-100 mg/kg)

exhibited anticancer activity against various cancers [337, 338]. In order to examine the

combined effect, we selected relatively low doses of sulforaphane and 17-AAG that

exhibit only moderate effects when they are used alone. Two weeks after subcutaneous

implantation of Mia Paca-2 cells, we injected 25 mg/kg 17-AAG three times per week or

25 mg/kg sulforaphane five times per week for four weeks. In addition, we administered

both drugs in a group of mice with the same dose regimen. The tumor growth rates were

compared across different treatment groups. As shown in Fig. 3.5A, single treatment with sulforaphane and 17-AAG suppressed the tumor growth by approximately 45% and 50%, respectively. In contrast, combination treatment with sulforaphane and 17-AAG led to about 70% inhibition of the tumor growth (Fig. 3.5A). Meanwhile, neither sulforaphane nor 17-AAG at the administered dose regimen had apparent toxicity as determined by body weight measurement (Fig. 3.5B). These data indicate that combination of sulforaphane and 17-AAG enhanced their anticancer efficacy.

3.5 Discussion

Concomitant use of chemoprevention agents with no or low toxicity has been suggested to be a potential strategy to enhance chemotherapy effect [339-341].

Sulforaphane has shown promise in this regard with several therapeutic agents. Co- administration of sulforaphane and doxorubicin in mouse fibroblasts enhanced the efficacy of doxorubicin [341]. Shankar et al. demonstrated that sulforaphane enhanced the therapeutic potential of tumor necrosis factor-related apoptosis-inducing ligand

(TRAIL) against prostate cancer [288]. Sulforaphane has also been suggested to have the potential to overcome tumor resistance [342]. Moreover, as a chemoprevention agent, sulforaphane possesses many advantages, including high bioavailability and low toxicity

[30]. Sulforaphane from broccoli extracts is efficiently and rapidly absorbed in human small intestine, and quickly distributed throughout the body [1, 65]. Oral administration of a single dose of 200 µmol broccoli sprout isothiocyanates (mainly sulforaphane)

achieves plasma concentrations of sulforaphane equivalents at micromolar level in the systemic circulation [66]. Clinical trials have shown that oral administration of 25 µmol isothiocyanates (primarily sulforaphane) at 8-h intervals for 7 days or consumption of broccoli sprout solution containing 400 µmol glucoraphanin (precursor of sulforaphane) nightly for 2 weeks caused no significant toxicity in human subjects [13, 28].

Although 17-AAG has been evaluated in a great deal of preclinical studies and clinical trials, several drawbacks of 17-AAG including toxicity, low water solubility, instability in solution, and low oral-bioavailability prevent its use as a single agent [343,

344]. In particular, hepatotoxicity is the most significant dose-limiting factor of 17-AAG

[313]. Given that different phase I and II clinical trials have reported a significant inconsistency in the adequacy of Hsp90 inhibition by 17-AAG as a single agent [313], it is more effective to combine 17-AAG with other agent(s) to elicit an improved anticancer response while reducing the dose-limiting toxicity [313]. For example, inhibition of

PI3K/Akt pathway by LY294002 was demonstrated to play a critical role in regulating the apoptotic response of glioma cells to 17-AAG, which may provide a strategy to treat malignant gliomas [345]. Combination of rapamycin and 17-AAG abolished Akt activation and potentiated mTOR blockade in breast cancer cells [346]. Administration of

17-AAG and carboplatin in specific sequences produced a more pronounced growth inhibitory effect in human ovarian cancer models [336].

Collectively, these studies provide a strong basis for investigating the

combinatorial efficacy of sulforaphane and 17-AAG against pancreatic cancer. Herein,

utilizing in vitro pancreatic cancer cells and in vivo pancreatic cancer xenograft model,

we have shown that sulforaphane potentiates the therapeutic efficacy of 17-AAG against pancreatic cancer. This finding provides a rationale for further preclinical and clinical investigation of sulforaphane or broccoli/broccoli sprout preparations as a supplement to

17-AAG to sensitize pancreatic cancer and elicit a more pronounced clinical response.

The chaperoning cycle of Hsp90 depends on an ordered assembly and disassembly of cochaperones that preferentially bind to a specific conformation of Hsp90 to modulate its ATPase activity [96, 110]. As revealed by the crystal structure, p50 Cdc37 facilitates the client loading to Hsp90 by inserting its C-terminal side chain into the nucleotide binding pocket of Hsp90; binding of ATP to Hsp90 would have to eject this side chain of p50 Cdc37 out of ATP pocket [96]. Our previous study has demonstrated that p50 Cdc37 can only bind ADP-bound, open conformation of Hsp90 but not ATP-bound

Hsp90 [156]. Hence, it is possible that p50 Cdc37 is removed from the early Hsp90 complex

after the binding of ATP to Hsp90 [156]. On the other hand, it was demonstrated that p23

is recruited to the ATP-bound, closed conformation of Hsp90 and locks it in an ATP-

dependent conformational state that has high affinity for client proteins [347]. The

unaltered level of Hsp90-p23 complex indicates that either the chaperone cycle still

proceeds to later stages without client folding or some client proteins are loaded onto

Hsp90 without the help of p50 Cdc37 but require p23 for maturation. The latter hypothesis

may hold true because the majority of p50 Cdc37 -associated clients are kinases [111]. Given that 17-AAG inhibits binding of ATP to Hsp90, it is reasonable to expect sulforaphane and 17-AAG to collaborate to achieve a more pronounced inhibition of Hsp90 chaperone function.

Our previous studies revealed that sulforaphane could interfere with Hsp90 activity through direct binding to Hsp90/p50 Cdc37 and blocking Hsp90-p50 Cdc37

association, resulting in degradation of Hsp90 client proteins [329]. The present study

extends these observations by demonstrating that sulforaphane-induced Hsp90-p50 Cdc37 disruption improves 17-AAG-induced inhibition of Hsp90 function, leading to synergistic depletion of Hsp90 client proteins. In addition, sulforaphane modulates a plethora of cellular activities of cancer cells by regulating a variety of molecules such as Bcl-2 family proteins, caspases, p21, histone deacetylase, VEGF, HIF-1α, c-Myc, NF-κB,

Stat3, Nrf2 [1, 29, 31, 41, 194]. All of them may contribute to the potentiated therapeutic efficacy of 17-AAG against pancreatic cancer. A very recent study found that sulforaphane induced hyperacetylation of Hsp90 by inhibiting HDAC6 enzyme activity in prostate cancer cells, thereby leading to destabilization of androgen receptor (AR), a client protein of Hsp90 [289]. Although there has been no evidence for this activity of sulforaphane in pancreatic cancer cells, hyperacetylation of Hsp90 could be another potential mechanism for the improved abrogation of Hsp90 function.

3.6 Conclusions

In conclusion, we have demonstrated that co-administration of 17-AAG with sulforaphane can enhance 17-AAG-induced inhibition of Hsp90 chaperone function, thereby potentiating anticancer efficacy in pancreatic cancer xenograft model. Our study identified the disruption of Hsp90-p50 Cdc37 interaction by sulforaphane as a potential

mechanism for the combinatorial effect against pancreatic cancer. These findings provide a rationale for further preclinical and clinical evaluation of sulforaphane or broccoli/broccoli sprout preparations combined with 17-AAG for better efficacy and lower dose-limiting toxicity of 17-AAG in pancreatic cancer.

A 120 Mia Paca-2

100

60 40 SF * 17 -AAG Cell viability viability (%) Cell 20 5 µM SF +17-AAG 0 0.000010 0.0001 0.001 0.01 0.1 1 10 100 Concentration of 17-AAG (µM)

B 140 Panc-1 120

100

60 SF 40 17 -AAG * Cell viability viability (%) Cell 20 5 µM SF + 17-AAG

0 0.000010 0.0001 0.001 0.01 0.1 1 10 100 Concentration of 17-AAG (µM)

Figure 3.1. Simultaneous treatment with sulforaphane and 17-AAG enhances anti- proliferative effect in pancreatic cancer cells

8 * 7

1 Fold increase in caspase-3inactivity increase Fold 0 Ctr 0.1 µM AAG 10 µM SF 10 µM SF + 0.1 µM AAG

Figure 3.2. Combination of sulforaphane and 17-AAG induces a more pronounced activation of caspase-3 in pancreatic cancer cells

A ATP Pull-down 15 µM SF - - + 5 µM 17-AAG - + - Hsp90

B IP: Hsp90 Supernatant 15 µM SF - - + - - + 5 µM 17-AAG - + - - + - p50 Cdc37

p23

Hsp90

Figure 3.3. Effect of sulforaphane and 17-AAG on ATP binding to Hsp90 and Hsp90- cochaperone association in pancreatic cancer cells

10 µM SF - + - + 0.1 µM AAG - - + + Akt p53 mutant

Raf-1

Cdk4

Hsp90

β-actin

120 Akt p53 100 Raf -1 80 Cdk4

60 * 40 * 20 *

Signal Strength (% Control) Strength Signal * 0 Ctrl SF AAG SF + AAG

Figure 3.4. Co-administration of sulforaphane and 17-AAG synergistically down- regulates Hsp90 client proteins in pancreatic cancer cells

A 700

) Control 3 600 25 mg/kg 17-AAG 500 25 mg/kg SF 400 17-AAG+SF 300 200 * 100 Tumor Volume (mm Volume Tumor 0 10 20 30 40 50 Days

B 30 Control 25 mg/kg 17-AAG )

g 25 mg/kg SF 17-AAG+SF 25

20 Body Weight ( Weight Body

15 10 20 30 40 50 Days

Figure 3.5. Sulforaphane and 17-AAG combination exhibits a potentiated anticancer activity in pancreatic cancer xenograft model

Chapter 4: Sulforaphane Targets Breast Cancer Stem Cells by Down-regulating

Wnt/ β-catenin Self-renewal Pathway

4.1 Abstract

The existence of cancer stem cells (CSCs) in breast cancer has profound implications for cancer prevention. In this study, we evaluated sulforaphane, a natural compound derived from broccoli/broccoli sprouts, for its efficacy to inhibit breast CSCs and its potential mechanism. Aldefluor assay and mammosphere formation assay were used to evaluate the effect of sulforaphane on breast CSCs in vitro . A NOD/SCID xenograft model was employed to determine whether sulforaphane could target breast

CSCs in vivo , as assessed by Aldefluor assay and tumor growth upon cell re-implantation in secondary mice. The potential mechanism was investigated utilizing Western blotting analysis and β-catenin reporter assay. Sulforaphane (1~5 µM) decreased aldehyde

dehydrogenase (ALDH)-positive cell population by 65%~80% in human breast cancer

cells ( P < 0.01), and reduced the size and number of primary mammospheres by 8~125- fold and 45%~75% ( P < 0.01), respectively. Daily injection with 50 mg/kg sulforaphane for two weeks reduced ALDH-positive cells by more than 50% in NOD/SCID xenograft tumors ( P = 0.003). Sulforaphane eliminated breast CSCs in vivo , thereby abrogating

tumor growth after re-implantation of primary tumor cells into the secondary mice ( P <

0.01). Western blotting analysis and β-catenin reporter assay showed that sulforaphane down-regulated Wnt/ β-catenin self-renewal pathway. In conclusion, sulforaphane inhibits

breast CSCs and down-regulates Wnt/ β-catenin self-renewal pathway. These findings

support the use of sulforaphane for chemoprevention of breast cancer stem cells and

warrant further clinical evaluation.

4.2 Introduction

Broccoli and broccoli sprouts contain large amounts of glucosinolates [2].

Numerous studies have substantiated the chemoprevention effect of increasing cruciferous vegetable intake against cancer, which has been attributed to the activity of various isothiocyanates that are enzymatically hydrolyzed from glucosinolates [1].

Sulforaphane was found to be converted from glucoraphanin, a major glucosinolate in broccoli/broccoli sprouts [12]. The chemoprevention properties of sulforaphane against cancer are through both “blocking” and “suppressing” effects [1]. The “blocking” function of sulforaphane is achieved through inhibiting Phase 1 metabolism enzymes that convert procarcinogens to carcinogens and inducing Phase 2 metabolism enzymes that promote excretion of carcinogens [1]. Subsequent studies revealed the “suppressing” effects of sulforaphane in modulating diverse cellular activities to inhibit the growth of transformed cells [1, 30]. The ability of sulforaphane to induce apoptosis and cell cycle arrest is associated with regulation of many molecules including Bcl-2 family proteins,

caspases, p21, cyclins and cdks [30]. Sulforaphane was also shown to suppress angiogenesis and metastasis by down-regulating VEGF, HIF-1, MMP-2 and MMP-9

[30].

Accumulating evidence has shown that many types of cancer, including breast cancer, are initiated from and maintained by a small population of cancer stem cells

(CSCs) [181, 182]. This minor population produces the tumor bulk through continuous self-renewal and differentiation, which may be regulated by similar signaling pathways occurring in normal stem cells [181, 182, 185, 203]. Several pathways including Wnt/ β-

catenin, Hedgehog, and Notch have been identified to be critical to the self-renewal

behavior of CSCs [203-205]. Furthermore, CSCs have been suggested to contribute to

tumor resistance/relapse because chemotherapy and radiation therapy are incapable of

eradicating them [182, 186, 187]. Thus, targeting these self-renewal pathways may

provide an effective strategy to target CSCs and thereby overcome tumor resistance and

reduce relapse [181]. Several dietary compounds, such as curcumin [264, 267], quercetin

and epigallocatechin-gallate [348], were found to be potentially against CSC self-

renewal.

Wnt/ β-catenin signaling is one of the key pathways that promote self-renewal of breast CSCs [181]. Activation of Wnt target genes are mediated by β-catenin, which translocates into nucleus and binds to the transcription factors T cell factor/lymphoid enhancer factor (TCF/LEF) [181, 223]. The level of intracellular β-catenin is modulated

by a multi-protein complex consisting of glycogen synthase kinase3 β (GSK3 β),

adenomatous polyposis coli, casein kinase1 α and axin [228]. GSK3 β promotes the

ubiquitin-proteasome degradation of β-catenin by phosphorylating three specific amino acids, Ser33/Ser37/Thr41, on β-catenin [228].

Sulforaphane was shown to target pancreatic tumor-initiating cells in a very recent report [285]. In the present study, we examined the efficacy of sulforaphane against breast CSCs in both breast cancer cell lines and breast cancer xenografts. We demonstrated that sulforaphane eliminated breast CSCs in vivo , which was reflected by the inhibition of tumor growth in recipient mice that were inoculated with tumor cells derived from sulforaphane-treated primary xenografts. Furthermore, since sulforaphane was reported to induce down-regulation of β-catenin in human cervical carcinoma HeLa

and hepatocarcinoma HepG2 cells [41], we investigated the suppressing activity of

sulforaphane on Wnt/β-catenin pathway.

4.3 Materials and Methods

4.3.1 Cell culture

Human breast cancer cell lines, MCF7 and SUM159, were obtained from

American Type Culture Collection and from Dr. Stephen Ethier (Karmanos Cancer

Center), respectively. The source of SUM159 cell line is primary breast anaplastic carcinoma. This cell line is ER negative, PR negative, and does not have Her2 overexpression. Both cell lines were tested and authenticated in their origin sources.

Authentication of these cell lines included morphology analysis, growth curve analysis, isoenzyme analysis, short tandem repeat analysis, and mycoplasma detection. Both cell

lines were passaged in our laboratory for fewer than six months after receipt. To maintain the integrity of collections, stocks of the earliest-passage cells have been stored, and cell lines have been carefully maintained in culture as described below. MCF7 was maintained in RPMI1640 medium (Invitrogen, Carlsbad, CA) supplemented with 10% fetal bovine serum (Fisher Scientific, Pittsburgh, PA), 1% antibiotic-antimycotic

(Invitrogen, Carlsbad, CA), and 5 µg/ml insulin (Sigma-Aldrich, St Louis, MO).

SUM159 was maintained in Ham’s F12 medium (Invitrogen, Carlsbad, CA) supplemented with 5% fetal bovine serum, 1% antibiotic-antimycotic, 5 µg/ml insulin,

1µg/ml hydrocortisone (Sigma-Aldrich, St Louis, MO), and 4 µg/ml gentamicin

(Invitrogen, Carlsbad, CA).

4.3.2 Reagents

Sulforaphane was obtained from LKT Laboratories (St. Paul, MN). Propidium iodide (PI) was from Invitrogen (Carlsbad, CA). LiCl was purchased from Fisher

Scientific (Pittsburgh, PA); BIO (GSK3 inhibitor IX) was from Calbiochem (EMD

Biosciences, San Diego, CA); and MG132 was obtained from Assay Designs (Stressgen,

Ann Arbor, MI).

Antibodies to β-catenin, phospho-β-catenin Ser33/Ser37/Thr41, phospho-GSK3 β

Ser9, and GSK3 β were purchased from Cell Signaling Technology (Danvers, MA).

Antibodies to cyclin D1 and β-actin were acquired from Santa Cruz Biotechnology (Santa

Cruz, CA).

4.3.3 MTS cell proliferation assay

MCF7 and SUM159 were seeded in 96-well microplates at a density of

3,000~5,000 cells per well. Cells were treated with increasing concentrations of sulforaphane as indicated. After 48 h incubation cell viability was assessed by MTS assay

(Promega, Madison, WI) according to manufacturer’s instruction. The number of living cells is directly proportional to the absorbance at 490 nm of a formazan product reduced from MTS by living cells.

4.3.4 Caspase-3 activity assay

Cells were treated with different concentrations of sulforaphane and collected after 24 h. The caspase-3 activity assay was based on the manufacturer’s instruction of

Caspase-3/CPP32 Fluorometric Assay Kit (Biovision Research Products, Mountain

View, CA). Cellular protein was extracted with the supplied lysis buffer, followed by determination of protein concentration using BCA Protein Assay Reagents (Pierce,

Rockford, IL). The cleavage of DEVD-AFC, a substrate of caspase-3, was quantified by using a fluorescence microtiter plate reader with a 400 nm excitation filter and a 505 nm emission filter.

4.3.5 Mammosphere formation assay

Stem/progenitor cells are enriched in mammospheres of breast cancer cells [349],

based on the unique ability of stem/progenitor cells to grow and form spheres in serum-

free medium [198]. Mammosphere culture was performed as previously described [193,

197] in a serum-free mammary epithelium basal medium (Lonza Inc., Walkersville, MD)

supplemented with B27 (Invitrogen), 1% antibiotic-antimycotic, 5 µg/ml insulin, 1µg/ml

hydrocortisone, 4 µg/ml gentamicin, 20 ng/mL EGF (Sigma-Aldrich, St Louis, MO), 20 ng/mL bFGF (Sigma-Aldrich, St Louis, MO), and 1:25,000,000 β-mercaptoethanol

(Sigma-Aldrich, St Louis, MO). Single cells prepared from mechanical and enzymatic

dissociation were plated in six-well ultra-low attachment plates (Corning, Acton, MA) at

a density of 500~1,000 cells per milliliter in primary culture and 100~500 cells per

milliliter in the following passages. Different concentrations of sulforaphane were added

to primary culture, while the second and third passages were grown in the absence of

drug. After 7 days of culture, the number of mammospheres was counted under Nikon

Eclipse TE2000-S microscope and the photos were acquired with MetaMorph 7.6.0.0.

4.3.6 Aldefluor assay

A cell population with a high Aldehyde dehydrogenase (ALDH) enzyme activity

was previously reported to enrich mammary stem/progenitor cells [193]. Aldefluor assay

was performed according to manufacturer’s guidelines (StemCell Technologies,

Vancouver, BC, Canada). Single cells obtained from cell cultures or xenograft tumors

were incubated in Aldefluor assay buffer containing an ALDH substrate, bodipy-

aminoacetaldehyde (BAAA, 1 µmol/L per 1,000,000 cells), for 40~50 min at 37 °C. As a

negative control, a fraction of cells from each sample was incubated under identical

condition in the presence of ALDH inhibitor, diethylaminobenzaldehyde (DEAB). Flow

cytometry was used to measure ALDH-positive cell population.

4.3.7 Primary NOD/SCID mouse model

All experimentation involving mice were conducted in accordance with standard

protocol approved by the University Committee on the Use and Care of Animals

(UCUCA) at University of Michigan. SUM159 cells (2,000,000) mixed with Matrigel

(BD Biosciences, San Jose, CA) were injected to the mammary fat pads of 5-week-old female NOD/SCID mice (The Jackson Laboratory, Bar Harbor, ME) as previously described [350]. Tumors were measured with a caliper, and the volume was calculated using V=1/2 (width 2 × length). Two weeks after cell injection, the mice were randomly separated into two groups, one group intraperitoneally injected with control (0.9% NaCl solution) and the other group with 50 mg/kg sulforaphane (dissolved in 0.9% NaCl solution) daily for two weeks.

4.3.8 Dissociation of tumors

At the end of drug treatment, the mice were humanely euthanized and tumors were harvested. Tumor tissues were dissociated mechanically and enzymatically to obtain single cell suspension as previously described [175]. Briefly, tumors were minced by scalpel and incubated in medium 199 (Invitrogen, Carlsbad, CA) mixed with collagenase/hyaluronidase (StemCell Technologies, Vancouver, BC, Canada) at 37 °C for 15~20 min. The tissues were further dissociated by pipet trituration, and then passed through 40 µm nylon mesh to produce single cell suspension, which was used for

Aldeflour assay and flow cytometry.

4.3.9 Secondary NOD/SCID mouse model

Living cells from the dissociated tumors were sorted out by Fluorescence- activated cell sorting (FACS). Two groups of mice (four in group 1 and three in group 2) were implanted with tumor cells separately. Each secondary NOD/SCID mouse was inoculated with 50,000 cells from control mouse tumors in one side of inguinal mammary

fat pad and another 50,000 cells from sulforaphane-treated tumors in the contralateral mammary fat pad. The growth of tumors was monitored; and tumor volumes were measured twice a week. Mice were humanely euthanized when the larger one of the two tumors reached 300~500 mm 3.

4.3.10 Western blotting analysis

Cells were treated with sulforaphane at varying concentrations for indicated time

periods in Figure legends. Cells were harvested, lysed in RIPA buffer (20 mM Tris-HCl,

150 mM NaCl, 1% NP-40, 5 mM EDTA, 1 mM Na 3VO 4, pH 7.5) supplemented with a

protease inhibitor cocktail (Pierce, Rockford, IL) and a phosphatase inhibitor

(Calbiochem, EMD Biosciences, San Diego, CA), and incubated on ice for 30 min. Cell

lysate was centrifuged at 14,000 rpm for 15 min, and the supernatant was recovered.

Protein concentration was determined with BCA Protein Assay Reagents (Pierce,

Rockford, IL). Equal amounts of protein were subject to SDS-PAGE, and transferred to

PVDF membrane (BioRad, Richmond, CA). The membrane was then incubated with

appropriate antibodies.

4.3.11 TOP-dGFP lentiviral β-catenin reporter assay

TCF/LEF-1 (TOP-dGFP, FOP-dGFP) lentiviral reporter system was kindly gifted by Dr. Wiessman at Ludwig Center, Stanford University School of Medicine [351]. Cells were infected with TOP-dGFP or control reporter FOP-dGFP with mutated TCF/LEF-1 binding sites. TOP-dGFP MCF7 and FOP-dGFP MCF7 cells were maintained in the same RPMI1640 medium as MCF7 cells. MCF7, TOP-dGFP MCF7 and FOP-dGFP

MCF7 cells were cultured in the same serum-free mammary epithelium basal medium as

mammospheres in six-well ultra-low attachment plates at a density of 1,000~1,500 cells per milliliter for 5 days. Single cells prepared from the primary spheres were incubated in medium containing 5 µM sulforaphane or/and 0.5 µM BIO for 48 h. After dissociation, single cell suspension was subject to flow cytometry analysis for dGFP-positive cell population. Parental MCF7 cells served as a control for auto-fluorescence. The photos of mammospheres were taken with Nikon Eclipse TE2000-S microscope and acquired with

MetaMorph 7.6.0.0.

4.3.12 Statistical analysis

Statistical differences were determined using two-tailed Student t-test. Data are

presented as mean ± SD (n ≥ 3).

4.4 Results

4.4.1 Sulforaphane inhibits proliferation and induces apoptosis of breast cancer cells

Sulforaphane was previously shown to inhibit proliferation [352] and induce apoptosis [353] in breast cancer cells. We first evaluated the anti-proliferative effects of sulforaphane in two human breast cancer cell lines, SUM159 and MCF7, by MTS assay.

Cells were treated with increasing concentrations of sulforaphane for 48 h; and the ratio of viable cells of treatment relative to control is plotted in Fig. 4.1A. Cell survival decreased as the concentration of sulforaphane increased, with the IC 50 around 10 µM for

SUM159 and 16 µM for MCF7. Caspase-3 fluorometric assay showed that sulforaphane

(10 µM) significantly ( P = 0.005) induced activation of caspase-3 (Fig. 4.1B).

4.4.2 Sulforaphane inhibits breast cancer stem/progenitor cells in vitro

It has been demonstrated that mammary stem/progenitor cells are enriched in non- adherent spherical clusters of cells, termed mammospheres [197]. These cells are capable of yielding secondary spheres and differentiating along multiple lineages [197]. In order to evaluate whether sulforaphane could suppress the formation of mammospheres in vitro , we exposed primary MCF7 and SUM159 spheres to varying concentrations of

sulforaphane and then cultured them two additional passages in the absence of drug. As

shown in Fig. 4.2A and 4.2B, sulforaphane inhibited the formation of primary spheres.

Not only the number of spheres declined by 45%~75% ( P < 0.01) (Fig. 4.2A), but also the size of spheres was reduced by 8~125-fold (Fig. 4.2B). Furthermore, a significant decrease in the number of sphere-forming cells in subsequent passages indicated a reduced self-renewal capacity of these stem/progenitor cells (Fig. 4.2C) [197]. MCF7 cells initially propagated in the presence of 5 µM sulforaphane barely produced secondary spheres, with no cells passaged to third generation (Fig. 4.2C). It is worth noting that the concentrations of sulforaphane that were capable of suppressing mammosphere formation (IC 50 around 0.5~1 µM for both SUM159 and MCF7 spheres)

were approximately 10-fold lower than those exhibiting anti-proliferative effects in MTS

assay (IC 50 around 10 µM for SUM159 and 16 µM for MCF7).

In breast carcinomas, a cell population with high aldehyde dehydrogenase

(ALDH) activity as assessed by the Aldefluor assay has been demonstrated to enrich tumorigenic stem/progenitor cells [193]. This cell population is capable of self-renewal and generating tumors resembling the parental tumor [193]. Since SUM159 has a

relatively high percentage of ALDH-positive cells, we selected SUM159 to examine whether sulforaphane inhibits the tumor-initiating ALDH-positive cells in vitro . As

shown in Fig. 4.3A, 1 µM sulforaphane significantly decreased the ALDH-positive

population of SUM159 cells by over 65% ( P = 0.008), while 5 µM produced greater than

an 80% reduction of ALDH-positive population ( P < 0.008). Representative flow

cytometry dot plots are presented in Fig. 4.3B. These data showed that sulforaphane

inhibited the ALDH-positive cells at similar concentrations to those inhibited

mammosphere formation and at 10-fold lower concentrations than those inhibited cancer

cells as determined by MTS assay.

Therefore, these findings demonstrate sulforaphane in reducing the breast cancer

stem/progenitor cell population in vitro . An interesting observation is that sulforaphane

was able to inhibit stem/progenitor cells at the concentrations (0.5~ 5 µM) that hardly

affected the bulk population of cancer cells, implying that sulforaphane is likely to

preferentially target stem/progenitor cells compared to the differentiated cancer cells.

4.4.3 Sulforaphane eliminates breast cancer stem cells in vivo

In order to determine whether sulforaphane could target breast CSCs in vivo , we

utilized a xenograft model of SUM159 cells in NOD/SCID mice. Two weeks after cell

inoculation, animals were daily injected with 50 mg/kg sulforaphane. After two weeks of

treatment, tumors in sulforaphane-treated mice were 50% of the size of 0.9% NaCl

solution control animals (P = 0.018) (Fig. 4.4A), while sulforaphane had no apparent

toxicity as determined by body weight (Fig. 4.4B). Tumors were isolated from animals

and the tumor cells were analyzed by Aldefluor assay. As shown in Fig. 4.4C and 4.4D,

sulforaphane reduced ALDH-positive population by more than 50% compared to that from control mice ( P = 0.003).

Although the decreased ALDH-positive cell population in sulforaphane-treated tumors suggests that sulforaphane may target breast cancer stem/progenitor cells, the ability of residual cancer cells to initiate tumors upon re-implantation in secondary mice is a more definitive assay [182]. Therefore, we examined the growth of secondary tumors in NOD/SCID mice inoculated with primary tumor cells obtained from primary xenografts. In order to avoid potential variations due to mouse heterogeneity, each recipient mouse was injected with 50,000 cells obtained from sulforaphane-treated tumors in one side of inguinal mammary fat pad and another 50,000 cells obtained from control tumors in the contralateral fat pad. The results showed that cancer cells from control animals exhibited rapid tumor re-growth, reaching a final tumor size ranging from

300 to 500 mm 3 in secondary NOD/SCID mice. However, the cancer cells obtained from sulforaphane-treated mice largely failed to produce any tumors in recipient mice up to 33 days after implantation (Fig. 4.5A). Fig. 4.5A & 4.5B showed that tumor cells derived from sulforaphane-treated mice only gave rise to one small tumor (6 mm 3) out of 7 inoculations at day 19, while control tumor cells yielded tumors as early as day 7 ( P <

0.01). All control inoculations produced tumors by day 15 (Fig. 4.5B). These results suggest that sulforaphane was able to eliminate breast CSCs in primary xenografts, thereby abrogating the re-growth of tumors in secondary mice. Taken together with the in vivo Aldefluor assay results, these findings suggest that sulforaphane targets breast CSCs with high potency.

4.4.4 Sulforaphane down-regulates Wnt/ β-catenin pathway in breast cancer cells

Next, we investigated the mechanisms that may contribute to the effects of sulforaphane on breast CSCs. Wnt/ β-catenin pathway is known to be an important

regulator of stem cell self-renewal [185]. Since sulforaphane was reported to down-

regulate β-catenin in human cervical carcinoma and hepatocarcinoma cell lines [41], we

examined whether β-catenin and Wnt/ β-catenin downstream targets are down-regulated by sulforaphane in human breast cancer cells. As shown in Fig. 4.6A, sulforaphane decreased the protein level of β-catenin by up to 85% in MCF7 and SUM159 cells; and the expression of cyclin D1, one of the Wnt/ β-catenin target genes, declined by up to

77% as well. To further confirm that the down-regulation of β-catenin protein level decreased its transcriptional activity, we utilized a TCF/LEF TOP-dGFP lentiviral reporter system. The β-catenin activates TCF/LEF in nucleus, driving the transcription of destabilized GFP (dGFP) gene; and the dGFP expression was analyzed by fluorescence microscopy and quantified by flow cytometry. As determined by flow cytometry, approximately 3% of transfected cells are dGFP-positive, and 5 µM sulforaphane reduced this population by 30% ~40% ( P = 0.002) (Fig. 4.6B).

The intracellular level of β-catenin is regulated by its phosphoryaltion status and subsequent proteasomal degradation. When β-catenin is phosphorylated at

Ser33/Ser37/Thr41 by GSK3 β, it is immediately subject to ubiquitin-proteasome degradation [228]. Phospharylation of GSK3 β at Ser9 may decrease the activity of

GSK3 β, thereby stabilizing β-catenin [229, 231]. Thus, we used a proteasome inhibitor,

MG132, to block proteasome function and observed an accumulation of phospho-β-

catenin (Ser33/Ser37/Thr41) in response to sulforaphane (Fig. 4.6C). The sulforaphane- induced β-catenin phosphorylation was reversed when LiCl, a GSK3 β inhibitor, was present (Fig. 4. 6C) [354]. As shown in Fig. 4.6B, 0.5 µM BIO, another specific GSK3 β inhibitor [354, 355], enhanced the dGFP-positive cell population by more than 5-fold (P

< 0.0001), and sulforaphane (5 µM) decreased this population by over 60% in the presence of BIO (P < 0.0001). Furthermore, our result demonstrated a decreased level of phospho-GSK3 β (Ser9) by up to 74% in cells with increasing concentrations of sulforaphane (Fig. 4.6C). LiCl was demonstrated to inactivate GSK3 β through Ser9

phosphorylation, which in turn reduce phosphorylation of β-catenin at Ser33/Ser37/Thr41

and its degradation [354, 355]. As shown in the bottom panel of Fig. 4.6C, sulforaphane

was able to attenuate LiCl-induced GSK3 β phosphorylation and β-catenin accumulation.

Taken together, these data suggest that the down-regulation of Wnt/ β-catenin self-

renewal pathway might contribute to the inhibitory effects of sulforaphane on breast

CSCs. This warrants further studies to establish the conclusive role of this down-

regulation in inhibition of breast CSCs by sulforaphane.

4.5 Discussion

The anticancer efficacy of sulforaphane, a natural compound derived from

broccoli/broccoli sprouts, has been evaluated in various cancers. For instance, oral or

intraperitoneal administration of sulforaphane inhibited the tumor growth in prostate PC-

3 and pancreatic Panc-1 xenografts [21, 23]. The risk of premenopausal breast cancer was

shown to be inversely associated with broccoli consumption [26]. The orally administered sulforaphane reached mammary gland and increased the detoxification enzyme activity [27]. Additionally, it has been suggested that sulforaphane may have the potential to act against tumor resistance and relapse/recurrence [342]. A very recent study demonstrated the effectiveness of sulforaphane in abrogating pancreatic tumor resistance to TRAIL by interfering with NF-κB induced anti-apoptotic signaling [285]. Another

study indicated that sulforaphane could overcome doxorubicin resistance and restore

apoptosis induction in cells [341]. These findings provide a strong rationale for

investigating the chemoprevention property of sulforaphane or broccoli/broccoli sprouts

in clinical trials.

Increasing evidence supports the cancer stem cell theory, which states that a

variety of cancers are driven and sustained by a small proportion of CSCs [185]. The

concept of CSCs has profound clinical implications for cancer therapeutics and

prevention [185, 240]. Recent studies indicate that CSCs have the capacity to drive tumor

resistance and relapse/recurrence [184, 356]. Lack of efficacy of current chemotherapies

in advance and metastatic disease requires novel approaches to specifically target CSC

population [185, 188, 189]. Thus, therapies that are directed against both differentiated

cancer cells and CSCs may provide advantages to treat these diseases. Researchers have

found that several dietary compounds are promising chemoprevention agents against

CSCs, such as curcumin [264, 267]. Therefore, based on the chemopreventive activity of

sulforaphane and the implications of CSC theory, we have utilized both in vitro and in vivo systems to determine whether sulforaphane acts against breast CSCs.

Several techniques have been developed to isolate and characterize breast CSCs in vitro . Mammosphere culture was first used to isolate and expand mammary stem/progenitor cells by Dontu et al. [197], based on the ability of stem/progenitor cells to grow in serum-free suspension, while differentiated cells fail to survive under the same condition [198]. By employing this technique, we have demonstrated that sulforaphane

(0.5~5 µM) significantly suppressed the mammospheres formation of both SUM159 and

MCF7 cells (Fig. 4.2). Another technique is to utilize cell makers, e.g., CD44 +CD24 -

/low lin - and ALDH-positive [175, 193, 198], to distinguish mammary stem/progenitor cells

from differentiated cancer cells. It has been reported that as few as 500 ALDH-positive

cells were able to generate a breast tumor within 40 days, while 50,000 ALDH-negative

cells failed to form tumor [193]. ALDH-positive and CD44 +CD24 -/low lin - were identified

a small overlap that has the highest tumorigenic capacity, generating tumors from as few

as 20 cells [193]. In contrast, ALDH-positive cells without the CD44 +CD24 -/low lin - marker were able to produce tumors from 1,500 cells, whereas 50,000 CD44 +CD24 -

/low lin -ALDH-negative cells did not [193]. Thus, we utilized Aldefluor assay to evaluate

the ability of sulforaphane to target breast cancer stem/progenitor cells. We have

demonstrated that sulforaphane (1~5 µM) could inhibit the tumor-initiating ALDH-

positive cells in vitro by 65% to 80% (Fig. 4.3). Of special note, concentrations of

sulforaphane which inhibit stem/progenitor cells in both mammosphere formation assay

and Aldefluor assay had only minimal effects on the bulk population of breast cancer cell

lines, which implies the preferential targeting of stem/progenitor cells by sulforaphane.

100

The injection of human breast cancer cells into the mammary fat pad of immune- deficient NOD/SCID mice provides a reliable and sensitive in vivo system for studying human breast cancer [175, 202]. We demonstrated that sulforaphane was able to target breast CSCs in vivo by using this xenograft model. Daily injection of sulforaphane for two weeks suppressed tumor growth in primary NOD/SCID mice and reduced ALDH- positive cell population of the tumors by ~50% (Fig. 4.4). More importantly, we found that the tumor cells derived from sulforaphane-treated mice were not able to form secondary tumors in recipient mice up to 33 days (Fig. 4.5). There are two possible reasons that may explain the difference between the 50% reduction of ALDH-positive population and the failure of tumor growth in secondary mice. One is that although

ALDH-positive cells are enriched with stem/progenitor cells, not all ALDH-positive cells have tumor-initiating capacity. Another possible reason is the experimental setting we used for the primary NOD/SCID mice. We inoculated 2,000,000 SUM159 cells into the primary NOD/SCID mice, and treated them with the drug after two weeks of cell inoculation, both of which could lead to an under-estimation of the effect of sulforaphane on ALDH-positive cell population. However, the ability of CSCs to self-renew and differentiate as determined by re-implantation of primary tumor cells in secondary animals is a more definitive functional assay [182]. These are consistent with the in vitro observation that sulforaphane preferentially targeted cancer stem/progenitor cells instead of bulk cell population. The preference of sulforaphane in killing CSCs may be significant for chemoprevention.

101

The well-known curcumin was shown to interfere with self-renewal pathways,

Wnt and Notch, in colon and pancreatic cancer cells respectively [264, 267]. Apple-

derived quercetin and green tea epigallocatechin-gallate were reported to regulate key

elements of Wnt and Notch pathways in human colon cancer cells [348]. Park et al. previously reported that β-catenin was down-regulated in HeLa and HepG2 cells [41]. In consistent with this study, we demonstrated that sulforaphane was able to down-regulate

Wnt/ β-catenin self-renewal pathway in breast cancer cells, and sulforaphane-induced β- catenin phosphorylation (Ser33/Ser37/Thr41) and proteasome degradation was possibly through activation of GSK3 β (Fig. 4.6). Myzak et al. reported that sulforaphane increased

β-catenin activity without altering its protein level in HDAC1-transfected HEK293 cells

[60]. The differences among the studies could arise from distinct cell lines and treatment

conditions.

As a chemoprevention agent, sulforaphane possesses many advantages, such as

high bioavailability and low toxicity [30]. Sulforaphane from broccoli extracts is

efficiently and rapidly absorbed in human small intestine, and distributed throughout the

body [1, 65]. Plasma concentrations of sulforaphane equivalents peaked 0.94~2.27 µM in

humans 1 h after a single dose of 200 µmol broccoli sprout isothiocyanates (mainly

sulforaphane) [66]. A recent pilot study detected an accumulation of sulforaphane in

human breast tissue, with 1.45 ± 1.12 pmol/mg for the right breast and 2.00 ± 1.95

pmol/mg for the left, in eight women who consumed broccoli sprout preparation

containing 200 µmol sulforaphane about 1 h before the surgery [27]. These

concentrations of sulforaphane are expected to be effective against breast CSCs, based on

102

our in vitro results. Although sulforaphane itself has not been evaluated in humans, broccoli sprouts were tested for toxicity in clinical trials [30]. A Phase I trial showed that broccoli sprouts caused no significant toxicity when administered orally at 8-h intervals for 7 days as 25 µmol isothiocyanates (mainly sulforaphane) [13]. In another study, it was well tolerated in 200 adults who consumed broccoli sprout solution containing 400

µmol glucoraphanin (precursor of sulforaphane) nightly for 2 weeks [28]. Additionally, sulforaphane at concentrations below 10 µM did not show significant effect on cell cycle arrest and apoptosis induction of human non-transformed T-lymphocytes [357].

4.6 Conclusions

In conclusion, we have demonstrated that sulforaphane was able to target breast

CSCs as determined by the mammosphere formation assay, Aldefluor assay, and tumor growth upon re-implantation in secondary mice. Furthermore, our study identified the down-regulation of Wnt/ β-catenin self-renewal pathway by sulforaphane as one of the possible mechanisms for its efficacy. These studies support the use of sulforaphane for breast cancer chemoprevention. These findings provide a strong rationale for preclinical and clinical evaluation of sulforaphane or broccoli/broccoli sprouts for breast cancer therapies.

103

A 120 100 80 60

40 SUM159 20 MCF7 Cell Viability (%) ViabilityCell 0 0.0010 0.1 10 Concentration of SF (µM)

B 2.5 P=0.005 2

1.5

1 Fold increase inincrease Fold

caspase-3 activity caspase-3 0.5

0 0 5 10 Concentration of SF (µM)

Figure 4.1. Sulforaphane inhibits proliferation and induces apoptosis in breast cancer cells

104

A Primary Mammospheres 120.0 MCF7 100.0 SUM159 80.0 ** 60.0 ** ** ** 40.0 ** **

Sphere formation Sphere 20.0

normalized to control (%) control to normalized 0.0 0 0.5 1 5 Concentration of SF (µM) B MammosphereSize 0 µM SF 0.5 µM SF 1 µM SF 5 µM SF

MCF7

SUM159

C 2nd Passage 3rd Passage 120.0 140.0 MCF7 MCF7 100.0 SUM159 120.0 SUM159 100.0 80.0 ** 80.0 * 60.0 ** 60.0 * 40.0 ** ** 40.0

Sphere formation Sphere 20.0 ** formation Sphere 20.0 ** ** ** normalized to control (%) control to normalized 0.0 (%) control to normalized 0.0 0 0.5 1 5 0 0.5 1 5 Concentration of SF (µM) Concentration of SF (µM)

Figure 4.2. Inhibitory effect of sulforaphane on mammosphere formation

105

A P=0.003 4 P=0.008 3.5 3 2.5 2 1.5 1 0.5 % ALDH-positive cells ALDH-positive % 0 0 1 5

Concentration of SF (µm)

B Control 1 µM SF 5 µM SF

3.01% 1.47% 0.49%

Figure 4.3. Inhibitory effect of sulforaphane on ALDH-positive cell population

106

A B 1st Generation 25 ) 500 1st Generation 3 400 Control 20 50 mg/kg SF 300 P=0.018 15 200 10 100 Control Body Body weight (g) 50 mg/kg SF Tumor volume (mm Tumor 0 5 13 18 23 28 33 13 18 23 28 33 Days after inoculation Days after inoculation

CD3.00 P=0.003 2.50 Control Treatment 2.00 1.50 1.00 0.50

% ALDH-positive cells ALDH-positive % 0.00

2.39% 1.08%

Figure 4.4. Sulforaphane decreases tumor size and ALDH-positive cell population in primary breast cancer xenografts

107

A 2nd Generation 1000

100 )

3 P=0.007 10 Control group 1 P=0 .0006 (mm Control group 2 1 SF group 1 Tumorvolume SF group 2 0.1 5 15 25 35 Days after inoculation

B 2nd Generation 120 100 80 60 40 Control 20

% Tumor-free Tumor-free % mice 50 mg/kg SF 0 0 10 20 30 40

Days after inoculation

Figure 4.5. Sulforaphane eradicates breast cancer stem cells in vivo as assessed by re- implantation in secondary mice

108

A SF (µM) 4 d SF (5 µm) MCF7 0 1 5 10 0 1 2 4 d β-catenin cyclin D1 β-actin SF (µM) 4 d SF (5 µm) SUM159 0 1 5 10 0 1 2 4 d β-catenin cyclin D1 β-actin

B TOP-dGFP TOP-dGFP+SF P P 18.00 <0.0001 <0.0001 16.00 14.00 12.00 10.00 8.00 3.04% 2.13% 6.00 P=0.002 4.00 TOP-dGFP+BIO TOP-dGFP+BIO+SF 2.00

% dGFP-positive cells dGFP-positive % 0.00

16.56% 6.92%

C 50 mM LiCl - - - + SF (µM) 4 d 10 µM MG132 - + + + 0 1 5 10 10 µM SF - - + + p-GSK3β p-β-catenin (33/37/41) GSK3β β-actin β-actin

50 mMLiCl - + - + 10 µM SF - - + + β-catenin

p-GSK3β GSK3β

β-actin

Figure 4.6. Sulforaphane down-regulates Wnt/ β-catenin self-renewal pathway

109

Chapter 5: Development of Broccoli Sprout Preparations to Deliver Optimal

Sulforaphane Levels for Cancer Chemoprevention

5.1 Abstract

Sulforaphane is a naturally occurring isothiocyanate in broccoli sprouts with cancer chemopreventive activity. The purpose of this study is to use different methods to develop three broccoli sprout preparations to compare their ability to deliver sulforaphane in vivo and to evaluate the pharmacokinetics and tissue distribution of sulforaphane and sulforaphane-GSH conjugate in mice after oral administration of the broccoli sprout preparations. The sulforaphane-rich sprout preparation generated by our two-step procedure (SM preparation) contained the highest amount of sulforaphane, which was 11 times and 5 times higher than the freeze-dried sprouts with and without plant enzyme activities (FR and ST preparations), respectively. Oral administration of SM preparation produced the highest plasma response among all the three preparations, with the peak plasma concentration of sulforaphane 6 times and 2.3 times higher, and the AUC 7.9 and

2.2 times higher, compared to the other two preparations. Consumption of 2.5 mg/g body weight of the sulforaphane-rich preparation resulted in rapid oral absorption and tissue distribution of sulforaphane in mice, achieving high levels of sulforaphane (336.6 ng/ml,

110

449.97 ± 90.45, 235.26 ± 33.84, 90.49 ± 14.34, 58.55 ± 10.73, 34.90 ± 11.47, and 19.40 ±

6.41 ng/g tissue) and sulforaphane-GSH (1425.6 ng/ml, 1857.97 ± 288.10, 51.85 ± 5.08,

536.30 ± 118.33, 190.76 ± 63.30, and 61.33 ± 2.16 ng/g tissue) in plasma, liver, kidney, lung, heart, muscle, and mammary fat pad, respectively. Oral administration of ST preparation resulted in high level of glucoraphanin and relatively high concentration of sulforaphane in the plasma, suggesting oral absorption of glucoraphanin and incomplete conversion of glucoraphanin in the intestinal tract. The urinary excretion of intact sulforaphane was 49.99%, 78.96%, and 48.56% of a single oral dose over 24 h after administration of SM, ST, and FR preparations, respectively. This study provides a broccoli sprout preparation that can serve as a good source of sulforaphane, and these data can be utilized to guide the dose regimen of sulforaphane for the use of broccoli sprout preparation in cancer chemoprevention.

5.2 Introduction

Numerous studies continue to support that dietary intake of cruciferous vegetables, especially broccoli and broccoli sprouts, may reduce the risk of different types of malignancies [1]. The cancer chemopreventive properties of these vegetables have been primarily attributed to isothiocyanates that occur naturally as the glucosinolate precursors in the plant [1, 2]. In particular, sulforaphane, a member of the isothiocyanate family, has received extensive attention for its potent chemopreventive activity [12, 18].

Sulforaphane has been shown to be not only effective in preventing chemically induced

111

cancers in animal models [12, 14-16], but also inhibit the growth of established tumors

[21, 22]. Early research focused on induction of Phase 2 enzymes and inhibition of Phase

1 enzymes by sulforaphane, which enhances the detoxification of carcinogens [1, 29].

Recent studies suggest that sulforaphane offers protection against tumor development during the “post-initiation” phase by controlling cell proliferation, differentiation, apoptosis, cell cycle, angiogenesis and metastasis [1, 30].

Sulforaphane is converted from glucoraphanin (the major glucosinolate in broccoli and broccoli sprouts) by myrosinase, a β-thioglucosidase [12]. Broccoli sprouts contain approximately 20 times more glucoraphanin than mature broccoli, which represents approximately 74% of all glucosinolates in the sprouts [13]. Myrosinase is physically separated from glucosinolates in the intact plant cells. Disruption of the plant during harvesting, processing, cooking, and chewing leads to loss of cellular compartmentalization and subsequent mixing of glucoraphanin and myrosinase to produce sulforaphane [3, 5]. Epithiospecifier protein (ESP) directs hydrolysis of glucoraphanin toward sulforaphane nitrile formation [7]. While sulforaphane has been shown to possess anticancer property, sulforaphane nitrile has not. The hydrolysis of glucoraphanin to sulforaphane can be also mediated by the microflora in the mammalian gastrointestinal tract [6].

Although glucoraphanin is abundant in broccoli and broccoli sprouts, the processing of broccoli/broccoli sprouts often results in decreased intake of sulforaphane.

Cooking procedures that inactivate plant myrosinase and ESP were shown to significantly reduce the bioavailability of sulforaphane [64], suggesting the incomplete

112

conversion of glucoraphanin by gut microflora. In addition, sulforaphane is relatively thermo-labile [10, 11]. It was shown to rapidly degrade when the temperature was above

60 °C [10]. Glucoraphanin in broccoli sprouts, on the other hand, are relatively stable.

The loss of glucoraphanin during cooking is primarily due to the leaching into the water

[8]. Steam cooking was reported to have minimal effects on glucosinolates of broccoli florets compared to boiling and microwave cooking [8].

After absorption, sulforaphane is primarily metabolized through mercapturic acid pathway in vivo [68]. The electrophilic central carbon of the –N=C=S group in

sulforaphane reacts with the sulfhydryl group of glutathione (GSH) to form a GSH

conjugate (sulforaphane-GSH), which is catalyzed by glutathione S-transferase (GST).

Sulforaphane-GSH is further metabolized to cysteinylglycine (sulforaphane-Cys-Gly),

cysteine (sulforaphane-Cys), and N-acetylcysteine conjugates (sulforaphane-NAC) [69,

70].

The purpose of this study is: (1) to use different methods to develop three broccoli

sprout preparations; (2) to compare their ability to deliver sulforaphane in vivo ; and (3) to

study the pharmacokinetics and tissue distribution of sulforaphane and sulforaphane-GSH

in mice after oral administration of the broccoli sprout preparations. This study will

provide information to optimize dose regimen of broccoli sprout preparation for

chemoprevention.

113

5.3 Materials and Methods

5.3.1 Reagents

Sulforaphane and sulforaphane-GSH were purchased from LKT Laboratories (St.

Paul, MN). Glucoraphanin was obtained from Cfm Oskar Tropitzsch (Marktredwitz,

Germany). Sinigrin hydrate and myrosinase (thioglucosidase) were from Sigma-Aldrich

(St. Louis, MO). Other chemicals and reagents used in the study were all analytical grade unless specified.

5.3.2 Broccoli sprout preparations

Broccoli sprouts were purchased from local Whole Foods Market (Ann Arbor,

MI). The procedure of making there different sprout preparations is described below and illustrated in Fig. 5.1. Fresh broccoli sprouts were quick-frozen in liquid nitrogen, and freeze-dried for 48 h. The dried sprouts were then ground into fine powder (fresh preparation, FR) with pestle and mortar. Fresh broccoli sprouts were quick-steamed for

10 min over boiling water to inactivate endogenous epithiospecifier protein and myrosinase, frozen in liquid nitrogen, and freeze-dried for 48 h. The dried sprouts were then ground into fine powder (steam preparation, ST). Approximately 5 units of myrosinase were incubated with 3 g of ST powder in water at 37 °C for 4 h with occasional stirring. The mixture was then frozen in liquid nitrogen and freeze-dried

(steam preparation followed by myrosinase treatment, SM). The three preparations were stored at -80 °C until used.

114

5.3.3 Determination of glucoraphanin in broccoli sprout preparations

The procedure for measurement of glucoraphanin in broccoli sprout preparations was modified from previously reported method described by Tian et al. [358]. Briefly, 50

mg of sprout preparation was sonicated for 30 min at 70 °C in 1 ml of 70% aqueous

methanol (1 mM sinigrin as internal standard). After cooling in an ice bath, the

supernatant of the extracts was collected by centrifugation at 4,000 rpm for 10 min. The

extraction procedure was repeated three times and the supernatants were combined. The

450 µl of each extract was transferred to a glass vial and dried under nitrogen gas. The

dried samples were reconstituted in 500 µl water, filtered through 0.2 µylon filters, and

diluted 1,000-fold in mobile phase B (acetonitrile containing 5 mM ammonium acetate)

for analysis.

All mass spectra were obtained using a 3200 Q-Trap linear ion trap quadrupole

mass spectrometer (Applied Biosystems, Carlsbad, CA) coupled to an Agilent 1200

Series HPLC system (Agilent Technologies, Santa Clara, CA). HPLC separation was

performed on a Luna 3-µm HILIC column (50 x 2.00 mm, 3 micron) (Phenomenex,

Torrance, CA). Mobile phase A (acetonitrile/water 50/40 containing 5 mM ammonium

acetate) was first kept at 0% for 2.5 min, then increased linearly to 50% in 2.5 min,

returned to 100% B (acetonitrile containing 5 mM ammonium acetate) and maintained

for 5 min. The flow rate was 400 µl/min. Negative ion tandem mass spectrometry was

conducted to detect glucoraphanin. The mass spectrometric conditions were as follows:

source temperature, 700 °C; curtain gas (CUR), 20 psi; ionspray voltage (IS), 4500 V;

desolvation gas temperature (TEM), 700 °C; ion source gas 1 (GS1), 40 psi; ion source

115

gas 2 (GS2), 60 psi; collision gas (CAD), 6 psi; entrance potential (EP), 5 eV; collision energy (CE), 45 eV. The transitions 436.10 > 96.9 and 358.10 > 96.9 were used to detect glucoraphanin and sinigrin, respectively. A dwell time of 200 msec was used for each transition.

5.3.4 Determination of sulforaphane in broccoli sprout preparations

The procedure for measurement of sulforaphane in broccoli sprout preparations

was modified from a previously reported method [359]. Briefly, 50 mg of sprout

preparation was extracted with 1 ml of ethyl acetate by vortexing for 1 min and

sonicating in ice bath for 10 min. After centrifugation at 4,000 rpm for 10 min, the

organic layer was collected. The extraction procedure was repeated three times, the

organic phases were combined and dried under nitrogen gas. The dried samples were

reconstituted in 1 ml of mixture of mobile phase A (water containing 5 mM ammonium

acetate and 0.1% formic acid) and B (acetonitrile containing 0.1% ammonium acetate)

(50:50), sonicated in ice bath for 10 min, filtered through 0.2 µylon filters, and diluted

1,000-fold for analysis.

HPLC separation was performed on an Agilent ZORBAX 5-µm extend-C18

column (50 x 2.1 mm) (Agilent Technologies, Santa Clara, CA). Mobile phase A (water

containing 5 mM ammonium acetate and 0.1% formic acid) was first kept at 95% for 1

min, then decreased to 10% and maintained for 2 min, returned to 5% B (acetonitrile

containing 0.1% ammonium acetate) and maintained for 3 min. The flow rate was 400

µl/min. Positive ion tandem mass spectrometry was conducted to detect sulforaphane.

The mass spectrometric conditions were as follows: source temperature, 400 °C; curtain

116

gas (CUR), 30 psi; ionspray voltage (IS), 5500 V; desolvation gas temperature (TEM),

400 °C; ion source gas 1 (GS1), 60 psi; ion source gas 2 (GS2), 40 psi; collision gas

(CAD), high; entrance potential (EP), 4 eV; collision energy (CE), 15 eV. The transition

178 > 114 was used to detect sulforaphane. A dwell time of 200 msec was used for the transition.

5.3.5 Determination of sulforaphane, sulforaphane-GSH and glucoraphanin in animal samples

Female CD-1 mice (20-25 g in body weight) were purchased from Charles River

Laboratories (Wilmington, MA). Plasma was obtained at pre-dose and 0.25, 0.5, 1, 2, 4,

6, 8, 12, and 24 h post-dose after oral gavage of sprout preparations (2.5 mg/g body weight, homogenized in ice-cold PBS). Blood was collected into heparinized syringes, and immediately separated by centrifugation at 2,000 g for 10 min at 4 °C. Tissue samples were obtained at pre-dose and 0.5, 1, 2, 4, 6, 8, 12, and 24 h post-dose. Plasma and tissue samples were stored at -80 °C until further analysis.

Mice were placed in metabolism cages overnight. Urine and feces were obtained at pre-dose and 1, 2, 4, 6, 8, 12, and 24 h post-dose after oral gavage of broccoli sprout preparations (2.5 mg/g body wt, homogenized in ice-cold PBS). Urine and feces samples were stored at -80 °C until further analysis.

Plasma, tissue homogenate, urine, or feces homogenate (50 µl) was extracted with

400 µl methanol containing 0.5% formic acid and votexed for 4 min, as described in a previous report [360]. The supernatant was collected after centrifugation at 13,000 rpm for 5 min. The 450 µl of supernatant was transferred to a clean centrifuge tube, and dried

117

in a speed vacuum at medium temperature. The dried sample was reconstituted in 200 µl of mixture of mobile phase A (water containing 0.1% formic acid and 5 mM ammonium acetate) and B (acetonitrile containing 0.1% formic acid), mixed for 3 min, and centrifuged at 13,000 rpm for 5 min. The supernatant was transferred to HPLC vial for analysis. Similarly to above, an Applied Biosystems 3200 Q-Trap linear ion trap quadrupole mass spectrometer coupled to an Agilent 1200 Series HPLC system was used for the detection and quantification of sulforaphane, sulforaphane-GSH, and glucoraphanin. The transition 485.3 > 179 was used to detect sulforaphane-GSH.

5.3.6 Pharmacokinetic analysis

The plasma concentration data were analyzed by non-compartmental pharmacokinetic analysis using WinNonlin software (Pharsight). The area under the plasma concentration versus time from time zero to the time of the last measured concentration (AUC0-t) and the AUC0-∞ were calculated by the software. Cmax and

Tmax were determined graphically from the plasma concentration versus time plots.

5.3.7 Statistical analysis

Statistical differences were determined using two-tailed Student t-test. Data are

presented as mean ± SD (n ≥ 3).

118

5.4 Results

5.4.1 Content of sulforaphane and glucoraphanin in three different broccoli sprout preparations

Steam cooking was shown to have minimal effects on glucosinolates of broccoli florets compared to high pressure boiling, conventional boiling, and microwave cooking

[8]. Thus, we developed three different broccoli sprout preparations and measured the content of sulforaphane and glucoraphanin in these preparations, as described in

Materials and Methods and illustrated in Fig. 5.1. Fresh preparation (FR) was made to preserve glucoraphanin with plant myrosinase and ESP activities. Steam preparation (ST) was made to preserve high content of glucoraphanin and inactivate plant myrosinase and

ESP activities. The preparation generated from two-step procedure (steaming followed by myrosinase treatment, SM) was made to preserve high content of sulforaphane.

The retention of sulforaphane and glucoraphanin in the broccoli sprout preparations was largely affected by the way the preparations were made (Table 5.1). The glucoraphanin content of ST preparation (22.737 ± 1.332 mg/g d.w.) was approximately

12- and 253-fold higher than that of FR (1.938 ± 0.100 mg/g d.w.) and SM (0.090 ±

0.009 mg/g d.w.), respectively. As expected, the glucoraphanin in SM preparation was fully converted to sulforaphane by incubation with pure myrosinase. The sulforaphane content of SM was the highest among all the three preparations, 4.252 ± 0.151 mg/g d.w., which is approximately 11 and 5 times higher than FR (0.392 ± 0.045 mg/g d.w.) and ST

(0.918 ± 0.049 mg/g d.w.), respectively.

119

5.4.2 Pharmacokinetic profile of sulforaphane, sulforaphane-GSH, and glucoraphanin after oral administration of the broccoli sprout preparations in mice

We next evaluated the pharmacokinetics of sulforaphane, sulforaphane-GSH, and

glucoraphanin in mice after oral administration of the broccoli sprout preparations and

compared the ability of the three different preparations to deliver sulforaphane in vivo .

Each preparation was administered into CD-1 mice by oral gavage. The plasma and tissue

samples were collected at 0, 0.25, 0.5, 1, 2, 4, 8, 12, 24 h. Given that glutathione

conjugate of sulforaphane (sulforaphane-GSH) are the major means of transport of this

bioactive substance through the body [73], we measured both sulforaphane and

sulforaphane-GSH in plasma and tissue samples.

The plasma concentration-time profile of sulforaphane, sulforaphane-GSH, and

glucoraphanin are shown in Fig. 5.2. Oral administration of all three preparations led to

an immediate increase in the levels of sulforaphane and sulforaphane-GSH in plasma,

suggesting a rapid oral absorption of sulforaphane. This was consistent with previous

reports that the absorption of sulforaphane was fast with first-order kinetics [270].

However, the “secondary peak” at 2 h after oral administration of ST preparation, but not

SM or FR preparations, suggest the conversion of glucoraphanin into sulforaphane by gut

microflora is relatively slow.

The summary of the pharmacokinetic parameters is depicted in Table 5.2. The peak plasma concentration of sulforaphane was 6 and 2.3 times higher when SM preparation (336.6 ng/ml) was administered, compared to FR (56.0 ng/ml) and ST (146.5 ng/ml) preparations, respectively. The peak plasma time was similar, 0.5 h for SM and

120

FR preparations and 0.25 h for ST. The AUC0-24 in mice after oral administration of SM preparation (854.9 ng*h/ml) was 7.9 and 2.2 times higher, compared to FR (108.3 ng*h/ml) and ST (386.4 ng*h/ml) preparations, respectively. Similarly, the AUC0-∞ in

mice after oral administration of SM preparation (858.5 ng*h/ml) was 7.8 and 2.2 times

higher, compared to FR (110.1 ng*h/ml) and ST (388.5 ng*h/ml) preparations,

respectively. The half-life of sulforaphane was similar among all the three groups, 2.1,

2.5, and 2.6 h for SM, ST, and FR.

As shown in Table 5.2, for sulforaphane-GSH conjugate, the Cmax of SM group

(1425.6 ng/ml) was 10.8 and 2.7 times higher than the Cmax of FR (132.3 ng/ml) and ST

(523.6 ng/ml) groups; and Tmax was similar in all the three groups. The AUC in mice

after oral administration of SM preparation (AUC0-24: 2380.5 ng*h/ml; AUC0-∞:

2383.4 ng*h/ml) was 12.6 and 2.7 times higher, in comparison to FR (AUC0-24: 188.3 ng*h/ml; AUC0-∞: 189.0 ng*h/ml) and ST (AUC0-24: 895.6 ng*h/ml; AUC0-∞: 897.0 ng*h/ml) preparations, respectively. The T1/2 of sulforaphane-GSH was similar among all the three groups, 1.9, 2.7, and 2.2 h for SM, ST, and FR. When the pharmacokinetic parameters of sulforaphane and sulforaphane-GSH were compared, the Cmax values of sulforaphane-GSH for SM, ST, and FR groups were approximately 4.2, 3.6, and 2.4 times higher than that of sulforaphane. The AUC of sulforaphane-GSH were approximately 2.8

(SM), 2.3 (ST), and 1.7 times (FR) higher than that of sulforaphane. The Tmax and T1/2 of sulforaphane and sulforaphane-GSH were similar.

Collectively, these data demonstrate that SM preparation can deliver the highest amount of sulforaphane and sulforaphane-GSH into plasma.

121

The mercapturic acid pathway is the major route of sulforaphane elimination [73].

After absorption sulforaphane is conjugated to GSH, further metabolized to mercapturic acids, and subsequently excreted in the urine. In this study, we determined excretion of the unchanged sulforaphane. As shown in Fig. 5.3, the cumulative urinary level of sulforaphane was almost linear within 2 h after administration of ST and FR preparations and 8 h for SM. Approximately 49.99%, 78.96%, and 48.56% of sulforaphane intake was excreted unchanged in urine over 24 h after oral administration of SM, ST, and FR preparations, respectively (Table 5.3). The level of sulforaphane-GSH in urine was very low. The feces recovery of sulforaphane was approximately 1% (Table 5.3).

We also observed that intact glucoraphanin was absorbed into blood circulation and excreted in urine. The peak plasma concentration of glucoraphanin was 932 and 339 times higher when ST preparation (2983.2 ng/ml) was administered, compared to SM

(3.2 ng/ml) and FR (8.8 ng/ml) preparations, respectively (Table 5.2). The AUC for ST group was approximately 2400-2700 and 140-170 times higher than SM and FR groups, respectively. In addition, approximately 19.97%, 18.96%, and 0.08% of glucoraphanin

(relative to glucoraphanin content of sprout preparations) was recovered unchanged in urine after oral administration of SM, ST, and FR preparations, respectively (Table 5.3).

5.4.3 Tissue distribution of sulforaphane and sulforaphane-GSH

Given that the plasma levels of sulforaphane and sulforaphane-GSH were very low for FR preparation, we only measured the two compounds in mice administered with

ST and SM preparations. As shown in Fig. 5.4-5.9, levels of sulforaphane in liver, kidney, lung, heart, muscle, and mammary fat pad tissues were highest at 0.5 h after

122

administration of SM preparation, reaching 449.97 ± 90.45, 235.26 ± 33.84, 90.49 ±

14.34, 58.55 ± 10.73, 34.90 ± 11.47, and 19.40 ± 6.41 ng/g tissue, respectively, and decreased thereafter. Likewise, tissue concentrations of sulforaphane-GSH reached their peak 0.5 h after oral dosing of SM preparation (1857.97 ± 288.10, 51.85 ± 5.08, 536.30 ±

118.33, 190.76 ± 63.30, and 61.33 ± 2.16 ng/g tissue for liver, kidney, lung, heart, and muscle, respectively) and decreased thereafter. Sulforaphane-GSH was under detection limit in mammary fat pad. For sulforaphane, the peak concentrations of SM group was approximately 1.8-3.3 times higher compared to ST group (134.05 ± 43.21, 101.21 ±

19.59, 37.39 ± 3.2, 27.14 ± 8.53, 19.15 ± 4.43, and 6.55 ± 1.29 ng/g tissue for liver, kidney, lung, heart, muscle, and mammary fat pad, respectively). For sulforaphane-GSH, the peak concentrations of SM group was 2.9-16.7 times higher compared to ST group

(461.12 ± 17.42, 3.10 ± 0.81, 107.67 ± 8.51, 37.45 ± 10.02, and 20.79 ± 6.79 ng/g tissue for liver, kidney, lung, heart, and muscle, respectively). These data will be further used for physiologically-based pharmacokinetic modeling.

5.5 Discussion

Although the precursor of sulforaphane, glucoraphanin, is known to be abundant in broccoli and broccoli sprouts, the processing of broccoli/broccoli sprouts often results in the loss of the chemopreventive compound. This can be attributed to three factors, the presence of ESP in the plant, inactivation of plant myrosinase by cooking, and the low stability of sulforaphane itself. Disruption of plant cellular structures can result in the

123

release of both myrosinase and ESP. While myrosinase catalyzes the conversion from glucoraphanin to sulforaphane, the naturally occurring ESP directs hydrolysis of glucoraphanin toward sulforaphane nitrile that does not possess anticancer properties [7].

However, cooking procedures that inactivate both plant myrosinase and ESP were shown to significantly reduce the bioavailability of sulforaphane up to 3-fold [64], indicating the incomplete conversion of glucoraphanin to sulforaphane by mammalian gut microflora after consumption. Furthermore, sulforaphane once formed is relatively unstable, especially at high temperature during processing of broccoli sprouts [10, 11].

Sulforaphane was rapidly degraded when the temperature was above 60 °C [10].

Glucoraphanin in broccoli sprouts, on the other hand, is relatively stable. It was suggested that thermal degradation is unlikely to be a major cause of glucosinolate loss when cooking time is less than 10 min [9]. The loss of glucoraphanin during cooking is primarily due to the leaching into the cooking water [8]. Steam cooking was reported to have minimal effects on glucosinolates of broccoli florets compared to boiling and microwave cooking [8].

Therefore, we used a two-step method to generate a sulforaphane-rich broccoli sprout preparation. First, we steamed broccoli sprouts to inactive both myrosinase and

ESP in the plant and minimize hydrolysis of glucoraphanin. Then we converted glucoraphanin to sulforaphane by adding myrosinase. This product was immediately freeze-dried to preserve high content of sulforaphane. We hypothesized that this sprout preparation (named SM) would be superior to the other two preparations (freeze-dried broccoli sprouts with and without plant enzyme activities, named FR and ST,

124

respectively), based on the following: (1) Upon consumption, mastication and digestion, the plant myrosinase and ESP in FR preparation could convert glucoraphanin to sulforaphane and sulforaphane nitrile, respectively; and (2) ST preparation could minimize hydrolysis of glucoraphanin by plant myrosinase, completely relying on gut microflora in vivo for conversion of glucoraphanin to sulforaphane after consumption.

We first examined the content of glucoraphanin and sulforaphane in the three

preparations. As expected, ST preparation preserved the highest amount of

glucoraphanin, while SM preparation had the highest amount of sulforaphane among all

the three preparations, which is approximately 11 and 5 times higher than FR and ST,

respectively (Table 5.1). Low content of both sulforaphane and glucoraphanin in FR

preparation suggests that glucoraphanin might be converted to other compounds (e.g.,

sulforaphane nitrile) during the processing and storage. The low stability of sulforaphane

might also contribute to this observation. The content of sulforaphane nitrile was not

determined because its authentic standard was not commercially available.

Next, we evaluated these sprout preparations for their ability to deliver

sulforaphane in mouse plasma after oral administration and examined the in vivo

pharmacokinetics and tissue distribution of sulforaphane and sulforaphane-GSH

conjugate. The peak plasma concentration and AUC value of sulforaphane in SM group

was 6- and 7.9-fold higher than those in FR group, respectively; and likewise, 2.3- and

2.2-fold higher than ST group, respectively (Table 5.2 and Fig. 5.2). The peak

concentrations of sulforaphane in mouse tissues after ingestion of SM preparation was

approximately 1.8 to 3.3 times higher compared to ST group (Fig. 5.4-5.9). Since

125

sulforaphane-GSH is the major means of transport of this bioactive substance through the body [73], we also measured sulforaphane-GSH in plasma and tissues. The peak plasma concentration and AUC of this compound was 10.8 and 12.6 times higher after administration of SM preparation, in comparison to FR preparation, respectively; and 2.7 times higher than ST group. The peak concentrations of sulforaphane-GSH in mouse tissues after ingestion of SM preparation was approximately 2.9 to 16.7 times higher compared to ST group.

The kinetics of sulforaphane absorption and excretion has been described in several publications. Sulforaphane was shown to efficiently and rapidly absorbed in human subjects given a single dose of broccoli extract, with peak plasma concentration occurring 1 h after feeding and a half life of 1.77 ± 0.13 h [1, 65, 66]. A human perfusion study showed that 74 ± 29% of sulforaphane from broccoli extracts can be absorbed in the jejunum and that a portion of that returns to the lumen of the jejunum as sulforaphane-GSH conjugate [65]. In our study, following oral administration of broccoli sprout preparations, peak plasma concentration of sulforaphane and sulforaphane-GSH conjugate occurred around 0.25-0.5 h (Table 5.2). The Cmax and AUC values of both compounds were proportionally increased from FR, ST, and SM preparations (Table 5.2).

In addition, we observed that although both sulforaphane and sulforaphane-GSH had similar Tmax value, the Cmax and AUC of sulforaphane were much lower than those of sulforaphane-GSH (Table 5.2). This observation is consistent with the results from the pharmacokinetic studies in rats and mice [286, 360]. Furthermore, our data have shown that tissue levels of sulforaphane in liver, kidney, lung, heart, muscle, and mammary fat

126

pad reached the peak at 0.5 h (Fig. 5.4-5.9), indicating that sulforaphane can be quickly distributed throughout the mouse body after ingestion of broccoli sprout preparations, which is in agreement with previous studies [66].

Notably, we observed consistent higher levels of sulforaphane and sulforaphane-

GSH at the time point of 2 h compared to 1 h in plasma and tissues after oral administration of ST preparation, but not SM preparation (Fig. 5.2 and Fig. 5.4-5.9). The

“secondary peaks” at 2 h are probably due to the presence of myrosinase produced by gut microflora. The conversion of glucoraphanin into sulforaphane in the colon is slow and incomplete [361].

The mercapturic acid pathway is the major route of elimination of sulforaphane

[73]. Minor routes of excretion could be defecation, exhalation, and perspiration [73].

After absorption sulforaphane is quickly conjugated to GSH, and further metabolized to mercapturic acids, which are subsequently excreted in the urine. Although we did not measure the level of mercapturic acids in our study, there are literature reports indicating that 37% of sulforaphane was recovered in urine as sulforaphane mercapturic acids in 24 h after consumption of fresh broccoli by human subjects [73]. Herein, we found that sulforaphane was excreted unchanged in urine after oral administration of broccoli sprout preparations in mice (49.99%, 78.96%, and 48.56% for SM, ST, and FR preparations, respectively) (Table 5.3). Based on this information, we postulate that ~37% of sulforaphane would be excreted in urine as mercapturic acid and 48-50% as unchanged parent compound; and the total amount would be about 85-87%. This estimation is in consistent with the observation that the urinary excretion of isothocyanates (primarily

127

sulforaphane) equivalents amounted to approximately 73.7% in 24 h in rats fed with broccoli sprout extract [72]. The higher urinary recovery of intact sulforaphane in ST group is probably due to the conversion of glucoraphanin to sulforaphane in intestinal tract.

In addition, we also observed that a large amount of intact glucoraphanin was absorbed into blood circulation and excreted in urine. The presence of intact glucoraphanin in the urine (5% of an oral dose) of orally dosed rats was firstly reported by Bheemreddy et al. , based on which the authors suggested that glucoraphanin was absorbed intact [362]. Herein, we found that oral administration of ST preparation resulted in high levels of intact glucoraphanin in plasma and that approximately 18.96% of glucoraphanin was excreted unchanged in urine. The low recovery of FR group

(0.08%) might be due to the hydrolysis of glucoraphanin during the homogenization of sprout preparation and oral administration.

5.6 Conclusions

In conclusion, as we hypothesized, the sulforaphane-rich broccoli sprout preparation (SM) generated by the two-step procedure contained the highest amount of sulforaphane and produced the highest plasma concentration of sulforaphane. The freeze- dried broccoli sprouts with plant enzyme activities (FR) resulted in the lowest plasma concentration. The pharmacokinetic analysis and tissue distribution indicate that sulforaphane was well and rapidly absorbed and distributed to various tissues following

128

oral administration of our broccoli sprout preparations in the mice. These data can be utilized to guide the dose regimen of sulforaphane for the use of broccoli sprout preparation in cancer chemoprevention.

129

Broccoli sprout Glucoraphanin SF preparations (mg/g d.w.) (mg/g d.w.) Steamed + myrosinase (SM) 0.090 ± 0.009 4.252 ± 0.151

Steamed (ST) 22.737 ± 1.332 0.918 ± 0.049

Fresh (FR) 1.938 ± 0.100 0.392 ± 0.045

Table 5.1. Glucoraphanin and sulforaphane content in three different broccoli sprout preparations

130

SF SF-GSH Glucoraphanin Broccoli sprout SM ST FR SM ST FR SM ST FR preparations AUC0-∞ (ng*hr/ml) 858.5 388.5 110.1 2383.4 897.0 189.0 3.29 8123.2 55.5

AUC0-last (ng*hr/ml) 854.9 386.4 108.3 2380.5 895.6 188.3 2.97 8106.9 44.2

Tmax(hr) 0.5 0.25 0.5 0.25 0.25 0.5 0.5 1 1

Cmax(ng/ml) 336.6 146.5 56.0 1425.6 523.6 132.3 3.2 2983.2 8.8

T1/2 (hr) 2.1 2.5 2.6 1.9 2.7 2.2 - - -

Table 5.2. Pharmacokinetic parameters of sulforaphane, sulforaphane-GSH, and glucoraphanin after a single oral dose of 2.5 mg/g body weight of three different broccoli sprout preparations in mice

131

SF Glucoraphanin Broccoli sprout SM ST FR SM ST FR preparations Recovery 49.99 78.96 48.56 19.97 18.96 0.08 in urine (%) Recovery 0.94 0.98 1.56 - - - in feces (%)

Table 5.3. Recovery of sulforaphane and glucoraphanin in urine and feces after a single oral dose of 2.5 mg/g body weight of three different broccoli sprout preparations in mice

132

Fresh Broccoli Sprouts

Freeze Inactivate Inactivate Steam Steam -dry ESP and ESP and myrosinase myrosinase FR preparation Freeze Freeze -dry -dry

ST preparation Convert Myrosinase glucoraphanin to SF

Freeze -dry

SM preparation

Figure 5.1. Development of three broccoli sprout preparations

133

AB 1000 10000 SM SM 1000 100 ST ST FR FR 100 10 10

1 Plasma (ng/ml) Plasma 1 Plasma (ng/ml) Plasma Concentration of SF in SF of Concentration 0.1 0.1 Concentration of SF-GSH in ofSF-GSH Concentration 0 10 20 30 0 10 20 30 Time (hr) Time (hr) C 10000 SM ST 1000 FR

100 (ng/ml) 10 Concentration of Concentration 1 Glucoraphanin in PlasmaGlucoraphanin 0 10 20 30 Time (hr) Figure 5.2. Plasma concentration profile of sulforaphane (A), sulforaphane-GSH (B), and glucoraphanin (C) after a single dose of 2.5 mg/g body weight of three different broccoli sprout preparations in mice

134

A 1000000 160000 140000 100000 120000 10000 SM 100000 ST 1000 80000 SM FR 60000 100 ST 40000 10 FR 20000 of SF SF permouse) (ng of of SF SF (ngof permouse) Cumulative Urinary Level Cumulative Cumulative Urinary Level Cumulative 1 0 0 10 20 30 0 10 20 30 Time (hr) Time (hr)

B 1000000 300000

100000 250000 10000 200000 SM 1000 150000 ST SM FR 100 100000 ST

(ng permouse) 10 permouse) (ng 50000

of Glucoraphanin of FR Glucoraphanin of 1 0 Cumulative Urinary Level Cumulative Urinary Level Cumulative 0 10 20 30 0 10 20 30 Time (hr) Time (hr)

Figure 5.3. Cumulative urinary levels of sulforaphane (A) and glucoraphanin (B) after a single dose of 2.5 mg/g body weight of three different broccoli sprout preparations in mice

135

A 600

500 SM ST 400

300

200 Liver (ng/g) Liver 100 Concentration of SF in ofSF Concentration 0 0 10 20 30 Time (hr) B 2500 SM 2000 ST

1500

1000 Liver (ng/g) Liver 500

0 Concentration of SF-GSH in ofSF-GSH Concentration 0.0 10.0 20.0 30.0 Time (hr) Figure 5.4. Tissue levels of sulforaphane (A) and sulforaphane-GSH (B) in the liver after a single dose of 2.5 mg/g body weight of broccoli sprout preparations in mice

136

A 300

250 SM ST 200

150

100 Kidney (ng/g) Kidney 50 Concentration of SF in ofSF Concentration 0 0 10 20 30 Time (hr)

B 60

50 SM ST 40

20 Kidney (ng/g) Kidney 10

0 Concentration of SF-GSH in ofSF-GSH Concentration 0 10 20 30 Time (hr)

Figure 5.5. Tissue levels of sulforaphane (A) and sulforaphane-GSH (B) in the kidney after a single dose of 2.5 mg/g body weight of broccoli sprout preparations in mice

137

A 120 SM 100 ST 80

40 Lung (ng/g) Lung

20 Concentration of SF in ofSF Concentration 0 0 10 20 30 Time (hr)

B 700 600 SM ST 500 400 300

Lung (ng/g) Lung 200 100 0 Concentration of SF-GSH in ofSF-GSH Concentration 0 10 20 30 Time (hr)

Figure 5.6. Tissue levels of sulforaphane (A) and sulforaphane-GSH (B) in the lung after a single dose of 2.5 mg/g body weight of broccoli sprout preparations in mice

138

A 80 70 SM 60 ST 50 40 30

Heart (ng/g) Heart 20 10 Concentration of SF in ofSF Concentration 0 0 10 20 30 Time (hr)

B 250.00 SM 200.00 ST

150.00

100.00 Heart (ng/g) Heart 50.00

Concentration of SF-GSH in ofSF-GSH Concentration 0.00 0 10 20 30 Time (hr)

Figure 5.7. Tissue levels of sulforaphane (A) and sulforaphane-GSH (B) in the heart after a single dose of 2.5 mg/g body weight of broccoli sprout preparations in mice

139

A 50 SM 40 ST

Muscle (ng/g) Muscle 10 Concentration of SF in ofSF Concentration 0 0 10 20 30 Time (hr)

B 70 SM 60 ST 50 40 30 20 Muscle (ng/g) Muscle 10 0 Concentration of SF-GSH in ofSF-GSH Concentration 0 10 20 30 Time (hr)

Figure 5.8. Tissue levels of sulforaphane (A) and sulforaphane-GSH (B) in the muscle after a single dose of 2.5 mg/g body weight of broccoli sprout preparations in mice

140

25 SM

20 ST

5 Concentration of SF in ofSF Concentration Mammary Fatpad (ng/g) Fatpad Mammary 0 0 10 20 30 Time (hr)

Figure 5.9. Tissue levels of sulforaphane in the mammary fat pad after a single dose of 2.5 mg/g body weight of broccoli sprout preparations in mice

141

Chapter 6: Conclusions and Further Work

Numerous studies have substantiated the chemopreventive activity of sulforaphane, a naturally occurring compound derived from broccoli and broccoli sprouts. In these studies, we have investigated the cancer chemoprevention activities of sulforaphane, revealed a novel molecular target of sulforaphane in pancreatic cancer, evaluated the effect of sulforaphane on breast cancer stem cells, and compared different broccoli sprout preparations for optimal delivery of sulforaphane for future chemoprevention studies.

We first found a novel molecular target of sulforaphane in pancreatic cancer. We have demonstrated that sulforaphane blocks Hsp90-p50 Cdc37 interaction, induced the degradation of Hsp90 client proteins, and inhibited pancreatic cancer cells in vitro and exhibited anticancer activity in pancreatic cancer xenograft in vivo . Furthermore, we traced its activity to a novel interaction site of Hsp90. Most Hsp90 inhibitors target the

ATP binding pocket. Our data showed that sulforaphane inhibits Hsp90 through an ATP- binding-independent manner. Sulforaphane blocked the interaction of Hsp90 with its cochaperone p50 Cdc37 by directly interacting with specific amino acid residues of Hsp90 and p50 Cdc37 . Proteolytic fingerprinting and LC-MS found sulforaphane interaction with

the N-terminal domain of Hsp90 and the central domain of p50 Cdc37 . LC-MS tryptic

142

peptide mapping and NMR spectra of full-length Hsp90 identified a covalent sulforaphane adduct in sheet 2 and the adjacent loop in Hsp90 N-terminal domain. These data suggest a novel mechanism of sulforaphane that disrupts protein-protein interactions in Hsp90 complex for its chemoprevention activity.

17-AAG competitively binds to N-terminal ATP pocket of Hsp90, and induces a conformational change in the Hsp90 molecule, leading to proteasomal degradation of its client proteins. 17-AAG has been evaluated in a great deal of preclinical studies and clinical trials; however, hepatotoxicity seems to be the most significant dose-limiting factor for its application as a single agent. A more appropriate strategy is to combine 17-

AAG with other agent(s) to lower 17-AAG dose and to achieve better anticancer effect.

Thus, we evaluated the combination efficacy of sulforaphane and 17-AAG in pancreatic cancer. Sulforaphane disrupted Hsp90-p50 Cdc37 interaction while 17-AAG inhibited ATP

binding to Hsp90. Concomitant use of sulforaphane and 17-AAG synergistically down-

regulated Hsp90 client proteins in pancreatic cancer cells, indicating an enhanced

abrogation of Hsp90 function. The therapeutic efficacy of 17-AAG was greatly

potentiated by sulforaphane both in vitro and in pancreatic cancer xenograft model. This finding provide a rationale for further evaluation of broccoli/broccoli sprout preparations combined with 17-AAG for better efficacy and lower dose-limiting toxicity in pancreatic cancer.

Cancer stem cells (CSCs) have been suggested to contribute to tumor resistance/relapse because chemotherapy and radiation therapy are incapable of eradicating them. Targeting these self-renewal pathways may provide an effective

143

strategy to target CSCs and thereby overcome tumor resistance and reduce relapse. We, therefore, evaluated sulforaphane for its efficacy to inhibit breast CSCs and its potential mechanism. Sulforaphane (1-5 µM) decreased aldehyde dehydrogenase (ALDH)-positive cell population by 65%-80% in human breast cancer cells, and reduced the size and number of primary mammospheres by 8- to 125-fold and 45%-75%, respectively, as evidenced by Aldefluor assay and mammosphere formation assay. Daily injection with

50 mg/kg sulforaphane to NOD/SCID xenograft mice for two weeks eliminated breast

CSCs in vivo , thereby abrogating tumor growth after re-implantation of primary tumor

cells into the secondary mice. We further demonstrated that sulforaphane down-regulated

Wnt/ β-catenin self-renewal pathway. The results suggest the down-regulation of Wnt/ β-

catenin self-renewal pathway by sulforaphane as one of the potential mechanisms for its

efficacy. These findings provide a strong rationale for the use of sulforaphane for breast

cancer chemoprevention and therapies in combination with chemotherapy.

All these studies support the preclinical and clinical evaluation of broccoli sprout

preparations for cancer chemoprevention. Therefore, we developed three different

broccoli sprout preparations, compared their ability to deliver chemopreventive

compounds in vivo , and evaluated the pharmacokinetics and tissue distribution of sulforaphane and sulforaphane-GSH conjugate in mice after oral administration of the broccoli sprout preparations. The sulforaphane-rich sprout preparation generated by our two-step procedure contained the highest amount of sulforaphane, 11 times and 5 times higher than the freeze-dried sprouts with and without plant enzyme activities, respectively; and produced the greatest plasma response among all the three preparations,

144

with the peak plasma concentration of sulforaphane 6 times and 2.3 times higher, and the

AUC 7.9 and 2.2 times higher, compared to the other two preparations. Consumption of

2.5 mg/g body weight of the sulforaphane-rich preparation resulted in rapid absorption and distribution of SF in mice, achieving high levels of sulforaphane (336.6 ng/ml,

449.97 ± 90.45, 235.26 ± 33.84, 90.49 ± 14.34, 58.55 ± 10.73, 34.90 ± 11.47, and 19.40 ±

6.41 ng/g tissue) and sulforaphane-GSH (1425.6 ng/ml, 1857.97 ± 288.10, 51.85 ± 5.08,

536.30 ± 118.33, 190.76 ± 63.30, and 61.33 ± 2.16 ng/g tissue) in plasma, liver, kidney, lung, heart, muscle, and mammary fat pad, respectively. These data will be further used for physiologically-based pharmacokinetic modeling. This study provides a broccoli sprout preparation that can serve as a good source of SF for further evaluation of the chemopreventive efficacy in mouse models, alone or in combination with clinical drugs, in the future.

145

Appendix: Green Tea Epigallocatechin-3-gallate Inhibits Hsp90 Function by

Impairing Hsp90 Association with Cochaperones in Pancreatic Cancer Cells

A.1 Abstract

(-)-Epigallocatechin-3-gallate [(-)-EGCG], the most abundant polyphenolic catechin in green tea, showed chemoprevention and anticancer activities. (-)-EGCG was reported to bind to the C-terminal domain of heat shock protein 90 (Hsp90). The purpose of this study is to investigate (-)-EGCG as a novel Hsp90 inhibitor to impair Hsp90 super-chaperone complex for simultaneous down-regulation of oncogenic proteins in pancreatic cancer cells. MTS assay showed that (-)-EGCG exhibited anti-proliferative activity against pancreatic cancer cell line Mia Paca-2 in vitro with IC50 below 50 µM. (-

)-EGCG increased caspase-3 activity up to 3-fold in a time- and concentration-dependent manner. Western Blotting analysis demonstrated that (-)-EGCG induced down-regulation of oncogenic Hsp90 client proteins by approximately 70%~95%, including Akt, Cdk4,

Raf-1, Her-2, and pERK. Co-immunoprecipitation showed that (-)-EGCG decreased the association of cochaperones p23 and Hsc70 with Hsp90 by more than 50%, while it had little effect on the ATP binding to Hsp90. Proteolytic fingerprinting assay confirmed

146

direct binding between (-)-EGCG and the Hsp90 C-terminal domain. These data suggest that the binding of (-)-EGCG to Hsp90 impairs the association of Hsp90 with its cochaperones, thereby inducing degradation of Hsp90 client proteins, resulting anti- proliferating effects in pancreatic cancer cells.

A.2 Introduction

Pancreatic cancer is the fourth leading cause of cancer death in the United States

[317]. The overall 5-year survival rate after diagnosis for pancreatic cancer patients is less than 5% [318]. Major therapeutic targets for pancreatic cancer include K-ras pathway downstream signaling (e.g., Raf-MEK ERK pathway), epidermal growth factor receptor

(EGFR), ErbB-2 (Her-2), phosphoinositide 3-OH kinase (PI3K)/Akt, and p53 mutant

[306, 311, 323, 324]. Due to the complexity of the disease, targeting multiple oncogenic pathways would be beneficial for chemoprevention of pancreatic cancer.

Heat shock protein 90 (Hsp90), a highly abundant molecular chaperone in the stress response, assists maturation of more than 200 proteins, which include transmembrane tyrosine kinases (Her-2, EGFR), metastable signaling proteins (Akt, K- ras, Raf-1), mutated signaling proteins (p53, v-Src), chimeric signaling proteins (Bcr-

Abl), cell cycle regulators (Cdk4, Cdk6), and steroid receptors (androgen, estrogen, and progesterone receptors) [97, 100, 102, 363] [96]. Many of these client proteins are mutated and/or over-expressed in pancreatic cancer [103, 104]. Based on crystal structure, Hsp90 protein consists of three highly conserved domains: an N-terminal ATP-

147

binding domain, a middle domain and a C-terminal dimerization domain [364]. The N- terminus of Hsp90 contains a specific ATP binding pocket, which has been well characterized [311, 365]. Benzoquinone ansamycins, such as geldanamycin (GA) [366] and its derivatives 17-allylamino-17-desmethoxygeldanamycin (17-AAG) [125, 366], competitively block ATP binding to this N-terminal ATP binding site on Hsp90 [119,

143], resulting in ubiquitination and proteasomal degradation of client proteins. Hsp90 requires an array of cochaperones to assemble a super-chaperone complex for its function. These cochaperones, including Cdc37, Hsc70, Hsp40, Hop, Hip, p23, pp5, and immunophilins, bind to and release from the complex at various stages to accomplish the folding and maturation of Hsp90 client proteins [143]. A newly synthesized client protein binds to Hsc70/Hsp40 complex, and then associates with the “open” state Hsp90 via the bridging cochaperone Hop, which interacts simultaneously with Hsp90 and Hsc70 [103]..

Upon ATP binding, Hsp90 then binds to p23 and immunophilins, converting the intermediate chaperone complex into a mature complex [143]. Upon ATP hydrolysis, the correctly-folded client protein is released from Hsp90 [108].

Recently, the C-terminal domain of Hsp90 has been shown to possess a second

ATP binding site [367]. Novobiocin, a coumarin antibiotic isolated from Streptomyces

species, binds Hsp90 at the C-terminal ATP binding site [367]. This binding induced an

alteration in Hsp90 conformation [367, 368], interfering Hsp90/Hsc70 and Hsp90/p23

interactions [368]. An allosteric regulation is suggested between the C-terminal and N-

terminal domains of Hsp90 such that the interaction of ligands with one site might affect

the occupancy of the other site [367, 369].

148

Green tea is one of the most widely consumed beverages in the world.

Epidemiological studies suggest an association between green tea consumption and cancer prevention effects [370]. The various polyphenolic catechins contained in green tea are thought to contribute to its chemoprevention against certain types of cancer. In particular, several studies indicate that (-)-epigallocatechin-3-gallate [(-)-EGCG], the most abundant catechin in green tea, is a potent chemoprevention and anticancer component [371]. However, the underlying mechanism of (-)-EGCG for its chemoprevention is not well defined. In 2005, Palermo et al. reported that (-)-EGCG could inhibit the transcriptional activity of aryl hydrocarbon receptor (AhR) through a mechanism involving direct binding to the C-terminal region of Hsp90 [372]. It remains unclear whether (-)-EGCG could inhibit Hsp90 function through direct binding and how

(-)-EGCG affect the chaperone function through this binding. The purpose of this study is to investigate (-)-EGCG as a novel Hsp90 inhibitor to impair Hsp90 super-chaperone complex for inhibiting its chaperoning function, which simultaneously down-regulates oncogenic proteins in pancreatic cancer cell line Mia Paca-2.

A.3 Materials and Methods

A.3.1 Drugs and antibodies

(-)-EGCG was purchased from Calbiochem (EMD Biosciences, Inc., San Diego,

CA), and dissolved in DMSO as a stock solution. The following antibodies were used for immunoblotting: Akt, phospho-ERK1/2 (Thr202/Tyr204), ERK1/2 (p44/42 MAPK) (Cell

149

Signaling, Beverly, MA), Hop (Assay Designs, Inc., Ann Arbor, MI), p23 (Abcam,

Cambridge, MA), Cdk4, Cdc37, Hsp90, Hsp70, Hsc70, Her-2, Raf-1, β-actin (Santa Cruz

Biotechnology, Santa Cruz, CA). Purified Hsp90 β N-terminus (N-Hsp90 β) (amino acids

1-246) was a gift from Dr. Dan Bolon (University of Massachusetts Medical School).

A.3.2 MTS cell proliferation assay

Human pancreatic cancer cells, Mia Paca-2, were seeded in 96-well microplates at a density of 3,000 to 5,000 cells per well. Cells were treated with increasing concentrations of (-)-EGCG as indicated, and after 24 h incubation cell viability was assessed by MTS assay (Promega, Madison, WI) according to the manufacturer’s instruction. The number of living cells in the culture is directly proportional to the absorbance at 490 nm by a formazan product bioreduced from MTS by living cells. T he anti-proliferative effect of (-)-EGCG was also tested on pancreatic cancer cell lines

(Panc-1, BxPC-3, and AsPC-1) with similar results, and thus only one cell line (Mia

Paca-2) was used for the following mechanistic studies.

A.3.3 Caspase-3 activity assay

Mia Paca-2 cells were treated with (-)-EGCG and collected at different time

points as indicated. The following Caspase-3 activity assay was based on the

manufacturer’s instruction of Caspase-3/CPP32 Fluorometric Assay Kit (Biovision

Research Products, Mountain View, CA). Cellular protein was extracted with the

supplied lysis buffer, followed by determination of protein concentration using BCA

Protein Assay Reagents (Pierce, Rockford, IL). The cleavage of DEVD-AFC, a substrate

of caspase-3, was quantified by using a fluorescence microtiter plate reader with a 400

150

nm excitation filter and a 505 nm emission filter. Results are reported as arbitrary fluorescence units (AFU) normalized to milligram of cellular protein.

A.3.4 Protein expression and purification

The expression plasmids pET15b-hHsp90 β, pET28a(+)-hHsp90 β (530-724) for human full-length Hsp90 β and Hsp90 β C-terminus (C-Hsp90 β) were kindly provided by

Dr. Thomas Ratajczak (University of Western Australia, Australia). The plasmids were transformed into E.coli strain Rosetta 2(DE3) (EMD Biosciences, Inc., San Diego, CA) following the protocol provided by manufacturer. Primary cultures of transformed cells were grown overnight, pelleted by centrifugation, resuspended in fresh culture medium, and grown for 1-2 h at 37°C until OD 600 reached 0.6. Protein expression was induced by

0.2 mM IPTG (isopropyl-beta-D-thiogalactopyranoside) (GE Healthcare, Piscataway, NJ)

for 2 h. His-tagged proteins were purified by affinity chromatography through mixed

with HisPurTM Cobalt Resin (Pierce, Rockford, IL), followed by dialysis against PBS.

The purity was assessed by SDS-PAGE, and the concentration was determined by BCA

assay (Pierce, Rockford, IL). Proteins were stored at -70°C after adding glycerol to 10%.

A.3.5 Western blotting

The procedure for Western blotting analysis was briefly described below. After

treated with (-)-EGCG for the indicated time periods, Mia Paca-2 cells were washed

twice with ice-cold PBS, collected in RIPA lysis buffer (20 mM Tris-HCl, 150 mM

NaCl, 1% NP-40, 5 mM EDTA, 1 mM Na 3VO 4, pH 7.5) supplemented with a protease

inhibitor mixture (Sigma-Aldrich, St. Louis, MO), and incubated on ice for 20 min.

Afterwards cell lysate was centrifuged at 14,000 x rpm for 10 min, and the supernatant

151

was recovered. Protein concentration was determined with BCA Protein Assay Reagents

(Pierce, Rockford, IL). Equal amounts of total protein were subject to SDS-PAGE, transferred to PVDF membrane (BioRad, Richmond, CA), and then probed with appropriate antibodies.

A.3.6 ATP-sepharose binding assay

Hsp90 Protein (200 µg) was extracted from treated Mia Paca-2 cells and incubated with 25 µl pre-equilibrated γ-phosphate-linked ATP-Sepharose (Jena

Bioscience GmbH, Jena, Germany) in 200 µl incubation buffer (10 mM Tris-HCl, 50 mM

KCl, 5 mM MgCl 2, 2 mM DTT, 0.01% NP-40, pH 7.5) overnight at 4 °C. The beads

were washed four times and bead-bound proteins were subsequently analyzed by SDS-

PAGE. For ATP binding assay with purified protein, 5 µg protein was pre-incubated with

(-)-EGCG on ice in 200 µl incubation buffer (10 mM Tris-HCl, 50 mM KCl, 5 mM

MgCl 2, 2 mM DTT, 0.01% NP-40, pH 7.5) for 1 h. Following incubation, ATP-

Sepharose was added and further incubated at 37 °C for 30 min with frequent agitation.

The beads were washed and bound proteins were subject to Western blotting.

A.3.7 Hsp90 co-immunoprecipitation

Mia Paca-2 cells were treated with (-)-EGCG for the indicated time period, and

then harvested. Cells were lysed in 20 mM Tris-HCl (pH 7.4), 25 mM NaCl, 2 mM DTT,

20 mM Na 2MoO 4, 0.1% NP-40, and protease inhibitors. After centrifugation, supernatant was recovered and protein concentrations were determined with BCA Protein Assay

Reagents (Pierce, Rockford, IL). Protein (500 µg) was first incubated with H9010 antibody (Axxora, San Diego, CA) followed by addition of protein A/G agarose (Santa

152

Cruz Biotechnology, Santa Cruz, CA). The bound proteins were resolved by SDS-PAGE and analyzed by Western blotting.

A.3.8 Trypsinolytic fingerprinting assay

The experiment was performed similarly to previously described [368]. Purified human N-Hsp90 β (1-246) and C-Hsp90 β (530-724) (0.5 µg) was incubated with DMSO,

(-)-EGCG or other compounds in assay buffer (10 mM Tris-HCl, 50 mM KCl, 5 mM

MgCl 2, 0.1 mM EDTA, pH 7.4) on ice for 1 h. The samples were digested on ice with different concentrations of trypsin for 6 min. The reactions were terminated by adding

SDS sample buffer followed by boiling for 3-5 min. The digested products from N-

Hsp90 β and C-Hsp90 β were analyzed by Western blotting with Hsp90 (N-17) antibody

(Santa Cruz Biotechnology, Santa Cruz, CA) and Hsp90 (AC88) antibody (Assay

Designs, Inc., Ann Arbor, MI), respectively.

A.3.9 Statistical analysis

Statistical analysis was performed using student t-test. Data are presented as mean

± SD (n = 3, p < 0.01).

A.4 Results

A.4.1 (-)-EGCG inhibits cell growth and induces apoptosis in pancreatic cancer cells

(Mia Paca-2)

First, we selected a human pancreatic cancer cell line, Mia Paca-2, to evaluate the therapeutic potential of (-)-EGCG. As shown in Fig. A.1A, (-)-EGCG exhibited a dose-

153

dependent inhibitory effect on Mia Paca-2 with IC50 less than 50 µM. One of the primary events in apoptosis is activation of caspase-3 [373]. Caspase-3 activity assay showed that (-)-EGCG induced activation of caspase-3 in a time- and dose-dependent manner (Fig. A.1B). Markedly, more than a 3-fold increase in caspase-3 activity was observed in Mia Paca-2 cells after incubated with 60 µM (-)-EGCG for 48 h in comparison with untreated cells.

A.4.2 (-)-EGCG decreases cellular levels of Hsp90 client proteins

Since the inhibition of Hsp90 will result in simultaneous down-regulation of multiple oncogenic proteins, we examined whether (-)-EGCG could decrease the levels of cancer-associated Hsp90 client proteins in pancreatic cancer cells. Mia Paca-2 cells were treated with either 80 µM (-)-EGCG for 0-24 h or 60 µM (-)-EGCG every 24 h up to 72 h. As shown in Figure 2A and 2B, (-)-EGCG caused a progressive decline in the protein levels of Her-2, Akt, Cdk4, Raf-1, and pERK in a time- and dose-dependent manner. Akt, Raf-1 and pERK were down-regulated by 35%~50% as early as 3 h after 80

µM (-)-EGCG incubation (Fig. A.2A). All five client proteins were decreased by approximately 70%-80% upon 24 h incubation with 80 µM (-)-EGCG (Fig. A.2A).

Moreover, although a 48 h treatment of Mia Paca-2 cells with 60 µM (-)-EGCG only moderately decreased the cellular levels of Akt, Raf-1, Her-2, Cdk4, and pERK, an additional 24 h treatment with another dose of 60 µM (-)-EGCG was able to completely abrogate the endogenous levels of Akt, Cdk4, and Her-2 and down-regulate Raf-1 and pERK by about 70% (Fig. A.2B). In contrast to ansamycin inhibitors of Hsp90 (e.g., GA,

17-AAG), (-)-EGCG did not induce the protein level of Hsp70 (Fig. A.2B).

154

A.4.3 (-)-EGCG impairs the association of cochaperones p23 and Hsc70 with Hsp90 in pancreatic cancer cells (Mia Paca-2)

We further characterized the effect of (-)-EGCG on the association between cochaperones and Hsp90. After incubation with different concentrations of (-)-EGCG for various time periods, Mia Paca-2 cells were harvested for extraction of total cellular protein. Hsp90 was immunoprecipitated using its antibody, and the amounts of Hsc70,

Hop, Cdc37, and p23 were detected by Western blotting in the precipitated Hsp90 complexes. The result showed that 24 h treatment with 60 µM (-)-EGCG significantly suppressed the interaction of Hsc70 and p23 with Hsp90 by approximately 60% and 55%, respectively, while this treatment had little effect on the amount of Cdc37 or Hop in the

Hsp90 complex (Fig. A.3). Higher concentrations of (-)-EGCG further reduced

Hsc70/Hsp90 and p23/Hsp90 associations (Fig. A.3).

A.4.4 (-)-EGCG directly binds the C-terminal region of Hsp90

In order to examine how (-)-EGCG impairs the association of p23 and Hsc70 with

Hsp90, we utilized proteolytic fingerprinting assay to investigate the region that is involved in the interaction between (-)-EGCG and Hsp90. In the absence of (-)-EGCG, the C-terminal region of Hsp90 β (C-Hsp90 β, amino acids 530-724) was highly sensitive to proteolytic enzyme digestion. While 30 µg/ml of trypsin yielded a single band close to

6 kD, 150 µg/ml of the enzyme was able to completely hydrolyze the C-Hsp90 β (Fig.

A.4A, lane 1-3). Consistent with the fact that GA does not interact directly with the C- terminal domain of Hsp90 β, GA had no effect on the trypsinolytic fingerprint of C-

Hsp90 β in comparison to control (Fig. A.4A, lane 7-9). However, incubation of (-)-

155

EGCG with Hsp90 blocked the trypsin hydrolysis of Hsp90 and produced a band representing complete C-Hsp90 β at a lower concentration of trypsin (Fig. A.4A, lane 4 &

5). (-)-EGCG protected C-terminus from cleavage by a higher concentration of trypsin

(Fig. A.4A, lane 4 & 6). As a positive control, binding of ATP to the C-Hsp90 β allowed

very limited trypsin digestion to occur (Fig. A.4B). In addition, the trypsinolytic patterns

were different between (-)-EGCG- and ATP-protected C-Hsp90 β. This suggests that the

binding site of (-)-EGCG on C-Hsp90 β may be different from the C-terminal ATP binding site.

In contrast, (-)-EGCG did not change the trypsinolytic fingerprint of N-Hsp90 β, either with the cleavage by 30 or 150 µg/ml trypsin compared to control (Fig. A.4C, lane

1-6). As expected, Geldanamycin (GA) remarkably protected the enzyme digestion by interacting with the N-terminal nucleotide binding pocket, producing a digest pattern similar to ATP-bound N-Hsp90 β (Fig. A.4C, lane 7-12).

Next, we examined the effect of (-)-EGCG on ATP binding capacity of cellular

Hsp90, recombinant full-length Hsp90 β, and purified C-Hsp90 β by utilizing ATP- sepharose pull-down assay. The results showed that (-)-EGCG treatment had little effect on the ATP binding to endogenous Hsp90 in pancreatic cancer cells (Fig. A.5A). (-)-

EGCG did not block the ATP binding to either recombinant full-length Hsp90 β or C-

Hsp90 β (Fig. A.5B & Fig. A.5C).

156

A.5 Discussion

In recent years, many studies have shown chemopreventive and chemotherapeutic effect of green tea against skin, lung, breast, colon, liver, stomach, and prostate cancers

[374]. Numerous studies have suggested that (-)-EGCG, the most abundant catechin in green tea, is the primary component for these activities [375]. (-)-EGCG induces apoptosis and cell cycle arrest in cancer cells without affecting normal cells [374, 376].

The majority in vitro studies have revealed that (-)-EGCG inhibited NF-κB activity,

MAPK pathway, activator protein-1 (AP-1) activity, and EGFR-mediated downstream signaling pathways [274]. Clinical trials further verified the cancer preventive effect of (-

)-EGCG [371, 377]. The purpose of the current study is to reveal a new chemopreventive mechanism of (-)-EGCG against pancreatic cancer cells.

Recently, Hsp90 has emerged as a target in cancer therapeutics based on the

Hsp90 super-chaperone complex status in cancer cells. First, Hsp90 is involved in the maturation and stabilization of a wide range of oncogenic client proteins that are crucial for oncogenesis and malignant progression [96, 378]. Second, Hsp90 comprises as much as 4-6% of total protein in tumor cells, in contrast with the 1-2% within normal cells

[379]. Finally, Hsp90 predominantly exists as multi-chaperone complex with high affinity for ATP and drug, whereas in normal cells most Hsp90 is present in an uncomplexed state [379]. hence, cancer cells are dependent on Hsp90 function for their survival and proliferation [379].

157

The first class of Hsp90 inhibitors, represented by GA, competitively binds to the

N-terminal ATP pocket of Hsp90, thus restraining Hsp90 in its ADP-bound conformation and preventing the subsequent “clamping” of Hsp90 around a client protein [119-121], resulting in proteasome-dependent degradation of the client. Another type of Hsp90 inhibitor, novobiocin, interacts with Hsp90 at the C-terminal ATP binding site with relatively weak activity [367]. Inhibition of Hsp90 by novobiocin was able to induce similar cellular responses as N-terminal inhibitors, i.e., destabilization of a range of

Hsp90 client proteins such as Her-2, Raf-1 and p53 mutant [367, 380]. In addition, novobiocin interferes with Hsp90/Hsc70 and Hsp90/p23 association [367, 368].

Furthermore, it was suggested that an allosteric regulation may correlate the C-terminal domain of Hsp90 with the N-terminus, where the interaction of ligands with one site might affect the occupancy of the other site [367, 369, 381].

(-)-EGCG was reported to bind the C-terminus of Hsp90 at the region of amino acids 538-728 on Hsp90; this interaction region was discovered by using affinity chromatography with immobilized (-)-EGCG-sepharose and various purified fragments and truncation mutants of Hsp90 [372]. Because of the similar binding region of novobiocin and (-)-EGCG on Hsp90, in the current study we aim to investigate whether

(-)-EGCG: (1) impairs Hsp90 association with its cochaperones, (2) interferes with

Hsp90 chaperoning function, and (3) exerts inhibitory effect on pancreatic cancer cells.

Indeed, the data suggest that binding of (-)-EGCG to Hsp90 impairs the association of

Hsp90 with its cochaperones (Hsc70 and p23), thereby inducing degradation of Hsp90 client proteins, resulting in anti-proliferating effects in pancreatic cancer cells.

158

Identification of the (-)-EGCG binding site on Hsp90 is the key to understand its effects on Hsp90 function. Proteolytic fingerprinting assay revealed that (-)-EGCG directly bound the purified Hsp90 β C-terminus but not N-terminus. Both ATP and (-)-

EGCG could prevent C-Hsp90 β from trypsin cleavage, although they exhibited different protection patterns. These data suggest that (-)-EGCG may bind to Hsp90 C-terminus differently from ATP binding. A very recent study by using mass spectrometry and chemical detection methods discovered that (-)-EGCG could form covalent adducts with the thiol group of cysteine residues in proteins through autoxidation [382]. Based on the amino acid sequence search (NP 005339), there are three cysteine residues within C- terminal fragment of human Hsp90 β (residues 530-724), Cys572, Cys597, and Cys598, all of which do not fall into the C-terminal ATP binding region (residues 663-676). The possible reactions between (-)-EGCG and Hsp90 β C-terminus need to be further

investigated.

To further validate these assays, we used geldanamycin (GA) as a negative control. Since GA competes with ATP binding to N-terminal pocket of Hsp90 rather than

C-terminus, the trypsin digestion of C-terminus of Hsp90 was not affected by GA. On the contrary, ATP and GA shielded the N-terminus of Hsp90 β from trypsin cleavage with a similar pattern and high efficiency, while (-)-EGCG did not alter the proteolytic fingerprint of Hsp90N-terminus. These data suggest that (-)-EGCG directly binds to the

C-terminal domain of Hsp90 β, specifically within the region of amino acids 530-724,

which is supported by the previous study using immobilized (-)-EGCG affinity

chromatography [372].

159

In the current study, co-immunoprecipitation of endogenous Hsp90 from cell lysates demonstrated that (-)-EGCG impaired the association of Hsc70 and p23 with

Hsp90. This effect of (-)-EGCG is very similar to that of novobiocin on

Hsp90/cochaperone association as previously reported [367, 368], which provides the biologic significance of (-)-EGCG binding to Hsp90 for its chemoprevention efficacy.

Considering the similarity between (-)-EGCG and novobiocn in Hsp90 binding, the influence of (-)-EGCG on association of Hsp90 with the cochaperones may be expected.

First, according to Marcu et al. [367], the association region of Hsc70 on Hsp90 overlaps with the C-terminal dimerization domain and contains the novobiocin binding site [383].

In addition, both N- and C-terminal regions of Hsp90 are necessary for interaction with the yeast homolog of p23, SBA1 [384]. Finally, Allan et al. [380] suggests that modulation within the Hsp90 C-terminus by novobiocin could significantly impact other regions in Hsp90, probably through allosteric effects [380]. This is consistent with the evidence that an allosteric regulation may influence the conformation of the C-terminus and N-terminus of Hsp90, where the interaction of ligands with one site might affect the occupancy of the other site [367, 369, 380, 381]. Therefore, (-)-EGCG is likely to alter the conformation and/or occupy the necessary residues of Hsp90 by directly binding to its

C-terminal domain, subsequently leading to the impairment of Hsp90/Hsc70 and

Hsp90/p23 interactions.

In order to assess whether (-)-EGCG affects ATP binding activity of Hsp90, we applied ATP-sepharose binding assay. However, (-)-EGCG treated Mia Paca-2 cells did not show altered ATP binding to Hsp90. ATP-sepharose pull-down assay with

160

recombinant full-length Hsp90 and C-Hsp90 β further confirmed this finding. During the

preparation of this manuscript, we located a very recent manuscript by Yin et al. [385],

which reported that (-)-EGCG inhibited the ATP binding to purified Hsp90 and Hsp90 C-

terminus. Presumably, experimental conditions may contribute to this discrenpancy. In

ATP-sepharose binding assay, sodium molybdate (Na 2MoO 4) is a common constituent in the incubation buffer, because molybdate can “freeze” Hsp90 complex in the presence of

ATP. However, we observed that mixing colorless (-)-EGCG and sodium molybdate together immediately appeared brown, which indicated a reaction occurring between these two compounds. Thus, the incubation buffer of the ATP binding assay used in the current study did not contain sodium molybdate.

As a result of direct binding to Hsp90 and interference with cochaperone association to Hsp90, (-)-EGCG exhibited a simultaneous down-regulation of oncogenic

Hsp90 client proteins in Mia Paca-2 cells. Consequently, the cell growth was inhibited and apoptosis was dramatically induced. Unlike ansamycin inhibitors of Hsp90 (e.g., GA,

17-AAG), (-)-EGCG did not significantly induce the increase of Hsp70 even after a prolonged treatment. This is in contrast to GA, since binding of ansamycin drugs usually induces a heat shock response through the release, activation, nuclear localization and trimerization of heat shock factor-1 (HSF-1) [386]. HSF1 binds to heat shock elements

(HSE) to trigger the expression of some stress-responsive proteins such as Hsp70 [386,

387]. This up-regulation of Hsp70 is believed to compromise the Hsp90-targeted drug efficacy by inhibiting apoptosis signaling [386, 387].

161

A.6 Conclusions

In summary, the data presented in this manuscript suggest that (-)-EGCG, a novel

Hsp90 inhibitor, impairs the association of Hsp90/Hsc70 and Hsp90/p23 by directly binding to the C-terminal region of Hsp90, inhibits Hsp90 chaperoning function, and simultaneously degrades multiple cancer-related Hsp90 client proteins. This finding provides a new mechanism for chemoprevention efficacy of (-)-EGCG.

162

Figure A.1. (-)-EGCG inhibites pancreatic cancer proliferation and increases caspase-3 activity

163

Figure A.2. Effect of (-)-EGCG on Hsp90 client proteins

164

Figure A.3. Influence of (-)-EGCG on Hsp90/cochaperones association

165

Figure A.4. (-)-EGCG binds to the C-terminus but not N-terminus of Hsp90

166

Figure A.5. Effect of (-)-EGCG on ATP binding to Hsp90

167

References

1. Clarke, J.D., R.H. Dashwood, and E. Ho, Multi-targeted prevention of cancer by sulforaphane. Cancer Lett, 2008. 269 (2): p. 291-304.

2. Zhang, Y., et al., A major inducer of anticarcinogenic protective enzymes from broccoli: isolation and elucidation of structure. Proc Natl Acad Sci U S A, 1992. 89 (6): p. 2399-403.

3. Fahey, J.W., A.T. Zalcmann, and P. Talalay, The chemical diversity and distribution of glucosinolates and isothiocyanates among plants. Phytochemistry, 2001. 56 (1): p. 5-51.

4. Herr, I. and M.W. Buchler, Dietary constituents of broccoli and other cruciferous vegetables: implications for prevention and therapy of cancer. Cancer Treat Rev. 36 (5): p. 377-83.

5. Ratzka, A., et al., Disarming the mustard oil bomb. Proc Natl Acad Sci U S A, 2002. 99 (17): p. 11223-8.

6. Shapiro, T.A., et al., Human metabolism and excretion of cancer chemoprotective glucosinolates and isothiocyanates of cruciferous vegetables. Cancer Epidemiol Biomarkers Prev, 1998. 7(12): p. 1091-100.

7. Matusheski, N.V., et al., Epithiospecifier protein from broccoli (Brassica oleracea L. ssp. italica) inhibits formation of the anticancer agent sulforaphane. J Agric Food Chem, 2006. 54 (6): p. 2069-76.

8. vallejo, F., F.A. tomas-Barberan, and C. Garcia-Viguera, Glucosinolates and vitamin C content in edible parts of broccoli florets after domestic cooking. Eur. Food Res. Technol., 2002. 215 : p. 310-316.

168

9. Rosa, E.A.S., et al., Glucosinolates in crop plants. Hortic. Rev., 1997. 19 : p. 99- 215.

10. Van Eylen, D., et al., Kinetics of the stability of broccoli (Brassica oleracea Cv. Italica) myrosinase and isothiocyanates in broccoli juice during pressure/temperature treatments. J Agric Food Chem, 2007. 55 (6): p. 2163-70.

11. Jin, Y., et al., Thermal degradation of sulforaphane in aqueous solution. J Agric Food Chem, 1999. 47 (8): p. 3121-3.

12. Fahey, J.W., Y. Zhang, and P. Talalay, Broccoli sprouts: an exceptionally rich source of inducers of enzymes that protect against chemical carcinogens. Proc Natl Acad Sci U S A, 1997. 94 (19): p. 10367-72.

13. Shapiro, T.A., et al., Safety, tolerance, and metabolism of broccoli sprout glucosinolates and isothiocyanates: a clinical phase I study. Nutr Cancer, 2006. 55 (1): p. 53-62.

14. Fahey, J.W., et al., Sulforaphane inhibits extracellular, intracellular, and antibiotic-resistant strains of Helicobacter pylori and prevents benzo[a]pyrene- induced stomach tumors. Proc Natl Acad Sci U S A, 2002. 99 (11): p. 7610-5.

15. Zhang, Y., et al., Anticarcinogenic activities of sulforaphane and structurally related synthetic norbornyl isothiocyanates. Proc Natl Acad Sci U S A, 1994. 91 (8): p. 3147-50.

16. Chung, F.L., et al., Chemoprevention of colonic aberrant crypt foci in Fischer rats by sulforaphane and phenethyl isothiocyanate. Carcinogenesis, 2000. 21 (12): p. 2287-91.

17. Xu, C., et al., Inhibition of 7,12-dimethylbenz(a)anthracene-induced skin tumorigenesis in C57BL/6 mice by sulforaphane is mediated by nuclear factor E2-related factor 2. Cancer Res, 2006. 66 (16): p. 8293-6.

18. Conaway, C.C., et al., Phenethyl isothiocyanate and sulforaphane and their N- acetylcysteine conjugates inhibit malignant progression of lung adenomas

169

induced by tobacco carcinogens in A/J mice. Cancer Res, 2005. 65 (18): p. 8548- 57.

19. Kuroiwa, Y., et al., Protective effects of benzyl isothiocyanate and sulforaphane but not resveratrol against initiation of pancreatic carcinogenesis in hamsters. Cancer Lett, 2006. 241 (2): p. 275-80.

20. Gills, J.J., et al., Sulforaphane prevents mouse skin tumorigenesis during the stage of promotion. Cancer Lett, 2006. 236 (1): p. 72-9.

21. Singh, A.V., et al., Sulforaphane induces caspase-mediated apoptosis in cultured PC-3 human prostate cancer cells and retards growth of PC-3 xenografts in vivo. Carcinogenesis, 2004. 25 (1): p. 83-90.

22. Jackson, S.J. and K.W. Singletary, Sulforaphane: a naturally occurring mammary carcinoma mitotic inhibitor, which disrupts tubulin polymerization. Carcinogenesis, 2004. 25 (2): p. 219-27.

23. Pham, N.A., et al., The dietary isothiocyanate sulforaphane targets pathways of apoptosis, cell cycle arrest, and oxidative stress in human pancreatic cancer cells and inhibits tumor growth in severe combined immunodeficient mice. Mol Cancer Ther, 2004. 3(10): p. 1239-48.

24. Dinkova-Kostova, A.T., et al., Protection against UV-light-induced skin carcinogenesis in SKH-1 high-risk mice by sulforaphane-containing broccoli sprout extracts. Cancer Lett, 2006. 240 (2): p. 243-52.

25. Hu, R., et al., Cancer chemoprevention of intestinal polyposis in ApcMin/+ mice by sulforaphane, a natural product derived from cruciferous vegetable. Carcinogenesis, 2006. 27 (10): p. 2038-46.

26. Ambrosone, C.B., et al., Breast cancer risk in premenopausal women is inversely associated with consumption of broccoli, a source of isothiocyanates, but is not modified by GST genotype. J Nutr, 2004. 134 (5): p. 1134-8.

27. Cornblatt, B.S., et al., Preclinical and clinical evaluation of sulforaphane for chemoprevention in the breast. Carcinogenesis, 2007. 28 (7): p. 1485-90.

170

28. Kensler, T.W., et al., Effects of glucosinolate-rich broccoli sprouts on urinary levels of aflatoxin-DNA adducts and phenanthrene tetraols in a randomized clinical trial in He Zuo township, Qidong, People's Republic of China. Cancer Epidemiol Biomarkers Prev, 2005. 14 (11 Pt 1): p. 2605-13.

29. Juge, N., R.F. Mithen, and M. Traka, Molecular basis for chemoprevention by sulforaphane: a comprehensive review. Cell Mol Life Sci, 2007. 64 (9): p. 1105- 27.

30. Zhang, Y. and L. Tang, Discovery and development of sulforaphane as a cancer chemopreventive phytochemical. Acta Pharmacol Sin, 2007. 28 (9): p. 1343-54.

31. Cheung, K.L. and A.N. Kong, Molecular targets of dietary phenethyl isothiocyanate and sulforaphane for cancer chemoprevention. AAPS J. 12 (1): p. 87-97.

32. Wogan, G.N., et al., Environmental and chemical carcinogenesis. Semin Cancer Biol, 2004. 14 (6): p. 473-86.

33. Maheo, K., et al., Inhibition of cytochromes P-450 and induction of glutathione S- transferases by sulforaphane in primary human and rat hepatocytes. Cancer Res, 1997. 57 (17): p. 3649-52.

34. Kensler, T.W., Chemoprevention by inducers of carcinogen detoxication enzymes. Environ Health Perspect, 1997. 105 Suppl 4 : p. 965-70.

35. Chen, C. and A.N. Kong, Dietary cancer-chemopreventive compounds: from signaling and gene expression to pharmacological effects. Trends Pharmacol Sci, 2005. 26 (6): p. 318-26.

36. Brooks, J.D., V.G. Paton, and G. Vidanes, Potent induction of phase 2 enzymes in human prostate cells by sulforaphane. Cancer Epidemiol Biomarkers Prev, 2001. 10 (9): p. 949-54.

37. Keck, A.S., Q. Qiao, and E.H. Jeffery, Food matrix effects on bioactivity of broccoli-derived sulforaphane in liver and colon of F344 rats. J Agric Food Chem, 2003. 51 (11): p. 3320-7.

171

38. Jones, S.B. and J.D. Brooks, Modest induction of phase 2 enzyme activity in the F-344 rat prostate. BMC Cancer, 2006. 6: p. 62.

39. Nguyen, T., P. Nioi, and C.B. Pickett, The Nrf2-antioxidant response element signaling pathway and its activation by oxidative stress. J Biol Chem, 2009. 284 (20): p. 13291-5.

40. Itoh, K., et al., Keap1 regulates both cytoplasmic-nuclear shuttling and degradation of Nrf2 in response to electrophiles. Genes Cells, 2003. 8(4): p. 379- 91.

41. Park, S.Y., et al., Induction of apoptosis by isothiocyanate sulforaphane in human cervical carcinoma HeLa and hepatocarcinoma HepG2 cells through activation of caspase-3. Oncol Rep, 2007. 18 (1): p. 181-7.

42. Choi, S., et al., D,L-Sulforaphane-induced cell death in human prostate cancer cells is regulated by inhibitor of apoptosis family proteins and Apaf-1. Carcinogenesis, 2007. 28 (1): p. 151-62.

43. Singh, S.V., et al., Sulforaphane-induced cell death in human prostate cancer cells is initiated by reactive oxygen species. J Biol Chem, 2005. 280 (20): p. 19911-24.

44. Shan, Y., et al., Effect of sulforaphane on cell growth, G(0)/G(1) phase cell progression and apoptosis in human bladder cancer T24 cells. Int J Oncol, 2006. 29 (4): p. 883-8.

45. Chiao, J.W., et al., Sulforaphane and its metabolite mediate growth arrest and apoptosis in human prostate cancer cells. Int J Oncol, 2002. 20 (3): p. 631-6.

46. Wang, L., et al., Targeting cell cycle machinery as a molecular mechanism of sulforaphane in prostate cancer prevention. Int J Oncol, 2004. 24 (1): p. 187-92.

47. Tang, L. and Y. Zhang, Dietary isothiocyanates inhibit the growth of human bladder carcinoma cells. J Nutr, 2004. 134 (8): p. 2004-10.

172

48. Parnaud, G., et al., Mechanism of sulforaphane-induced cell cycle arrest and apoptosis in human colon cancer cells. Nutr Cancer, 2004. 48 (2): p. 198-206.

49. Jackson, S.J. and K.W. Singletary, Sulforaphane inhibits human MCF-7 mammary cancer cell mitotic progression and tubulin polymerization. J Nutr, 2004. 134 (9): p. 2229-36.

50. Gamet-Payrastre, L., et al., Sulforaphane, a naturally occurring isothiocyanate, induces cell cycle arrest and apoptosis in HT29 human colon cancer cells. Cancer Res, 2000. 60 (5): p. 1426-33.

51. Singh, S.V., et al., Sulforaphane-induced G2/M phase cell cycle arrest involves checkpoint kinase 2-mediated phosphorylation of cell division cycle 25C. J Biol Chem, 2004. 279 (24): p. 25813-22.

52. Herman-Antosiewicz, A., et al., Induction of p21 protein protects against sulforaphane-induced mitotic arrest in LNCaP human prostate cancer cell line. Mol Cancer Ther, 2007. 6(5): p. 1673-81.

53. Bertl, E., H. Bartsch, and C. Gerhauser, Inhibition of angiogenesis and endothelial cell functions are novel sulforaphane-mediated mechanisms in chemoprevention. Mol Cancer Ther, 2006. 5(3): p. 575-85.

54. Asakage, M., et al., Sulforaphane induces inhibition of human umbilical vein endothelial cells proliferation by apoptosis. Angiogenesis, 2006. 9(2): p. 83-91.

55. Rose, P., et al., Broccoli and watercress suppress matrix metalloproteinase-9 activity and invasiveness of human MDA-MB-231 breast cancer cells. Toxicol Appl Pharmacol, 2005. 209 (2): p. 105-13.

56. Rayet, B. and C. Gelinas, Aberrant rel/nfkb genes and activity in human cancer. Oncogene, 1999. 18 (49): p. 6938-47.

57. Baldwin, A.S., Control of oncogenesis and cancer therapy resistance by the transcription factor NF-kappaB. J Clin Invest, 2001. 107 (3): p. 241-6.

173

58. Jeong, W.S., et al., Modulatory properties of various natural chemopreventive agents on the activation of NF-kappaB signaling pathway. Pharm Res, 2004. 21 (4): p. 661-70.

59. Xu, C., et al., Suppression of NF-kappaB and NF-kappaB-regulated gene expression by sulforaphane and PEITC through IkappaBalpha, IKK pathway in human prostate cancer PC-3 cells. Oncogene, 2005. 24 (28): p. 4486-95.

60. Myzak, M.C., et al., A novel mechanism of chemoprotection by sulforaphane: inhibition of histone deacetylase. Cancer Res, 2004. 64 (16): p. 5767-74.

61. Myzak, M.C., et al., Sulforaphane inhibits histone deacetylase activity in BPH-1, LnCaP and PC-3 prostate epithelial cells. Carcinogenesis, 2006. 27 (4): p. 811-9.

62. Xu, C., et al., ERK and JNK signaling pathways are involved in the regulation of activator protein 1 and cell death elicited by three isothiocyanates in human prostate cancer PC-3 cells. Carcinogenesis, 2006. 27 (3): p. 437-45.

63. Shapiro, T.A., et al., Chemoprotective glucosinolates and isothiocyanates of broccoli sprouts: metabolism and excretion in humans. Cancer Epidemiol Biomarkers Prev, 2001. 10 (5): p. 501-8.

64. Conaway, C.C., et al., Disposition of glucosinolates and sulforaphane in humans after ingestion of steamed and fresh broccoli. Nutr Cancer, 2000. 38 (2): p. 168- 78.

65. Petri, N., et al., Absorption/metabolism of sulforaphane and quercetin, and regulation of phase II enzymes, in human jejunum in vivo. Drug Metab Dispos, 2003. 31 (6): p. 805-13.

66. Ye, L., et al., Quantitative determination of dithiocarbamates in human plasma, serum, erythrocytes and urine: pharmacokinetics of broccoli sprout isothiocyanates in humans. Clin Chim Acta, 2002. 316 (1-2): p. 43-53.

67. Hu, R., et al., In vivo pharmacokinetics and regulation of gene expression profiles by isothiocyanate sulforaphane in the rat. J Pharmacol Exp Ther, 2004. 310 (1): p. 263-71.

174

68. Zhang, Y., Cancer-preventive isothiocyanates: measurement of human exposure and mechanism of action. Mutat Res, 2004. 555 (1-2): p. 173-90.

69. Kassahun, K., et al., Biotransformation of the naturally occurring isothiocyanate sulforaphane in the rat: identification of phase I metabolites and glutathione conjugates. Chem Res Toxicol, 1997. 10 (11): p. 1228-33.

70. Al Janobi, A.A., et al., Quantitative measurement of sulforaphane, iberin and their mercapturic acid pathway metabolites in human plasma and urine using liquid chromatography-tandem electrospray ionisation mass spectrometry. J Chromatogr B Analyt Technol Biomed Life Sci, 2006. 844 (2): p. 223-34.

71. Zhang, Y., et al., Induction of GST and NQO1 in cultured bladder cells and in the urinary bladders of rats by an extract of broccoli (Brassica oleracea italica) sprouts. J Agric Food Chem, 2006. 54 (25): p. 9370-6.

72. Munday, R., et al., Inhibition of urinary bladder carcinogenesis by broccoli sprouts. Cancer Res, 2008. 68 (5): p. 1593-600.

73. Vermeulen, M., et al., Bioavailability and kinetics of sulforaphane in humans after consumption of cooked versus raw broccoli. J Agric Food Chem, 2008. 56 (22): p. 10505-9.

74. Huang, C.S., et al., Lycopene inhibits matrix metalloproteinase-9 expression and down-regulates the binding activity of nuclear factor-kappa B and stimulatory protein-1. J Nutr Biochem, 2007. 18 (7): p. 449-56.

75. Bardeesy, N. and R.A. DePinho, Pancreatic cancer biology and genetics. Nat Rev Cancer, 2002. 2(12): p. 897-909.

76. Borja-Cacho, D., et al., Molecular targeted therapies for pancreatic cancer. Am J Surg, 2008. 196 (3): p. 430-41.

77. Yanyan Li, T.Z., Hasan Korkaya, Suling Liu, Hsiu-Fang Lee, Bryan Newman, Yanke Yu, Shawn G. Clouthier, Steven J. Schwartz, Max S. Wicha, Duxin Sun, Sulforaphane, a dietary component of broccoli/broccoli sprouts, inhibits breast cancer stem cells. Clin Cancer Res, 2010.

175

78. Yokoyama, Y., Y. Nimura, and M. Nagino, Advances in the treatment of pancreatic cancer: limitations of surgery and evaluation of new therapeutic strategies. Surg Today, 2009. 39 (6): p. 466-75.

79. Mehta, S.P., Palliative chemotherapy for pancreatic malignancies. Surg Clin North Am. 90 (2): p. 365-75.

80. Burris, H.A., 3rd, et al., Improvements in survival and clinical benefit with gemcitabine as first-line therapy for patients with advanced pancreas cancer: a randomized trial. J Clin Oncol, 1997. 15 (6): p. 2403-13.

81. Oettle, H., et al., A phase III trial of pemetrexed plus gemcitabine versus gemcitabine in patients with unresectable or metastatic pancreatic cancer. Ann Oncol, 2005. 16 (10): p. 1639-45.

82. Herrmann, R., et al., Gemcitabine plus capecitabine compared with gemcitabine alone in advanced pancreatic cancer: a randomized, multicenter, phase III trial of the Swiss Group for Clinical Cancer Research and the Central European Cooperative Oncology Group. J Clin Oncol, 2007. 25 (16): p. 2212-7.

83. Colucci, G., et al., Gemcitabine alone or with cisplatin for the treatment of patients with locally advanced and/or metastatic pancreatic carcinoma: a prospective, randomized phase III study of the Gruppo Oncologia dell'Italia Meridionale. Cancer, 2002. 94 (4): p. 902-10.

84. Poplin, E., et al., Phase III, randomized study of gemcitabine and oxaliplatin versus gemcitabine (fixed-dose rate infusion) compared with gemcitabine (30- minute infusion) in patients with pancreatic carcinoma E6201: a trial of the Eastern Cooperative Oncology Group. J Clin Oncol, 2009. 27 (23): p. 3778-85.

85. Berlin, J.D., et al., Phase III study of gemcitabine in combination with fluorouracil versus gemcitabine alone in patients with advanced pancreatic carcinoma: Eastern Cooperative Oncology Group Trial E2297. J Clin Oncol, 2002. 20 (15): p. 3270-5.

86. Di Marco, M., et al., Metastatic pancreatic cancer: is gemcitabine still the best standard treatment? (Review). Oncol Rep. 23 (5): p. 1183-92.

176

87. Stathopoulos, G.P., et al., A multicenter phase III trial comparing irinotecan- gemcitabine (IG) with gemcitabine (G) monotherapy as first-line treatment in patients with locally advanced or metastatic pancreatic cancer. Br J Cancer, 2006. 95 (5): p. 587-92.

88. Zerbi, A., et al., Intraoperative radiation therapy adjuvant to resection in the treatment of pancreatic cancer. Cancer, 1994. 73 (12): p. 2930-5.

89. Wong, H.H. and N.R. Lemoine, Pancreatic cancer: molecular pathogenesis and new therapeutic targets. Nat Rev Gastroenterol Hepatol, 2009. 6(7): p. 412-22.

90. Li, D., et al., Pancreatic cancer. Lancet, 2004. 363 (9414): p. 1049-57.

91. Van Cutsem, E., et al., Phase III trial of gemcitabine plus tipifarnib compared with gemcitabine plus placebo in advanced pancreatic cancer. J Clin Oncol, 2004. 22 (8): p. 1430-8.

92. Yamanaka, Y., et al., Coexpression of epidermal growth factor receptor and ligands in human pancreatic cancer is associated with enhanced tumor aggressiveness. Anticancer Res, 1993. 13 (3): p. 565-9.

93. Moore, M.J., et al., Erlotinib plus gemcitabine compared with gemcitabine alone in patients with advanced pancreatic cancer: a phase III trial of the National Cancer Institute of Canada Clinical Trials Group. J Clin Oncol, 2007. 25 (15): p. 1960-6.

94. Seo, Y., et al., High expression of vascular endothelial growth factor is associated with liver metastasis and a poor prognosis for patients with ductal pancreatic adenocarcinoma. Cancer, 2000. 88 (10): p. 2239-45.

95. Smith, J.R. and P. Workman, Targeting CDC37: an alternative, kinase-directed strategy for disruption of oncogenic chaperoning. Cell Cycle, 2009. 8(3): p. 362- 72.

96. Kamal, A., M.F. Boehm, and F.J. Burrows, Therapeutic and diagnostic implications of Hsp90 activation. Trends Mol Med, 2004. 10 (6): p. 283-90.

177

97. Welch, W.J., The role of heat-shock proteins as molecular chaperones. Curr Opin Cell Biol, 1991. 3(6): p. 1033-8.

98. Welch, W.J. and J.R. Feramisco, Purification of the major mammalian heat shock proteins. J Biol Chem, 1982. 257 (24): p. 14949-59.

99. Wiech, H., et al., Hsp90 chaperones protein folding in vitro. Nature, 1992. 358 (6382): p. 169-70.

100. McClellan, A.J., et al., Diverse cellular functions of the Hsp90 molecular chaperone uncovered using systems approaches. Cell, 2007. 131 (1): p. 121-35.

101. Zhao, R., et al., Navigating the chaperone network: an integrative map of physical and genetic interactions mediated by the hsp90 chaperone. Cell, 2005. 120 (5): p. 715-27.

102. Falsone, S.F., et al., A proteomic snapshot of the human heat shock protein 90 interactome. FEBS Lett, 2005. 579 (28): p. 6350-4.

103. Pearl, L.H., C. Prodromou, and P. Workman, The Hsp90 molecular chaperone: an open and shut case for treatment. Biochem J, 2008. 410 (3): p. 439-53.

104. Didelot, C., et al., Anti-cancer therapeutic approaches based on intracellular and extracellular heat shock proteins. Curr Med Chem, 2007. 14 (27): p. 2839-47.

105. Pearl, L.H. and C. Prodromou, Structure and mechanism of the Hsp90 molecular chaperone machinery. Annu Rev Biochem, 2006. 75 : p. 271-94.

106. Dutta, R. and M. Inouye, GHKL, an emergent ATPase/kinase superfamily. Trends Biochem Sci, 2000. 25 (1): p. 24-8.

107. Hawle, P., et al., The middle domain of Hsp90 acts as a discriminator between different types of client proteins. Mol Cell Biol, 2006. 26 (22): p. 8385-95.

108. Terasawa, K., M. Minami, and Y. Minami, Constantly updated knowledge of Hsp90. J Biochem, 2005. 137 (4): p. 443-7. 178

109. Richter, K., et al., Conserved conformational changes in the ATPase cycle of human Hsp90. J Biol Chem, 2008. 283 (26): p. 17757-65.

110. Wandinger, S.K., K. Richter, and J. Buchner, The Hsp90 chaperone machinery. J Biol Chem, 2008. 283 (27): p. 18473-7.

111. Pearl, L.H., Hsp90 and Cdc37 -- a chaperone cancer conspiracy. Curr Opin Genet Dev, 2005. 15 (1): p. 55-61.

112. Arlander, S.J., et al., Chaperoning checkpoint kinase 1 (Chk1), an Hsp90 client, with purified chaperones. J Biol Chem, 2006. 281 (5): p. 2989-98.

113. Mandal, A.K., et al., Cdc37 has distinct roles in protein kinase quality control that protect nascent chains from degradation and promote posttranslational maturation. J. Cell Biol., 2007. 176 : p. 319-328.

114. Roe, S.M., et al., The Mechanism of Hsp90 regulation by the protein kinase- specific cochaperone p50(cdc37). Cell, 2004. 116 (1): p. 87-98.

115. Siligardi, G., et al., Regulation of Hsp90 ATPase activity by the co-chaperone Cdc37p/p50cdc37. J Biol Chem, 2002. 277 (23): p. 20151-9.

116. Whitesell, L., et al., Inhibition of heat shock protein HSP90-pp60v-src heteroprotein complex formation by benzoquinone ansamycins: essential role for stress proteins in oncogenic transformation. Proc Natl Acad Sci U S A, 1994. 91 (18): p. 8324-8.

117. Supko, J.G., et al., Preclinical pharmacologic evaluation of geldanamycin as an antitumor agent. Cancer Chemother. Pharmacol., 1995. 36 : p. 305-315.

118. Schulte, T.W., Neckers, L.M., The benzoquinone ansamycin 17-allylamino-17- demethoxygeldanamycin binds to HSP90 and shares important biologic activities with geldanamycin. . Cancer Chemother. Pharmacol., 1998. 42 : p. 273-279.

119. Roe, S.M., et al., Structural basis for inhibition of the Hsp90 molecular chaperone by the antitumor antibiotics radicicol and geldanamycin. J Med Chem, 1999. 42 (2): p. 260-6. 179

120. Blagg, B.S. and T.D. Kerr, Hsp90 inhibitors: small molecules that transform the Hsp90 protein folding machinery into a catalyst for protein degradation. Med Res Rev, 2006. 26 (3): p. 310-38.

121. Neckers, L., Chaperoning oncogenes: Hsp90 as a target of geldanamycin. Handb Exp Pharmacol, 2006(172): p. 259-77.

122. Mimnaugh, E.G., C. Chavany, and L. Neckers, Polyubiquitination and proteasomal degradation of the p185c-erbB-2 receptor protein-tyrosine kinase induced by geldanamycin. J Biol Chem, 1996. 271 (37): p. 22796-801.

123. Neckers, L., T.W. Schulte, and E. Mimnaugh, Geldanamycin as a potential anticancer agent: its molecular target and biochemical activity. Invest New Drugs, 1999. 17 (4): p. 361-73.

124. Workman, P., et al., Drugging the cancer chaperone HSP90: combinatorial therapeutic exploitation of oncogene addiction and tumor stress. Ann N Y Acad Sci, 2007. 1113 : p. 202-16.

125. Solit, D.B., et al., Phase I trial of 17-allylamino-17-demethoxygeldanamycin in patients with advanced cancer. Clin Cancer Res, 2007. 13 (6): p. 1775-82.

126. Smith, V., et al., Comparison of 17-dimethylaminoethylamino-17-demethoxygeldanamycin (17DMAG) and 17-allylamino-17-demethoxygeldanamycin (17AAG) in vitro: effects on Hsp90 and client proteins in melanoma models. Cancer Chemother Pharmacol, 2005. 56 (2): p. 126-37.

127. Sydor, J.R., et al., Development of 17-allylamino-17-demethoxygeldanamycin hydroquinone hydrochloride (IPI-504), an anti-cancer agent directed against Hsp90. Proc Natl Acad Sci U S A, 2006. 103 (46): p. 17408-13.

128. Banerji, U., et al., Phase I pharmacokinetic and pharmacodynamic study of 17- allylamino, 17-demethoxygeldanamycin in patients with advanced malignancies. J Clin Oncol, 2005. 23 (18): p. 4152-61.

129. Pacey, S., et al., Hsp90 inhibitors in the clinic. Handb Exp Pharmacol, 2006(172): p. 331-58.

180

130. Ronnen, E.A., et al., A phase II trial of 17-(Allylamino)-17- demethoxygeldanamycin in patients with papillary and clear cell renal cell carcinoma. Invest New Drugs, 2006. 24 (6): p. 543-6.

131. Heath, E.I., et al., A phase II trial of 17-allylamino-17- demethoxygeldanamycin in patients with hormone-refractory metastatic prostate cancer. Clin Prostate Cancer, 2005. 4(2): p. 138-41.

132. Solit, D.B. and N. Rosen, Hsp90: a novel target for cancer therapy. Curr Top Med Chem, 2006. 6(11): p. 1205-14.

133. Modi, S., et al., Combination of trastuzumab and tanespimycin (17-AAG, KOS- 953) is safe and active in trastuzumab-refractory HER-2 overexpressing breast cancer: a phase I dose-escalation study. J Clin Oncol, 2007. 25 (34): p. 5410-7.

134. da Rocha Dias, S., et al., Activated B-RAF is an Hsp90 client protein that is targeted by the anticancer drug 17-allylamino-17-demethoxygeldanamycin. Cancer Res, 2005. 65 (23): p. 10686-91.

135. Grbovic, O.M., et al., V600E B-Raf requires the Hsp90 chaperone for stability and is degraded in response to Hsp90 inhibitors. Proc Natl Acad Sci U S A, 2006. 103 (1): p. 57-62.

136. Dai, C. and L. Whitesell, HSP90: a rising star on the horizon of anticancer targets. Future Oncol, 2005. 1(4): p. 529-40.

137. Hollingshead, M., et al., In vivo antitumor efficacy of 17-DMAG (17- dimethylaminoethylamino-17-demethoxygeldanamycin hydrochloride), a water- soluble geldanamycin derivative. Cancer Chemother Pharmacol, 2005. 56 (2): p. 115-25.

138. Peng, C., et al., Inhibition of heat shock protein 90 prolongs survival of mice with BCR-ABL-T315I-induced leukemia and suppresses leukemic stem cells. Blood, 2007. 110 (2): p. 678-85.

139. Soga, S., et al., Development of radicicol analogues. Curr Cancer Drug Targets, 2003. 3(5): p. 359-69.

181

140. Proisy, N., et al., Inhibition of Hsp90 with synthetic macrolactones: synthesis and structural and biological evaluation of ring and conformational analogs of radicicol. Chem Biol, 2006. 13 (11): p. 1203-15.

141. Shiotsu, Y., et al., Novel oxime derivatives of radicicol induce erythroid differentiation associated with preferential G(1) phase accumulation against chronic myelogenous leukemia cells through destabilization of Bcr-Abl with Hsp90 complex. Blood, 2000. 96 (6): p. 2284-91.

142. Yamamoto, K., et al., Total synthesis as a resource in the discovery of potentially valuable antitumor agents: cycloproparadicicol. Angew Chem Int Ed Engl, 2003. 42 (11): p. 1280-4.

143. Neckers, L., Development of small molecule Hsp90 Inhibitors: utilizing both forward and reverse chemical genomics for drug identification. Current Medicinal Chemistry, 2003. 10 (9): p. 733-739.

144. Zhang, T., et al., A novel Hsp90 inhibitor to disrupt Hsp90/Cdc37 complex against pancreatic cancer cells. Mol Cancer Ther, 2008. 7(1): p. 162-70.

145. Gray, P.J., Jr., et al., Targeting the oncogene and kinome chaperone CDC37. Nat Rev Cancer, 2008. 8(7): p. 491-5.

146. Holmes, J.L., et al., Silencing of HSP90 cochaperone AHA1 expression decreases client protein activation and increases cellular sensitivity to the HSP90 inhibitor 17-allylamino-17-demethoxygeldanamycin. Cancer Res, 2008. 68 (4): p. 1188-97.

147. Powers, M.V., P.A. Clarke, and P. Workman, Dual targeting of HSC70 and HSP72 inhibits HSP90 function and induces tumor-specific apoptosis. Cancer Cell, 2008. 14 (3): p. 250-62.

148. Smith, J.R., et al., Silencing the cochaperone CDC37 destabilizes kinase clients and sensitizes cancer cells to HSP90 inhibitors. Oncogene, 2008.

149. Cortajarena, A.L., F. Yi, and L. Regan, Designed TPR modules as novel anticancer agents. ACS Chem Biol, 2008. 3(3): p. 161-6.

182

150. Yi, F. and L. Regan, A novel class of small molecule inhibitors of Hsp90. ACS Chem Biol, 2008. 3(10): p. 645-54.

151. Stepanova, L., et al., Induction of human Cdc37 in prostate cancer correlates with the ability of targeted Cdc37 expression to promote prostatic hyperplasia. Oncogene, 2000. 19 (18): p. 2186-93.

152. Pascale, R.M., et al., Role of HSP90, CDC37, and CRM1 as modulators of P16(INK4A) activity in rat liver carcinogenesis and human liver cancer. Hepatology, 2005. 42 (6): p. 1310-9.

153. Stepanova, L., et al., Mammalian p50(Cdc37) is a protein kinase-targeting subunit of Hsp90 that binds and stabilizes Cdk4. Genes Dev., 1996. 10 : p. 1491- 1502.

154. Stepanova, L., et al., The oncoprotein kinase chaperone CDC37 functions as an oncogene in mice and collaborates with both c-myc and cyclin D1 in transformation of multiple tissues. Mol Cell Biol, 2000. 20 (12): p. 4462-73.

155. Yu, Y., et al., Withaferin A targets heat shock protein 90 in pancreatic cancer cells. Biochem Pharmacol, 2009.

156. Zhang, T., et al., Characterization of celastrol to inhibit HSP90 and CDC37 interaction. J Biol Chem, 2009.

157. de Graaf, H., et al., Issues in the chemotherapy of breast cancer. Neth J Med, 1998. 53 (3): p. 97-108.

158. Moulder, S. and G.N. Hortobagyi, Advances in the treatment of breast cancer. Clin Pharmacol Ther, 2008. 83 (1): p. 26-36.

159. Berry, D.A., et al., Effect of screening and adjuvant therapy on mortality from breast cancer. N Engl J Med, 2005. 353 (17): p. 1784-92.

160. Hortobagyi, G.N., The status of breast cancer management: challenges and opportunities. Breast Cancer Res Treat, 2002. 75 Suppl 1 : p. S61-5; discussion S57-9. 183

161. Fisher, B., et al., 1-Phenylalanine mustard (L-PAM) in the management of primary breast cancer. A report of early findings. N Engl J Med, 1975. 292 (3): p. 117-22.

162. Bonadonna, G., et al., Combination chemotherapy as an adjuvant treatment in operable breast cancer. N Engl J Med, 1976. 294 (8): p. 405-10.

163. Nissen-Meyer, R., et al., Surgical adjuvant chemotherapy: results with one short course with cyclophosphamide after mastectomy for breast cancer. Cancer, 1978. 41 (6): p. 2088-98.

164. Canellos, G.P., et al., Combination chemotherapy for advanced breast cancer: response and effect on survival. Ann Intern Med, 1976. 84 (4): p. 389-92.

165. Jones, S.E., B.G. Durie, and S.E. Salmon, Combination chemotherapy with adriamycin and cyclophosphamide for advanced breast cancer. Cancer, 1975. 36 (1): p. 90-7.

166. Clavel, M. and G. Catimel, Breast cancer: chemotherapy in the treatment of advanced disease. Eur J Cancer, 1993. 29A (4): p. 598-604.

167. Stockler, M., et al., Systematic reviews of chemotherapy and endocrine therapy in metastatic breast cancer. Cancer Treat Rev, 2000. 26 (3): p. 151-68.

168. Stockler, M., N. Wilcken, and A. Coates, Chemotherapy for metastatic breast cancer--when is enough enough? Eur J Cancer, 1997. 33 (13): p. 2147-8.

169. Clark, G.M., et al., Survival from first recurrence: relative importance of prognostic factors in 1,015 breast cancer patients. J Clin Oncol, 1987. 5(1): p. 55-61.

170. Campos, S.M., Evolving treatment approaches for early breast cancer. Breast Cancer Res Treat, 2005. 89 Suppl 1 : p. S1-7.

171. Nabholtz, J.M., et al., Docetaxel (Taxotere) in combination with anthracyclines in the treatment of breast cancer. Semin Oncol, 2000. 27 (2 Suppl 3): p. 11-8.

184

172. Jassem, J., et al., Doxorubicin and paclitaxel versus fluorouracil, doxorubicin, and cyclophosphamide as first-line therapy for women with metastatic breast cancer: final results of a randomized phase III multicenter trial. J Clin Oncol, 2001. 19 (6): p. 1707-15.

173. Hortobagyi, G.N., Developments in chemotherapy of breast cancer. Cancer, 2000. 88 (12 Suppl): p. 3073-9.

174. Bonnet, D. and J.E. Dick, Human acute myeloid leukemia is organized as a hierarchy that originates from a primitive hematopoietic cell. Nat Med, 1997. 3(7): p. 730-7.

175. Al-Hajj, M., et al., Prospective identification of tumorigenic breast cancer cells. Proc Natl Acad Sci U S A, 2003. 100 (7): p. 3983-8.

176. Singh, S.K., et al., Identification of a cancer stem cell in human brain tumors. Cancer Res, 2003. 63 (18): p. 5821-8.

177. O'Brien, C.A., et al., A human colon cancer cell capable of initiating tumour growth in immunodeficient mice. Nature, 2007. 445 (7123): p. 106-10.

178. Li, C., et al., Identification of pancreatic cancer stem cells. Cancer Res, 2007. 67 (3): p. 1030-7.

179. Schatton, T., et al., Identification of cells initiating human melanomas. Nature, 2008. 451 (7176): p. 345-9.

180. Yang, Z.F., et al., Significance of CD90+ cancer stem cells in human liver cancer. Cancer Cell, 2008. 13 (2): p. 153-66.

181. Liu, S., G. Dontu, and M.S. Wicha, Mammary stem cells, self-renewal pathways, and carcinogenesis. Breast Cancer Res, 2005. 7(3): p. 86-95.

182. Korkaya, H., et al., Regulation of mammary stem/progenitor cells by PTEN/Akt/beta-catenin signaling. PLoS Biol, 2009. 7(6): p. e1000121.

185

183. Zhou, B.B., et al., Tumour-initiating cells: challenges and opportunities for anticancer drug discovery. Nat Rev Drug Discov, 2009. 8(10): p. 806-23.

184. Sakariassen, P.O., H. Immervoll, and M. Chekenya, Cancer stem cells as mediators of treatment resistance in brain tumors: status and controversies. Neoplasia, 2007. 9(11): p. 882-92.

185. Reya, T., et al., Stem cells, cancer, and cancer stem cells. Nature, 2001. 414 (6859): p. 105-11.

186. Shafee, N., et al., Cancer stem cells contribute to cisplatin resistance in Brca1/p53-mediated mouse mammary tumors. Cancer Res, 2008. 68 (9): p. 3243- 50.

187. Hambardzumyan, D., M. Squatrito, and E.C. Holland, Radiation resistance and stem-like cells in brain tumors. Cancer Cell, 2006. 10 (6): p. 454-6.

188. Williams, S.D., et al., Treatment of disseminated germ-cell tumors with cisplatin, bleomycin, and either vinblastine or etoposide. N Engl J Med, 1987. 316 (23): p. 1435-40.

189. Lippman, M.E., High-dose chemotherapy plus autologous bone marrow transplantation for metastatic breast cancer. N Engl J Med, 2000. 342 (15): p. 1119-20.

190. Bednar, F. and D.M. Simeone, Pancreatic cancer stem cells and relevance to cancer treatments. J Cell Biochem, 2009. 107 (1): p. 40-5.

191. Ischenko, I., et al., Cancer stem cells: how can we target them? Curr Med Chem, 2008. 15 (30): p. 3171-84.

192. Lapidot, T., et al., A cell initiating human acute myeloid leukaemia after transplantation into SCID mice. Nature, 1994. 367 (6464): p. 645-8.

193. Ginestier, C., et al., ALDH1 is a marker of normal and malignant human mammary stem cells and a predictor of poor clinical outcome. Cell Stem Cell, 2007. 1(5): p. 555-67. 186

194. Li, Y., et al., Sulforaphane, a dietary component of broccoli/broccoli sprouts, inhibits breast cancer stem cells. Clin Cancer Res. 16 (9): p. 2580-90.

195. Kakarala, M., et al., Targeting breast stem cells with the cancer preventive compounds curcumin and piperine. Breast Cancer Res Treat. 122 (3): p. 777-85.

196. Hirsch, H.A., et al., Metformin selectively targets cancer stem cells, and acts together with chemotherapy to block tumor growth and prolong remission. Cancer Res, 2009. 69 (19): p. 7507-11.

197. Dontu, G., et al., In vitro propagation and transcriptional profiling of human mammary stem/progenitor cells. Genes Dev, 2003. 17 (10): p. 1253-70.

198. Charafe-Jauffret, E., et al., Cancer stem cells in breast: current opinion and future challenges. Pathobiology, 2008. 75 (2): p. 75-84.

199. Marshall, G.P., 2nd, B.A. Reynolds, and E.D. Laywell, Using the neurosphere assay to quantify neural stem cells in vivo. Curr Pharm Biotechnol, 2007. 8(3): p. 141-5.

200. Hemmati, H.D., et al., Cancerous stem cells can arise from pediatric brain tumors. Proc Natl Acad Sci U S A, 2003. 100 (25): p. 15178-83.

201. Visvader, J.E. and G.J. Lindeman, Cancer stem cells in solid tumours: accumulating evidence and unresolved questions. Nat Rev Cancer, 2008. 8(10): p. 755-68.

202. Dick, J.E., Breast cancer stem cells revealed. Proc Natl Acad Sci U S A, 2003. 100 (7): p. 3547-9.

203. Liu, S., et al., Hedgehog signaling and Bmi-1 regulate self-renewal of normal and malignant human mammary stem cells. Cancer Res, 2006. 66 (12): p. 6063-71.

204. Dontu, G., et al., Role of Notch signaling in cell-fate determination of human mammary stem/progenitor cells. Breast Cancer Res, 2004. 6(6): p. R605-15.

187

205. Smalley, M.J. and T.C. Dale, Wnt signalling in mammalian development and cancer. Cancer Metastasis Rev, 1999. 18 (2): p. 215-30.

206. Turashvili, G., et al., Wnt signaling pathway in mammary gland development and carcinogenesis. Pathobiology, 2006. 73 (5): p. 213-23.

207. Polakis, P., Wnt signaling and cancer. Genes Dev, 2000. 14 (15): p. 1837-51.

208. Yamaguchi, T.P., Heads or tails: Wnts and anterior-posterior patterning. Curr Biol, 2001. 11 (17): p. R713-24.

209. Akiyama, T., Wnt/beta-catenin signaling. Cytokine Growth Factor Rev, 2000. 11 (4): p. 273-82.

210. Li, Y., et al., Evidence that transgenes encoding components of the Wnt signaling pathway preferentially induce mammary cancers from progenitor cells. Proc Natl Acad Sci U S A, 2003. 100 (26): p. 15853-8.

211. Woodward, W.A., et al., WNT/beta-catenin mediates radiation resistance of mouse mammary progenitor cells. Proc Natl Acad Sci U S A, 2007. 104 (2): p. 618-23.

212. Kawaguchi-Ihara, N., et al., Promotion of the self-renewal capacity of human acute leukemia cells by Wnt3A. Anticancer Res, 2008. 28 (5A): p. 2701-4.

213. Khan, N.I., K.F. Bradstock, and L.J. Bendall, Activation of Wnt/beta-catenin pathway mediates growth and survival in B-cell progenitor acute lymphoblastic leukaemia. Br J Haematol, 2007. 138 (3): p. 338-48.

214. Ysebaert, L., et al., Expression of beta-catenin by acute myeloid leukemia cells predicts enhanced clonogenic capacities and poor prognosis. Leukemia, 2006. 20 (7): p. 1211-6.

215. Chien, A.J., et al., Activated Wnt/beta-catenin signaling in melanoma is associated with decreased proliferation in patient tumors and a murine melanoma model. Proc Natl Acad Sci U S A, 2009. 106 (4): p. 1193-8.

188

216. Schulenburg, A., et al., CD44-positive colorectal adenoma cells express the potential stem cell markers musashi antigen (msi1) and ephrin B2 receptor (EphB2). J Pathol, 2007. 213 (2): p. 152-60.

217. Yang, W., et al., Wnt/beta-catenin signaling contributes to activation of normal and tumorigenic liver progenitor cells. Cancer Res, 2008. 68 (11): p. 4287-95.

218. Teng, Y., et al., Wnt/beta-catenin signaling regulates cancer stem cells in lung cancer A549 cells. Biochem Biophys Res Commun.

219. Nelson, W.J. and R. Nusse, Convergence of Wnt, beta-catenin, and cadherin pathways. Science, 2004. 303 (5663): p. 1483-7.

220. He, T.C., et al., Identification of c-MYC as a target of the APC pathway. Science, 1998. 281 (5382): p. 1509-12.

221. Tetsu, O. and F. McCormick, Beta-catenin regulates expression of cyclin D1 in colon carcinoma cells. Nature, 1999. 398 (6726): p. 422-6.

222. Mann, B., et al., Target genes of beta-catenin-T cell-factor/lymphoid-enhancer- factor signaling in human colorectal carcinomas. Proc Natl Acad Sci U S A, 1999. 96 (4): p. 1603-8.

223. Clevers, H., Wnt/beta-catenin signaling in development and disease. Cell, 2006. 127 (3): p. 469-80.

224. Lin, S.Y., et al., Beta-catenin, a novel prognostic marker for breast cancer: its roles in cyclin D1 expression and cancer progression. Proc Natl Acad Sci U S A, 2000. 97 (8): p. 4262-6.

225. Orsulic, S., et al., E-cadherin binding prevents beta-catenin nuclear localization and beta-catenin/LEF-1-mediated transactivation. J Cell Sci, 1999. 112 ( Pt 8) : p. 1237-45.

226. Schweizer, L. and H. Varmus, Wnt/Wingless signaling through beta-catenin requires the function of both LRP/Arrow and frizzled classes of receptors. BMC Cell Biol, 2003. 4: p. 4. 189

227. Takahashi-Yanaga, F. and T. Sasaguri, GSK-3beta regulates cyclin D1 expression: a new target for chemotherapy. Cell Signal, 2008. 20 (4): p. 581-9.

228. Liu, C., et al., Control of beta-catenin phosphorylation/degradation by a dual- kinase mechanism. Cell, 2002. 108 (6): p. 837-47.

229. Pap, M. and G.M. Cooper, Role of glycogen synthase kinase-3 in the phosphatidylinositol 3-Kinase/Akt cell survival pathway. J Biol Chem, 1998. 273 (32): p. 19929-32.

230. Yost, C., et al., The axis-inducing activity, stability, and subcellular distribution of beta-catenin is regulated in Xenopus embryos by glycogen synthase kinase 3. Genes Dev, 1996. 10 (12): p. 1443-54.

231. Cohen, P. and S. Frame, The renaissance of GSK3. Nat Rev Mol Cell Biol, 2001. 2(10): p. 769-76.

232. Cohen, M.M., Jr., The hedgehog signaling network. Am J Med Genet A, 2003. 123A (1): p. 5-28.

233. Clement, V., et al., HEDGEHOG-GLI1 signaling regulates human glioma growth, cancer stem cell self-renewal, and tumorigenicity. Curr Biol, 2007. 17 (2): p. 165-72.

234. Lewis, M.T. and J.M. Veltmaat, Next stop, the twilight zone: hedgehog network regulation of mammary gland development. J Mammary Gland Biol Neoplasia, 2004. 9(2): p. 165-81.

235. Pasca di Magliano, M. and M. Hebrok, Hedgehog signalling in cancer formation and maintenance. Nat Rev Cancer, 2003. 3(12): p. 903-11.

236. Wang, Z., et al., Emerging role of Notch in stem cells and cancer. Cancer Lett, 2009. 279 (1): p. 8-12.

237. Wilson, A. and F. Radtke, Multiple functions of Notch signaling in self-renewing organs and cancer. FEBS Lett, 2006. 580 (12): p. 2860-8.

190

238. Peacock, C.D. and D.N. Watkins, Cancer stem cells and the ontogeny of lung cancer. J Clin Oncol, 2008. 26 (17): p. 2883-9.

239. Fan, X. and C.G. Eberhart, Medulloblastoma stem cells. J Clin Oncol, 2008. 26 (17): p. 2821-7.

240. Kakarala, M. and M.S. Wicha, Implications of the cancer stem-cell hypothesis for breast cancer prevention and therapy. J Clin Oncol, 2008. 26 (17): p. 2813-20.

241. Scoville, D.H., et al., Current view: intestinal stem cells and signaling. Gastroenterology, 2008. 134 (3): p. 849-64.

242. Farnie, G. and R.B. Clarke, Mammary stem cells and breast cancer--role of Notch signalling. Stem Cell Rev, 2007. 3(2): p. 169-75.

243. Mumm, J.S. and R. Kopan, Notch signaling: from the outside in. Dev Biol, 2000. 228 (2): p. 151-65.

244. Wu, J.Y. and Y. Rao, Fringe: defining borders by regulating the notch pathway. Curr Opin Neurobiol, 1999. 9(5): p. 537-43.

245. Borggrefe, T. and F. Oswald, The Notch signaling pathway: transcriptional regulation at Notch target genes. Cell Mol Life Sci, 2009. 66 (10): p. 1631-46.

246. Weng, A.P., et al., c-Myc is an important direct target of Notch1 in T-cell acute lymphoblastic leukemia/lymphoma. Genes Dev, 2006. 20 (15): p. 2096-109.

247. Satoh, Y., et al., Roles for c-Myc in self-renewal of hematopoietic stem cells. J Biol Chem, 2004. 279 (24): p. 24986-93.

248. Palomero, T., et al., NOTCH1 directly regulates c-MYC and activates a feed- forward-loop transcriptional network promoting leukemic cell growth. Proc Natl Acad Sci U S A, 2006. 103 (48): p. 18261-6.

191

249. Ronchini, C. and A.J. Capobianco, Induction of cyclin D1 transcription and CDK2 activity by Notch(ic): implication for cell cycle disruption in transformation by Notch(ic). Mol Cell Biol, 2001. 21 (17): p. 5925-34.

250. Rangarajan, A., et al., Notch signaling is a direct determinant of keratinocyte growth arrest and entry into differentiation. EMBO J, 2001. 20 (13): p. 3427-36.

251. Oswald, F., et al., NF-kappaB2 is a putative target gene of activated Notch-1 via RBP-Jkappa. Mol Cell Biol, 1998. 18 (4): p. 2077-88.

252. Smith-Warner, S.A., et al., Fruits, vegetables and lung cancer: a pooled analysis of cohort studies. Int J Cancer, 2003. 107 (6): p. 1001-11.

253. Lamartiniere, C.A., et al., Genistein chemoprevention: timing and mechanisms of action in murine mammary and prostate. J Nutr, 2002. 132 (3): p. 552S-558S.

254. Li, Y. and F.H. Sarkar, Gene expression profiles of genistein-treated PC3 prostate cancer cells. J Nutr, 2002. 132 (12): p. 3623-31.

255. Chinni, S.R., et al., Indole-3-carbinol (I3C) induced cell growth inhibition, G1 cell cycle arrest and apoptosis in prostate cancer cells. Oncogene, 2001. 20 (23): p. 2927-36.

256. Li, Y., X. Li, and F.H. Sarkar, Gene expression profiles of I3C- and DIM-treated PC3 human prostate cancer cells determined by cDNA microarray analysis. J Nutr, 2003. 133 (4): p. 1011-9.

257. Choudhuri, T., et al., Curcumin induces apoptosis in human breast cancer cells through p53-dependent Bax induction. FEBS Lett, 2002. 512 (1-3): p. 334-40.

258. Mukhopadhyay, A., et al., Curcumin downregulates cell survival mechanisms in human prostate cancer cell lines. Oncogene, 2001. 20 (52): p. 7597-609.

259. Gupta, S., T. Hussain, and H. Mukhtar, Molecular pathway for (-)- epigallocatechin-3-gallate-induced cell cycle arrest and apoptosis of human prostate carcinoma cells. Arch Biochem Biophys, 2003. 410 (1): p. 177-85.

192

260. Hastak, K., et al., Role of p53 and NF-kappaB in epigallocatechin-3-gallate- induced apoptosis of LNCaP cells. Oncogene, 2003. 22 (31): p. 4851-9.

261. Park, C.H., et al., The inhibitory mechanism of curcumin and its derivative against beta-catenin/Tcf signaling. FEBS Lett, 2005. 579 (13): p. 2965-71.

262. Satoskar, R.R., S.J. Shah, and S.G. Shenoy, Evaluation of anti-inflammatory property of curcumin (diferuloyl methane) in patients with postoperative inflammation. Int J Clin Pharmacol Ther Toxicol, 1986. 24 (12): p. 651-4.

263. Shao, Z.M., et al., Curcumin exerts multiple suppressive effects on human breast carcinoma cells. Int J Cancer, 2002. 98 (2): p. 234-40.

264. Jaiswal, A.S., et al., Beta-catenin-mediated transactivation and cell-cell adhesion pathways are important in curcumin (diferuylmethane)-induced growth arrest and apoptosis in colon cancer cells. Oncogene, 2002. 21 (55): p. 8414-27.

265. Yan, C., et al., Gene expression profiling identifies activating transcription factor 3 as a novel contributor to the proapoptotic effect of curcumin. Mol Cancer Ther, 2005. 4(2): p. 233-41.

266. Ryu, M.J., et al., Natural derivatives of curcumin attenuate the Wnt/beta-catenin pathway through down-regulation of the transcriptional coactivator p300. Biochem Biophys Res Commun, 2008. 377 (4): p. 1304-8.

267. Wang, Z., et al., Notch-1 down-regulation by curcumin is associated with the inhibition of cell growth and the induction of apoptosis in pancreatic cancer cells. Cancer, 2006. 106 (11): p. 2503-13.

268. Kakarala, M., et al., Targeting breast stem cells with the cancer preventive compounds curcumin and piperine. Breast Cancer Res Treat, 2009.

269. Kim, J., et al., Suppression of Wnt signaling by the green tea compound (-)- epigallocatechin 3-gallate (EGCG) in invasive breast cancer cells. Requirement of the transcriptional repressor HBP1. J Biol Chem, 2006. 281 (16): p. 10865-75.

193

270. Kawasaki, B.T., et al., Targeting cancer stem cells with phytochemicals. Mol Interv, 2008. 8(4): p. 174-84.

271. Bose, M., et al., Inhibition of tumorigenesis in ApcMin/+ mice by a combination of (-)-epigallocatechin-3-gallate and fish oil. J Agric Food Chem, 2007. 55 (19): p. 7695-700.

272. Ju, J., et al., Inhibition of intestinal tumorigenesis in Apcmin/+ mice by (-)- epigallocatechin-3-gallate, the major catechin in green tea. Cancer Res, 2005. 65 (22): p. 10623-31.

273. Peng, G., et al., Green tea polyphenol (-)-epigallocatechin-3-gallate inhibits cyclooxygenase-2 expression in colon carcinogenesis. Mol Carcinog, 2006. 45 (5): p. 309-19.

274. Shimizu, M., et al., (-)-Epigallocatechin gallate and polyphenon E inhibit growth and activation of the epidermal growth factor receptor and human epidermal growth factor receptor-2 signaling pathways in human colon cancer cells. Clin Cancer Res, 2005. 11 (7): p. 2735-46.

275. Ahmad, N., S. Gupta, and H. Mukhtar, Green tea polyphenol epigallocatechin-3- gallate differentially modulates nuclear factor kappaB in cancer cells versus normal cells. Arch Biochem Biophys, 2000. 376 (2): p. 338-46.

276. Afaq, F., et al., Inhibition of ultraviolet B-mediated activation of nuclear factor kappaB in normal human epidermal keratinocytes by green tea Constituent (-)- epigallocatechin-3-gallate. Oncogene, 2003. 22 (7): p. 1035-44.

277. Guo, S., et al., Microarray-assisted pathway analysis identifies mitogen-activated protein kinase signaling as a mediator of resistance to the green tea polyphenol epigallocatechin 3-gallate in her-2/neu-overexpressing breast cancer cells. Cancer Res, 2006. 66 (10): p. 5322-9.

278. Kim, S.J., et al., Epigallocatechin-3-gallate suppresses NF-kappaB activation and phosphorylation of p38 MAPK and JNK in human astrocytoma U373MG cells. J Nutr Biochem, 2007. 18 (9): p. 587-96.

194

279. Sarkar, F.H., et al., Cellular signaling perturbation by natural products. Cell Signal, 2009. 21 (11): p. 1541-7.

280. Stearns, M.E., et al., Combination Therapy with Epigallocatechin-3-Gallate and Doxorubicin in Human Prostate Tumor Modeling Studies. Inhibition of Metastatic Tumor Growth in Severe Combined Immunodeficiency Mice. Am J Pathol.

281. Tang, S.N., et al., The dietary bioflavonoid quercetin synergizes with epigallocathechin gallate (EGCG) to inhibit prostate cancer stem cell characteristics, invasion, migration and epithelial-mesenchymal transition. J Mol Signal. 5: p. 14.

282. Park, C.H., et al., Quercetin, a potent inhibitor against beta-catenin/Tcf signaling in SW480 colon cancer cells. Biochem Biophys Res Commun, 2005. 328 (1): p. 227-34.

283. Shan, B.E., M.X. Wang, and R.Q. Li, Quercetin inhibit human SW480 colon cancer growth in association with inhibition of cyclin D1 and survivin expression through Wnt/beta-catenin signaling pathway. Cancer Invest, 2009. 27 (6): p. 604- 12.

284. Zhou, W., et al., Dietary polyphenol quercetin targets pancreatic cancer stem cells. Int J Oncol. 37 (3): p. 551-61.

285. Kallifatidis, G., et al., Sulforaphane targets pancreatic tumour-initiating cells by NF-kappaB-induced antiapoptotic signalling. Gut, 2009. 58 (7): p. 949-63.

286. Shen, G., et al., Chemoprevention of familial adenomatous polyposis by natural dietary compounds sulforaphane and dibenzoylmethane alone and in combination in ApcMin/+ mouse. Cancer Res, 2007. 67 (20): p. 9937-44.

287. Chaudhuri, D., S. Orsulic, and B.T. Ashok, Antiproliferative activity of sulforaphane in Akt-overexpressing ovarian cancer cells. Mol Cancer Ther, 2007. 6(1): p. 334-45.

288. Shankar, S., S. Ganapathy, and R.K. Srivastava, Sulforaphane enhances the therapeutic potential of TRAIL in prostate cancer orthotopic model through

195

regulation of apoptosis, metastasis, and angiogenesis. Clin Cancer Res, 2008. 14 (21): p. 6855-66.

289. Gibbs, A., et al., Sulforaphane destabilizes the androgen receptor in prostate cancer cells by inactivating histone deacetylase 6. Proc Natl Acad Sci U S A, 2009. 106 (39): p. 16663-8.

290. Whitesell, L. and S.L. Lindquist, HSP90 and the chaperoning of cancer. Nat Rev Cancer, 2005. 5(10): p. 761-72.

291. Yao, H., et al., Sulforaphane inhibited expression of hypoxia-inducible factor- 1alpha in human tongue squamous cancer cells and prostate cancer cells. Int J Cancer, 2008. 123 (6): p. 1255-61.

292. Xiong, H.Q., Molecular targeting therapy for pancreatic cancer. Cancer Chemother Pharmacol, 2004. 54 Suppl 1 : p. S69-77.

293. Buchler, P., et al., Hypoxia-inducible factor 1 regulates vascular endothelial growth factor expression in human pancreatic cancer. Pancreas, 2003. 26 (1): p. 56-64.

294. Karagöz, E.G., et al. , The N-terminal domain of human Hsp90 triggers binding to the cochaperone p23 . Submitted, 2010.

295. Tugarinov, V., P.M. Hwang, and L.E. Kay, Nuclear magnetic resonance spectroscopy of high-molecular-weight proteins. Annu Rev Biochem, 2004. 73 : p. 107-46.

296. Butz, J., E. Wickstrom, and J. Edwards, Characterization of mutations and loss of heterozygosity of p53 and K-ras2 in pancreatic cancer cell lines by immobilized polymerase chain reaction. BMC Biotechnol, 2003. 3: p. 11.

297. Mohiuddin, M., et al., Influence of p53 status on radiation and 5-flourouracil synergy in pancreatic cancer cells. Anticancer Res, 2002. 22 (2A): p. 825-30.

196

298. Alcock, R.A., et al., Farnesyltransferase inhibitor (L-744,832) restores TGF-beta type II receptor expression and enhances radiation sensitivity in K-ras mutant pancreatic cancer cell line MIA PaCa-2. Oncogene, 2002. 21 (51): p. 7883-90.

299. Neckers, L., Development of small molecule Hsp90 inhibitors: utilizing both forward and reverse chemical genomics for drug identification. Curr Med Chem, 2003. 10 (9): p. 733-9.

300. Kumar, A. and G. Sabbioni, New biomarkers for monitoring the levels of isothiocyanates in humans. Chem Res Toxicol. 23 (4): p. 756-65.

301. Sreeramulu, S., et al., The human Cdc37.Hsp90 complex studied by heteronuclear NMR spectroscopy. J Biol Chem, 2009. 284 (6): p. 3885-96.

302. Bertelsen, E.B., et al., Solution conformation of wild-type E. coli Hsp70 (DnaK) chaperone complexed with ADP and substrate. Proc Natl Acad Sci U S A, 2009. 106 (21): p. 8471-6.

303. Sprangers, R. and L.E. Kay, Quantitative dynamics and binding studies of the 20S proteasome by NMR. Nature, 2007. 445 (7128): p. 618-22.

304. Koorstra, J.B., et al., Pancreatic carcinogenesis. Pancreatology, 2008. 8(2): p. 110-25.

305. Barton, C.M., et al., Abnormalities of the p53 tumour suppressor gene in human pancreatic cancer. Br J Cancer, 1991. 64 (6): p. 1076-82.

306. Schlieman, M.G., et al., Incidence, mechanism and prognostic value of activated AKT in pancreas cancer. Br J Cancer, 2003. 89 (11): p. 2110-5.

307. Almoguera, C., et al., Most human carcinomas of the exocrine pancreas contain mutant c-K-ras genes. Cell, 1988. 53 (4): p. 549-54.

308. Vivanco, I. and C.L. Sawyers, The phosphatidylinositol 3-Kinase AKT pathway in human cancer. Nat Rev Cancer, 2002. 2(7): p. 489-501.

197

309. Reichert, M., et al., Phosphoinositide-3-kinase signaling controls S-phase kinase- associated protein 2 transcription via E2F1 in pancreatic ductal adenocarcinoma cells. Cancer Res, 2007. 67 (9): p. 4149-56.

310. Altomare, D.A., et al., Frequent activation of AKT2 kinase in human pancreatic carcinomas. J Cell Biochem, 2002. 87 (4): p. 470-6.

311. Prodromou, C., et al., Identification and structural characterization of the ATP/ADP-binding site in the Hsp90 molecular chaperone. Cell, 1997. 90 (1): p. 65-75.

312. Stebbins, C.E., et al., Crystal structure of an Hsp90-geldanamycin complex: targeting of a protein chaperone by an antitumor agent. Cell, 1997. 89 (2): p. 239- 50.

313. Usmani, S.Z., R. Bona, and Z. Li, 17 AAG for HSP90 inhibition in cancer--from bench to bedside. Curr Mol Med, 2009. 9(5): p. 654-64.

314. Dickinson, S.E., et al., Inhibition of activator protein-1 by sulforaphane involves interaction with cysteine in the cFos DNA-binding domain: implications for chemoprevention of UVB-induced skin cancer. Cancer Res, 2009. 69 (17): p. 7103-10.

315. Hong, F., M.L. Freeman, and D.C. Liebler, Identification of sensor cysteines in human Keap1 modified by the cancer chemopreventive agent sulforaphane. Chem Res Toxicol, 2005. 18 (12): p. 1917-26.

316. Jiang, Y., et al., Split Renilla luciferase protein fragment-assisted complementation (SRL-PFAC) to characterize Hsp90-Cdc37 complex and identify critical residues in protein/protein interactions. J Biol Chem. 285 (27): p. 21023- 36.

317. Jemal, A., et al., Cancer statistics, 2007. CA Cancer J Clin, 2007. 57 (1): p. 43-66.

318. Ghaneh, P., E. Costello, and J.P. Neoptolemos, Biology and management of pancreatic cancer. Gut, 2007. 56 (8): p. 1134-52.

198

319. Ougolkov, A.V. and D.D. Billadeau, Targeting GSK-3: a promising approach for cancer therapy? Future Oncol, 2006. 2(1): p. 91-100.

320. Li, L., F.S. Braiteh, and R. Kurzrock, Liposome-encapsulated curcumin: in vitro and in vivo effects on proliferation, apoptosis, signaling, and angiogenesis. Cancer, 2005. 104 (6): p. 1322-31.

321. Kunnumakkara, A.B., et al., Curcumin potentiates antitumor activity of gemcitabine in an orthotopic model of pancreatic cancer through suppression of proliferation, angiogenesis, and inhibition of nuclear factor-kappaB-regulated gene products. Cancer Res, 2007. 67 (8): p. 3853-61.

322. Solit, D.B. and G. Chiosis, Development and application of Hsp90 inhibitors. Drug Discov Today, 2008. 13 (1-2): p. 38-43.

323. Korc, M., et al., Overexpression of the epidermal growth factor receptor in human pancreatic cancer is associated with concomitant increases in the levels of epidermal growth factor and transforming growth factor alpha. J Clin Invest, 1992. 90 (4): p. 1352-60.

324. Yamanaka, Y., et al., Overexpression of HER2/neu oncogene in human pancreatic carcinoma. Hum Pathol, 1993. 24 (10): p. 1127-34.

325. Bos, J.L., ras oncogenes in human cancer: a review. Cancer Res, 1989. 49 (17): p. 4682-9.

326. Warne, P.H., P.R. Viciana, and J. Downward, Direct interaction of Ras and the amino-terminal region of Raf-1 in vitro. Nature, 1993. 364 (6435): p. 352-5.

327. Van Aelst, L., et al., Complex formation between RAS and RAF and other protein kinases. Proc Natl Acad Sci U S A, 1993. 90 (13): p. 6213-7.

328. Cheng, J.Q., et al., Amplification of AKT2 in human pancreatic cells and inhibition of AKT2 expression and tumorigenicity by antisense RNA. Proc Natl Acad Sci U S A, 1996. 93 (8): p. 3636-41.

199

329. Li Y, K.G., Seo YH, Zhang T, Jiang Y, et al. , Sulforaphane inhibits Hsp90 function by disrupting Hsp90-p50 Cdc37 complex in pancreatic cancer cells through direct interaction with specific residues. Submitted.

330. Hieronymus, H., et al., Gene expression signature-based chemical genomic prediction identifies a novel class of HSP90 pathway modulators. Cancer Cell, 2006. 10 (4): p. 321-30.

331. Sullivan, W., et al., Nucleotides and two functional states of hsp90. J Biol Chem, 1997. 272 (12): p. 8007-12.

332. Johnson, J.L. and D.O. Toft, A novel chaperone complex for steroid receptors involving heat shock proteins, immunophilins, and p23. J Biol Chem, 1994. 269 (40): p. 24989-93.

333. Beliakoff, J., et al., Hormone-refractory breast cancer remains sensitive to the antitumor activity of heat shock protein 90 inhibitors. Clin Cancer Res, 2003. 9(13): p. 4961-71.

334. Solit, D.B., et al., 17-Allylamino-17-demethoxygeldanamycin induces the degradation of androgen receptor and HER-2/neu and inhibits the growth of prostate cancer xenografts. Clin Cancer Res, 2002. 8(5): p. 986-93.

335. Solit, D.B., et al., Inhibition of heat shock protein 90 function down-regulates Akt kinase and sensitizes tumors to Taxol. Cancer Res, 2003. 63 (9): p. 2139-44.

336. Banerji, U., et al., An in vitro and in vivo study of the combination of the heat shock protein inhibitor 17-allylamino-17-demethoxygeldanamycin and carboplatin in human ovarian cancer models. Cancer Chemother Pharmacol, 2008. 62 (5): p. 769-78.

337. Matsui, T.A., et al., Sulforaphane induces cell cycle arrest and apoptosis in murine osteosarcoma cells in vitro and inhibits tumor growth in vivo. Oncol Rep, 2007. 18 (5): p. 1263-8.

200

338. Liang, H., B. Lai, and Q. Yuan, Sulforaphane Induces Cell-Cycle Arrest and Apoptosis in Cultured Human Lung Adenocarcinoma LTEP-A2 Cells and Retards Growth of LTEP-A2 Xenografts in Vivo. J Nat Prod, 2008.

339. Lampe, J.W., Sulforaphane: from chemoprevention to pancreatic cancer treatment? Gut, 2009. 58 (7): p. 900-2.

340. Khoshyomn, S., et al., Synergistic action of genistein and cisplatin on growth inhibition and cytotoxicity of human medulloblastoma cells. Pediatr Neurosurg, 2000. 33 (3): p. 123-31.

341. Fimognari, C., et al., Sulforaphane increases the efficacy of doxorubicin in mouse fibroblasts characterized by p53 mutations. Mutat Res, 2006. 601 (1-2): p. 92-101.

342. Myzak, M.C. and R.H. Dashwood, Chemoprotection by sulforaphane: keep one eye beyond Keap1. Cancer Lett, 2006. 233 (2): p. 208-18.

343. McCollum, A.K., et al., P-Glycoprotein-mediated resistance to Hsp90-directed therapy is eclipsed by the heat shock response. Cancer Res, 2008. 68 (18): p. 7419-27.

344. Messaoudi, S., et al., Recent advances in Hsp90 inhibitors as antitumor agents. Anticancer Agents Med Chem, 2008. 8(7): p. 761-82.

345. Premkumar, D.R., et al., Synergistic interaction between 17-AAG and phosphatidylinositol 3-kinase inhibition in human malignant glioma cells. Mol Carcinog, 2006. 45 (1): p. 47-59.

346. Roforth, M.M. and C. Tan, Combination of rapamycin and 17-allylamino-17- demethoxygeldanamycin abrogates Akt activation and potentiates mTOR blockade in breast cancer cells. Anticancer Drugs, 2008. 19 (7): p. 681-8.

347. McLaughlin, S.H., et al., The co-chaperone p23 arrests the Hsp90 ATPase cycle to trap client proteins. J Mol Biol, 2006. 356 (3): p. 746-58.

201

348. Pahlke, G., et al., Impact of quercetin and EGCG on key elements of the Wnt pathway in human colon carcinoma cells. J Agric Food Chem, 2006. 54 (19): p. 7075-82.

349. Ponti, D., et al., Isolation and in vitro propagation of tumorigenic breast cancer cells with stem/progenitor cell properties. Cancer Res, 2005. 65 (13): p. 5506-11.

350. Luo, M., et al., Mammary epithelial-specific ablation of the focal adhesion kinase suppresses mammary tumorigenesis by affecting mammary cancer stem/progenitor cells. Cancer Res, 2009. 69 (2): p. 466-74.

351. Reya, T., et al., A role for Wnt signalling in self-renewal of haematopoietic stem cells. Nature, 2003. 423 (6938): p. 409-14.

352. Azarenko, O., et al., Suppression of microtubule dynamic instability and turnover in MCF7 breast cancer cells by sulforaphane. Carcinogenesis, 2008. 29 (12): p. 2360-8.

353. Pledgie-Tracy, A., M.D. Sobolewski, and N.E. Davidson, Sulforaphane induces cell type-specific apoptosis in human breast cancer cell lines. Mol Cancer Ther, 2007. 6(3): p. 1013-21.

354. Hedgepeth, C.M., et al., Activation of the Wnt signaling pathway: a molecular mechanism for lithium action. Dev Biol, 1997. 185 (1): p. 82-91.

355. Klein, P.S. and D.A. Melton, A molecular mechanism for the effect of lithium on development. Proc Natl Acad Sci U S A, 1996. 93 (16): p. 8455-9.

356. Tang, C., C.L. Chua, and B.T. Ang, Insights into the cancer stem cell model of glioma tumorigenesis. Ann Acad Med Singapore, 2007. 36 (5): p. 352-7.

357. Fimognari, C., et al., Isothiocyanates as novel cytotoxic and cytostatic agents: molecular pathway on human transformed and non-transformed cells. Biochem Pharmacol, 2004. 68 (6): p. 1133-8.

358. Tian, Q., R.A. Rosselot, and S.J. Schwartz, Quantitative determination of intact glucosinolates in broccoli, broccoli sprouts, Brussels sprouts, and cauliflower by 202

high-performance liquid chromatography-electrospray ionization-tandem mass spectrometry. Anal Biochem, 2005. 343 (1): p. 93-9.

359. Agrawal, S., et al., Simultaneous determination of sulforaphane and its major metabolites from biological matrices with liquid chromatography-tandem mass spectroscopy. J Chromatogr B Analyt Technol Biomed Life Sci, 2006. 840 (2): p. 99-107.

360. Keum, Y.S., et al., Pharmacokinetics and pharmacodynamics of broccoli sprouts on the suppression of prostate cancer in transgenic adenocarcinoma of mouse prostate (TRAMP) mice: implication of induction of Nrf2, HO-1 and apoptosis and the suppression of Akt-dependent kinase pathway. Pharm Res, 2009. 26 (10): p. 2324-31.

361. Combourieu, B., et al., Identification of new derivatives of sinigrin and glucotropaeolin produced by the human digestive microflora using 1H NMR spectroscopy analysis of in vitro incubations. Drug Metab Dispos, 2001. 29 (11): p. 1440-5.

362. Bheemreddy, R.M. and E.H. Jeffery, The metabolic fate of purified glucoraphanin in F344 rats. J Agric Food Chem, 2007. 55 (8): p. 2861-6.

363. Westerheide, S.D. and R.I. Morimoto, Heat shock response modulators as therapeutic tools for diseases of protein conformation. J Biol Chem, 2005. 280 (39): p. 33097-100.

364. Ali, M.M., et al., Crystal structure of an Hsp90-nucleotide-p23/Sba1 closed chaperone complex. Nature, 2006. 440 (7087): p. 1013-7.

365. Scheibel, T., et al., ATP-binding properties of human Hsp90. J Biol Chem, 1997. 272 (30): p. 18608-13.

366. Workman, P., Altered states: selectively drugging the Hsp90 cancer chaperone. Trends Mol Med, 2004. 10 (2): p. 47-51.

203

367. Marcu, M.G., et al., The heat shock protein 90 antagonist novobiocin interacts with a previously unrecognized ATP-binding domain in the carboxyl terminus of the chaperone. J Biol Chem, 2000. 275 (47): p. 37181-6.

368. Yun, B.G., et al., Novobiocin induces a distinct conformation of Hsp90 and alters Hsp90-cochaperone-client interactions. Biochemistry, 2004. 43 (25): p. 8217-29.

369. Marcu, M.G., T.W. Schulte, and L. Neckers, Novobiocin and related coumarins and depletion of heat shock protein 90-dependent signaling proteins. J Natl Cancer Inst, 2000. 92 (3): p. 242-8.

370. Landis-Piwowar, K.R., et al., A novel prodrug of the green tea polyphenol (-)- epigallocatechin-3-gallate as a potential anticancer agent. Cancer Res, 2007. 67 (9): p. 4303-10.

371. Fujiki, H., Two stages of cancer prevention with green tea. J Cancer Res Clin Oncol, 1999. 125 (11): p. 589-97.

372. Palermo, C.M., C.A. Westlake, and T.A. Gasiewicz, Epigallocatechin gallate inhibits aryl hydrocarbon receptor gene transcription through an indirect mechanism involving binding to a 90 kDa heat shock protein. Biochemistry, 2005. 44 (13): p. 5041-52.

373. Khan, N., et al., Targeting multiple signaling pathways by green tea polyphenol (- )-epigallocatechin-3-gallate. Cancer Res, 2006. 66 (5): p. 2500-5.

374. Yang, C.S., P. Maliakal, and X. Meng, Inhibition of carcinogenesis by tea. Annu Rev Pharmacol Toxicol, 2002. 42 : p. 25-54.

375. Nagle, D.G., D. Ferreira, and Y.D. Zhou, Epigallocatechin-3-gallate (EGCG): chemical and biomedical perspectives. Phytochemistry, 2006. 67 (17): p. 1849-55.

376. Ahmad, N., et al., Green tea constituent epigallocatechin-3-gallate and induction of apoptosis and cell cycle arrest in human carcinoma cells. J Natl Cancer Inst, 1997. 89 (24): p. 1881-6.

204

377. Jatoi, A., et al., A phase II trial of green tea in the treatment of patients with androgen independent metastatic prostate carcinoma. Cancer, 2003. 97 (6): p. 1442-6.

378. Powers, M.V. and P. Workman, Inhibitors of the heat shock response: biology and pharmacology. FEBS Lett, 2007. 581 (19): p. 3758-69.

379. Chiosis, G. and L. Neckers, Tumor selectivity of Hsp90 inhibitors: the explanation remains elusive. ACS Chem Biol, 2006. 1(5): p. 279-84.

380. Allan, R.K., et al., Modulation of chaperone function and cochaperone interaction by novobiocin in the C-terminal domain of Hsp90: evidence that coumarin antibiotics disrupt Hsp90 dimerization. J Biol Chem, 2006. 281 (11): p. 7161-71.

381. Garnier, C., et al., Binding of ATP to heat shock protein 90: evidence for an ATP- binding site in the C-terminal domain. J Biol Chem, 2002. 277 (14): p. 12208-14.

382. Ishii, T., et al., Covalent modification of proteins by green tea polyphenol (-)- epigallocatechin-3-gallate through autoxidation. Free Radic Biol Med, 2008. 45 (10): p. 1384-94.

383. Young, J.C., W.M. Obermann, and F.U. Hartl, Specific binding of tetratricopeptide repeat proteins to the C-terminal 12-kDa domain of hsp90. J Biol Chem, 1998. 273 (29): p. 18007-10.

384. Fang, Y., et al., SBA1 encodes a yeast hsp90 cochaperone that is homologous to vertebrate p23 proteins. Mol Cell Biol, 1998. 18 (7): p. 3727-34.

385. Yin, Z., E.C. Henry, and T.A. Gasiewicz, (-)-Epigallocatechin-3-gallate is a novel Hsp90 inhibitor. Biochemistry, 2009. 48 (2): p. 336-45.

386. Kaur, J. and R. Ralhan, Induction of apoptosis by abrogation of HSP70 expression in human oral cancer cells. Int J Cancer, 2000. 85 (1): p. 1-5.

205

387. Schmitt, E., et al., Heat shock protein 70 neutralization exerts potent antitumor effects in animal models of colon cancer and melanoma. Cancer Res, 2006. 66 (8): p. 4191-7.

206