Copyright Undertaking

By reading and using the thesis, the reader understands and agrees to the following terms:

1. The reader will abide by the rules and legal ordinances governing copyright regarding the use of the thesis.

2. The reader will use the thesis for the purpose of research or private study only and not for distribution or further reproduction or any other purpose.

3. The reader agrees to indemnify and hold the University harmless from and against any loss, damage, cost, liability or expenses arising from copyright infringement or unauthorized usage.

IMPORTANT If you have reasons to believe that any materials in this thesis are deemed not suitable to be distributed in this form, or a copyright owner having difficulty with the material being included in our database, please contact [email protected] providing details. The Library will look into your claim and consider taking remedial action upon receipt of the written requests.

Pao Yue-kong Library, The Hong Kong Polytechnic University, Hung Hom, Kowloon, Hong Kong

http://www.lib.polyu.edu.hk

CHARACTERIZATION OF THE TISSUE-SELECTIVITY

OF TRADITIONAL CHINESE MEDICINE

(TCM)-DERIVED PHYTOESTROGEN AND THE

POSSIBLE MECHANISMS INVOLVED

ZHOU LIPING

Ph.D

The Hong Kong Polytechnic University

2017

The Hong Kong Polytechnic University

Department of Applied Biology and Chemical Technology

Characterization of the Tissue-selectivity of Traditional

Chinese Medicine (TCM)-derived Phytoestrogen and the

Possible Mechanisms Involved

ZHOU Liping

A thesis submitted in partial fulfillment of the requirements for

the degree of Doctor of Philosophy

September 2016

Certificate Originality

I hereby declare that this thesis is my own work and that, to the best of my knowledge and belief, it reproduces no material previously published or written, nor material that has been accepted for the award of any other degree or diploma, except where due acknowledgement has been made in the text.

(Signed)

ZHOU Liping (Name of student)

II Abstract

Icariin is the most abundant flavonoid in Herba Epemedii (HEP), a commonly used

Chinese herb for treatment of bone disease, that has been reported to exert estrogenic effects. Danggui Buxue Tang (DBT), consisting of Radix Astragali and Radix

Angelicae Sinensis, is the most popular traditional Chinese Medicine (TCM) decoction prescribed for management of menopausal symptoms in China and its chemical composition has been well characterized. Recent studies found that DBT may contain phytoestrogens and majority of the activities of DBT are mediated by these phytoestrogens. This study aimed to investigate the estrogenic effects of icariin and DBT in different estrogen-sensitive tissues and the possible mechanisms involved.

In the first part of my study, the tissue-selectivity of icariin was evaluated by using mature ovariectomized (OVX) Sprague Dawley rats and four estrogen-responsive cell lines (human breast cancer MCF-7 cell, human endometrial Ishikawa cell, human neuroblastoma SH-SY5Y cell and human osteosarcoma MG-63 cell). Upon treatment for 12 weeks, icariin dose-dependently increased the bone mineral density (BMD) and improved the trabecular bone properties at distal femur, proximal tibia and lumbar spine in OVX rats. Moreover, icariin reversed the changes in mRNA expression of tyrosine hydroxylase (TH) and dopamine transporter (DAT) in striatum of OVX rats

III but did not induce estrogenic responses in uterus and breast. In addition, icariin significantly stimulated cell proliferation or differentiation but failed to induce the expression of estrogen-responsive genes and estrogen response element-dependent luciferase activities in MCF-7, Ishikawa cell, SH-SY5Y cell and MG-63 cell. These results indicate that icariin did not induce estrogen-dependent transcriptional events in these cell lines.

In the second part of my study, the tissue-selectivity of DBT was investigated using both mature OVX rats and estrogen receptor (ER)-positive cell lines. The results showed that DBT reversed the decrease in estradiol level as well as the accompanying increase in follicle-stimulating hormone (FSH) and luteinizing hormone (LH) level in rats induced by ovariectomy. In addition, DBT increased BMD and improved micro-architecture of both long bone and vertebra in OVX rats. DBT also up-regulated mRNA expression of TH and suppressed mRNA expression of DAT in striatum of OVX rats without inducing estrogenic responses in uterus and breast tissue of OVX rats. Our in vitro studies showed that DBT directly stimulated cell proliferation and differentiation as well as ERE luciferase activities in MCF-7,

Ishikawa cell, SH-SY5Y cell and MG-63 cell, suggesting the direct estrogenic effects of DBT. Co-treatment of DBT-treated MG-63 cells with ICI182,780 (ER antagonist,

10-6M), U0126 (mitogen-activated protein kinase MAPK inhibitor, 10-6M) and

LY294002 (phosphoinositide 3-kinase PI3K inhibitor, 10-6M) completely or partially blocked the stimulating effects of DBT on cell proliferation and cell differentiation.

These results indicate that ER, MAPK and PI3K pathways might be involved in mediating the actions of DBT in bone cells.

It is also of concern to determine if the use of herbal medicine will be safe for postmenopausal women who are simultaneously prescribed with selective estrogen receptor modulators (SERMs) such as tamoxifen and raloxifene for treatment of breast cancer and osteoporosis as many of their activities are mediated by the same receptors. In the third part of my study, the potential interactions between DBT and

SERMs (tamoxifen and raloxifene) were thereby investigated. Upon co-treatment for three months, DBT did not alter the estrogenic effects of tamoxifen and raloxifene in the uterus, breast tissues, brain and bone of OVX rats. In addition, DBT at 0.1mg/ml did not interact with the estrogenic effects of SERMs in MCF-7 cells. However, DBT at 0.5mg/ml antagonized the effects of tamoxifen on cell proliferation in MCF-7 cell and enhanced the stimulating effects of tamoxifen and raloxifene on cell proliferation of SH-SY5Y cell as well as the alkaline phosphatase (ALP) activity in Ishikawa and

MG-63 cells. These results suggest that DBT does not interact with western drugs

(tamoxifen and raloxifene) in estrogen-sensitive tissues in vivo while the in vitro interactive effects appear to be dose-dependent and tissue-specific. This discrepancy

V between in vivo and in vitro studies might be due to the fact that DBT applied in the cells was without biological activation and such concentrations are too high to be achieved in vivo.

In conclusion, both icariin and DBT selectively exerted protective effects in bone and brain without causing side effects in uterus and breast in vivo. Our study provide evidence to support our hypothesis that phytoestrogens derived from TCM selectively exert estrogenic effects in estrogen-sensitive tissues and might be useful for the management of postmenopausal syndromes.

List of Publications 1. LP Zhou#, KW Tsim, MS Wong. Danggui Buxue Tang decoction (DBT) protects

rats against ovariectomy-induced osteoporosis by suppressing bone turnover. The

11th International Postgraduate symposium on Chinese Medicine. 2015. Hong

Kong

2. M.X. Ho#, L.P Zhou, M.S. Wong. Icariin Exerts Anabolic Effects on Osteoblasts

via Rapid Estrogen Receptor α Signaling Pathways. Annual Meeting of American

Society of Bone and Mineral Research. 2015. Seattle, USA.

3. Zhou LP#, Poon CCW, Tsim KWK, MS Wong. A Phytoestrogen-rich DBT

formula modulated hypothalamic-pituitary-ovary axis in mature ovariectomized

rats. ENDO 2016, Annual Meeting of the Endocrine Society. 2016. Boston, USA.

4. Zhou LP#, Dong XL, Cao SS, Tsim KWK, MS Wong. Danggui Buxue Tang

Decoction (DBT), a phytoestrogen-rich formula, increased bone mineral density

and improved bone properties in mature ovariectomized rats. ENDO 2016, Annual

Meeting of the Endocrine Society. 2016. Boston, USA.

5. Zhang Y#, Zhou LP, Ho MX, Wong KC, MS Wong. 8-prenylgenistein, a

derivative of genistein, exerted osteoprotective effects without uterotrophic

activity. ENDO 2016, Annual Meeting of the Endocrine Society. 2016. Boston,

USA.

6. Dong F, Dong X, Zhou L, Xiao H, Ho PY, Wong MS, Wang Y. 2016

Doxorubicin-loaded biodegradable self-assembly zein nanoparticle and its

anti-cancer effect: Preparation, in vitro evaluation, and cellular uptake. Colloids

Surf B Biointerfaces.140:324-31.

7. Zhang Yan, Zhou Li-Ping, Ho Mingxian, Li Xiao-Pi, Zhao Yong-Jian, Mok

Kam-Wah, Qiu Zuo-Cheng, Shi Qi, Wang Yong-Jun, Wong Man-Sau. VII

8-Prenylgenistein, a prenylated genistein derivative, exerted tissue selective

osteoprotective effects in ovariectomized mice. (Submitted)

8. Protective effects of Danggui Buxue Tang (DBT), a phytoestrogen-containing

Traditional Chinese Medicine, against osteoporosis in ovariectomy rats and

possible involvement of ER, MAPK and PI3K pathway. (in preparation)

9. A phytoestrogen rich DBT formula modulates circulating levels of reproductive

hormones in mature ovariectomized rats and exerts direct estrogenic actions in

estrogen sensitive cells. (in preparation)

10. Ginsenoside Rg1 improves the spatial learning and memory ability of

Aß25-35-induced AD mice possibly via ER and IGF-IR pathways. (in preparation)

VIII

Acknowledgement

Three years ago, I went to The Hong Kong Polytechnic University to pursue my study as a PhD student. It is one of the most important and brave decisions I have made. Yes, it is full of challenges and tortures while it is also a period of harvest. Thanks to all the help from my surroundings, I managed to finish my study. At the end of my PhD study, I would like to take this opportunity to express my gratitude to those who helped during my three years’ study.

My deepest gratitude goes first and foremost to my supervisor, Professor Wong Man

Sau. I appreciate her constant encouragement and invaluable advices throughout my

PhD study. Without her professional guidance and generous support, I would never have been able to finish my study. Her enthusiastic attitude towards science and her high work efficiency will be the goals for all my life.

Second, I would like acknowledge Professor Karl Tsim of The Hong Kong University of Science and Technology. He offered great help and provided useful recommendations for my experiments.

Heartfelt thanks to my teammates Dr. Dong Xiaoli, Miss. Cao Sisi, Dr. Gao Quangui,

Dr. Xiao Huihui, Dr. Zhang Yan, Dr. Christina Poon, Miss. Yu Wenxuan, Dr. Wong Ka

Chun, Mr. Ho Mingxian, Mr. Qiu Zuocheng for their generous support and

IX collaboration. It is a great and enjoyable period to work with these kind and friendly people.

I also thank all the technical and support staff in the Department of Applied Biology and Chemical Technology for their kind and professional assistance.

Special thanks go to my beloved family for their consideration and support all these years. I also owe my sincere gratitude to my friends. Thanks for their help and encouragement during my study.

Finally, I would like to give my deepest gratitude to all the people who offered me support and help.

Table of contents

Title page……………………………………………………………………………...I

Certificate of originality……………………………………………………………..II

Abstract……………………………………………………………………………...III

List of publications……………………………………………………………….VII

Acknowledgement………………………………………………………………...... IX

Table of Contents………………………………………………………………….XI

List of Figures and Tables………………………………………………………XVIII

Abbreviations………………………………………………………………….....XXV

Chapter 1 Background and Introduction…………………………………………...1

1.1 Women’s health and postmenopausal syndrome……………………...... 2

1.1.1 Definition of postmenopausal syndrome………………………………………2

1.1.2 Symptoms of postmenopausal syndrome and influences on women………….2

1.2 Estrogen and menopause……...………………………………………………….3

1.2.1 Physiological actions of estrogen and manifestation of estrogen disruption.....5

1.2.1.1 Uterus………………………………………………………………………….5

1.2.1.2 Bone…………...………………………………………………………………6

1.2.1.3 Central Nervous System...……………………...... 7

1.2.1.4 Breast…………………...……………………...……………...... 10

1.2.2 Estrogen receptor and tissue-selective responses to ligands…………………10

1.2.2.1 Subtypes of estrogen receptors……………………...... 11

1.2.2.2 Distribution of ERs and tissue-selective responses to ligands……………….15

1.2.2.3 Estrogen signaling……………………...... 15

1.2.3 Biosynthesis of estrogen……………………...... 20

1.2.4 Neuroendocrine regulation of estrogen……………………...... 21

1.3 Current therapy for postmenopausal syndrome……………...... 25

1.3.1 Hormone replacement therapy (HRT) ……………………...…………...... 25

1.3.2 Selective estrogen receptor modulators (SERMs) ……………………...... 26

1.3.3 Phytoestrogens……………………...……………………...... 26

1.3.3.1 Structural classification of phytoestrogens……………………...... 27

1.3.3.2 Effects of phytoestrogen in management of postmenopausal symptoms…….27

1.3.3.3 Controversy on phytoestrogens……………………...... 30

1.4 Traditional Chinese Medicine (TCM) as a source of Phytoestrogens……..…31

1.4.1 Icariin……………………...……………………...………………...... 32

1.4.1.1 Introduction and chemical structure of icariin……………………………….31

1.4.1.2 Physiological effects of icariin in estrogen-sensitive tissues………………...31

1.4.2 Danggui Buxue Tang decoction (DBT) ……………………...... 36

1.4.2.1 Clinical use of Danggui Buxue Tang (DBT) decoction…………………….36

XII

1.4.2.2 Physiological effects of DBT……………………...... 36

1.4.3 Hypothesis and objectives………………………………………………….…38

1.5 Concerns about the potential interactions between phytoestrogen-TCM and prescribed drugs………...... 39

Chapter 2 Hypothesis and Objectives……………………………………………...41

Chapter 3 Methodology………………………………………………...... 44

3.1 Authentication, extraction and quality control of DBT extract…………………..45

3.1.1 Preparation of DBT extract…………………………………………………….45

3.1.2 LC-MS analysis of DBT extract………………………………………………..45

3.2 In vivo study………………………....……………………...... 48

3.2.1 Experimental designs…………………………………………………………..48

3.2.2 Ovariectomy and Sham operation……………………………………………..52

3.2.3 Drugs preparation and animal feeding………………………………………..52

3.2.4 Sample collection………………………………………………………………55

3.2.5 Real-time PCR…………………………………………………………………55

3.2.6 Hematoxylin-Eosin (HE) staining……………………………………………...57

3.2.7 Bone mineral density (BMD) and Micro-CT analysis…………………………57

3.2.8 Measurement of bone biochemical markers……………………………………61

3.2.9 Measurement of serum level of reproductive hormones……………………….61

XIII

3.2.10 Statistical analysis…………………………………………………………….62

3.3 In vitro study……………………………………………….……………………..62

3.3.1 Experimental design……………………………………………………………62

3.3.2 Cell culture……………………………………………………………………..63

3.3.3 Drug preparation and treatment………………………………………………...65

3.3.4 MTS Assay……………………………………………………………………..65

3.3.5 ALP Assay……………………………………………………………………...66

3.3.6 Real-time PCR…………………………………………………………………67

3.3.7 ERE-dependent luciferase activity assay………………………………………68

3.3.8 Blocking effects of signaling inhibitors on actions of DBT……………………69

3.3.9 Interactive effects between DBT and SERMs………………………………….69

3.3.10 Statistical analysis…………………………………………………………….69

Chapter 4 Characterization of tissue selectivity of icariin in mature ovariectomized (OVX) rats and estrogen-sensitive cell lines……………………..72

4.1 Introduction………………………………………………………………………73

4.2 Results……………………………………………………………………………76

4.2.1 Characterization of estrogenic effects of icariin in OVX rats………………….76

4.2.1.1 Effects of icariin on body weight gain of OVX rats………………………….76

4.2.1.2 Effects of icariin on uterus of OVX rats……………………………………..78

XIV

4.2.1.3 Effects of icariin on breast tissue of OVX rats………………………………81

4.2.1.4 Effects of icariin on brain tissue of OVX rats……………………………….83

4.2.1.5 Effects of icariin on bone of OVX rats…………...... 86

4.2.2 Characterization of estrogenic effects of icariin in estrogen sensitive cell lines.95

4.2.2.1 Effects of icariin in MCF-7 cells……………………………………………..95

4.2.2.2 Effects of icariin in Ishikawa cell…………………………………………….99

4.2.2.3 Effects of icariin in SH-SY5Y cell………………………………………….102

4.2.2.4 Effects of icariin in MG-63 cells………………………………………….105

4.3 Discussion………………………………………………………………………110

Chapter 5 Characterization of tissue selectivity of DBT in mature ovariectomized

(OVX) rats and estrogen-sensitive cell lines……………………………………...119

5.1 Introduction……………………………………………………………………..122

5.2 Results…………………………………………………………………………..124

5.2.1 Quality control and standardization of DBT extract………………………….124

5.2.2 Characterization of the estrogenic effects of DBT in OVX rats………………126

5.2.2.1 Effects of DBT on body weight gain in OVX rats …………………………126

5.2.2.2 Effects of DBT on serum reproductive hormones in OVX rats…………….128

5.2.2.3 Effects of DBT on mRNA expression of aromatase in subcutaneous adipose tissue of OVX rats………………………………………………………………….130

5.2.2.4 Effects of DBT on uterus in OVX rats…………...... 132

5.2.2.5 Effects of DBT on breast tissue in OVX rats………………………………135

5.2.2.6 Effects of DBT on brain tissue in OVX rats………………………………..137

5.2.2.7 Effects of DBT on bone in OVX rats ………………………………………139

5.2.3 Characterization of estrogenic effects of DBT in estrogen sensitive cells……146

5.2.3.1 Effects of DBT in MCF-7 cells…………………...... 146

5.2.3.2 Effects of DBT in Ishikawa cells……………………………………………151

5.2.3.3 Effects of DBT in SH-SY5Y cells…………………………………………..153

5.2.3.4 Effects of DBT in MG-63 cells and possible mechanisms involved……….155

5.3 Discussion…………………………………………………………...... 161

Chapter 6 Characterization of the potential interactions between DBT and

SERMs (Tamoxifen and Raloxifene) in mature ovariectomized (OVX) rats and estrogen-sensitive cell lines………………………………………………………..172

6.1 Introduction……………………………………………………………………..173

6.2 Results…………………………………………………………………………..176

6.2.1 Characterization of interactive effects between DBT and SERMs in OVX rats…………………………………………………………………………………..176

6.2.1.1 Interactive effects between DBT and SERMs on body weight gain of OVX rats ………………………………………………………………………………….176

XVI

6.2.1.2 Interactive effects between DBT and SERMs on serum reproductive hormones in OVX rats…………………………………………………………………………178

6.2.1.3 Interactive effects between DBT and SERMs on uterus in OVX rats……...180

6.2.1.4 Interactive effects between DBT and SERMs on breast tissue in OVX rats..185

6.2.1.5 Interactive effects between DBT and SERMs on brain tissue in OVX rats...187

6.2.1.6 Interactive effects between DBT and SERMs on bone in OVX rats……….190

6.2.2 Characterization of interactive effects between DBT and SERMs in estrogen sensitive cells………………………………………………………………………..200

6.2.2.1 Interactive effects between DBT and SERMs in MCF-7 cells……………...200

6.2.2.2 Interactive effects between DBT and SERMs in Ishikawa cells……………202

6.2.2.3 Interactive effects between DBT and SERMs in SH-SY5Y cells…………..204

6.2.2.4 Interactive effects between DBT and SERMs in MG-63 cells……………...206

6.3 Discussion……………………………………………………………...... 208

Chapter 7 Discussion and Conclusion…………..…...... 217

7.1 Discussion………………………………………………………………………218

7.2 Conclusion……………………………………………………………………...231

7.3 Limitation and recommendations for further research……………………….....232

Reference

XVII

List of Figures and Tables

Figure 1-1 Stages of reproductive aging in postmenopausal women…………………4

Figure 1-2 Working model for actions of estrogen in bone…………………………..4

Figure 1-3 Distribution of ERs and their corresponding functions in the brain………9

Figure 1-4 Structures of estrogen receptors ER and ER with their respective domains and functional regions………………………………………………………14

Figure 1-5 Distribution of estrogen receptors………………………………………..19

Figure 1-6 Mechanisms of estrogen signaling……………………………………….19

Figure 1-7 Biosynthesis of estrogens from cholesterol………………………………23

Figure 1-8 Neuroendocrine regulations of estradiol and progesterone from the ovary………………………………………………………………………………….24

Figure 1.9 Chemical structure of the so far known phytoestrogens………………….28

Figure 1-10 Herba Epimedii and chemical structure of icariin………………………35

Figure 3-1 Flow chart for the animal experiment to test the tissue-selectivity of icariin in vivo……………………………………………………………………………….50

Figure 3.2 Flow chart for the animal experiment to test the tissue-selectivity of DBT alone and interactions between DBT and SERMs in vivo………………….51

Figure 4.1 Effects of icariin on body weight gain of OVX rats……………………..77

Figure 4.2 Effects of icariin on uterus index and mRNA expression of

XVIII estrogen-responsive genes in uterus of OVX rats……………………………………80

Figure 4.3 Effects of icariin on morphology of mammary gland in OVX rats………82

Figure 4.4 Effects of icariin on mRNA expression of estrogen-responsive genes in striatum of OVX rats………………………………………………………………..85

Figure 4.5 Effects of icariin on micro-architecture and bone mineral density at distal femur, proximal tibia and lumbar spine of OVX rats………………………………..89

Figure 4.6 Effects of icariin on the bone turnover biomarkers of OVX rats…………94

Figure 4.7 Effects of icariin on cell proliferation, mRNA expression of estrogen-responsive genes and ERE-dependent luciferase activity in MCF-7 cells…………………………………………………………………………………...98

Figure 4.8 Estrogenic effects of icariin on cell proliferation, ALP activity,

ERE-dependent luciferase activity and expression of estrogen-responsive genes in

Ishikawa cell………………………………………………………………………100

Figure 4.9 Effects of icariin on cell proliferation, mRNA expression of estrogen-responsive genes and ERE luciferase activity in SH-SY5Y cells………104

Figure 4.10 Effects of icariin on cell proliferation and ALP activity in MG-63 cells106

Figure 4.11 Effects of icariin on mRNA expression of estrogen-responsive genes and

ERE-luciferase activity in MG-63 cells…………………………………………108

Figure 5.1 Chemical standardization of DBT by HPLC fingerprint analysis………125

XIX

Figure 5.2 Effect of DBT on body weight gain of OVX rats……………………….127

Figure 5.3 Effects of DBT on serum reproductive hormones in OVX rats……...... 129

Figure 5.4 Effects of DBT on mRNA expression of aromatase in adipose tissue in

OVX rats……………………………………………………………………………131

Figure 5.5 Estrogenic effects of DBT on uterus index, endometrial morphology and mRNA expression of estrogen-responsive genes in uterus of OVX rats…………...133

Figure 5.6 Estrogenic effects of DBT on the morphology of breast tissues in OVX rats…………………………………………………………………………………..136

Figure 5.7 Estrogenic effects of DBT on mRNA expression of Tyrosine hydroxylase

(TH) and Dopamine Transporter (DAT) in striatum of OVX rats…………………..138

Figure 5.8 Effects of DBT on micro-architecture and bone mineral density at distal femur, proximal tibia and lumbar spine as well as the bone turnover biomarkers of

OVX rats……………………………………………………………………………142

Figure 5.9 Effects of DBT on bone turnover biomarkers of OVX rats…………….145

Figure 5.10 Effects of DBT on cell proliferation, ERE-dependent luciferase activity, and mRNA expression of estrogen-responsive genes in MCF-7 cells...... 148

Figure 5.11 Effects of DBT on cell proliferation, ALP activity, ERE-dependent luciferase activity and expression of estrogen-responsive genes in Ishikawa cell….152

Figure 5.12 Effects of DBT on cell proliferation, ERE-dependent luciferase activity

XX and expression of estrogen-responsive genes in SH-SY5Y cell……………..……...154

Figure 5.13 Effects of DBT on cell proliferation in MG-63 cells…………………..156

Figure 5.14 Effects of DBT on ERE-dependent luciferase activity and mRNA expression of estrogen-responsive genes in MG-63 cells…………………………..158

Figure 5.15 Blocking effects of ICI182,780, U0126 and LY294002 on actions of DBT in MG-63 cells………………………………………………………………………160

Figure 6.1 Effects of DBT, tamoxifen, raloxifene and their combinations on body weight gain of OVX rats……………………………………………………………177

Figure 6.2 Effects of DBT, tamoxifen, raloxifene and their combinations on serum reproductive hormones in OVX rats………………………………………………..179

Figure 6.3 Effects of DBT, tamoxifen, raloxifene and their combinations on uterus index, endometrial morphology and mRNA expression of estrogen-responsive genes in uterus of OVX rats……………………………………………………………….183

Figure 6.4 Effects of DBT, tamoxifen, raloxifene and their combinations on histology of breast in OVX rats……………………………………………………………….186

Figure 6.5 Effects of DBT, tamoxifen, raloxifene and their combinations on mRNA expression of estrogen-responsive genes in striatum of OVX rats…………………189

Figure 6.6 Effects of DBT, tamoxifen, raloxifene and their combinations on micro-architecture and bone mineral density at distal femur, proximal tibia and lumbar

XXI spine in OVX rats…………………………………………………………………..193

Figure 6.7 Effects of DBT, tamoxifen, raloxifene and their combinations on the bone turnover biomarkers in OVX rats…………………………………………………...199

Figure 6.8 Interactive effects of DBT and tamoxifen or raloxifene on cell proliferation in MCF-7 cells………………………………………………………………………201

Figure 6.9 Interactive effects of DBT and tamoxifen or raloxifene on ALP activity in

Ishikawa cells……………………………………………………………………….203

Figure 6.10 Interactive effects of DBT and tamoxifen or raloxifene on cell proliferation in SH-SY5Y cells……………………………………………………..205

Figure 6.11 Interactive effects of DBT and tamoxifen or raloxifene on ALP activities in MG-63 cells……………………………………………………………………....207

Figure 7.1 Regulations of DBT on sex hormone profile and hypothalamus-pituitary-gonadal axis in OVX rats………………………………….229

Figure 7.2 Possible mechanisms for the tissue-selective effects of phytoestrogens..230

Table 1-1 Characteristics of estrogen receptors……………………………………...13

Table 3-1 Chemical information of I.S. and four standard components in DBT extracts……………………………………………………………………………….47

Table 3-2 Ingredients of the phytoestrogen-free diet and the icariin diet……………54

Table 3-3 Sequence of primers for estrogen-responsive genes (rat)…………………59

XXII

Table 3-4 Variables for assessment of trabecular bone microarchitecture…………...60

Table 3-5 Specific estrogen-responsive parameters for each cell line……………….70

Table 3-6 Culture conditions for each cell line………………………………………70

Table 3-7 Sequences of primers for the estrogen-responsive genes (human)………..71

Table 4.1 Effects of icariin on trabecular bone properties at distal femur of OVX rats……………………………………………………………………………………91

Table 4.2 Effects of icariin on trabecular bone properties at proximal tibia of OVX rats……………………………………………………………………………………91

Table 4.3 Effects of icariin on trabecular bone properties at lumbar spine of OVX rats……………………………………………………………………………………92

Table 4.4 Summary of the tissue-selective estrogenic effects of icariin in four estrogen sensitive tissues in comparison to 17ß-estradiol……………………………………120

Table 5.1 Chemical composition of DBT extract…………………………………...125

Table 5.2 Effects of DBT on trabecular bone properties at distal femur of OVX rats…………………………………………………………………………………..143

Table 5.3 Effects of DBT on trabecular bone properties at proximal tibia of OVX rats…………………………………………………………………………………..143

Table 5.4 Effects of DBT on trabecular bone properties at lumbar spine of OVX rats…………………………………………………………………………………..143

XXIII

Table 5.5 Summary of the tissue-selective estrogenic effects of DBT in four estrogen sensitive tissues……………………………………………………………………171

Table 6.1 Effects of DBT, tamoxifen, raloxifene and their combinations on trabecular bone properties at distal femur of OVX rats………………………………………..195

Table 6.2 Effects of DBT, tamoxifen, raloxifene and their combinations on trabecular bone properties at proximal tibia of OVX rats…………………………………...... 195

Table 6.3 Effects of DBT, tamoxifen, raloxifene and their combinations on trabecular bone properties at lumbar spine of OVX rats……………………………………….196

Table 6.4 Summary of potential interactions between DBT and tamoxifen or raloxifene in four estrogen sensitive tissues………………………………………...216

XXIV

Abbreviations

AchE Acetylcholinesterase

AD Alzheimer’s Disease

ALP Alkaline Phosphatase

AP-1 Activator Protein-1

BMD Bone Mineral Density

BS Bone Surface

BV/TV Bone Volume/Total Volume

C3 Complement Component 3

CAM Complementary Alternative Medicine

cAMP Cyclic Adenosine Monophosphate

cs-FBS Charcoal-stripped FBS

DAT Dopamine Transporter

DFS Disease-Free Survival

DMEM Dulbecco’s Modified Eargle Medium

DPD Deoxypyridinoline

E2 17ß-estradiol

ER Estrogen Receptor

ERE Estrogen Responsive Element

ERK Extracellular signal-regulated Kinase

FBS Fetal Bovine Serum

FSH Follicle-Stimulating Hormone

GH Growth Hormone

GnRH Gonadotropin-Releasing Hormone

GPER G-Protein Coupled Estrogen Receptor

XXV

GPR30 G-Protein Receptor 30

H&E Hematoxylin-Eosin

HRT Hormone Replacement Therapy

HUVEC Human Umbilical Vein Endothelial Cell

IGF-I Insulin-like Growth Factor 1

LH Luteilizing Hormone

MAPK Mitogen-Activated Protein Kinase

MEM Modified Eargle Medium

MENQOL Menopause-specific Quality of Life mRNA Messenger RNA

MTS 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxypenyl)-2-(

4-sulfophenyl)-2H-tetrazolium

NF-kB Nuclear Factor kB

OPG Osteoprotegerin

OS Overall Survival

OVX Ovariectomy

PD Parkinson’s Disease

PI3K Phosphatidylinositide 3-kinases

PLB Passive Lysis Buffer

PMS Phenazine Methosulfate

PR Progesterone Receptor

RANKL Receptor Activator Nuclear Factor Kappa-B Ligand

RBC Red Blood Cell

SERM Selective Estrogen Receptor Modulator

Tb.N Trabecular Bone Number

XXVI

Tb.Sp Trabecular Bone Separation

Tb.Th Trabecular Bone Thickness

TCM Traditional Chinese Medicine

VEGF Vascular Endothelial Growth Factor

WBC White Blood Cell

WHI Women’s Health Institute

WHO World Health Organization

XXVII

Chapter 1

Background

and

Introduction

1.1 Women’s health and postmenopausal syndrome

1.1.1 Definition of postmenopausal syndrome

Postmenopause refers to the period after menopause until the end of life. World

Health Organization (WHO) has defined menopause as the permanent cessation of menstrual cycle for more than 12 months resulted from dysfunction of ovary

(Organization, 1996). Menopause occurs at the ages of 45 to 55 with a mean age of

52.3 year and indicates the loss of reproductive ability and aging in women (Coulam et al., 1986). It does not occur at a specific age but to some extent can be predicted by a series of preceding changes (Brambilla et al., 1989; Parazzini, 2007).

1.1.2 Symptoms of postmenopausal syndrome and influences on women

Menopause happens in both male and female but exerts more pronounced influences on women who possess much higher basic hormone level (Nakajima, 1999). Varieties of manifestations ranging from uncomfortable experience as hot flashes, night sweat, vaginal dryness, asthenia, palpitation, headache, bone and joint pain and breast tenderness to degenerative disorders like osteoporosis, neurodegenerative diseases are involved and suffered by 70% women from the onset of menopause until the end of their lives (Beyene, 1986; Ohta et al., 1998). Psychological symptoms including fatigue and loss of confidence, attention-deficit disorder, amnesia, irritability even depression come along with these physical symptoms (Avis et al., 2005; Obermeyer et

2 al., 2007). Severity of symptoms associated with menopause is determined by biological, psychological, social and cultural status of the women (Beyene, 1986;

Lock, 2002; Sievert et al., 2007).

Life qualities of postmenopausal women are seriously deteriorated from the beginning of the postmenopausal period to the end of women’s life, accounting for one third of women’s life. In Daly’s study (Daly et al., 1993), life quality of postmenopausal women is evaluated by scoring, in which “0” stands for death while “1” equals to perfect health. Scoring for the postmenopausal women varied with the symptoms they were experiencing. For those women with mild symptom, their average scoring for life quality was 0.65 while for their counterparts with deteriorated experiences, the scoring was only 0.3. The differences between scoring reflected the severity level of substantial influences menopause brings to women.

1.2 Estrogen and menopause

Estrogen is the most affected hormone during menopause. It fluctuates slightly during the early phases of menopause but declines sharply in the following phases as shown in figure 1.1 (Hansen et al., 2012). Estrogens belong to steroids hormone and is the primary sex hormone in female that regulate the reproductive functions and help in maintenance of secondary sex characteristics in female.

Figure 1.1 Stages of reproductive aging in postmenopausal women (Soules et al.,

2001)

Figure 1.2 Working model for actions of estrogen in bone (Khosla et al., 2012)

1.2.1 Physiological actions of estrogen and manifestation of estrogen disruption

While estrogens are produced in both man and woman, they are present at a significantly higher level in mature women and regulate the development and reproductive functions in ovary, uterus and breast where estrogens act as a sex hormone. Besides these tissues, estrogens also play crucial roles in bone and central nervous system where they serve not just as a sex hormone.

1.2.1.1 Uterus

Uterus is an estrogen-sensitive organ where fetus develops. In uterus, estrogen acts as the reproductive hormone to enhance the endometrium and the uterine lining for implantation of fertilized egg. Estrogen stimulates the muscles in the uterus to develop and contract. Contractions help during the delivery of a child and placenta and help the uterine wall to cast off dead tissue during menstruation. During menopause, due to the extremely low level of estrogen, uterine shrink in both size and weight and the menstrual cycle becomes irregular and finally ceases, indicating the disappearance of reproductive ability in female. Supplement with exogenous estrogen attenuates these changes caused by estrogen deficiency. That is why estrogen is used in treatment of menopause-related symptoms in uterus (Mirkin et al., 2014). However, long-term exposure to exogenous estrogen causes hyperplasia of endometrium and increases the risk of endometrial cancer (Yager, 2015).

1.2.1.2. Bone

Remodeling involving osteoclastic bone resorption and osteoblastic bone formation continuously takes place in the skeleton. Overall bone building activity occurs when osteoblast activity and/or number increases with or without a decrease in osteoclast activity and/or number; on the other hand, bone mineral density will decrease when any osteoblast activity and/or number decrease, especially with an increase in osteoclast activity and/or number. Estrogen plays a crucial role in the regulation of bone metabolism in both women and men via actions on osteocytes, osteoblasts and

T-cells (Wang et al., 2012). By activating the Fas-mediated cell death via activation of

ERα and regulating the RANKL/OPG ratio (Kameda et al., 1997; Kousteni et al.,

2002; Shevde et al., 2000), estradiol induces apoptosis of osteoclasts while protects osteoblast and prolongs lifespan of osteoblast via stimulation of anti-apoptotic proteins and repressing pro-inflammatory and pro-osteoclastic cytokines as well

(Chang et al., 2009; Kousteni et al., 2001; Weitzmann et al., 2002), resulting in a net increase in bone building. Actions of estrogen in bone and corresponding target cells are summarized in figure 1.2.

Protective effects of estrogen on cortical and trabecular bone as well as different actions in several bone cell types have been confirmed. The decreased serum level of estrogen due to ovarian dysfunction in menopause and following increase in

6 follicle-stimulating hormone level cause the imbalance between activities of osteoblast and osteoclast, leading to bone loss and high risk of bone fracture (Sun et al., 2006). This is the postmenopausal osteoporosis characterized as low bone mass and deterioration in bone micro-architecture at long bone, hip and spine (Kanis et al.,

1994), which has become a major public health concern. Osteoporotic fractures were reported as a public health problem with high morbidity and mortality and brings heavy burden on the postmenopausal women aged 50 and over (Rahmani et al., 2009), making it an urgent problem to be settled.

1.2.1.3 Central nervous system

Estrogens are known to induce a plethora of effects including the control of reproduction in central nervous system. Besides, estrogens have been reported to be referred to the cognition and memory in the central nervous system. This potential

“non-reproductive” actions of estrogen in brain attracted special interests among scientists following findings that estrogen receptors including the newly discovered

G-protein estrogen receptor (GPER) are widely and abundantly expressed in the brain, such as hippocampus, cortex, striatum and amygdala as shown in figure 1.3

(Acevedo-Rodriguez et al., 2015; Almey et al., 2015; Almey et al., 2016; Bean et al.,

2014; Xu et al., 2015). Although widespread, distributions of ERs are regionally specific, indicating the association with the functions in each region of brain. The

7 higher incidence of Alzheimer’s disease (AD) in postmenopausal women than their man counterparts of a similar age has been recognized to be related to estrogen deficiency (Baum, 2005). Lower level of estrogen could be implicated for dementia in postmenopausal women (Rocca et al., 2014). In addition, estrogen also protect the menopausal women from Parkinson’s disease (PD), a common aging-related neurodegenerative disease mainly affects the mid-brain. According to a clinical trial conducted from the year 1950 through 1987, unilateral and bilateral oophorectomy before menopause may be associated to increased risk of PD (Rocca et al., 2008).

Observations from epidemiological studies suggest that moderate exposure to exogenous estrogen reduces risk of memory impairment, irreversible damage of neurons (Rocca et al., 2008), and improves cognitive performance of postmenopausal women (Sundermann et al., 2006).

Figure 1.3 Distribution of ERs and their corresponding functions in the brain

(Brinton et al., 2015)

1.2.1.4 Breast

Together with growth hormone (GH) and its secretory product insulin-growth factor I

(IGF-I), estrogen stimulates breast development in puberty and breast maturation during pregnancy for lactation as well as the cessation of the flow of milk (Ruan et al.,

1999). Among these three hormones, it is estrogen that primarily and directly induces the pigmentation of nipple and ductal component in breast development. Estrogen also stimulates the accumulation of fat tissue and growth of connective tissue in breast

(Brisken et al., 2010). When menopause starts, sensitivity of mammary glands to estrogen decreases and the proliferation of the epithelial cells as well as the ductal cells cease, leading to the shrank of breast and the cessation of milk-flowing (Haslam et al., 2002). However, development and maintenance of adipose cells do not require the estrogen. Supplement with estrogen attenuates these symptoms caused by estrogen deficiency. Unfortunately, according to the study of WHI, long-term exposure to exogenous estrogen significantly increases the risk of breast cancer in postmenopausal women (Hinds et al., 2010).

1.2.2 Estrogen receptor and tissue-selective responses to ligands

Majority of actions of estrogen in human body are mediated by estrogen receptor (ER)

(Nilsson et al., 2001) and three types of estrogen receptors have been discovered so far (Ho et al., 2002; Jensen et al., 1973).

1.2.2.1 Subtypes of estrogen receptors

Estrogen receptor α was the first discovered and therefore the most thoroughly studied estrogen receptor (Brooks et al., 1973). Four decades later, ERß was identified and shown to possess distinct, non-redundant roles (Kuiper et al., 1996). Recently,

G-protein coupled receptor 30 (GPR30), which belongs to G-protein coupled receptor, has been identified as a non-classical membrane-bound ER and brings novel idea to the investigation of estrogen signaling (Filardo et al., 2000). Table 1.1 shows the characteristics of these three estrogen receptors.

So far, three different isoforms of ERα have been identified and two of them are slightly shorter due to the lack of the N-terminal domain as shown in figure 1.4. But they are still able to heterodimerize with the full length ERα and weaken its activity.

Among all the five identified isoforms of ERß as shown in figure 1.4, four of them have been demonstrated to be different from the full-length ERß in the ligand-binding domain, resulting in the different ligand-binding ability (Smith et al., 2010). The isoforms of ERß that are unable to bind to any ligands or co-activators to activate transcription could dimerize with ERα and silence the signaling mediated by ERα

(Chen et al., 2005). This is the antagonism between ERα and ERß, which has been reported to be involved in many pathological events like prostate cancer (Liu et al.,

2012). ERα promotes cell proliferation and survival as well as E2-mediated gene

11 expression while ERß exerts inhibitory effects on actions of ERα (Nelson et al., 2014;

Williams et al., 2008). At the molecular level, they signal in opposite ways when complexed with the natural hormone estradiol from an AP1 site: activation of estradiol via ERα vs repression of transcription of estradiol mediated by ERß at AP1

(activator protein-1) site (Paech et al., 1997). GPER, a membrane ER mainly localizing at the endoplasmic reticulum, is a multi-pass G-protein with relatively high affinity to estrogen by direct binding (Revankar et al., 2005). Interestingly, GPR30 translocalizes between membranes and cell surface after being activated by estradiol

(Funakoshi et al., 2006; Mo et al., 2013). Although GPR30 has been reported to activate the rapid signaling pathways, physiological functions mediated by it still need further investigation.

Table 1.1 Characteristics of estrogen receptors (Olde et al., 2009)

Receptor ERα ERß GPER characteristic

Receptor Nuclear steroid hormone receptor G-protein coupled receptor superfamily superfamily superfamily

Type Nuclear Membrane-bound G

protein-coupled

Structure DNA-binding domain, 7 transmembrane α-helical

ligand-binding domain, regions, 4 extracellular and 4

N-terminal domain cytosolic segments

Chromosome region 6q25.1 14q23.2 7P22.3

Number of isoforms 3 5 1

Size 595aa 530aa 375aa

Distribution in Uterus, Colon, salivary Central and peripheral human tissues epididymis, gland, vascular nervous system, uterus,

breast, liver, endometrium, ovaries, mammary gland,

kidney, white lung, bladder, testes, spermatogonial cells,

adipose tissue, prostate, ovary, gastrointestinal system,

prostate, ovary, testes, skeleton pancreas, kidney, liver,

testes, skeleton and brain adrenal and pituitary glands,

and brain bone, cardiovascular system,

immune cells

Tamoxifen activity Partial agonist antagonist agonist

ER-estrogen receptor; GPER-G protein coupled estrogen receptor; aa-amino acid

Figure 1.4 Structures of estrogen receptors ERα and ERß with their respective domains and functional regions (Shanle et al., 2010) ERα and ERß are the members of the nuclear factors which share a common structural architecture. The structures of both estrogen receptors contain six conserved functional domain labeled A to F according to their functional properties.

1.2.2.2 Distribution of ERs and tissue-selective responses to ligands

ERs widely distribute throughout the human body (Figure 1.5). ERα and ERß are co-expressed in majority of the ER-positive tissues including brain, breast, uterus and bone while only ERα or ERß is found in liver or gastrointestinal tract, respectively.

Moreover, even in the same tissue, the expression level of ERα and ERß is different

(Kuiper et al., 1997). Similar to that of ERα and ERß, GPR30 also distributes widely throughout the whole body. So far, GPR30 has been identified in central nervous system, liver, breast tissue, bone and ovary (Brailoiu et al., 2007; Heino et al., 2008;

Wang et al., 2007). However, expression level as well as intracellular localization of

GPR30 varies between tissues and cell types.

The wide distribution and different expression level in certain tissue of ERs result in the distinct responses to estrogen from tissue to tissue. In other word, ligands of ER, like estrogen and selective estrogen receptor modulators (SERMs), selectively exert their estrogenic effects depending on expression of ER in the target tissue. Some responses only appear in certain and partially or not appear in other tissues. These responses are generally termed tissue-selective responses. For the ligands themselves, it is called tissue-selective effect or tissue-selectivity.

1.2.2.3 Estrogen signaling

In terms of the outcome of cellular events, like the regulation of gene transcription or

15 activation of signaling factors, the estrogen-dependent signaling can be classified into genomic and non-genomic signaling. For the genomic signaling, estrogen-ER complex binds either directly or indirectly to DNA. Mechanisms of each signaling are showed in Figure 1.6.

Direct genomic signaling is regarded as the classical estrogen signaling pathway.

Intracellular ERα or ERß upon binding with their ligands are activated and translocated to nucleus, and subsequently lead to the alteration of gene transcription via interaction with estrogen response elements (EREs) in the promoters of targets genes (Muyan et al., 2012). Binding of ER with ligand also triggers recruitment of a variety of co-regulators in a complex that changes chromatin structure and facilitates recruitment of the RNA polymerase II transcriptional machinery. In this direct genomic signaling pathway, estrogen-ER complex is often considered as an activator to promote gene expression (Lodish, 2008).

About one third of the estrogen responsive genes have been demonstrated to lack

ERE-like sequences. In these genes, ligand-activated ERs do not directly bind to DNA, but through the protein-protein interaction with other transcription factors at their responsive elements. The subsequent result depends on the type of ligand and the subtype of ER. Activator protein (AP-1), Sp-1, nuclear factor kB (NF-kB), all belong to this kind of transcription factors that facilitate estrogen signaling (Dahlman-Wright

16 et al., 2012; Tsai et al., 2013a).

However, some of the estrogen-induced changes are too rapid that cannot lead to targeted gene transcription and subsequent synthesis of certain protein. That is the non-genomic signaling which is usually mediated by steroid (Ribeiro et al., 2014;

Stournaras et al., 2014) and are usually completed by a series of activation of protein-kinase cascades and ended with indirect changes in gene expression because of the phosphorylation of transcription factors (Jing et al., 2015). The newly found membrane ER, e.g. GPER has been shown to mediate part of this rapid signaling of estrogen (Tsai et al., 2013b). Other variants of ERα and ERß located at cell surface and GPR30 are also associated with this signaling. Binding of estrogen to these membrane ERs usually are followed by mobilization of intracellular calcium, stimulation of adenylate cyclase activity and cyclic adenosine monophosphate

(cAMP), activation of the mitogen-activated protein kinase (MAPK) signaling pathway, activation of the phosphoinositol 3-kinase (PI3K) signaling pathway and activation of membrane tyrosine kinase receptors (Jing et al., 2015; Tsai et al., 2013b).

Besides the genomic and non-genomic ligand-dependent signaling pathway, ERs can also be activated without stimulation of 17ß-estradiol or other suitable ligands by the phosphorylation of ERs on multiple amino acid residues, including serine118, 167,

236, 305. ERs are activated by phosphorylation at these sites in responses to the

17 activation of several kinases which include extracellular signal-regulated kinase/mitogen-activated protein kinase (ERK/MAPK), protein kinase B (Akt) and protein kinase A (Chalupka, 2011; Lannigan, 2003; Murphy et al., 2006). Other residues, like serine 104, serine 106, serine 305, can also be phosphorylated in response to the activations of different protein kinases (Thomas et al., 2008).

Phosphorylation of the residues is believed to be related to some endocrine conditions.

For example, phosphorylation of serine 118 is related to the resistance of breast cancer to treatment with Tamoxifen treatment and may lead to a not so good prognosis (Thomas et al., 2008) while phosphorylation of serine167 may be correlated significantly with the improved disease-free survival (DFS) and overall survival (OS) of breast cancer patients (Chen et al., 2013). Phosphorylation is one of the usual post-transcriptional modifications of peptide that makes peptide active. However, there will still be a long way to go in term of its interpretation.

Recently, it becomes clear that estrogen exerts physiological function through both the genomic and non-genomic signaling pathways. Although it has been regarded as the classical pathway, genomic signaling contributes a small part to the complexity of estrogen signaling. It is possible that estrogens use distinct pathways depending on the cell types and lead to cell-specific regulation.

Figure 1.5 Distribution of estrogen receptors (Burns et al., 2012)

Figure 1.6 Mechanisms of estrogen signaling (Vrtacnik et al., 2014)

I. Direct genomic signaling pathway;

II. Indirect genomic signaling pathway;

III. Non-genomic signaling pathway

IV. Ligand-independent signaling pathway

1.2.3. Biosynthesis of estrogen

Natural estrogen belongs to steroid hormone and naturally exists in three major types:

estrone (E1), estradiol (E2) and estriol (E3) as shown in Figure 1.7. In sexually mature

female, estradiol (E2) is the predominant estrogen in terms of absolute concentration and estrogenic activities (Hall, 2011). Similar to all steroid hormones, estrogens are synthesized from cholesterol as the common precursor. Estrogen are mainly produced and secreted by ovary in sexually mature female, where cholesterol is converted into androstenedione and testosterone in the theca interna and then converted into estradiol by estrone and testosterone under the catalysis of aromatase in granulosa cells

(Channing et al., 1976). Besides, some extragonadal tissues including adrenal gland, liver, breast and adipose cell can also produce estrogen in small amount mainly via bioconversion of estrogen from androstenedione and cholesterol under the catalysis of aromatase (Figure 1.7), which is of special importance for postmenopausal women

(Gruber et al., 2002; Nelson et al., 2001; Pesonen, 1960; Simpson, 2002; YoungLai,

1973).

Among the enzymes involved in the biosynthesis of estrogen, aromatase is the rate-limiting enzyme and therefore called estrogen synthetase or estrogen synthase

(Gray et al., 1995). Crucially, it transforms androstenedione to estrone and testosterone to estradiol as shown in figure 1.7. With aging and subsequent

20 dysfunction of ovary, estrogen from the secondary sources accounts for ever-growing proportion of estrogen in circulation and finally these extragonadal tissues turn into the main source of estrogen for postmenopausal women. Thereby, aromatase-producing organs, such as adrenal gland, and the tissues where aromatase function including liver, adipose tissue, attract enthusiastic attentions from scientists in the field of estrogen. This should be of special significance for postmenopausal women in terms of the management of symptoms induced by estrogen-deficiency.

The expression level of steroid-converting enzymes differs in and between tissues and consequently leads to differences in metabolism of steroid (Ishida et al., 2002). These differences could be enhanced under some special conditions, like treatment for menopause-associated symptoms, finally resulting in tissue-selective effects.

1.2.4 Neuroendocrine regulation of estrogen

Endocrine system of female is regulated and controlled by hypothalamus-pituitary-ovary axis (figure 1.8) as one major part of the neuroendocrine axis. As the advanced center of this axis, hypothalamus secretes gonadotropin-releasing hormone (GnRH) which stimulates the secretion and production of FSH and LH from pituitary after binding with the GnRH receptor in anterior pituitary. FSH and LH, two gonadotropins, regulate functions of ovary including production and secretion of estrogen as well as progesterone. On the

21 contrary, estrogen also influences FSH and LH by feedback regulation. With aging, ovary gradually loses its function and leads to decrease of production and section of estrogen. Then menopause happens. In response to the decrease in estrogen especially the sharp decrease in the beginning of menopause, serum level of FSH and LH increase by several fold after menopause. Moreover, FSH rises disproportionately more than LH because of several possible factors like the loss of estrogen feedback as well as possible changes in their structures and their half-lives in circulation

(Brodowska et al., 2012; Navarro et al., 2012b; WIDE et al., 1984). These hormonal changes directly induce a series of symptoms suffered by 70% women until the end of their lives (Jouyandeh et al., 2013).

Estrone (E1) Estradiol (E2) Estriol (E3)

Figure 1.7 Biosynthesis of estrogens from cholesterol (Häggström, 2014)

Figure 1.8 Neuroendocrine regulations of estradiol and progesterone from the ovary (Kong et al., 2014) Endocrine system of female is regulated and controlled by hypothalamus-pituitary-ovary axis which plays crucial role in the development and regulation of the body’s system. The Gonadotropin-releasing hormone (GnRH) is secreted from the hypothalamus. FSH and LH, produced and secreted by pituitary, regulate functions of ovary including production and secretion of estrogen and progesterone in ovary. On the contrary, estrogen also regulates FSH and LH in a negative-feedback manner.

1.3 Current therapy for postmenopausal syndrome

Menopause does not really require medical treatment since it is a biological process in nature. All the clinical treatments just focus on relieving their symptoms and preventing any chronic conditions that may occur during the postmenopausal years, such as heart disease and osteoporosis.

1.3.1 Hormone replacement therapy (HRT)

Hormone replacement therapy (HRT) has been traditionally regarded as the gold standard method for management of postmenopausal syndrome. Estrogen alone or combination of estrogen and progesterone (Hekimoglu et al., 2010) are likely to be the most effective treatments for postmenopausal hot flushes, vaginal dryness and osteoporosis (Cohen et al., 2003; Maclennan et al., 2004). Besides, according to results of the retrospective studies, the earlier the women receive HRT, the later dementia happens, indicating that HRT may delay the pathogenesis of Alzheimer’s disease (AD) (Bagger et al., 2005).

However, because of the possible side effects of the long-term treatment with HRT, especially the increased risk of thromboembolic accidents, endometrial cancer, breast cancer and stroke (Schwartz et al., 2015), use of HRT has declined in clinical setting.

Therefore, new drugs that reach the designated effects for management of menopause but without causing undesirable effects are in urgent needs.

1.3.2. Selective estrogen receptor modulators (SERMs)

SERMs are ER ligands that are full or partial agonists in some tissue

(Dahlman-Wright et al., 2006), like bone, but antagonists in other tissue, such as breast and uterus, which makes it possible for SERMs to exert like estrogen without side effects of estrogen (Komm et al., 2014). Effects of SERMs in terms of treatment of breast cancer and postmenopausal osteoporosis have been clinically confirmed. In addition, evidences from clinical study showed the protective effects of SERMs against Alzheimer’s disease (O'Neill et al., 2004a; O'Neill et al., 2004b). Both tamoxifen and raloxifene alone have been reported to protect neurons in either

Parkinson’s animal models (Bourque et al., 2012; Morissette et al., 2008; Tian et al.,

2009) or Alzheimer’s disease models (Breuer et al., 2000; Du et al., 2004), indicating the potential neuroprotective effects of SERMs. However, both clinical and preclinical studies demonstrated that tamoxifen and raloxifene exerts as a partial agonist in uterus and has the potential to stimulate the proliferation of endometrium and growth of endometrial cancer (DeMichele et al., 2008; Wickerham et al., 2002).

1.3.3 Phytoestrogens

Due to the side effects of HRT, increasing women turn to the use of complementary and alternative medicine (CAM) for help during the past few decades (Buhling et al.,

2014; Peng et al., 2014). Phytoestrogens derived from the natural plants are the most

26 popular alternative approach among the CAMs (Borrelli et al., 2010). They are groups of non-steroidal compounds that are derived from natural plants with structural and functional similarities to that of estrogen (Brzezinski et al., 1999). That’s why they have been termed as “phytoestrogen”.

1.3.3.1 Structural classification of phytoestrogens

Based on the chemical structures, phytoestrogens are classified as isoflavones, lignans and coumestans. The main classes of phytoestrogens are the isoflavones (genistein, daidzein, glycitein, euuol and biochanin A), the lignans (enterolactone, enterodiol), the coumestanes (coumestrol), the flavonoids (quercetin, kaempferol), the stilbenes

(resveratrol) and the mycotocins (zearalenol) (Benassayag et al., 2002) (Figure 1.9).

The aforementioned compounds are all polyphenols and structurally resemble the natural estrogens (Adlercreutz, 2002) (Miksicek, 1995). Due to the competitive binding to ER with estrogen, phytoestrogen exert both the agonistic and antagonistic effects depending on endogenous level of estrogen and tissue types. Because of this, phytoestrogens are usually regarded as natural estrogen receptor modulators (An et al.,

2016).

Figure 1.9 Chemical structure of the so far known phytoestrogens (Benassayag et al., 2002)

1.3.3.2 Effects of phytoestrogen in management of postmenopausal symptoms

Users of soy supplementation containing 90mg of isoflavone for 16weeks faced a

49.8% reduction of daily hot flush, suggesting that soy has effect in releasing hot flush (Carmignani et al., 2010). Similar result was found in genistein users. Oral taking of genistein for 12 weeks reduced hot flushes by 22% when compared with placebo (Crisafulli et al., 2004). Not only the frequency, intensity of hot flushes was also found to be relieved after daily taking 104mg of isoflavone for 12 months in 169 postmenopausal women with medium or severe-climacteric symptoms (Drews et al.,

2007). Isoflavones were found to protect vagina in a time-dependent manner.

Short-term use manifests at least protection against vagina atrophy, like weaker dryness compared with placebo group (Manonai et al., 2006) and this protective effect becomes more apparent with prolonged usage (Umland et al., 2000). Conversely, some studies failed to observe the relief of vasomotor symptoms after daily receiving soy isoflavone for two years (Chalupka, 2011). Despite conflicting results, there are abundant evidences that soy extract confers overall benefits for relief of vasomotor symptoms in postmenopausal women.

Beneficial effects of phytoestrogen on human bone health are clear. Asian populations that regularly consume much more soy have a much lower incidence of osteoporotic fracture than their Caucasians counterparts (Tham et al., 1998). Recently, Femarelle, a

29 commercial phytoestrogen supplement combined of isoflavone and lignin, has been demonstrated to significantly protect bone against osteoporosis in the same degree as

HRT in postmenopausal women (Labos et al., 2013). However, results of animal studies differ. 6-month-old ovariectomized rats were given isoflavones at dosages of

0.3 and 0.8 mg/g of diet for 8 weeks. However, estrogen rather than isoflavone was found to protect trabecular bone loss (Cai et al., 2005). In addition, calcium and bone balance were unaffected by soy isoflavones. Lack of effects on trabecular bone loss was found in a 3-year study on monkeys (Register et al., 2003).

1.3.3.3 Controversy on phytoestrogens

Although phytoestrogens seem to exert some effects regarding treatment of menopause, concerns about their safety and tolerability have been raised since they are not FDA-approved. Studies on monkeys reported that soy isoflavonoid helped the clearance of endogenous estradiol from circulation and breast tissue and reduced the proliferation in uterine and breast after exposure to soy isoflavone for 36 months in

OVX monkeys (Wood et al., 2007; Wood et al., 2006), indicating that it was safe to use isoflavonoids. Moreover, isoflavone appears to be breast cancer protective and has been recommended moderate lifelong dietary soy consumption as part of a healthy lifestyle for breast cancer women by the North American Menopause Society in 2011

(Clarkson et al., 2011). However, long-term studies with isoflavones, i.e. five years,

30 reported the abnormal uterine bleeding and proliferative endometrium in women

(Villaseca, 2012).

Above all, phytoestrogen may provide a safe and partially effective alternative to HRT and SERMs; however, as these products are not yet FDA-approved, further studies pertaining its mechanism and pharmaco-vigilence are needed.

1.4. Traditional Chinese Medicine (TCM) as a source of phytoestrogens

Traditional Chinese Medicine (TCM) has been used for management of menopausal symptoms with a long history of safe use. Recently, with increasing concerns about the side effects of HRT, many postmenopausal women have turned to TCM for help.

In Chinese medicine, kidney is the root of congenital foundation and problems of menopause are attributed to dysfunction of kidney in both male and female. The TCM concept of kidneys is different from the western medical definition. The TCM concept of kidney is a way of describing a set of interrelated parts than an anatomical organ.

In TCM the kidneys are associated with "the gate of Vitality" or "Ming Men" which involves all physiological functions that include the kidney-urinary system plus the endocrine systems and especially the adrenal glands. Thus, principles of TCM for management of menopause are to prescribe Chinese herbs that can tonify the kidney so as to help women in this special period of life. Recent studies demonstrated that many TCMs prescribed for postmenopausal women exert estrogen-like activities

(Zhang et al., 2005).

1.4.1 Icariin

Herba Epimedii (HEP) is an important medicinal plant used in various Chinese

Medicine formulas for thousands of years as well as in modern Chinese herb products

(Li et al., 2015). In Chinese Medicine, kidney is the root of congenital foundation and problems of menopause are attributed to dysfunction of kidney in both male and female (LI et al., 2013a). Herba Epimedii has traditionally been used to tonify kidney in kidney-deficient conditions to achieve and keep balance of kidney (Leung et al.,

2013; Yan et al., 2008). Among more than 260 components of Epimedium, icariin is the most abundant and therefore the main effective constituent (Guo et al., 1995).

Moreover, icariin has been chosen as the chemical marker for quality control of

Herba Epimedii in Chinese Pharmacopeia (Li et al., 2015). It belongs to flavonoids with a prenyl group at C-8 (Figure 2.0).

Icariin exerts extensive pharmacological effects including anti-osteoporosis, anti-oxidation, anti-apoptosis, anti-cancer (Fan et al., 2016).

Osteoprotective effect

For icariin, the extensively investigated effect is its anti-osteoporosis effect. A randomized double-blind placebo-controlled clinical trial conducted in postmenopausal women has shown that a daily oral intake of a preparation containing

60 mg of icariin for 2 years showed a higher BMD in femoral neck and lumbar spine when compared with their placebo counterparts (Zhang et al., 2007). At animal level, treatment with icariin for 12 weeks increased femoral BMD as well as trabecular thickness and number, indicating its potential to increase bone formation and reduce bone absorption. Wnt/beta-catenin and BMP signaling was proven to probably mediate this osteoprotective effects (Li et al., 2013b). Our previous work in OVX mice demonstrated that icariin stored bone loss induced by OVX in a similar way to estradiol but without inducing uterotrophic effect (Mok et al., 2010). At the cellular level, icariin showed osteoblast-promoting effects in UMR106 cells, MC3T3-E1 cell and primary osteoblast from both rat and mouse (Mok et al., 2010) (Cao et al., 2012) (Hsieh et al.,

2010; Zhang et al., 2012a) by stimulating cell proliferation and osteogenic differentiation. On the other hand, icariin exerted osteoclast-suppressing effects by reducing TRAP activity and bone resorption in RAW 264.7 cell (Cui et al., 2014).

Icariin showed a more potent effect than genistein, another potent bone protective isoflavonoid phytoestrogen in promoting osteoblast differentiation and mineralization of primary osteoblast (Ma et al., 2011). Such effect was believed to be associated to the prenyl group at C-8. These results indicate that icariin might be a potential alternative approach for treatment of postmenopausal osteoporosis.

Neuroprotection

Oral administration of icariin improved spatial memory impairment in in Alzheimer’s disease mice model (5xFAD) by attenuating amyloid ß-protein fragment 42-induced neurotoxicity (Urano et al., 2010) as well as in ß-amyloid-induced AD rat model (Nie et al., 2010) and APP (Amyloid precursor protein, APP) transgenic mice. In addition, icariin has been found to be a promising anti-depression drug in mice (Pan et al., 2005) and remarkably increased the social interaction time in social defeat mice model (Wu et al., 2011). These results indicate the beneficial effects of icariin in nervous system.

Reproductive effects

The primary traditional function of icariin is in reproductive system. Icariin improves sexual function of male rats by increasing sperm counts and testosterone level

(Makarova et al., 2007) and increasing erectile function of castrated Wistar rats (Liu et al., 2005). Oral administration of icariin prolonged estrus cycle of female SD rats for several days (Kang et al., 2012).

Furthermore, icariin has been demonstrated to have some other effects including cardiovascular protective effect (Meng et al., 2015), anti-cancer effect (Fan et al.,

2016), anti-inflammation effect (Ma et al., 2015) and immunoactivation (Teng et al.,

2008).

Figure 1.10 Herba Epimedii and chemical structure of icariin

Icariin, C-8 prenylated flavonol glycoside, is the most abundant constituent of a

Chinese herb-Herba Epimedii and has been used as the chemical marker for quality control of Herba Epimedii in Chinese Pharmacopeia (Li et al., 2015).

1.4.2 Danggui Buxue Tang decoction

1.4.2.1 Clinical use of Danggui Buxue Tang (DBT) decoction

Danggui Buxue Tang (DBT) decoction is a traditional Chinese medicine formula mainly for management of menopause-related symptoms used in China with a long history of safe use. According to LI Dongyuan, who for the first time described DBT in Neiwaishang Bianhuo Lun in the year 1247, this formula should contain two commonly used Chinese herbs, Astragali Radix and Sinensis Radix Angelica, and the ratio of these two herbs should be 5:1, which has been used as standard formula in modern Chinese Medicine (Zhang et al., 2014). DBT has been proven to be effective in alleviating menopausal symptoms and improving their well-beings by raising “Qi” and nourishing “Blood”, in which the lack of “Qi” and “Blood” in Chinese medicine are believed to be the cause of menopause (Zheng et al., 2012a). Thus, disorders related to “Qi” and (or) “Blood” deficiency, fever syndrome of blood deficiency, all kinds of anaemia, allergic purpura caused by deficient qi-blood, fever and headache due to blood deficiency during period and after delivery, all are indications of treatment with DBT.

1.4.2.2 Physiological effects of DBT

Modern pharmacological studies provide further evidences for the actions of DBT.

DBT was reported to increase cell populations in S-phase and expression of vascular

36 endothelial growth factor (VEGF) in human umbilical vein endothelial cells

(HUVEC), indicating its potential role in angiogenesis (Lei et al., 2003).

Polysaccharide extract of DBT was demonstrated to obviously promote the level of red blood cell (RBC), white blood cell (WBC) and hemoglobin in blood-deficient mice model induced by cyclophosphamide and hydracetin, confirming the hematopoietic effect of DBT (Miao et al., 2002). A 3-month randomized, double-blind clinical study reported that DBT at 6g/day significantly reduced the vasomotor symptoms of Hong Kong Chinese postmenopausal women by 30-50% by using the

Greene Climacteric Scale and the vasomotor domain of the Menopause-Specific

Quality of Life (MENQOL) (Haines et al., 2014). Compared with the baseline, hot flushes and night sweat significantly decreased after three- month administration and the effect of DBT on night sweat even lasted until one month after drug withdrawal.

In addition, DBT has been proved to be effective in improving cardiovascular circulation (Chiu et al., 2007), stimulating the immune function (Gao et al., 2008), protecting bone against osteoporosis in both in vivo and in vitro model (Choi et al.,

2011; Xie et al., 2012), increasing antioxidant activity (Gong et al., 2016; Zierau et al.,

2014) and promoting hematopoietic functions (Zheng et al., 2010). Particularly, DBT markedly increased estrogen response element (ERE)-dependent reporter activities and estrogen receptor antagonist ICI182,780 completely blocked the stimulating

37 effects of DBT on cell proliferation and cell differentiation in human osteosarcoma

MG-63 cell (Choi et al., 2011), suggesting that DBT might act via phytoestrogen.

These results indicate that DBT may be a potential alternative for management of menopause.

1.5 Concerns about the potential interactions between phytoestrogen-containing TCM and SERMs

With the growing interest in taking phytoestrogen-containing Chinese Medicine as alternative approaches for management of menopausal symptoms, concerns have been raised for the safety of using herbal medicine as alternatives to HRT since majority of their effects are mediated by the similar receptors and pathways. It is possible that phytoestrogen might carry similar benefit-risk profile as that to estrogen and SERMs.

This is of particular meaning to menopausal women with breast cancer or osteoporosis taking TCM while prescribed standard treatment with either HRT or

SERMs for prevention of recurrence or treatment of menopausal symptoms.

Indeed, due to the increasing popularity of taking herbal medicine among patients prescribed with western drugs, potential herb-drug interaction has become significant challenge for the medical communities. For example, orally feeding with Biochanin A, an isoflavone, reduced bioavailability of tamoxifen and its metabolite in female rats

(Singh et al., 2012). Supplement with either genistein or 8-prenylnaringenin

38 significantly overcame effects of tamoxifen in MCF-7 cells, which was suggested to be better avoided during breast cancer treatment (van Duursen et al., 2013), which was confirmed in a recent study performed in athymic nude mice (Du et al., 2012a).

While, a clinical study conducted in Taiwan demonstrated that consumption of

Chinese herbal products containing coumestrol, genistein, or daidzein was negatively correlated with the incidence of endometrial cancer risk among the tamoxifen-treated female breast cancer survivors (Hu et al., 2015). Thus, there is an urgent demand to characterize the potential interactions that might occur between phytoestrogen or phytoestrogen-containing TCM and prescribed western drugs, so as to provide evidences for the medical professionals to critically assess and make logical decision regarding the benefits versus risks of using herbal products.

It will be ideal for management of postmenopausal syndrome if the alternatives with effects of HRT but with less even no undesirable adverse effects of HRT could be identified. Present study aims to investigate the tissue-selective effects of two representatives of phytoestrogens derived from TCMs, icariin and DBT, in four target tissues of estrogen, hoping that phytoestrogens exerting tissue-selective effects in target tissues with minimal side effects in other estrogen sensitive tissues will be identified. We hope that present study will provide new ideas and scientific evidences for the safe use of Traditional Chinese Medicine alone or in combination with western

39 drugs in treatment of postmenopausal symptoms as well as the development of modern TCMs.

Chapter 2

Hypothesis and

Objectives

2.1. Hypothesis

Due to the severe side effects induced by HRT, increasing number of postmenopausal women consider the use of herbal medicine for treatment of their symptoms. However, concerns have been raised for the safety of using herbal medicine, especially those containing phytoestrogens, as alternative approaches for HRT. As many effects of phytoestrogens are similar to those of estrogen and are mediated by estrogen receptors and similar pathways, phytoestrogen may carry similar risk-benefit profile as estrogen.

Thus, it would be important to identify phytoestrogen that can exert tissue selective estrogenic effects with minimal side effects as alternative regimen for management of menopausal symptoms. As one main source of phytoestrogens, TCM prescribed for management of postmenopausal symptoms has been demonstrated to exert estrogen-like effects. We hypothesized that phytoestrogens derived from Traditional

Chinese Medicine (TCM) selectively exert estrogenic effects in estrogen-sensitive tissues and are useful for management of postmenopausal syndromes. Icariin from

Herba Epimedii (HEP) and Danggui Buxue Tang (DBT) were chosen as sources of phytoestrogens to test our hypothesis in my study.

In addition, the present study used DBT as the representative of herbal medicine to investigate the potential interactions between TCM and SERMs and hypothesized that

DBT did not interact with SERMs in target tissues of estrogen.

2.2. Objectives

The present study aims to characterize the tissue-selective effects of TCM-derived phytoestrogens in target tissues of estrogen and the possible mechanisms involved by using preclinical in vivo and in vitro models. The objectives are:

1. To determine if icariin selectively exert estrogenic effects in estrogen-sensitive

tissues in mature ovariectomized OVX rats model and four estrogen

receptor-positive cell lines;

2. To determine if DBT selectively exert estrogenic effects in estrogen-sensitive

tissues in mature ovariectomized OVX rats model and four estrogen

receptor-positive cell lines;

3. To determine the possible interactions between DBT and SERMs (tamoxifen or

raloxifene) by performing the OVX rat model and the four estrogen

receptor-positive cell lines.

Chapter 3

Methodology

3.1 Authentication, extraction and quality control of DBT extract

3.1.1 Preparation of DBT extract

DBT extract was kindly provided by Prof. Karl Tsim of The Hong Kong University of

Science and Technology. Briefly, the two herbs for preparing Danggui Buxue Tang,

Radix Astragali (RA, Huangqi) and Radix Angelica Sinensis (RAS, Danggui), were purchased from Shanxi province and Gansu province, respectively. Exact amount of

RA and RAS were weighed according to a ratio of 5:1 and then mixed well. The mixture was boiled in 8 volumes of water (v/w) for 2 hours and extracted twice; procedure of extraction seriously followed the ancient recipe that had been shown to have the best extracting condition. And then the extraction was dried by lyophilization

(Choi et al., 2011) and stored at 4℃. For each herb, HPLC assay was performed to make sure that its quality fulfilled the requirement of the China Pharmacopoeia and/or the Hong Kong Chinese Materia Medica Standard.

3.1.2 LC-MS analysis of DBT extract

The chemical analysis of DBT extract was performed by Prof. Tsim’s group.

DBT extracts and chemicals

All chemicals were over 95% purity and dissolved in methanol to give stock solutions at 1 mg/ml and stored at -20℃. Polydatin was used as internal standard (I.S.) as shown in Table 3.1. I.S. working solution of 4 ug/ml was prepared by diluting proper

45 amount of IS stock solution in 50% methanol.

LC-MS analysis

UPLC-MS analysis was carried out on an Agilent 1200 series system combined with

Agilent QQQ-MS/MS (6410A). Chromatographic separation was performed by gradient elution with 0.1% formic acid in water (mobile phase A) and 0.1% formic acid in acetonitrile (mobile phase B) on an Agilent ZORBAX SB-C18 column (1.8 um i.d., 50 mm* 4.6 mm). The injection volume was 10ul. MS parameters were set as following: drying gas temperature, 325 ℃; drying gas flow, 10 L/min; nebulizer pressure, 35 psig; capillary voltage, 4000 V. Data were acquired under MRM model in

ESI+ mode except for ligustilide.

Table 3.1 Chemical information of I.S. and four standard components in DBT extracts.

Compound Chemical structure Folumar Molecular weight

Ferulic acid C10H10O4 194.1

Calycoisin C16H12O5 284.1

Formononetin C16H12O4 268.1

Z-ligustilide C12H14O2 190.1

Polydatin C20H22O8 390.1

3.2 In vivo study

3.2.1 Experimental design

The present experiment was conducted under the animal license issued by Department of Health, Hong Kong Special Administrative Region Government, and the Hong

Kong Polytechnic University Animal Subjects Ethics Sub-committee (animal license

No. 13-104; ASESC Case: 13/14 and 13/18).

6-month old female SD rats were subjected to ovariectomy to mimic estrogen deficient conditions or sham operation. After two weeks’ recovery, the OVX rats were orally treated with vehicle, 17ß-estradiol as positive control and icariin at different dosages or DBT for 12 consecutive weeks. To remove the influences of estrogen-like compositions in diets, rats were paired-fed with phytoestrogen-free AIN 93 diet.

During the whole treatment, the body weight of rat was measured every two weeks.

As mentioned in Chapter 1, in TCM, the kidneys are associated with "the gate of

Vitality" or "Ming Men" which involves all physiological functions that include the kidney-urinary system plus the endocrine systems. In postmenopausal condition, the most relevant hormone is estrogen, deficiency of which has been believed to be the direct cause of menopause. Thus, four estrogen-sensitive tissues and also the most affected tissues during menopause including uterus, breast, brain and bone were selected to evaluate the tissue-selective estrogenic effects of icariin and DBT in OVX

48 rats upon treatment. The experimental flow is shown in Figure 3.1 (tissue-selective effects of icariin) and 3.2 (tissue selective effects of DBT). In addition, OVX rats were also co-treated with DBT and tamoxifen or raloxifene, representatives of herbal medicine and western drugs, to investigate the potential herb-drug interactions in the four estrogen sensitive tissues as mentioned above (Figure 3.2).

Figure 3.1 Flow chart for the animal experiment to test the tissue-selectivity of icariin in vivo

Figure 3.2 Flow chart for the animal experiment to test the tissue-selectivity of

DBT alone and interactions between DBT and SERMs in vivo

3.2.2 Ovariectomy and Sham operation

Six-month old Sprague Dawley rats were purchased from The Chinese University of

Hong Kong. After acclimation for one week, rats were given ovariectomy to remove the bilateral ovaries or sham operation. Briefly, operation was performed under anesthesia of ketamine (50mg/kg) and xylazine (10mg/kg). Dorsal surgical area was shaved and swabbed with surgical swab. Two short incisions 1.5cm to the spine were made under the ribcage. The ovaries were pulled out through the incisions in the abdominal muscle by grasping the periovarian fat. After ligating the junction between oviduct and uterus with synthetic absorbable suture, bilateral ovaries were cut and the uterus was returned to the abdominal cavity. The ovary itself must not be touched otherwise small pieces may become detached and would re-implant and carry on normal function. Incisions on both the muscle wall and skin were sutured. Swab some antiseptic cream on the incisions on skin and place the rats back to the cages. Check health status of rats two or three hours later and give antiseptic cream on the incisions for at least three days after the surgery. Heater was used to keep rats warm during and several hours after the surgery.

For sham operation, the procedure was as same as that of OVX operation except the removal of bilateral ovaries.

3.2.3 Drugs preparation and animal feeding

17ß-estradiol (Sigma, Cat#E8875) was dissolved in distilled water in forms of suspension and orally given to OVX rats as positive control in animal experiments of present study. Icariin was purchased from Shanghai Ronghe Biomedical Development

Co., Ltd. (Cat#25810) and added into diets at three dosages (50, 500 and 3000ppm).

During the preliminary experiment, rats were allowed to freely take diets and the daily amount of intake was recorded for five days. The mean daily intake of rats was established as the daily amount of diet that was supplied in the whole experiment.

OVX rats were subjected to free intake of icariin-containing diet at the dosages of

50ppm (0.05g/kg), 500ppm (0.5g/kg) and 3000ppm (3.0g/kg). Rats in sham, OVX and 17ß-estradiol (1mg/kg.day) treatment groups were paired-given phytoestrogen-free AIN93 diet. The ingredients of the diets are shown in Table 3.1.

DBT powder was dissolved in distilled water. Tamoxifen (Sigma, Cat#T5648) and raloxifene (Sigma, Cat#R1402) were dissolved in distilled water in forms of suspension. After recovery, OVX rats were orally given vehicle, 17ß-estradiol

(2mg/kg.day), DBT (3g/kg.day), tamoxifen (1mg/kg.day), taloxifene (3mg/kg.day) as well as combination of DBT plus tamoxifen and DBT plus raloxifene for 12 consecutive weeks. During the whole treatment, rats were paired-given phytoestrogen-free AIN93 diet. The ingredients of AIN93 diet are shown in Table 3.2.

Table 3.2 Ingredients of the phytoestrogen-free diet and the Icariin-containing

diets

Product D00031602 D15061901 D15061902 D15061903 (AIN93) Control diet 50ppm Icariin 500ppm Icariin 3000ppm Icariin gm% kcal% gm% kcal% gm% kcal% gm% kcal% Protein 14 15 14 15 14 15 14 15 Carbohydrate 73 76 73 76 73 76 73 76 Fat 4 9 4 9 4 9 4 9 Total 100 100 100 100 kcal/gm 3.8 3.8 3.8 3.8 Ingredient Casein 140 560 140 560 140 560 140 560 L-Cystine 1.8 7 1.8 7 1.8 7 1.8 7 Corn starch 495.692 1983 495.6 1983 495.69 1983 495.69 1983 92 2 2 Maltodextrin 10 125 500 125 500 125 500 125 500 Sucrose 100 400 100 400 100 400 100 400 Cellulose, BW200 50 0 50 0 50 0 50 0 Soybean oil 40 360 40 360 40 360 40 360 Corn oil 0 0 0 0 0 0 0 0 t-Butyljydroquinone 0.008 0 0.008 0 0.008 0 0.008 0

Mineral Mix 35 0 35 0 35 0 35 0 S10022M Vitamin Mix 10 40 10 40 10 40 10 40 V10037 Choline Bitartrate 2.5 0 2.5 0 2.5 0 2.5 0 Icariin 0 0 0.05 0 0.5 0 3.0 0 FD&C Yellow 0 0 0 0 0 0 0 0 Dye#5 FD&C Red Dye #40 0 0 0 0 0 0 0 0 FD&C Blue Dye #1 0 0 0 0 0 0 0 0 Total 1000 3850 1000. 3850 1000.5 3850 1003 3850 05 Icariin (g/kg) 0.050 0.500 3.0

3.2.4 Sample collection

Upon treatment, one day before sacrifice, the rats were housed in metabolic cages

(one rat/per cage) and urine of 24 hours was collected in the morning of sacrifice day.

Supernatant was collected after centrifugation at 4000rpm/min for 15mins and aliquots were stored at -80℃ for further measurement. At sacrifice, blood was drawn from the abdominal aorta under the anesthesia with ketamine (50mg/kg) and xylazine

(10mg/kg). After centrifugation at 4000rpm at 4℃ for 15mins, serum was collected and aliquots of serum were stored at -80℃ for further measurement. Uterus was freshly collected and weighed. After that, half of uterus was immediately fixed in 4% paraformaldehyde for 6 hours for paraffin slides and the other half was stored at -80℃ for further detection. Striatum was collected and stored at -80℃ for real time-PCR assay. The whole left leg and spine were collected for micro-CT. The second breast tissue together with its surrounding skin was collected in 4% paraformaldehyde and fixed for 6 hours for H&E staining.

3.2.5 Real-time PCR

Uterus, striatum and abdominal adipose tissue were freshly collected at sacrifice and immediately stored at -80 ℃ until further assay. mRNA expressions of estrogen-responsive genes including estrogen receptor (ER), progesterone receptor

(PR) and complement components 3 (C3) in uterus, tyrosine hydroxylase (TH) and

55 dopamine transporter (DAT) in striatum, aromatase in adipose tissues were determined by Real-time PCR reaction.

Total RNA extraction

Total RNA was extracted from animal samples by Trizol reagent (Life Technologies,

Cat#15596) according to manufacturers’ instructions.

Reverse transcription PCR

2ug of the total RNA was used to generate cDNA in a 20ul of RT-PCR reaction system by using High-Capacity cDNA Reverse Transcription Kits (Applied

Biosystems, Cat#4368814) following the manufacturers’ instructions.

The 200ul PCR tubes containing the 20ul of reaction system were placed into

GeneAmp PCR System 9600 Thermal Cycler for RT-PCR at 25℃ for 10 min, 37 ℃ for 2h and 85℃ for 5min. The cDNA product was stored at -20℃ and used as template for real-time PCR reaction.

Quantitative Real-time PCR

2ul of the cDNA diluted by 10 times, 0.4ul of forward and reverse primers (as shown in Table 3.3), respective, 7.2ul of DNase and RNase-free water and 10ul of SsoFastTM

EvaGreen® Supermix (BIO-RAD, Cat#172-5213) were mixed well to get the 20ul of reaction system for real-time PCR. The iCycler with Iq5 Multicolor Real-time PCR

Detection System (Bio-Rad, IQ5) was used to perform and monitor the real-time

56 quantitative PCR reaction. Conditions for each primer were optimized and the optimal condition was chosen for each primer (as shown in Table 3.3). For each gene, standard curve was established to determine the relative quantity of mRNA and the melting curve was used to assess the specificity of the amplification.

3.2.6 Hematoxylin-Eosin (HE) staining

Uterus and breast were removed from sham and OVX rats treated with vehicle,

17ß-estradiol, icariin or DBT and fixed with 4% paraformaldehyde. After dehydration

(Leica HI1220), tissues were embedded in paraffin. Then 8um-thick tissue sections were produced for each sample. To observe the structural changes within the uterus and breast in response to treatment with icariin and DBT, HE staining was performed.

At minimum of 5 sections from each sample were observed using 60 magnification and photographed using a photoscope (Olympus BX51). Preparation of dyes and protocol of H&E staining was listed in Appendices.

3.2.7 Bone mineral density (BMD) and Micro-CT analysis

Micro-CT is applied in small animals to assess changes of trabecular and cortical bone morphology for preclinical studies. By utilizing differences in X-ray attenuation properties of bone, Micro-CT produces 3D images of very high resolution, with voxel sizes down to 1um or even smaller. It allows the quantitative analysis of properties of bone, like density and strength. Bone properties of trabecular bone at proximal tibia

57 and distal femur as well as lumbar vertebra were determined by Micro-CT (Viva 40,

Scanco Medical, Switerland) in the present study. The source energy selected for this study was 70 KVp and 114 μA with resolution of 21um. Approximately 200 slices were done for each scan. The distal/proximal were defined as 4.2mm and 2.2mm away from femur/tibia head (Wong et al., 2013). Scanning was done at the metaphyseal area located 0.63 mm below the lowest point of the epiphyseal growth plate and extending 2.0 mm in the proximal direction. Bone mineral density (BMD) and following bone morphometric properties (Table 3.4), bone volume over total volume (BV/TV), trabecular bone number (Tb.N), trabecular bone thickness (Tb.Th), trabecular bone separation (Tb.Sp) and bone surface over bone volume (BS/BV) were evaluated by contoured VOI images.

Table 3.3 Sequence of primers for estrogen-responsive genes (for rat used in in vivo study)

Genes Primer sequences Tm (℃)

GAPDH Forward: TGCCACTCAGAAGACTGTGG 59.1

Reverse: GCATGCAGGGATGATGTTCTA

PR Forward: CTGGTTCCGCCACTCATCAA 57.2

Reverse: TCAGGCTCATCCAGGAGTACTGA

ER Forward: AAGAGAAGGACCACATCCACC 57.8

Reverse: GGAATGTGCTGAAGTGGAGC

C3 Forward: GCTGTGCCTTATGTCATTGTCC 56.3

Reverse: ATTTCTCCCACTGTTCGGTCTG

TH Forward: ACACAGCGGAAGAGATTGCT 55.5

Reverse: CCCAGAGATGCAAGTCCAAT

DAT Forward: TTGGGTTTGGAGTGCTGATTGC 59.0

Reverse: AGAAGACGACGAAGCCAGAGG

CYP19A1 Forward: CTCCTCCTGATTCGGAATTGT 49.3

Reverse: TCTGCCATGGGAAATGAGAG

Table 3.4 Variables for assessment of trabecular bone microarchitecture using micro-CT

Variable Standard units Description

TV (total volume) mm3 Volume of the entire target region

BV (bone volume) mm3 Volume of the region segmented as bone

BS (bone surface) mm2 Surface of the region segmented as bone

BV/TV % Ration of the segmentaed bone volume

(bone volume fraction) to the total volume of the target region

Tb.N 1/mm3 Measure of the average number of

(trabecular number) trabecular per unit length

Tb.Th 1/mm Mean thickness of trabeculae, assessed

(trabecular thickness) using direct 3D methods

Tb.Sp mm Mean distance between trabeculae,

(trabecular separation) assessed using direct 3D methods;

3.2.8 Measurement of bone biochemical markers

Serum osteocalcin is a specific product of osteoblast and is a biomarker for bone formation (Nowacka-Cieciura et al., 2016) while urinary deoxypyridinoline (DPD) is a breakdown product of collagen during bone resorption and is a biomarker for bone resorption (Pang et al., 2010). Serum level of osteocalcin was measured with ELISA kit (Alfa Aesar, A Johnson Matthey Company, UK) by strictly following manufacturers’ instructions. Urinary DPD was determined with an enzyme immunoassay DPD EIA kit (QUIDEL Corporation, USA) by strictly following manufacturers’ instruction. Urinary level of DPD was normalized by creatinine.

Notes: Before the large scale of measurement, preliminary experiment should be performed to establish the dilution factor for samples.

3.2.9 Measurement of serum level of reproductive hormones

Blood was drawn from the abdominal aorta at sacrifice and serum was collected after centrifugation at 3500rpm for 15mins. Aliquots of serum were stored at -80℃for further measurement. To investigate the effects of drugs on hypothalamic pituitary gonad axis, the serum level of estradiol, follicle stimulating hormone (FSH) and luteinizing hormone were determined (LH).

Serum level of estradiol was measured by Estradiol EIA Kit according to manufacturer’s instruction (CayMan, Cat#582251). The assay is based on the

61 competition between estradiol and an estradiol-acetylcholinesterase (Matheny et al.,

2013) conjugate (Estradiol Tracer) for a limited amount of Estradiol Antiserum. The amount of estradiol tracer that is able to bind to estradiol antiserum will inversely proportional to the concentration of estradiol in the well.

The serum levels of FSH and LH were measured by enzyme-linked immunosorbent assay (ELISA) kits (Cloud Clone Corp, Cat#CEA830Ra for FSH and CEA441Ra for

LH) according to the manufacturer’s instructions.

For each hormone, due to the complicated hormonal status caused by ovariectomy, preliminary experiment should be performed to determine the dilution factors and detection limitation for each measurement before the large scale measurement.

3.2.10 Statistical analysis

Data was reported as mean ± SEM. Significance of differences between groups of in vivo experiment were determined by one-way ANOVA followed by Tukey’s test for post hoc comparison. Interactions between herb and drugs were performed by two-way ANOVA followed by Bonferroni test as a post test. A value of p＜0.05 was considered statistically significant. All graphs in this study were plotted by using

GraphPad Prism Version5.0.

3.3 In vitro study

3.3.1 Experimental design

To characterize the direct estrogenic effects of icariin and DBT four estrogen receptor-positive cell lines, human breast cancer MCF-7 cell, human endometrial cancer Ishikawa cell, human osteosarcoma MG-63 cell and human neuroblastoma

SH-SY5Y cell, which are consistent with the target tissues of estrogens investigated in the in vivo study, were employed in the in vitro experiment of the present study. Cells were subjected to treatment with vehicle, 17ß-estradiol, icariin or DBT of various concentrations. Effects on cell proliferation, mRNA expression of specific estrogen responsive genes and estrogen response element (ERE)-dependent luciferase activity were determined as listed in Table 3.5.

The potential interactions between herb medicine and western drugs were also investigated by co-treatment of cells with DBT and tamoxifen or raloxifene. Effects on cell proliferation or ALP activity in the four estrogen sensitive cell lines mentioned above were used to assess the interactive effects between DBT and SERMs.

3.3.2 Cell culture

MCF-7 cells (ATCC No. HTB-22 TM) were routinely cultured in high-glucose (4.5g/L)

Dulbecco’s Modified Eagle Medium (DMEM, Gibco) supplemented with antibiotics

P/S (100 U/ml penicillin, 100ug/ml Streptomycin, Gibco) and 5% Fetal Bovine Serum

(FBS, Gibco); for the Ishikawa cells (kindly provided by Dr. Li-Hui Wei of Peking

University People’s Hospital) and SH-SY5Y cells (kindly provided by Professor Chen

Wenfang of Qingdao University), the routine culture medium was high-glucose

(4.5g/L) Dulbecco’s Modified Eagle Medium (DMEM, Gibco) supplemented with

100 U/ml penicillin, 100ug/ml streptomycin (Gibco) and 10% FBS; for MG-63 cells

(ATCC No. CRL-1427TM), the routine culture medium was Minimum Essential

Medium (MEM, Gibco) supplemented with 100 U/ml penicillin, 100ug/ml streptomycin (Gibco) and 10% FBS as listed in Table 3.6, column 1&2. The cells were cultured on 100mm culture dishes and harvested with trypsin (0.05%

Trypsin-EDTA, Gibco, Cat#15400-54). Sub-cultured was performed when the cells reached approximately 80-90% confluence.

Special medium and FBS used for investigation of estrogenic effects in estrogen-sensitive cell lines

Since phenol red has been reported to exert estrogenic effects in estrogen-sensitive cell (Berthois et al., 1986), phenol red-free medium (PRF-DMEM, Gibco,

Cat#21063-029; PRF-MEM, Gibco, Cat#51200-038) was used for each cell line during the treatment as listed in Table 3.6, column 3. Activated charcoal was used to remove low-molecular-weight lipophilic compounds from fetal bovine serum including hormones, retinoids and fatty acid ligands of nuclear receptor transcription factors. In my study, the charcoal stripped FBS (cs-FBS, Gibco, Cat#12676-011) was applied in cell culture to get rid of the influences of these compounds.

Seeding of cell for drug treatment

Cells were seeded at different densities depending on the requirement of different assays (see appendices). After the confluence reached 80-90%, the plate was rinsed with PBS and the culture medium was changed into phenol red-free medium supplement with cs-FBS. After incubation for 24 hours, the cells were subjected to drug treatments for 48 hours.

3.3.3 Drug preparation and treatment

17ß-estradiol was dissolved in absolute ethanol while icariin, tamoxifen and raloxifene were all dissolved in DMSO. All of the stocks came in powder form. After dissolved in their solvent respectively and filter sterilized, subsequent dilution was performed using absolute ethanol. DBT was dissolved in distilled water and filter sterilized. Subsequent dilution of DBT solution was performed by using phenol red free medium for each cell line.

Concentrations of each drug used in present study were determined based on the reference of previous studies. 17ß-estradiol (E2, 10-8M or 10-7M) was used as positive control in this study.

3.3.4 MTS assay

Cells were seeded in 96-well plate and cultured in routine medium. 24 hours later, medium was changed into phenol-red free medium (shown in Table 3.6, column 3) for

65 another 24 hours. Then icariin in various concentrations (10-12-10-6M), DBT in various concentrations (0.05, 0.1, 0.25, 0.5, 1.0 and 2.0mg/ml) and 17ß-estradiol

(10-8M for MCF-7, Ishikawa and SH-SY5Y cell; 10-7M for MG-63 cell) were applied to cells for 48 hours. Upon treatment, medium was discarded and 100ul of MTS working solution (see appendices) was added into each well. After incubation at 37℃ for 0.5-4 hours, absorbance at 490nm was measured by iMarkTM Microplate

Absorbance Reader Model 680 (Bio-Rad) with MTS working solution as blank.

Results were expressed as ratio relative to control.

The concentrations of icariin that exerted the most potent effect in each cell line were applied in following experiments.

3.3.5 ALP Assay

Ishikawa and MG-63 cells were seeded in 24-well plate. 24 hours later, medium was changed into phenol-red free medium for another 24 hours and followed by treatment with vehicle, 17ß-estradiol, icariin and DBT at different concentrations for 48 hours.

Upon treatment, cells were rinsed by PBS for three times and lysed with 100ul of 1X passive lysis buffer (PLB, Promega, Cat#E194A) per well for 15mins at room temperature. Then cellular lysate was collected and centrifuged at 3000rpm for 5min.

20ul of the supernatant was transferred to 96-well plate to determine ALP activity in microplate reader by following manufacturer’s instructions (Wako, Cat#29158601).

Another 20ul of the cellular lysate was used to determine the protein concentration by

Protein Assay Dye Concentrate (BioRad, Cat#500-0006).

The ALP activity was normalized by the total protein content and expressed as

OD405nm/OD595nm. Results were expressed as ratio relative to control.

3.3.6 Real-time PCR

Cell treatment

Cells were seeded in 6-well plate and cultured in routine medium for 24 hours. Then cells were cultured in phenol red free medium with charcoal stripped FBS for another

24 hours followed by treatment with vehicle, estradiol, icariin and DBT in their optimal concentration determined by MTS assay or ALP assay for 48 hours.

Total RNA extraction and reverse transcription PCR

Upon treatment, medium was discarded and the plate was rinsed with PBS. Then 1ml of Trizol (Invitrogen, Cat#15596018) for each well was used to extract the total RNA from cells by following manufacturer’s instructions.

2ug of total RNA was used to generate cDNA in a 20ul of RT-PCR reaction system by using High-Capacity cDNA Reverse Transcription Kits (Applied Biosystems,

Cat#4368814) following the manufacturer’s instruction.

Quantitative Real time PCR

Specific estrogen-responsive genes of each cell line and the optimal reaction

67 conditions for each primer were listed in Table 3.7. Real time PCR was conducted as described in the in vivo study.

3.3.7 ERE-dependent luciferase activity assay

Transient transfection

To investigate whether icariin could induce estrogen response element-dependent transcription, these ER-positive cell lines were transfected with 0.4µg ERETkluc plasmid together with 0.01µg of an inactive control plasmid pRL-TK, a Renilla luciferase control vector, by LipofectamineTM 2000 reagent (Invitrogen,

Cat#11668-019) in PRF medium without antibiotics and FBS (Lau et al., 2009).

Cell treatment

Six hours after transfection, the medium for transfection was discarded. Cells were subjected to vehicle, estradiol, icariin and DBT in phenol red-free medium containing cs-FBS for 24 hours. Upon treatment, the medium was discarded and cells were lysed with 100ul of Passive Lysis Buffer (PLB) for 20mins at room temperature after rinsed with PBS. Then the lysate was collected and centrifuged. After centrifugation, 20ul of the cell lysate was collected for luciferase activity measurement.

Luciferase activity assay

Luciferase activity was measured by a Dual Luciferase® Reporter Assay System

(Promega, E1960) and the signal detected by a TD-20/20 Luminometer (Turner

Design, USA) following the manufacturer’s instructions. Results were expressed as ratio relative to control.

3.3.8 Blocking effects of signaling inhibitors on actions of DBT

To investigate the possible mechanisms that mediate the actions of DBT, the MG-63 cells were co-treated with DBT at its optimal concentrations and three signaling pathway blockers including ICI182,780 (ER antagonist, 10-6M, TOCRIS, Cat#1047),

U0126 (MAPK inhibitor, 10-6M, TOCRIS, Cat#1144) and LY294002 (PI3K inhibitor,

10-6M, TOCRIS, Cat#1130) for 48 hours. MTS or ALP activity assay was performed to evaluate the blocking effects on actions of DBT in MG-63 cells.

3.3.9 Interactive effects between DBT and SERMs

To investigate the potential interactive effects between DBT and tamoxifen, raloxifene, the four cell lines were co-treated with DBT and tamoxifen (10-12, 10-10, 10-8 and

10-6M) or raloxifene (10-12, 10-10, 10-8 and 10-6M) for 48 hours. MTS assay or ALP activity assay was performed to evaluate the combined effects of them.

3.3.10 Statistical analysis

Inter-group difference in in vitro study were determined by independent t-test A value of p＜0.05 was considered statistically significant. All graphs in this study were plotted by using GraphPad Prism Version5.0.

Table 3.5 Specific estrogen-responsive parameters for each cell line

Cell line Variables to be detected

MCF-7 cell MTS; pS2, IGF-IR, ER mRNA expression; ERE luciferase

activity assay

Ishikawa cell MTS; ALP activity; ALP and ER mRNA expression; ERE

luciferase activity assay

SH-SY5Y cell MTS; TH, DAT and ER mRNA expression; ERE luciferase

activity assay

MG-63 cell MTS; ALP activity; OCN, ALP, OPG, RANKL, ER mRNA

expression; ERE luciferase activity assay

Table 3.6 Culture conditions for each cell line

Cell line Routine medium Phenol red-free medium

MCF-7 5%FBS high glucose DMEM 1% cs-FBS PRF-DMEM

Ishikawa 10% FBS high glucose DMEM 1% cs-FBS PRF-DMEM

MG-63 10% FBS MEM 5% cs-FBS PRF-MEM

SH-SY5Y 10% FBS high glucose DMEM 1% cs-FBS PRF-DMEM

Table 3.7 Sequences of primers for the estrogen-responsive genes (for human used in in vitro study)

Genes Primer sequences Tm (℃)

GAPDH Forward: ACCACAGTCCATGCCTACAC 59.1

Reverse: TTCACCACCCTGTTGCTGTA

ER Forward: GGGAATGATGAAAGGTGGGAT 57.8

Reverse: GGCTGTTCTTCTTAGAGCGTT

ALP (Ishikawa Forward: CCTAAAAGGGCAGAAG 56.3

cell) Reverse: GCTGTAGTCTCTGGGTACTCA

pS2 Forward: ATGGCCACCATGGAGAACAAGG 55.2

Reverse: CATAAATTCACACTCCTCTTCTGG

IGF-IR Forward: GACAACCAGAACTTGCAGCA 59.3

Reverse: GATTCTTCGACGTGGTGGTG

ALP Forward: AGCCCTTCACTGCCATCCTGT 57.7

Reverse: ATTCTCTCGTTCACCGCCCAC

OCN Forward: CAAAGGTGCAGCCTTTGTGTC 59.4

Reverse: TCAAGTCCGGATTGAGCTCA

OPG Forward: GGCAACACAGCTCACAAGAA 49.8

Reverse: CGCTGTTTTCACAGAGGTCA

RANKL Forward: AGGAGCAAGCTTGAAGCTCAG 53.4

Reverse: TCTCCTGAAGTTTCATGATGTCG

Chapter 4

Characterization of the tissue selectivity of icariin in mature ovariectomized (OVX) rats

and estrogen-sensitive cell lines

4.1 Introduction

HRT has been used as an effective regimen for management of menopause related symptoms. However, recent studies indicate that HRT especially long-term HRT is associated with the increased risk of reproductive system cancer and stroke. Therefore, alternative approaches are urgently needed. Phytoestrogens are natural compounds derived from plants and herbs exerting estrogen-like properties. Its effectiveness in treatment of symptoms related to menopause has been determined and might be an alternative approach for HRT regarding the management of menopause-related symptoms.

Icariin, a C-8 prenylate flavonol glycoside, is a phytoestrogen extracted from a

Chinese herb-Herba Epimedii which is frequently prescribed in TCM formula for treatment of kidney disease (Chen et al., 2010). There are more than 260 chemical moieties detected in the genus Epimedium and among all these components, icariin is the most abundant and the main effective constituent (Gao et al., 1998). Icariin has been chosen as the chemical marker for quality control of Herba Epimedii in Chinese

Pharmacopeia (Li et al., 2015). According to the reported studies, icariin has extensive pharmacological activities including anti-osteoporosis (Zhang et al., 2007), neuroprotection (Urano et al., 2010), immunoprotection (Teng et al., 2008), anti-oxidation, anti-apoptosis, anti-cancer (Fan et al., 2016). Our previous work

73 demonstrated that icariin restored bone loss in OVX mice in a similar way as estradiol but without inducing uterotrophic effect (Mok et al., 2010). At cellular level, icariin promoted osteogenic proliferation, ALP activity and mRNA expression in

ER-independent manner in UMR106 cells. In particular, icariin induced phosphorylation of ER at Ser 118 (Urano et al., 2010). These results suggest that icariin may be phytoestrogen and exert estrogenic effects in vivo and in vitro.

The present study aimed to evaluate the tissue-selective effects of icariin at different dosages by using established preclinical in vivo and in vitro models. Four estrogen-responsive tissues including breast, uterus, brain and bone, which are most frequently affected during menopause, were studied in present study. Icariin was added into diet at different concentrations (50ppm, 500ppm and 3000ppm) for rats to take. In the in vivo experiment, the mature female ovariectomized SD rats, the most frequently used model for studying postmenopausal symptoms, were employed. Upon treatment, estrogenic effects of icariin in the four tissues mentioned above were evaluated by specific estrogen-responsive parameters for each tissue.

To be consistent with the in vivo study, four estrogen sensitive cell lines including human breast cancer MCF-7 cell, human endometrial cancer Ishikawa cell, human osteosarcoma MG-63 cell and human neuroblastoma SH-SY5Y cell were employed in the in vitro study. The responses of cell viability or ALP activity, mRNA expression of

74 estrogen-sensitive gene and ERE luciferase activity in the four cell lines to icariin in different concentrations were determined to evaluate the direct estrogenic effects of icariin in vitro.

4.2 Results

4.2.1 Characterization of estrogenic effects of icariin in OVX rats

4.2.1.1 Effects of icariin on body weight gain of OVX rats

Estrogen influences body weight homeostasis via regulation of the food intake and adipose tissue distribution (Brown et al., 2010). Estrogen deficiency has been reported to be associated to increased probability of obesity in post-menopausal women (Carr,

2003). To investigate the effects of icariin on body weight, the weight of rats was measured every two weeks during the whole treatment. As shown in figure 4.1, rats experienced dramatic increase in body weight after OVX due to possible uncontrolled food intake (p<0.01 vs sham rats). Exposure to 17ß-estradiol significantly suppressed the increase in body weight (p<0.001 vs OVX rat). This is consistent with our previous results in mice (Wong et al., 2013). Similar to the effect of estrogen, treatments with icariin at all dosages tested in the present study significantly suppressed the body weight gain of OVX rats (Figure 4.1, p<0.001 vs OVX rats), indicating icariin exerted estrogen-like effect on body weight gain of OVX rats.

However, the potency of icariin was much lower than estradiol.

30 **

20 ^^^ ^^^ ^^^ 10

^^^ 0 sham OVX E2 50 500 3000 Body weight gain (% gain weight the Body initial) of Icariin (ppm)

Figure 4.1 Effects of icariin on body weight gain of OVX rats

After recovery from the OVX operation, the OVX rats were orally administrated with vehicle, 17ß-estradiol (1mg/kg.day) and icariin (50ppm, 500ppm and 3000ppm) for 12 consecutive weeks. Body weight of rats was measured every two weeks during the whole experiment. Percentages of the changes in body weight over their baseline body weight were regarded as the body weight gain of rats. Sham: sham operated + vehicle treatment; OVX: OVX operated + vehicle treatment; E2: OVX operated + 17ß-estradiol treatment; 50: OVX operated + icariin 50ppm; 500: OVX operated + icariin 500ppm; 3000: OVX operated + icariin 3000ppm. Data was expressed as mean ± SEM. **p<0.01 vs sham; ^^^p<0.001 vs OVX. n=8 or 9.

4.2.1.2 Effects of icariin on uterus of OVX rats

Uterus is one target tissue of estrogen where it serves as sex hormone. Uterus is also the tissue that is frequently affected in menopause and the tissue where side effects of

HRT usually are observed. To investigate the estrogenic effects and the potential side effects of icariin on uterus, several estrogen-responsive parameters in uterus were determined. Weight of uterus significantly decreased in rats in response to OVX

(Figure 4.2A, p<0.001 vs sham rats) while treatment of OVX rats with 17ß-estradiol markedly restored the decline in uterus weight by about five fold (Figure 4.2A, p<0.001 vs OVX). Complement component 3 (C3) is a protein of the immune system and is regulated by estrogen in tissue-specific manner (Sundstrom et al., 1989). In addition, the mRNA expression of three estrogen-responsive genes, complement component 3 (C3), progesterone receptor (PR) and estrogen receptor (ER) in rat uterus in response to treatment of 17ß-estradiol and icariin were determined by real-time PCR. As shown in Figure 4.2B-4.4D, mRNA expression of C3, ER and PR in uterus markedly decreased in response to OVX (p<0.01, p<0.001 vs sham) and treatment with estradiol significantly attenuated the decline in mRNA expression of

C3 and PR (p<0.05, p<0.001 vs OVX) without significant effect on ER mRNA expression level. However, treatment with icariin at all dosages did not show any effect on either uterus weight or mRNA expression level of estrogen-responsive genes

78 in uterus of OVX rats, indicating that icariin acted differently from estrogen in uterus of OVX rats.

A B

20 2.0

15 ^^ 1.5 ^^^

1.0 10 C3/GAPDH 0.5 5 *** *** 0.0 0

sham OVX E2 50 500 3000 sham OVX E2 50 500 3000 Uterus index (mg/g body weight) body (mg/g Uterus index Icariin (ppm) Icariin (ppm)

C D

20 10 ^ 8 15

6 10

4 PR/GAPDH ER/GAPDH ** 5 2

0 0 sham OVX E 50 500 3000 sham OVX E2 50 500 3000 2 Icariin (ppm) Icariin (ppm)

Figure 4.2 Effects of icariin on uterus index and mRNA expression of estrogen-responsive genes in uterus of OVX rats Uterus was collected from sham and OVX rats treated with vehicle, 17ß-estradiol and icariin at 50, 500 and 3000ppm for 12 weeks. A. Uterus index: Uterine weight were divided by body weight of rats; B. mRNA expression of component complement 3(C3); C. mRNA expression of progesterone receptor (PR) and D. mRNA expression of estrogen receptor (ER) were determined by Real-time PCR as described in Methods (Chapter 3). Sham: sham operated + vehicle treatment; OVX: OVX operated

+ vehicle treatment; E2: OVX operated + 17ß-estradiol treatment; 50: OVX operated + icariin 50ppm; 500: OVX operated + icariin 500ppm; 3000: OVX operated + icariin 3000ppm. Data was expressed as mean ± SEM. **p<0.01, ***p<0.001 vs sham; ^p<0.05, ^^p<0.01, ^^^p<0.001 vs OVX. n=8 or 9.

4.2.1.3 Effects of icariin on breast tissue of OVX rats

Breast is another estrogen target tissue. Estrogen plays an important role in breast development during puberty and breast maturation during pregnancy. However, due to the high sensitivity of breast tissue to estrogen, undesirable adverse effects are usually observed in breast tissues in postmenopausal women prescribed with HRT.

To investigate the estrogenic effects and potential side effects of icariin on breast tissue, the morphology of breast was visualized by H&E staining. As shown in figure

4.3, the number of the mammary gland (indicated by the red arrow) dramatically decreased and mammary gland became atrophic in OVX rats. As expected, treatment with 17ß-estradiol attenuated these changes in mammary gland in OVX rats.

Treatment with icariin at all dosages did not cause any changes in breast morphology compared with OVX rats, indicating no estrogenic effects of icariin on breast in OVX rats.

Figure 4.3 Effects of icariin on morphology of mammary gland in OVX rats

Upon treatment, the second breast of Sham rats and OVX rats treated with vehicle, 17ß-estradiol or icariin were collected and immediately fixed in 4% paraformaldehyde. H&E staining was performed to visualize the morphology of mammary gland of rats. Sham: sham operated + vehicle treatment; OVX: OVX operated + vehicle treatment;

E2: OVX operated + 17ß-estradiol treatment; 50: OVX operated + icariin 50ppm; 500: OVX operated + icariin 500ppm; 3000: OVX operated + icariin 3000ppm. n=8 or 9.

4.2.1.4 Effects of icariin on brain tissue of OVX rats

Epidemiological studies have reported that supplement with exogenous estrogen significantly reduces the risk of Parkinson’s disease (PD) in postmenopausal women

(Currie et al., 2004), suggesting a potential regulation of estrogen in PD. Striatum is a main part of substantia nigra-striatum dopaminergic system that regulates the reward and movement. In present study, striatum was collected to investigate the estrogenic effects of icariin on central nervous system in OVX rats. Tyrosine hydroxylase (TH) is the rate-limiting enzyme in the biosynthesis of dopamine (Kaufman, 1995) and dopamine transporter (DAT) is the key factor for the clearance of dopamine from synapse. Estrogens regulate the expression level of these two proteins (Al-Sweidi et al., 2011; Chaube et al., 2011).

As shown in figure 4.4A, compared with sham group, mRNA expression of tyrosine hydroxylase (TH) significantly declined in OVX rats (p<0.01 vs Sham), which was significantly reversed by 17ß-estradiol (p<0.001 vs OVX). The mRNA expression of dopamine transporter (DAT) dramatically increased in response to OVX to compensate for the decreased dopamine level in response to OVX in rats (Figure 4.4B, p<0.001 vs Sham). Exposure to 17ß-estradiol significantly decreased the DAT mRNA expression in striatum of OVX rats (Figure 4.4B, p<0.001 vs OVX), indicating a protective effect on dopaminergic cells. Similarly to estradiol, icariin at 500 and

3000ppm also significantly increased TH mRNA expression and decreased DAT mRNA expression in striatum of OVX rats (Figure 4.4A and B, p<0.05, p<0.001 vs

OVX). Moreover, the inhibitory effects of icariin at 500 and 3000ppm on DAT mRNA expression were comparable to that of estradiol (Figure 4.4B). However, icariin at

50ppm showed the trend to restore the OVX-induced changes in mRNA expression of

TH and DAT in striatum while the effects did not reach statistical significance. These results suggested icariin exerted estrogen-like neuroprotective in striatum of OVX rats.

A B

2.5 0.8 ^^^ *** ^ 2.0 0.6 ^ 1.5 ^^^ 0.4 ^^ ^^^

** 1.0 TH/GAPDH 0.2 DAT/GAPDH 0.5

0.0 0.0 sham OVX E2 50 500 3000 sham OVX E2 50 500 3000 Icariin (ppm) Icariin (ppm)

Figure 4.4 Effects of icariin on mRNA expression of estrogen-responsive genes in striatum of OVX rats Upon treatment, striatum was freshly collected. Total RNA was extracted and mRNA expression of TH and DAT were measured by real time-PCR. A. mRNA expression of TH; B. mRNA expression of DAT. Sham: sham operated + vehicle treatment; OVX:

OVX operated + vehicle treatment; E2: OVX operated + 17ß-estradiol treatment; 50: OVX operated + icariin 50ppm; 500: OVX operated + icariin 500ppm; 3000: OVX operated + icariin 3000ppm. Data was expressed as mean ± SEM. **p<0.01, ***p<0.001 vs sham; ^p<0.05, ^^p<0.01, ^^^p<0.001 vs OVX. n=8 or 9.

4.2.1.5 Effects of icariin on bone of OVX rats

BMD and bone properties

Estrogens play important role in keeping the balance of bone remodeling between osteoclast and osteoblast in women (Wang et al., 2012). In the present study, bone was chosen as another major target tissue to investigate the estrogenic effects of icariin.

Bone mineral density (BMD) and trabecular bone microarchitecture and bone properties measured by micro-CT were used to evaluate effects of icariin on bone.

In response to OVX, rats experienced serious osteoporosis as the BMD, BS, BV/TV,

Tb.N, Tb.Th significantly decreased while Tb.Sp significantly increased (Figure 4.5A,

B and C, Table 4.1, 4.2 and 4.3, p<0.05, p<0.01, p<0.001 vs Sham). And these deteriorations of bone could be directly observed from the microarchitecture of bone at distal femur, proximal tibia and lumbar spine (Figure 4.5D, E and F). These changes in bone were significantly reversed by 17ß-estradiol (Figure 4.5, table 4.1,

4.2 and 4.3, p<0.05, p<0.01, p<0.001 vs OVX). Similarly, treatment with icariin at 50,

500 and 3000ppm also significantly restored the changes in BMD, bone microarchitecture and bone properties induced by OVX as icariin significantly increased BMD, BS, BV/TV, Tb.N, Tb.Th and significantly decreased Tb.Sp (Figure

4.5, Table 4.1, 4.2 and 4.3, p<0.05, p<0.01, p<0.001 vs OVX). In terms of increasing

BMD , icariin at 3000ppm exerted the most potent effects on long bone (Figure 4.5A and B, p<0.001 vs OVX) while the protective effects of icariin at 3000ppm on lumbar

86 spine appeared to be weaker than these of icariin at 50 and 500ppm (Figure 4.5 C p<0.05, p<0.01, p<0.001 vs OVX). These results suggested that icariin exerted estrogen-like bone protective effects as it increased BMD and improved bone microarchitecture and bone properties at distal femur, proximal tibia and lumbar spine in OVX rats. Moreover, such protective effect on bone might be related to dosage of icariin.

A B

800 800

600 600

*** *** 400 *** 400 *** *** *** *** **

200 *** 200 BMD (mg HA/ccm) (mg BMD BMD (mg HA/ccm) (mg BMD ***

0 0 Sham OVX E2 50 500 3000 Sham OVX E2 50 500ppm 3000 Icariin (ppm) Icariin (ppm)

800

600 ^ ^ 400 *

200 BMD (mg HA/ccm) (mg BMD

0 Sham OVX E2 50 500 3000 Icariin (ppm)

Figure 4.5 Effects of icariin on bone mineral density and micro-architecture at distal femur, proximal tibia and lumbar spine in OVX rats Upon treatment, the whole left leg and spine were collected for micro-CT scanning. Bone mineral density and microarchitecture of proximal tibia and distal femur as well as lumbar vertebra were determined by Micro-CT. A. BMD of distal femur; B. BMD of proximal tibia; C. BMD of spine; D. Microarchitecture of distal femur; E. Microarchitecture of proximal tibia; F. Microarchitecture of lumbar spine. Sham:

89 sham operated + vehicle treatment; OVX: OVX operated + vehicle treatment; E2: OVX operated + 17ß-estradiol treatment; 50: OVX operated + icariin 50ppm; 500: OVX operated + icariin 500ppm; 3000: OVX operated + icariin 3000ppm. Data was expressed as mean ± SEM. *p<0.05, **p<0.01, ***p<0.001 vs sham; ^p<0.05, ^^p<0.01, ^^^p<0.001 vs OVX. n=8 or 9.

Table 4.1 Effects of icariin on trabecular bone properties at distal femur of OVX rats BS (mm2) BV/TV Tb.N (1/mm) Tb.TH (mm) Tb.Sp (mm)

Sham 239.3±4.49 0.484±0.017 4.73±0.07 0.149±0.017 0.109±0.005

OVX 86.7±8.85*** 0.121±0.014*** 0.66±0.19*** 0.091±0.015 1.565±0.081***

E2 182.0±7.73^^^ 0.305±0.018^^^ 3.38±0.15^^^ 0.140±0.011 0.214±0.015^^^ icariin-50 139.2±4.76^^ 0.226±0.016^^ 1.15±0.05 0.166±0.008^^ 1.033±0.031^^^ icariin-500 158.4±6.67^^^ 0.236±0.013^^^ 1.16±0.06^ 0.161±0.009^^ 0.657±0.037^^^ icariin-3000 164.9±6.89^^^ 0.246±0.026^^^ 1.17±0.07^ 0.157±0.009^^ 0.909±0.036^^^

Table 4.2 Effects of icariin on trabecular bone properties at proximal tibia of OVX rats BS (mm2) BV/TV Tb.N(1/mm) Tb.TH (mm) Tb.Sp (mm)

Sham 219.6±8.68 0.517±0.023 4.97±0.05 0.189±0.010 0.097±0.005

OVX 54.2±6.97*** 0.076±0.010*** 1.23±0.16*** 0.111±0.015*** 0.856±0.144***

E2 155.2±9.86^^^ 0.255±0.023^^^ 3.38±0.19^^^ 0.213±0.024^^^ 0.214±0.014^^ icariin-50 101.3±5.63 0.113±0.008 0.833±0.05^^ 0.138±0.009^^ 0.212±0.127^^^ icariin-500 128.0±8.67^^^ 0.134±0.014 0.812±0.06^^ 0.134±0.005^^^ 0.236±0.074^^^ icariin-3000 135.6±10.67^^^ 0.168±0.024^ 0.921±0.07^^^ 0.133±0.007^^^ 0.307±0.088^^

Table 4.3 Effects of icariin on trabecular bone properties at lumbar spine of OVX rats BS (mm2) BV/TV Tb.N (1/mm) Tb.TH (mm) Tb.Sp (mm)

Sham 89.9±2.73 0.442±0.012 4.12±0.09 0.160±0.003 0.136±0.006

OVX 62.6±5.18** 0.093±0.016*** 0.62±0.17*** 0.114±0.009 1.313±0.026***

E2 90.4±3.33^^ 0.351±0.010^^^ 3.71±0.06^^^ 0.235±0.039 0.176±0.006^^^ icariin-50 85.2±3.40^ 0.138±0.011 0.95±0.04^ 0.160±0.015^^ 0.993±0.036^^^ icariin-500 81.8±4.55 0.200±0.039^^ 1.07±0.09^^ 0.170±0.011^^ 0.888±0.063^^^ icariin-3000 81.3±4.69 0.151±0 1.15±0.12^^^ 0.162±0.012^^ 0.929±0.065^^^

Trabecular bone properties of proximal tibia and distal femur as well as lumbar vertebra were also determined by Micro-CT. Sham: sham operated + vehicle treatment; OVX: OVX operated + vehicle treatment; E2: OVX operated + 17ß-estradiol treatment; 50: OVX operated + icariin 50ppm; 500: OVX operated + icariin 500ppm; 3000: OVX operated + icariin 3000ppm. Data was expressed as mean ± SEM. *p<0.05, **p<0.01, ***p<0.001 vs sham; ^p<0.05, ^^p<0.01, ^^^p<0.001 vs OVX. n=8 or 9. BS: Bone surface; BV/TV: Bone volume/total volume; Tb.N: Trabecular bone number; Tb.Th: Trabecular bone thickness; Tb.Sp: Trabecular separation.

Biochemical markers for bone turnover

Serum osteocalcin is a specific product of osteoblast and is regarded as a biomarker for bone formation (Nowacka-Cieciura et al., 2016) while urinary deoxypyridinoline

(DPD) is a breakdown product of collagen during bone resorption and is a biomarker for bone resorption (Pang et al., 2010). These two biomarkers are usually used to evaluate the effect on bone turnover rate. As shown in Figure 4.6A and B, significant increase was found in both serum osteocalcin and urinary deoxypyridinoline in OVX rats (p<0.001 vs Sham), indicating an increase in bone turnover rate, which were significantly suppressed by treatment with 17ß-estradiol (p<0.001 vs OVX). Icariin at

50 and 500ppm exerted estrogen-like but weaker suppressive effects on the two biomarkers as they significantly decreased serum osteocalcin and urinary DPD in

OVX rats (Figure 4.6 A and B, P<0.05, p<0.01, p<0.001 vs OVX). Icariin at 3000ppm significantly reduced the serum osteocalcin level (Figure 4.6A, p<0.05 vs OVX) but not urinary DPD in OVX rats. These results showed that icariin could suppress the increase in serum osteocalcin and urinary deoxypyridinoline induced by estrogen deficiency in OVX rats, indicating a suppressive effect on bone turnover.

Our results suggested that icariin protected bone from estrogen deficiency-induced osteoporosis in estrogen-like manner possible via suppressing bone turnover.

Moreover, such protective effects of icariin on bone appeared to be related to dosage.

A B

25 200 20 *** ^^ ^^^ *** ^^ 150 15 ^ ^ 100 10 ^^^

5 50 ^^^ osteocalcin (ng/ml) osteocalcin

0 0 sham OVX E 50 500 3000 (nmol/mmol) DPD/Creatinine 2 sham OVX E2 50 500 3000 Icariin (ppm) Icariin (ppm)

Figure 4.6 Effects of icariin on the bone turnover biomarkers of OVX rats

Serum level of osteocalcin and urinary deoxypyridinoline (DPD) were measured by using commercial kits. A. Serum level of osteocalcin; B. Urinary deoxypyridinoline. Data was expressed as mean ± SEM. *p<0.05, **p<0.01, ***p<0.001 vs sham; ^p<0.05, ^^p<0.01, ^^^p<0.001 vs OVX. n=8 or 9.

4.2.2 Characterization of estrogenic effects of icariin in estrogen sensitive cell lines

4.2.2.1 Effects of icariin in MCF-7 cells

MCF-7 cell is an estrogen receptor-positive cell line which responses sensitively to estrogen. To evaluate the direct estrogenic effects of icariin in MCF-7 cell, cell proliferation, ERE-dependent luciferase activity and expression level of several estrogen-responsive genes were determined. Our results confirmed that estradiol could significantly stimulate the cell proliferation in MCF-7 cells (Figure 4.7A, p<0.001 vs control). Similarly, icariin also significantly increased cell proliferation in MCF-7 cells in dose-dependent manner and icariin at 10-6M exerted the most potent effect on stimulating cell proliferation of MCF-7 cells (Figure 4.7A, p<0.001 vs control), which was as apparent as that of estradiol. Moreover, the stimulatory effects of icariin on cell proliferation of MCF-7 cells could be completely blocked by co-treatment with ER antagonist ICI182,780 (Figure 4.7B, p<0.001 vs icariin treatment), indicating the stimulatory effects of icariin on cell proliferation in MCF-7 cell was dependent on ER. In contrast, treatment with icariin (10-8-10-6M) for 48 hours did not alter the ERE luciferase activity (Figure 4.7C), indicating actions of icariin might be independent to ERE in MCF7-cell. pS2 is estrogen-regulated protein and reflects the function of ER (Won Jeong et al.,

2012). In the present study, pS2 mRNA expression was significantly up-regulated by

95 more than 50% upon treatment with both estradiol and icariin for 48 hours (Figure

4.7E, p<0.001 vs control), indicating an estrogenic effects of icariin in MCF-7 cells.

However, no significant changes were found in mRNA expression of ER upon treatment with either estradiol or icariin (Figure 4.7D). Insulin-like growth factor-1 receptor is an important receptor related to cell-cycle progression that is regulated by estrogen (Huang et al., 2015). In present study, estradiol rather than icariin significantly up-regulated the mRNA expression of IGF-IR in MCF-7 cells (Figure

4.7F, p<0.05 vs control). These results suggested that icariin directly exerted estrogenic effects in MCF-7 cell possibly via selective regulation of estrogen-sensitive target genes. Moreover, our results demonstrated that the estrogenic effects of icariin in MCF-7 cells might be ERE-independent.

1.5 *** *** *** *** ** *** *** ** ** 1.0

0.5

0.0

Cell viablity (Relative to control) (Relative viablity Cell -8 controlE2 (10 M)-12 -11 -10 -9 -8 -7 -6 -5 Icariin log[M]

no antagonist ICI

^^^ ^^^ ^^^ 1.5 ^^^ ^^^ *** *** *** ** *** 1.0 *** 0.5

0.0 -8 Cell viablity (Relative to control) (Relative viablity Cell control E2 (10 M) -12 -10 -8 -6 Icariin log [M]

C D

1.5

2.0 ** 1.5 1.0

1.0 0.5 0.5

0.0 0.0 -8 -8 control E2(10 M） -8 -7 -6 ER/GAPDH (relative to control) (relative ER/GAPDH control E2 (10 M) -8 -7 -6 Icariin log[M] Icariin log[M] ERE luciferase activity (relatibe to (relatibe control) activity ERE luciferase

E F

2.5 2.0 2.0 *** * 1.5 *** *** *** 1.5 1.0 1.0

0.5 0.5

0.0 0.0 control E (10-8M) -8 -7 -6 -8 pS2/GAPDH (relative to control) (relative pS2/GAPDH 2 control E2 (10 M) -8 -7 -6 Icariin log[M] to control) (relative IGF-IR/GAPDH Icariin log[M]

Figure 4.7 Effects of icariin on cell proliferation, mRNA expression of estrogen-responsive genes and ERE-dependent luciferase activity in MCF-7 cells MCF-7 cells were cultured and subjected to vehicle, estradiol (10-8M) and icariin (10-12-10-6M) in phenol red-free DMEM containing 1% of cs-FBS for 48 hours as described in Methods (Chapter 3). After treatment, cell proliferation, gene expression and ERE-luciferase activity were determined by MTS assay, Real time-PCR and Dual Luciferase Reporter Assay System, respectively. A. Cell proliferation; B. Blocking effects of ICI182,780; C. ERE-dependent luciferase activity; D. mRNA expression of ER; E. mRNA expression of pS2; F. mRNA expression of IGF-IR; Results were from two independent experiments and expressed as mean ± SEM (n=4). *p<0.05, **p<0.01, ***p<0.001 vs control.

4.2.2.2 Effects of icariin in Ishikawa cells

Ishikawa cell derived from a specimen of a well differentiated endometrial adenocarcinoma (Nishida et al., 1985) has been shown to contain ER and to respond to estrogens (Holinka et al., 1989; Holinka et al., 1986b). Alkaline phosphatase (ALP) enzyme activity in Ishikawa cell is specifically stimulated by estrogens and anti-estrogens completely block the induction of ALP activity by estradiol (Holinka et al., 1986a; Littlefield et al., 1990). Thus, in present study, the ALP activity was detected for the evaluation of the direct estrogenic effects of icariin in Ishikawa cell.

As shown in Figure 4.8A and B, icariin exerted potent stimulatory effects on promoting the ALP activity and the mRNA expression of ALP in Ishikawa cells, indicating the direct estrogenic effects of icariin in Ishikawa cells (Figure 4.8A and B, p<0.05, p<0.01, p<0.001 vs control). The stimulatory effects of icariin on ALP activity were demonstrated to be dependent on its concentrations as the potency of both lower and higher concentrations were not as apparent as the intermediate concentrations and icariin at 10-8M appeared to exert the most potent stimulatory effect on ALP activity. Icariin significantly up-regulated the mRNA expression of ER

(Figure 4.8C, P<0.01, p<0.001 vs control) in Ishikawa cells while showed no significant effect on ERE-dependent luciferase activity (Figure 4.8 D), suggesting that the estrogenic effects of icariin were ERE-independent.

1.5 *** * * 1.0

0.5

0.0 -8

ALP activity (relative to control) (relative activity ALP controlE2 (10 M)-12 -11 -10 -9 -8 -7 -6 -5 Icariin log[M]

B C

10 2.0 *** ** 8 *** 1.5 **

6 1.0 4 0.5 2 * **

0 0.0 -8 -8

control E (10 M) -9 -8 -7 to control) (relative ER/GAPDH control E (10 M) -9 -8 -7

ALP/GAPDH (relative to control) (relative ALP/GAPDH 2 2 Icariin log[M] Icariin log[M]

2.0

** 1.5

1.0

0.5

0.0 -8 control E2 (10 M) -9 -8 -7 Icariin log[M]

ERE luciferase activity (relatibe to (relatibe control) activity ERE luciferase

Figure 4.8 Estrogenic effects of icariin on cell proliferation, ALP activity, ERE-dependent luciferase activity and expression of estrogen-responsive genes in Ishikawa cell Ishikawa cells were cultured and were subjected to vehicle, estradiol (10-8M) and

100 icariin (10-12-10-6M) in phenol red-free DMEM containing 1% of cs-FBS for 48 hours as described in Methods (Chapter 3). Upon treatment, ALP activity, gene expression and ERE-dependent luciferase activity were measured by ALP kit, Real time-PCR and Dual Luciferase Reporter Assay System, respectively. A. ALP activity; B. mRNA expression of ALP; C. mRNA expression of ER; D. ERE-dependent luciferase activity. Results were from two independent experiments and expressed as mean ± SEM (n=4). *p<0.05, **p<0.01, ***p<0.001 vs control.

101

4.2.2.3 Effects of icariin in SH-SY5Y cell

The SH-SY5Y cell line is a thrice cloned sub-line of SK-N-SH cells which were originally derived from the bone marrow biopsy of a neuroblastoma patient in 1970’s

(Biedler et al., 1973). Since this cell line was demonstrated to possess many biochemical and functional properties of neurons, SH-SY5Y cell has been widely employed as model for studying on the central nervous system (Joshi et al., 2006). In the present study, SH-SY5Y cell line was used to investigate the direct estrogenic effects of icariin in the central nervous system.

In response to exposure to 17ß-estradiol, cell proliferation of SH-SY5H cell significantly increased (Figure 4.9A, p<0.001 vs control). Similarly, icariin also exerted positive effect on cell proliferation of SH-SY5Y cell at the intermediate concentrations tested in the present study (10-10-10-7M) (Figure 4.9A, p<0.05, p<0.01, p<0.001 vs control). Icariin at 10-10M and 10-8M induced 1.23-fold and 1.28 fold increase in cell proliferation of SH-SY5Y cells, respectively. However, icariin at lower concentrations (10-12-10-11M) did not exert significant stimulatory effect on cell proliferation of SH-SY5Y cell and icariin at the higher concentration, i.e.10-5M, even decreased the cell proliferation in SH-SY5Y cells. In addition, icariin at 10-10 to 10-8M significantly up-regulated the mRNA expression of ER (Figure 4.9B, p<0.01) in

SH-SY5Y cells and such stimulatory effect of icariin on mRNA expression was

102 comparable to that of estradiol in SH-SY5Y cell. Moreover, icariin at 10-9M even stimulated the ER mRNA expression to much higher level than estradiol. However, it did not induce ERE-dependent luciferase activity in SH-SY5Y cells (Figure 4.9C).

These results suggested that icariin exerted direct estrogenic effects in SH-SH5Y cell and its estrogenic effects were ERE-independent.

For the other two estrogen-regulate genes in the neurons, TH and DAT, we intended to study their mRNA expression level upon treatment with icariin for 48 hours. However, due to the low basal expression level of these two genes in undifferentiated SH-SY5Y cells without retinoic acid (RA) and the phorbol ester12-O-tetradecanoyl-phorbol-13-acetate (TPA) induction (Ikeda et al., 1994;

Presgraves et al., 2004), their responses to the effects of icariin could not be measured.

103

1.5 *** ** *** * * * 1.0 ***

0.5

0.0 control E (10-8M) -12 -11 -10 -9 -8 -7 -6 -5 Cell viablity (Relative to control) (Relative viablity Cell 2 Icariin log [M]

B C

2.5 ** 2.0 *** 2.0 ** ** ** 1.5 1.5 1.0 1.0

0.5 0.5

0.0 0.0 -8 -8 control E2 (10 M) -10 -9 -8

ER/GAPDH (relative to control) (relative ER/GAPDH control E (10 M) -10 -9 -8 2 Icariin log[M] Icariin log[M] ERE luciferase activity (relatibe to (relatibe control) activity ERE luciferase

Figure 4.9 Effects of icariin on cell proliferation, ERE-luciferase activity and mRNA expression of estrogen-responsive genes in SH-SY5Y cells SH-SY5Y cells were cultured and subjected to vehicle, estradiol (10-8M) and icariin (10-12-10-6M) in phenol red-free DMEM containing 1% of cs-FBS for 48 hours as described in Methods (Chapter 3). Upon treatment, cell proliferation, gene expression and ERE luciferase activity were measured by MTS assay, real time-PCR and dual luciferase reporter assay system, respectively. A. Cell proliferation; B. mRNA expression of ERE; C. ERE-dependent luciferase activity. Results were from two independent experiments and expressed as mean ± SEM (n=4). *p<0.05, **p<0.01, ***p<0.001 vs control.

104

4.2.2.4 Effects of icariin in MG-63 cells

Cell proliferation and ALP activity of MG-63 cells

Human osteosarcoma MG-63 cell line was employed in present study to evaluate the direct estrogenic effects of icariin in bone. Estrogens modulate the cell proliferation and differentiation in osteoblast (Nasu et al., 2000). According to our results, in response to estradiol, cell proliferation and ALP activity of MG-63 cells significantly increased by more than 20% (Figure 4.10A and B, p<0.01, p<0.001 vs control).

Icariin was also potent in promoting both cell proliferation and differentiation in

MG-63 cell. Treatment of MG-63 cells with icariin, especially at intermediate concentrations, i.e. 10-10-10-8M, for 48 hours significantly promoted cell proliferation by about 1.15 fold and increased ALP activity by around 1.3 fold (Figure 4.10A and B, p<0.001, p<0.05 vs control).

105

1.5

*** *** *** *** *** *** 1.0

0.5

0.0 -7 Cell viablity (Relative to control) (Relative viablity Cell controlE2(10 M)-12 -11 -10 -9 -8 -7 -6 -5 Icariin log [M]

2.0

** * 1.5 ** **

1.0

0.5

0.0 -7 ALP activity (relative to control) (relative activity ALP controlE2(10 M)-12 -11 -10 -9 -8 -7 -6 -5 Icariin log [M]

Figure 4.10 Effects of icariin on cell proliferation and ALP activity in MG-63 cells MG-63 cells were cultured and subjected to treatment with vehicle, estradiol (10-7M) and icariin (10-12-10-6M) in phenol red-free MEM containing 5% of cs-FBS for 48 hours as described in Methods (Chapter 3). Upon treatment, cell proliferation and ALP activity, were measured by MTS assay, ALP kit. A. Cell proliferation; B. ALP activity; Results were from two independent experiments and expressed as mean ± SEM (n=4). *p<0.05, **p<0.01, ***p<0.001 vs control.

106 mRNA expression of estrogen responsive genes in MG-63 cells

Moreover, estrogens were reported to up-regulate the transcription of osteoblast-specific genes, such as alkaline phosphatase (ALP) (Holzer et al., 2002), osteocalcin (OCN) (Hauschka et al., 1989) and osteoprotegerin (OPG) (Lacey et al.,

1998) while down-regulate the transcription of RANKL mRNA that plays an essential role in controlling the process of bone resorption (Fohr et al., 2000).

In the present study, the mRNA expression of OPG, OCN, ALP were significantly up-regulated in MG-63 cells upon treatment with estradiol and icariin (Figure 4.11A,

B and C, p<0.05, p<0.01, p<0.001 vs control) while the mRNA expression of RANKL was significantly inhibited by E2 and icariin to 70% (Figure 4.11D, p<0.01, p<0.05 vs control). Moreover, the inhibitory effect of icariin on RANKL mRNA expression was more potent than estradiol. A 1.7-fold increase was found in OPG/RANKL (Figure

4.11E, p<0.05 vs control) ratio in MG-63 cells upon treatment with icariin. However, icariin at the optimal concentrations did not alter the ERE-dependent luciferase activity in MG-63 cells (Figure 4.11F), indicating the estrogenic effects of icariin in

MG-63 cells might be independent to ERE. These results suggested that icariin exerted stimulatory effects on bone formation by up-regulating mRNA expression of

OPG, OCN, ALP and suppressive effects on osteoclastogenesis by down-regulating the ratio of OPG and RANKL gene expression in osteoblast.

107

A B

1.5 * 2.5

*** 2.0 1.0 *** 1.5

1.0 0.5 0.5

0.0 0.0 -7 -7 control E2 (10 M) -10 -9 control E (10 M) -10 -9

ALP/GAPDH (relative to control) (relative ALP/GAPDH 2 OCN/GAPDH (relative to control) (relative OCN/GAPDH Icariin log[M] Icariin log[M]

C D

2.0 1.5 ** 1.5 * * 1.0 * 1.0 *** **

0.5 0.5

0.0 0.0 -7 -7

control E2 (10 M) -10 -9 control E2 (10 M) -10 -9 OPG/GAPDH (relative to control) (relative OPG/GAPDH

Icariin log[M] to control) (relative RANKL/GAPDH Icariin log[M]

E F

2.5 2.0 ** * ** 2.0 * 1.5

1.5 1.0 1.0 0.5 0.5 0.0 0.0 control E (10-7M) -10 -9 -7 2 control E2 (10 M) -10 -9 Icariin log[M] OPG/RANKL (Relative to control) (Relative OPG/RANKL Icariin log[M] to (relatibe control) activity ERE luciferase

Figure 4.11 Effects of icariin on mRNA expression of estrogen-responsive genes and ERE-luciferase activity in MG-63 cells MG-63 cells were cultured and subjected to treatment with vehicle, estradiol (10-7M)

108 and icariin (10-12-10-6M) in phenol red-free MEM containing 5% of cs-FBS for 48 hours as described in Methods (Chapter 3). Upon treatment, gene expression and ERE luciferase activity were measured by Real time-PCR and Dual Luciferase Reporter Assay System, respectively. A. mRNA expression of OCN; B. mRNA expression of ALP; C. mRNA expression of OPG; D. mRNA expression of RANKL; E. OPG/RANKL; F. ERE-dependent luciferase activity. Results were from two independent experiments and expressed as mean ± SEM (n=4). *p<0.05, **p<0.01, ***p<0.001 vs control.

109

4.3 Discussion

The structural and functional similarities of phytoestrogens to mammalian estrogen with less reported side effects as compared with the synthetic estrogens makes it a potential alternative approach for the management of post-menopausal syndrome. It is hoped that the phytoestrogens exerting tissue selective estrogenic effects in target tissues with minimal side effects in other estrogen sensitive tissues will be identified to develop as alternative regimen for treatment of menopause associated symptoms.

As one of main sources of phytoestrogens, Traditional Chinese Medicine has attracted attention from medical communities in the field of estrogen. TCM has been used for management of menopausal symptoms with a long history of safe use. In Chinese medicine, problems of menopause are attributed to kidney deficiency (LI et al.,

2013a). Kidneys are believed to constitute a functional system that involves in regulation and maintenance of growth, maturation and aging. Thus, principles of

TCM for management of menopause are to tonify kidney. Therefore, kidney-tonifying herbs are used for the treatment of menopausal symptoms in forms of TCM formula or the herb itself (Li et al., 2015).

Herba Epimedii (HEP) is an important kidney-tonifying medicinal plant used to tonify kidney and keep balance of kidney in various TCM formulas for thousands of years as well as the modern Chinese herb products (Chen et al., 2010). Among more

110 than 260 detected components in this herb, icariin is the most abundant constituent of

HEP and thus the chemical marker for quality control of HEP in Chinese

Pharmacopeia (Li et al., 2015). Icariin has been reported to have extensive activities including anti-oxidation, anti-cancer, anti-inflammation, cardiovascular protective effect, immnuoactivation (Fan et al., 2016; Ma et al., 2011; Meng et al., 2015; Teng et al., 2008).

In the present study, icariin, a flavonoid, was chosen as one representative compound to study the tissue selective effect of phytoestrogens derived from the Chinese medicinal herbs in hopes that phytoestrogen exert protective effects in target tissues without undesirable adverse effects in other target tissues of estrogen could be identified.

The estrogenic effects of icariin in four estrogen sensitive tissues including bone, brain, uterus and breast were investigated in the mature ovariectomized rats.

Moreover, the direct estrogenic effects of icariin were also determined in four estrogen receptor-positive cell lines including human osteosarcoma MG-63, human neuroblastoma SH-SY5Y, human endometrial cancer Ishikawa and human breast cancer MCF-7 cells. Our results clearly demonstrated that icariin selectively exerted estrogenic effects in estrogen sensitive tissues in both in vivo and in vitro models as shown in Table 4.4.

111

In the present study, icariin was added into the phytoestrogen-free AIN93 diet at three dosages, 50, 500 and 3000ppm, and the animals were paired-fed with the icariin-containing diets for 12 consecutive weeks. Our results showed that long-term treatment with icariin significantly prevented the OVX rats from damages in bone and central nervous system induced by estrogen-deficiency. Observed form the micro-architecture of bone at three sites of bone scanned in present study, the bones of rats become osteoporotic in response to OVX as reflected by the significant reduction of BMD, Tb.Th, Tb.N, BS and BV/TV and significant induction of Tb.Sp as compared to Sham rats. Treatment with 17ß-estradiol dramatically increased BMD and alleviated the deterioration in bone architecture of OVX rats. The effect of estradiol against estrogen deficiency-induced osteoporosis has been reported in both human (Black et al., 2016; Ju et al., 2015) and animal studies (Jing et al., 2014; Niu et al., 2012; Song et al., 2015) as well as our own studies in both OVX rat and mice models (Mok et al., 2010; Wong et al., 2013). Similar to estradiol, treatment with icariin at all dosages significantly restored changes in BMD and bone properties as increasing BMD, BS, BV/TV, Tb.N, Tb.Th and decreasing Tb.Sp in OVX rats . The changes induced by icariin could be obviously observed in the microarchitecture of bone at all the three sites tested in present study. This result was in agreement with previous findings in both postmenopausal women (Zhang et al., 2007) and animal

112 models (Li et al., 2013b) as well as our own previous study in same model (Mok et al.,

2010). In terms of increasing BMD and improving trabecular bone properties, icariin at 3000ppm exerted the most potent effects, which was even more potent than estradiol. Moreover, the increase in two biomarkers for bone turnover, serum osteocalcin and urinary deoxypyridinoline, by estrogen deficiency were also decreased in OVX rats treated with icariin, suggesting that it exerted suppressive effect on bone turnover. In particular, in terms of effects on bone turnover, icariin at

500ppm exerted the most potent effects while icariin at 3000 ppm did not alter serum level of osteocalcin in OVX rats. These results indicated that icariin selectively protected bone from estrogen deficiency-induced osteoporosis in rats possibly via suppression of bone turnover, in a way similar to that of estradiol and 500ppm might be an approiate dose exerting the most potent bone protective effects in OVX rats.

Central nervous system is another target tissue of estrogens where they act as not only sex hormone. Estrogens are known to be related to a series of activities in brain including control of reproduction, cognition and memory. According to studies conducted from 1950 to 1987, unilateral and bilateral oophorectomy before menopause increased the risk of Parkinson’s disease (PD), a neurodegenerative disease (Rocca et al., 2008). This result was confirmed by recent epidemiological studies that postmenopausal women are at increased risk of PD (Ascherio et al., 2003;

113

Ragonese et al., 2004). Epidemiological studies also found that the supplement with exogenous estrogen significantly reduced the risk of PD in postmenopausal women

(Currie et al., 2004), suggesting a potential preventive effect of estrogen on PD.

Similar correlation was also found between estrogen and another neurodegenerative disease, Alzheimer’s disease (AD). The higher incidence of AD in postmenopausal women than their man counterparts of a similar age has been recognized to be related to estrogen deficiency (Baum, 2005). In addition, dementia in postmenopausal women may be predicted by the decrease in estrogen level (Rocca et al., 2014). Moderate exposure to exogenous estrogens reduced the risk of memory impairment, irreversible damage of neurons (Erickson et al., 2007) and improved cognitive performance in postmenopausal women (Sundermann et al., 2006). Furthermore, the earlier the women receive HRT, the later dementia happens, indicating that HRT may delay the onset of AD (Bagger et al., 2005). These results suggest that estrogen protects women from damages in the central nervous system associated with estrogen deficiency.

In the present study, mRNA expression level of tyrosine hydroxylase (TH) and dopamine transporter (DAT) in striatum were determined to evaluate the effects of icariin on central nervous system. TH is the key enzyme responsible for catalyzing the conversion of L-tyrosine to L-3,4-dihydroxyphenylalanine (L-DOPA) and is the rate-limiting enzyme (Kaufman, 1995). DA is the key factor for the clearance of

114 dopamine from synapse. Expression levels of these two proteins respond to estrogen modulation (Al-Sweidi et al., 2011; Chaube et al., 2011) and thereby often are used for evaluation of estrogenic effects in central nervous system. Our results showed that mRNA expression of TH dramatically decreased and DAT gene expression significantly decreased in rats in response to OVX, indicating the estrogen deficiency caused damages in striatum as the biosynthesis of dopamine decreased and release of dopamine increased. Treatment of OVX rats with 17ß-estradiol significantly restored mRNA expression of TH and DAT in striatum to comparable levels as in sham rats, confirming the protective effects of estrogens in central nervous system. Icariin dramatically up-regulated the TH mRNA expression and down-regulated DAT mRNA expression at higher dosages, i.e. 500 and 3000ppm. Icariin at 50ppm appeared to have a trend in stimulating mRNA expression of TH and inhibiting DAT mRNA expression but the changes did not reach statistical significance. These results indicated that icariin exerted estrogen-like protective effect on the central nervous system and protected against estrogen deficiency-induced damages in OVX rats.

The positive effects of icariin on bone and central nervous system have been confirmed by performing the human osteoblast MG-63 and human neuroblastoma

SH-SY5Y cells, respectively. Icariin significantly promoted both the cell proliferation and cell differentiation in MG-63 cells. The mRNA expression level of genes for bone

115 formation including osteocalcin, ALP and OPG were all up-regulated in MG-63 cells upon treatment with icariin for 48 hours while mRNA expression of RANKL was inhibited by more than 30%. Moreover, there was a significant increase in

OPG/RANKL ratio in MG-63 cells upon treatment with icariin. This was in agreement with our previous findings that flavonoids from Herba Epimedii significantly stimulated cell proliferation rate, ALP activity and OPG/RANKL mRNA expression in rat osteoblastic UMR106 cells (Xiao et al., 2014) and indicated that icariin might be one of the flavonoids that exerted bone protective effects. These results suggested that icariin exerted stimulatory effects on bone formation and suppressive effects on osteoclastogenesis via its action on osteoblasts. However, icariin did not alter ERE-dependent luciferase activity in MG-63 cells, indicating that the protective effects of icariin in osteoblast might be independent of ERE-dependent transcription. Regarding its actions on neuronal cells, icariin significantly promoted the cell proliferation in SH-SY5Y cells at a wide range of concentrations. Similar stimulatory effect of icariin in neurons has been reported in a study performed in human neural stem cells (NSCs) (Yang et al., 2016). In addition, mRNA expression level of ER in SH-SY5Y cells was markedly increased upon treatment with icariin.

However, icariin did not alter ERE-dependent luciferase activity in SH-SY5Y cells.

116

These results demonstrated that the direct estrogenic effects of icariin in SH-SY5Y cells might be ERE-independent.

Uterus and breast are main target tissues of estrogen which are usually affected during menopause and they are also the target tissues where side effects of HRT or estrogen are most frequently reported (Hinds et al., 2010). Thus, besides the positive effects in bone and brain of icariin, it is of special importance to investigate the possible side effects in uterus and breast as icariin has potential to exert estrogen-like benefit-risk profile as a phytoestrogen. In uterus, 17ß-estradiol significantly restored the estrogen-deficiency induced decrease in uterus weight in OVX rats. However, treatment with icariin at all dosages neither increased uterus weight nor induced hyperplasia in endometrium in OVX rats. The mRNA expression of complement component 3 (C3), a protein that plays a crucial role in the complement system, in endometrial tissue is known to be induced by estradiol and thus serves as an estrogen-sensitive target gene in endometrium (Wu et al., 2009). Our results showed that OVX caused the significant decrease in C3 mRNA expression in uterus but treatment with 17ß-estradiol reversed the decrease in C3 mRNA expression in rats.

However, long-term treatment of OVX rats with icariin at all dosages did not cause any changes in C3 mRNA expression compared with OVX rats. Moreover, mRNA expression of PR and ER, the other two estrogen responsive genes, were not altered in

117 uterus of OVX rats in response to treatment with icariin. In breast, icariin did not cause alterations in either number or morphology of mammary gland in OVX rats compared to the OVX rats. These results indicated that long-term treatment with icariin did not induce undesirable estrogenic effects in uterus or breast in OVX rats.

Results from the in vitro experiment clearly showed that icariin directly induced estrogenic responses in two estrogen receptor (ER) positive cell lines as it promoted cell proliferation in MCF-7 cells ER-dependently and increased ALP activity in

Ishikawa cells. Moreover, the pS2 gene expression in MCF-7 cells as well as ALP and

ER mRNA expression level in Ishikawa cells were markedly increased upon treatment with icariin. However, icariin did not alter ERE-dependent luciferase activities in both MCF-7 and Ishikawa cells, indicating the direct estrogenic effects of icariin in these cells were ERE-independent. Results from in vitro demonstrated the stimulatory effects of icariin in breast and endometrial cell lines while the in vivo study did not observe any stimulatory effects of icariin in either breast or uterus. The discrepancy between results from in vitro and in vivo may be caused by the too high concentrations applied in vitro which were much higher than that detected in animal blood.

In conclusion, the results of the present study demonstrated that long-term treatment with icariin selectively protected bone and brain from damages resulting from

118 estrogen deficiency in a way similar to estrogen without inducing undesirable side effects in uterus and breast. Icariin directly exerted estrogenic effects in estrogen receptor positive cell lines and the estrogenic effects in vitro were found to be

ERE-independent. Tresent study demonstrated the tissue-selective effects of icariin.

Based on our results, we are confident to recommend icariin, a flavonoid phytoestrogen, as a safe and practical alternative to HRT in management and prevention of menopause related symptoms.

119

Table 4.4 Summary of the tissue-selective estrogenic effects of icariin in four estrogen sensitive tissues in comparison to 17ß-estradiol

17ß-estradiol icariin Uterus Uterus Weight Increased No change Gene expression Increased No change Morphology hyperplasia No change Ishikawa cell ALP activity Stimulatory effect Stimulatory effect at 10-10M to 10-8M ALP ER gene expression ALP: increased (+++)ER: increased (+) ALP: increased (+)ER: increased (+) ERE luciferase activity Increased No change Breast Breast Morphology Hyperplasia No change MCF-7 cell Cell proliferation Stimulatory effect Stimulatory effect at 10-12 to 10-6M Estrogen responsive genes IGF-IR, pS2: increased (+);ER: no change pS2: increased (+)ER, IGF-IR: no change ERE luciferase activity Increased No change Bone Bone BMD Increased (+++) Increased (+++) Bone property Improved (+++) Improved (++) OCN, DPD OCN decreased (-) OCN decreased (-) at 50 and 500ppm DPD decreased (---) DPD decreased (-) at all dosages MG-63 cell Cell proliferation Stimulatory effect Stimulatory effect at 10-12 to 10-8M ALP activity Stimulatory effect Stimulatory effect at 10-10 to 10-7M Estrogen responsive genes OCN, ALP, OPG, OPG/RANKL: increased OCN, ALP, OPG, OPG/RANKL: increased (+);RANKL: decreased (-) (+)RANKL: decreased (-) ERE luciferase activity Increased No change CNS Striatum Estrogen responsive genes TH increased (++) TH increased (+) at 500 and 3000ppm DAT decreased (++) DAT decreased (+) at 500 and 3000ppm SH-SY5Y cell Cell proliferation Stimulatory effect Stimulatory effect at 10-10 to 10-7M Estrogen responsive genes ER: increased (+) ER: increased (+) ERE luciferase activity Increased No change Increase (+++) > (++) > (+); Decrease (---) > (--) > (-)

120

Chapter 5

Characterization of the tissue selectivity of DBT

in mature ovariectomized (OVX) rats and

estrogen-sensitive cell lines as well as the

possible mechanisms involved

121

5.1 Introduction

DBT, a Traditional Chinese Medicine formula consisting of two common herbs, Radix

Astragali and Radix Angelica Sinensis at a special ratio of 5:1, has been prescribed for postmenopausal women for a long history with safe use. According to Chinese

Medicine, by raising “Qi” and nourishing the “Blood”, DBT attenuates the symptoms of postmenopausal women and improves their health. A clinical trial conducted in

Hong Kong demonstrated that DBT alleviated menopausal symptoms including hot flashes, night sweat in postmenopausal women and no side effect has been reported

(Wang et al., 2013). In addition, DBT has been proved to be effective in improving cardiovascular circulation (Chiu et al., 2007), stimulating the immune function (Gao et al., 2008), protecting bone against osteoporosis in both in vivo and in vitro model, increasing antioxidant activity (Gong et al., 2016; Zierau et al., 2014) and promoting hematopoietic functions (Zheng et al., 2010). Recent study reported that DBT exerted estrogen-like activities in phosphorylation of ERs. In particular, DBT dramatically increased estrogen response element-dependent receptor activities and its stimulating effects on cell proliferation and cell differentiation in MCF-7 cell and MG-63 cell could be completely blocked by ER antagonist ICI182,780, indicating the possible involvement of ER in actions of DBT. These results suggest that DBT may contain phytoestrogen via which its actions are mediated.

122

Present study aimed to evaluate the tissue-selective effects of DBT by using established preclinical in vivo and in vitro models. Four estrogen-responsive tissues including breast, uterus, brain and bone, which are most frequently affected during menopause, were studied in present study. In the in vivo experiment, the mature female ovariectomized SD rats, the most frequently used model for studying postmenopausal symptoms, were employed. Upon treatment, estrogenic effects of

DBT in the four tissues were evaluated by various specific estrogen-responsive parameters for each tissue.

To be consistent with the in vivo study, four estrogen sensitive cell lines including human breast cancer MCF-7 cell, human endomentrial cancer Ishikawa cell, human osteosarcoma MG-63 cell and human neuroblastoma SH-SY5Y cell were employed in the in vitro study. To investigate whether DBT exerts direct estrogenic effects in these four cell lines, their effects on cell proliferation or cell differentiation, ALP activity

(ALP), expression of specific estrogen-responsive genes and estrogen response element (ERE) luciferase activity in response to various concentrations of DBT were determined. In addition, the possible mechanisms mediating actions of DBT in bone cells were investigated in human osteoblastic MG-63 cells by using specific blockers for ER-related signaling pathways.

123

5.2 Results

5.2.1 Quality control and standardization of DBT extract

For each herb, high performance liquid chromatography (HPLC) analysis was performed to ensure that its quality fulfilled the requirement of the China

Pharmacopoeia and/or the Hong Kong Chinese Materia Medica Standard. Constitutes of DBT extract were determined by Liquid Chromatography Mass Spectrometry

(LC-MS). The four peaks detected in HPLC are the four main metabolites, ferulic acid (1), calycoisin (2), formononetin (Rocca et al.) and Z-ligustilide (4), of the two herbs, Radix Astragali and Radix Angelica Sinensis, in DBT formula. The amounts of the metabolites in DBT extract measured by LC-MS were listed in table 5.1. These chemical characterizations were used to prepare the standardized extracts for all the biochemical analysis of DBT used in the present study.

124

Figure 5.1 Chemical standardization of DBT by HPLC fingerprint analysis

1. Feculic acid; 2. Calycoisin; 3. Formononetin; 4. Z-ligustilide

Table 5.1 Chemical composition of DBT extract

Marker Name Content (mg/g) Minimum Content (mg/g) Ferulic acid 0.809 0.351 Calycoisin 0.693 0.186 Formononetin 0.164 0.155 Z-ligustilide 0.212 0.204

125

5.2.2 Characterization of the estrogenic effects of DBT in OVX rats

5.2.2.1 Effects of DBT on body weight gain in OVX rats

In response to OVX, body weight of rat showed a dramatic increase (Figure 5.2, p<0.001 vs Sham) while treatment with 17ß-estradiol significantly suppressed the increase in body weight (p<0.001 vs OVX), possibly via suppressing the food intake and increasing energy expenditure (Gao et al., 2007a; Picherit et al., 2000). This was consistent with our own and previous findings by others that 17ß-estradiol could fully abolish the estrogen deficiency-induced body weight gain in OVX rats (Gao et al.,

2007a; Mok et al., 2010; Picherit et al., 2000). However, long-term treatment with

DBT did not alter the body weight gain of OVX rats, indicating that the actions of

DBT on body weight were different from those of estrogen in OVX rats (Figure 5.2).

126

10 ***

-5

-10 ^^^ Sham OVX E2 DBT Body weight gain (% gain weight the Body initial) of

Figure 5.2 Effect of DBT on body weight gain of OVX rats

Six-month old OVX Sprague Dawley rats were orally administrated with vehicle, 17ß-estradiol (2mg/kg.day) and DBT (3g/kg.day) for 12 consecutive weeks. Body weight of rats was measured every two weeks during the whole experiment. Percentages of the changes in body weight over their baseline body weight were regarded as the body weight gain of rats. Sham: sham operated + vehicle treatment; OVX: OVX operated + vehicle treatment; E2: OVX operated + 17ß-estradiol treatment; DBT: OVX operated + DBT treatment. Data was expressed as mean ± SEM. ***p<0.001 vs sham; ^^^p<0.001 vs OVX. n=8 or 9.

127

5.2.2.2 Effects of DBT on serum reproductive hormones of OVX rats

Endocrine system of female is regulated and controlled by hypothalamus-pituitary-ovary axis as one major part of the neuroendocrine axis. As the advanced center of this axis, hypothalamus secretes gonadotropin-releasing hormone (GnRH) which stimulates the secretion and production of Follicle stimulating hormone (FSH) and Luteinizing hormone (LH) from pituitary. The two gonadotropins regulate production and secretion of estrogen from ovary via feedback regulation.

Serum level of estradiol dramatically decreased in rats upon ovariectomy and was accompanied by the markedly increased serum FSH and LH levels (Figure 5.3A, B and C, p<0.01, 0<0.001, p<0.001 vs Sham). As expected, exposure to 17ß-estradiol significantly suppressed the increase in FSH and LH level in OVX rats (Fig 5.3A, B and C, p<0.001 vs OVX). Long-term treatment with DBT significantly increased the estradiol level of OVX rats by three fold and suppressed the increase in serum level of

FSH and LH (Fig 5.3B and C, p<0.001, p<0.001 vs OVX). Moreover, the suppressive effects of DBT on FSH and LH were comparable to that of estradiol. These results indicated that DBT might regulate the hypothalamus-pituitary-gonadal axis and restore the ovariectomy-induced disturbances in the regulation of reproductive hormones in OVX rats.

128

A B

20 400 350 ^^^ 15 *** 300 250 200 10 100 ^^^ 80 ^^^ 60 ^^^ 5 40 20 **

0 FSH (ng/ml) of level serum 0

Estradiol concentration (pg/ml) concentration Estradiol sham OVX E2 DBT sham OVX E2 DBT

6000 ***

4000 ^^^ ^^^

2000

serum level of LH (pg/ml) of level serum 0 sham OVX E2 DBT

Figure 5.3 Effects of DBT on serum reproductive hormones in OVX rats

Blood was obtained from sham or ovariectomized (OVX) (six-month old) Sprague Dawley rats treated with vehicle, 17ß-estradiol or DBT for 12 weeks. Serum level of estradiol, follicular stimulating hormone (FSH) and luteinizing hormone were measured by commercial kits. A. Serum level of estradiol; B. Serum level of FSH; C. Serum level of LH. Sham: sham operated + vehicle treatment; OVX: OVX operated + vehicle treatment; E2: OVX operated + 17ß-estradiol treatment; DBT: OVX operated + DBT treatment. Data was expressed as mean ± SEM. **p<0.01, ***p<0.001 vs sham; ^^^p<0.001 vs OVX. n=8 or 9.

129

5.2.2.3 Effects of DBT on mRNA expression of aromatase in subcutaneous adipose tissue of OVX rats

Aromatase is the key enzyme involved in the biosynthesis of estrogen and is encoded by gene CYP19A1 (Rankinen et al., 2006). To investigate how DBT increased the serum level of estradiol in OVX rats, mRNA expression of aromatase in adipose tissue was determined by real-time PCR. As shown in figure 5.4, aromatase mRNA expression in adipose tissue was not altered in OVX rats while 17ß-estradiol significantly suppressed the mRNA expression of aromatase in the adipose tissue of

OVX rats (p<0.05 vs OVX). Our result was consistent with a previous finding that estradiol was found to reduce the expression of aromatase in pituitary level in OVX rats (Galmiche et al., 2006). However, treatment with DBT significantly up-regulated mRNA expression of aromatase in adipose tissue by 54% when compared to that of

OVX rats (Figure 5.4, p<0.05 vs OVX), suggesting the extra-gonadal tissue might become a source of estrogen in DBT-treated OVX rats.

130

1.5 ^

1.0

0.5 ^

0.0

gene expression level CYP19A1 level expression gene sham OVX E2 DBT

Figure 5.4 Effects of DBT on mRNA expression of aromatase in adipose tissue in OVX rats Upon treatment, subcutaneous adipose tissue was also collected the abdomen for detection of mRNA expression of aromatase by performing Real time-PCR. Sham: sham operated + vehicle treatment; OVX: OVX operated + vehicle treatment; E2: OVX operated + 17ß-estradiol treatment; DBT: OVX operated + DBT treatment. Data was expressed as mean ± SEM. **p<0.01, ***p<0.001 vs sham; ^^^p<0.001 vs OVX. n=8 or 9.

131

5.2.2.4 Effects of DBT on uterus in OVX rats

As shown in figure 5.5A, uterine weight dramatically decreased in rats upon OVX.

Treatment of OVX rats with 17ß-estradiol markedly increased uterine weight by about five fold (p<0.001 vs OVX rats). However, DBT did not alter the uterus index in

OVX rats. In addition, the mRNA expression of three estrogen-responsive genes, complement component 3 (C3), progesterone receptor (PR) and estrogen receptor (ER) in rat uterus in response to treatment of 17ß-estradiol and DBT were determined by real-time PCR to further evaluate the estrogenic effects of DBT on uterus in OVX rats.

As shown in Figure 5.5B-5.5D, mRNA expression of C3, ER and PR decreased in rat uterus in response to OVX. Both DBT and estradiol appeared to up-regulate the mRNA expression of C3, ER and PR in OVX rats but the changes were not statistically significant. The morphology of endometrium was visualized by H&E staining. As shown in Figure 5.5E indicated by the red arrows, atrophy was observed in endometrium and the epithelial cells were flattened in rats in response to OVX;

17ß-estradiol, but not DBT, markedly attenuated the atrophic status in endometrium even induced hyperplasia in endometrium of OVX rats. These results further confirmed that DBT did not cause estrogenic side effect on uterus in OVX rats.

132

A B

2.5

2.0 20 ^^^ 1.5

1.0 10 C3/GAPDH

0.5 ***

0.0 0 sham OVX E2 DBT sham OVX E2 DBT Uterus index (mg/g body weight) body (mg/g Uterus index

C D

15 15

10 10

5 5

PR/GAPDH ER/GAPDH

0 0 sham OVX E2 DBT sham OVX E2 DBT

Figure 5.5 Estrogenic effects of DBT on uterus index, endometrial morphology and mRNA expression of estrogen-responsive genes in uterus of OVX rats Upon treatment, uterus was collected freshly and weighted. A. Uterus index, the ratio

133 of uterine weight (Giunta et al.) over body weight (g) was calculated as uterus index to evaluate the estrogenic effects of DBT in uterus; B. mRNA expression of component complement 3(C3); C. mRNA expression of progesterone receptor (PR); D. mRNA expression of estrogen receptor (ER); mRNA expression of three estrogen-responsive genes, C3, PR and ER were determined by Real time-PCR; E. Morphology of endometrium was visualized by H&E staining. Sham: sham operated + vehicle treatment; OVX: OVX operated + vehicle treatment; E2: OVX operated + 17ß-estradiol treatment; DBT: OVX operated + DBT treatment. Data was expressed as mean ± SEM. ***p<0.001 vs sham; ^^^p<0.001 vs OVX. n=8 or 9.

134

5.2.2.5 Effect of DBT on breast tissue in OVX rats

As shown in Figure 5.6 indicated by the red arrows, number of the mammary gland of

OVX rats dramatically decreased and mammary gland became atrophic when compared to Sham rats. As expected, treatment with 17ß-estradiol attenuated these changes in mammary gland in OVX rats. Treatment of OVX rats with DBT also slightly stimulated hyperplasia in mammary gland, indicating the mild estrogenic effect of DBT on breast of OVX rats.

135

Figure 5.6 Estrogenic effects of DBT on the morphology of breast tissues in OVX rats Upon treatment, the second breast of Sham rats and OVX rats treated with vehicle, 17ß-estradiol or DBT were collected and immediately fixed in 4% paraformaldehyde. H&E staining was performed to visualize the morphology of mammary gland of rats. Sham: sham operated + vehicle treatment; OVX: OVX operated + vehicle treatment; E2: OVX operated + 17ß-estradiol treatment; DBT: OVX operated + DBT treatment. n=8 or 9.

136

5.2.2.6 Effects of DBT on brain tissue in OVX rats

In the present study, the mRNA expression of tyrosine hydroxylase (TH) markedly decreased by 64% in OVX rats (Figure 5.7A, p<0.001 vs Sham rats). The changes in

TH mRNA expression induced by OVX in rats was significantly reversed by treatment with 17ß-estradiol (Figure 5.7A, p<0.001 vs OVX rats). Similarly, DBT also reversed the decrease in TH mRNA expression in OVX rats (Figure 5.7A, p<0.001 vs OVX rats). Moreover, the increase in TH mRNA expression induced by

DBT in OVX rats was higher than that induced by estrogen. The mRNA expression of dopamine transporter (DAT) dramatically increased by OVX in rats, indicating the enhanced release of dopamine in OVX rats (Figure 5.7B, p<0.001 vs Sham rats). Both

17ß-estradiol and DBT significantly suppressed the increase in DAT mRNA expression level in striatum of OVX rats (Figure 5.7B, p<0.001 vs OVX). The DAT mRNA expression was suppressed by DBT to lower level than that by estradiol

(Figure 5.7B). The stimulatory effect of DBT on mRNA expression of TH and inhibitory effect on DAT mRNA expression in striatum of OVX rats suggested that

DBT protected the dopaminergic system from damages induced by estrogen deficiency and improved the metabolism of dopamine in estrogen deficient conditions in OVX rats. In terms of effects on TH and DAT mRNA expression, DBT appeared to be more potent than estradiol.

137

A B

0.8 4 *** ^^^ ^^^ 0.6 3

0.4 2 ^^^

TH/GAPDH *** 0.2 DAT/GAPDH 1 ^^^

0.0 0 sham OVX E2 DBT sham OVX E2 DBT

Figure 5.7 Estrogenic effects of DBT on mRNA expression of Tyrosine hydroxylase (TH) and Dopamine Transporter (DAT) in striatum of OVX rats Upon treatment, striatum of Sham and OVX rats treated with vehicle, 17ß-estradiol or DBT for 12 weeks was freshly collected. mRNA expression of TH and DAT in striatum were determined by Real time-PCR. A. mRNA expression of TH; B. mRNA expression of DAT. Sham: sham operated + vehicle treatment; OVX: OVX operated + vehicle treatment; E2: OVX operated + 17ß-estradiol treatment; DBT: OVX operated + DBT treatment. Data was expressed as mean ± SEM. ***p<0.001 vs sham; ^^^p<0.001 vs OVX. n=8 or 9.

138

5.2.2.7 Effects of DBT on bone of OVX rats

BMD and bone properties

As shown in figure 5.8 and table 5.2, 5.3 and 5.4, bone mineral density (BMD) and microarchitecture properties including bone surface (BS), bone volume to total volume ratio (BV/TV), trabecular number (Tb.N) and Trabecular Thickness (Tb.Th) dramatically decreased while trabecular separation (Tb.Sp) markedly increased at distal femur, proximal tibia and spine in response to OVX (p<0.05, p<0.01, p<0.001 vs Sham rats), indicating the osteoporotic status of both long bone and spine in OVX rats. Such changes in BMD and bone properties could be directly reflected from the microarchitecture of bone (Figure 5.8D, E and F). Exposure to 17ß-estradiol significantly restored the deterioration caused by estrogen deficiency in OVX rats

(Figure 5.8A, B and C, Table 5.2, 5.3 and 5.4, P<0.01, P<0.001 VS ovx). Treatment of

OVX rats with DBT also significantly attenuated changes in BMD and bone properties as reflected DBT significantly increased BMD (Figure 5.8A, B and C, p<0.001 vs OVX), BS, BV/TV, Tb.N and Tb.Sp while decreased Tb.Sp (Table 5.2, 5.3 and 5.4, p<0.05, p<0.01, p<0.001 vs OVX). The protective effects of DBT on bone could be observed from the microarchitecture of bone at the three sites tested (Figure

5.8D for distal femur, 5.8E for proximal tibia and 5.8F for spine). In addition, the beneficial effects of DBT on trabecular bone properties of spine were comparable to

139 that of 17ß-estradiol, which restored Tb.Th back to about 80% of sham group (Table

5.4). These results suggested that DBT exerted potent protective effects on both long bone and spine against osteoporosis resulting from estrogen deficiency in OVX rats in estrogen-like manner.

140

A B

500 500

400 400 ^^^ ^^^ 300 300 ^^ ^^^ 200 200 *** ***

100 100

BMD (mg HA/ccm) (mg BMD BMD (mg HA/ccm) (mg BMD

0 0 sham OVX E2 DBT sham OVX E2 DBT

500

400 ^^^ ^^^ 300 ***

200

100 BMD (mg HA/ccm) (mg BMD

0 sham OVX E2 DBT

D E

141

Figure 5.8 Effects of DBT on bone mineral density and micro-architecture at distal femur, proximal tibia and lumbar spine of OVX rats Upon treatment, whole left leg and spine from Sham and OVX rat treated with vehicle, 17ß-estradiol or DBT were collected for micro-CT scanning. Bone mineral density (BMD) and trabecular bone microarchitecture at proximal tibia and distal femur as well as lumbar vertebra were determined by Micro-CT as described in Chapter 3 (Methods). A. BMD of distal femur; B. BMD of proximal tibia; C. BMD of spine; D. Microarchitecture of distal femur; E. Microarchitecture of proximal tibia; F. Microarchitecture of lumbar spine. Sham: sham operated + vehicle treatment; OVX: OVX operated + vehicle treatment; E2: OVX operated + 17ß-estradiol treatment; DBT: OVX operated + DBT treatment. Data was expressed as mean ± SEM. *p<0.05, **p<0.01, ***p<0.001 vs sham; ^p<0.05, ^^p<0.01, ^^^p<0.001 vs OVX. n=8 or 9.

142

Table 5.2 Effects of DBT on trabecular bone properties at distal femur of OVX rats BS (mm2) BV/TV Tb.N (1/mm) Tb.TH (mm) Tb.Sp (mm)

Sham 239.3±4.49 0.484±0.017 4.73±0.07 0.102±0.003 0.109±0.005

OVX 86.7±8.85*** 0.121±0.014*** 1.66±0.19*** 0.067±0.002*** 0.565±0.081***

E2 182.0±7.73^^^ 0.305±0.018^^^ 3.38±0.15^^^ 0.086±0.003^ 0.214±0.015^^^

DBT 141.1±13.25^^ 0.207±0.032^ 2.64±0.30^^^ 0.076±0.002 0.353±0.041^^^

Table 5.3 Effects of DBT on trabecular bone properties at proximal tibia of OVX rats BS (mm2) BV/TV Tb.N (1/mm) Tb.TH (mm) Tb.Sp (mm)

Sham 228.8±9.43 0..517±0.023 4.97±0.05 0.104±0.004 0.097±0.005

OVX 55.0±7.67*** 0.076±0.010*** 1.23±0.16*** 0.062±0.001*** 0.856±0.144***

E2 155.1±10.26^^^ 0.255±0.023^^^ 3.38±0.19^^^ 0.076±0.002^^^ 0.214±0.014^^^

DBT 110.4±18.62^ 0.172±0.047 2.46±0.31^^^ 0.070±0.004 0.420±0.061^^^

Table 5.4 Effects of DBT on trabecular bone properties at lumbar spine of OVX rats BS (mm2) BV/TV Tb.N (1/mm) Tb.Th (mm) Tb.Sp (mm)

Sham 89.9±2.73 0.442±0.012 4.12±0.09 0.107±0.001 0.136±0.006

OVX 62.6±5.18** 0.209±0.016*** 2.62±0.17*** 0.079±0.001*** 0.313±0.026***

E2 90.4±3.33^^ 0.350±0.010^^^ 3.71±0.06^^^ 0.094±0.002^ 0.176±0.006^^^

DBT 81.3±4.78^ 0.307±0.019^^ 3.40±0.09^^^ 0.090±0.004 0.205±0.010^^^

Bone morphometric properties, bone volume over total volume (BV/TV), trabecular bone number (Tb.N, mm-1), trabecular bone thickness (Tb.Th, mm), trabecular bone separation (Tb.Sp, mm) and bone surface (BS, mm2) were evaluated by contoured VOI images. Sham: sham operated + vehicle treatment; OVX: OVX operated + vehicle treatment; E2: OVX operated + 17ß-estradiol treatment; DBT: OVX operated + DBT treatment. Data was expressed as mean ± SEM. *p<0.05, **p<0.01, ***p<0.001 vs sham; ^p<0.05, ^^p<0.01, ^^^p<0.001 vs OVX. n=8 or 9. BS: Bone surface; BV/TV: Bone volume/total volume; Tb.N: Trabecular bone number; Tb.Th: Trabecular bone thickness; Tb.Sp: Trabecular separation.

143

Biochemical markers for bone turnover

Two biochemical markers, the serum level of osteocalcin and urinary level of DPD, were also measured to further investigate the effects of DBT on bone turnover in

OVX rats. As shown in Figure 5.9A and B, both serum osteocalcin and urinary deoxypyridinoline were significantly increased in OVX rats (p<0.01 vs Sham rats), indicating an increase in bone turnover rate. Both 17ß-estradiol and DBT significantly suppressed the increase in serum osteocalcin and urinary DPD levels in OVX rats

(p<0.05, p<0.001 vs OVX rats). Moreover, DBT showed more potent suppressive effect on serum osteocalcin in OVX rats than estradiol. Our results suggested that

DBT protected bone from osteoporosis induced by estrogen deficiency in OVX rats in estrogen-like manner possibly via suppressing bone turnover rate.

144

A B

20 ** 150 ^ *** ^ 15 100

10 ^^^ 50

5 ^^^ osteocalcin (ng/ml) osteocalcin

0 0 sham OVX E2 DBT (nmol/mmol) DPD/Creatinine sham OVX E2 DBT

Figure 5.9 Effects of DBT on bone turnover biomarkers of OVX rats

Serum level of osteocalcin and urinary DPD were measured by commercial kits. A. Serum level of osteocalcin; B. Urinary deoxypyridinoline. Sham: sham operated + vehicle treatment; OVX: OVX operated + vehicle treatment; E2: OVX operated + 17ß-estradiol treatment; DBT: OVX operated + DBT treatment. Data was expressed as mean ± SEM. *p<0.05, **p<0.01, ***p<0.001 vs sham; ^p<0.05, ^^p<0.01, ^^^p<0.001 vs OVX. n=8 or 9.

145

5.2.3 Characterization of estrogenic effects of DBT in estrogen sensitive cells

5.2.3.1 Effects of DBT on MCF-7 cells

MCF-7 cell is an estrogen receptor-positive cell line which responds to estrogen in proliferation and gene expression. To evaluate the direct estrogenic effects of DBT in

MCF-7 cell, cell proliferation, ERE-dependent luciferase activity and mRNA expression level of several estrogen-regulated genes were determined. Our results showed that DBT significantly increased cell proliferation at all the concentrations tested (Figure 5.10A, p<0.05, p<0.01 vs control). The stimulatory effect of DBT on cell proliferation was comparable to the effect of estradiol. DBT at intermediate concentrations, i.e. 0.1 and 0.5mg/ml, were chosen for subsequent studies. DBT significantly promoted ERE-dependent luciferase activity in MCF-7 cell (Figure

5.10B, p<0.05, p<0.001 vs control). DBT at 0.5mg/ml even increased the

ERE-dependent luciferase activities by more than 4 times in which the effects were comparable to those of estradiol. The results indicate that DBT mimicked estrogen and directly stimulated ERE-dependent transcription in MCF-7 cell. pS2 is an estrogen-regulated protein and reflects the function of ER (Won Jeong et al.,

2012) while insulin-like growth factor-1 receptor (IGF-IR) is an important receptor related to cell-cycle progression that is regulated by estrogen (Huang et al., 2015).

Together with ER, gene expression of pS2 and IGF-IR were measured by Real time-PCR for the assessment of estrogenic effects of DBT in MCF-7 cells. Our results

146 showed that ER mRNA expression were increased by more than one fold in response to treatment with DBT at both 0.1 and 0.5mg/ml in MCF-7 cells (Figure 5.10C, p<0.05 vs control). For the other two estrogen responsive genes in MCF-7 cells, DBT showed the trend to stimulate the mRNA expression of IGF-IR (Figure 5.10D) and significantly stimulated the mRNA expression of pS2 by more than three time (Figure

5.10E, p<0.05 vs control). These results suggested that the direct estrogenic effects of

DBT in MCF-7 cells might be mediated via regulation on mRNA expression of estrogen-responsive genes.

147

1.5 * ** ** ** ** * * 1.0

0.5

0.0 C E 0.05 0.1 0.25 0.5 1.0 2.0 Cell viablity (Relative to control) (Relative viablity Cell 2 DBT (mg/ml)

B C

4 6 * *** *** 3 * * 4 * 2

(relatibe to control) (relatibe ERE luciferase activity ERE luciferase 0 0 -8 -8 control E2 (10 M) 0.1 0.5 to (relative control) ER/GAPDH control E2 (10 M) 0.1 0.5 DBT (mg/ml) DBT (mg/ml)

D E

2.0 6 * 1.5 * ** 4 1.0

0.5 2

0.0 0 -8 -8 control E2 (10 M) 0.1 0.5 control E (10 M) 0.1 0.5 pS2/GAPDH (relative to (relative control) pS2/GAPDH 2 IGF-IR/GAPDH (relative to control) (relative IGF-IR/GAPDH DBT (mg/ml) DBT (mg/ml)

Figure 5.10 Effects of DBT on cell proliferation, ERE-dependent luciferase activity and mRNA expression of estrogen-responsive genes in MCF-7 cells MCF-7 cells were cultured and subjected to treatment with vehicle, 17ß-estradiol or DBT at various concentrations in phenol red-free DMEM containing 1% cs-FBS for 48 hours. A. Cell proliferation were measure by MTS assay; B. ERE-dependent

148 luciferase activity were determined using Dual Luciferase Reporter Assay System; C. mRNA expression of ER; D. mRNA expression of IGF-IR; E. mRNA expression of pS2 were determined by Real time-PCR. Results were from two independent experiments and expressed as mean ± SEM (n=4). *p<0.05, **p<0.01, ***p<0.001 vs control.

149

5.2.3.2 Effects of DBT in Ishikawa cell

Besides cell proliferation, ERE-dependent luciferase activity and mRNA expression of estrogen-responsive genes, alkaline phosphatase (ALP) activity was also used to evaluate the estrogenic effects of DBT in Ishikawa cells, which is of particular importance for studying estrogenic effects in endometrial cancer (Holinka et al.,

1986a). As shown in Figure 5.11A and B, DBT at 0.05 to 2mg/ml did not alter cell proliferation but was able to significantly increase ALP activities in Ishikawa cells at the lower concentrations, i.e. 0.05, 0.1 and 0.25mg/ml, tested in present study (p<0.05, p<0.01vs control). Such results suggested that DBT exerted the estrogenic effects in

Ishikawa cells. Referred to the results of both cell proliferation and ALP activity, DBT at 0.5 and 1.0mg/ml were chosen for the subsequent studies. Moreover, DBT at both doses dramatically increased ERE-dependent luciferase activities in Ishikawa cells

(Figure 5.11C, p<0.05, p<0.01 vs control), suggesting that DBT could directly activate ERE-dependent transcription in Ishikawa cells. In particular, the stimulatory effects of DBT at 0.5 and 1.0mg/ml were comparable to the effects of estradiol, indicating the potent estrogenic effects of DBT on ERE luciferase activity in Ishikawa cells. Furthermore, DBT appeared to induce mRNA expression of ER and ALP in

Ishikawa cells (Figure 5.11D and E). However, only DBT at 0.5mg/ml up-regulated the mRNA expression of ER by three fold with significance (Figure 5.11D, p<0.05 vs

150 control). These results suggested that DBT directly activate the expression of estrogen-regulated gene in Ishikawa cells.

151

A B

1.5 2.0 *** ** 1.0 1.5 ** * *** ** * * ** 1.0 0.5 0.5

0.0 0.0 C E 0.05 0.1 0.25 0.5 1.0 2.0 Cell viablity (Relative to control) (Relative viablity Cell 2 C E 0.05 0.1 0.25 0.5 1.0 2.0 ALP activity (relative to control) (relative activity ALP 2 DBT (mg/ml) DBT (mg/ml)

C D

2.0 ** 10 *** * 1.5 ** 8

6 1.0 * 4 0.5

(relatibe to control) (relatibe ERE luciferase activity ERE luciferase 0.0 0 control E (10-8M) 0.5 1.0 -8 2 to control) (relative ER/GAPDH control E2 (10 M) 0.5 1.0 DBT (mg/ml) DBT (mg/ml)

15 *** 10

0 control E (10-8M) 0.5 1.0

ALP/GAPDH (relative to control) (relative ALP/GAPDH 2 DBT (mg/ml)

Figure 5.11 Effects of DBT on cell proliferation, ALP activity, ERE-dependent luciferase activity and expression of estrogen-responsive genes in Ishikawa cell Ishikawa cells were cultured and subjected to treatment with vehicle, 17ß-estradiol or DBT at various concentrations in phenol red-free DMEM containing 1% cs-FBS for 48 hours. Upon treatment, A. Cell proliferation was measured by MTS assay; B. ALP activity was measure by ALP kit; C. ERE-dependent luciferase activity was determined by Dual Luciferase Reported Assay System; D. mRNA expression of ER; E. mRNA expression of ALP were determined by Real-time PCR. Results were from two independent experiments and expressed as mean ± SEM (n=4). *p<0.05, **p<0.01, ***p<0.001 vs control.

152

5.2.3.3 Effects of DBT in SH-SY5Y cell

The effects of DBT on cell proliferation, ERE-dependent luciferase activity and ER gene expression were determined in SH-SY5Y cells to evaluate the direct estrogenic effects of DBT in central nervous system. Our results showed that DBT significantly promoted cell proliferation in SH-SY5Y cell at all concentrations tested in the present study (Figure 5.12A, p<0.05, p<0.01 vs control). DBT at 0.25 and 0.5 mg/ml appeared to exert the most potent effects in SH-SY5Y cells and were chosen for the subsequent studies. DBT at both doses significantly increased ERE-dependent luciferase activities (Figure 5.12B, p<0.01, p<0.05 vs control) and ER mRNA expression (Figure 5.12C, p<0.001 vs control) in SH-SY5Y cell. ER mRNA expression was increased by more than three times in SH-SY5Y cells upon DBT treatment and the effects were far more potent than that of estradiol treatment group

(Figure 5.12C). The results indicated DBT could activate ERE-dependent transcription and might exert estrogenic effects via the up-regulation of ER expression in SH-SY5Y cells.

However, as for the other two estrogen-responsive genes, TH and DAT, due to the extreme low base expression, no results were obtained.

153

2.0 *** ** 1.5 ** * * ** * 1.0

Cell viablity Cell 0.5 (Relative to control) to (Relative 0.0 -8 controlE2 (10 M)0.05 0.1 0.25 0.5 1.0 2.0 DBT (mg/ml)

B C

Figure 5.12 Effects of DBT on cell proliferation, ERE-dependent luciferase activity and expression of estrogen-responsive genes in SH-SY5Y cell SH-SY5Y cells were cultured and subjected to treatment with vehicle, 17ß-estradiol or DBT at various concentrations in phenol red-free DMEM containing 1% cs-FBS for 48 hours. Upon treatment, A. Cell proliferation was measured by MTS assay; B. ERE-dependent luciferase activity was determined by Dual Luciferase Reported Assay System; C. mRNA expression of ER was determined by Real-time PCR. D Results were from two independent experiments and expressed as mean ± SEM (n=4). *p<0.05, **p<0.01, ***p<0.001 vs control.

154

5.2.3.4 Effects of DBT in MG-63 cells and possible mechanisms involved

Cell proliferation and ALP activity

As shown in Figure 5.13A and B, 17ß-estradiol significantly stimulated cell proliferation and ALP activity in MG-63 cells (p<0.05, p<0.001 vs control). Similarly,

DBT also significantly promoted the cell proliferation in MG-63 cells at all concentrations tested while only DBT at higher concentrations, i.e. 1.0 and 2.0mg/ml, significantly increased ALP activity in MG-63 cells (p<0.01, p<0.001 vs control) by more than two fold (p<0.001vs control). In particular, DBT at 1.0 and 2.0mg/ml appeared to exert more potent stimulatory effects than estradiol on cell proliferation and ALP activity in MG-63 cells. These results indicated the direct estrogenic effects of DBT in MG-63 cells.

By referring to the results of both cell proliferation and ALP activity, two doses of

DBT, 1.0 and 2.0mg/ml, were chosen for the subsequent studies in MG-63 cells.

155

3 *** *** 2 *** ** ** ** ***

Cell viablity Cell (Relative to control) to (Relative 0 -7 ControlE2(10 M) 0.05 0.1 0.25 0.5 1.0 2.0 DBT (mg/ml)

2.5 ***

2.0 ***

1.5 *

1.0

ALP activity ALP 0.5 (relative to control) to (relative 0.0 -7 Control E2 (10 M) 0.05 0.1 0.25 0.5 1.0 2.0 DBT (mg/ml)

Figure 5.13 Effects of DBT on cell proliferation in MG-63 cells

MG-63 cells were cultured and subjected to vehicle, estradiol (10-7M) and DBT in phenol red-free MEM containing 5% of charcoal-stripped fetal bovine serum (cs-FBS) for 48 hours. Upon treatment, A. Cell proliferation was measured by MTS assay; B. ALP activity was measured by ALP assay. Results were from two independent experiments and expressed as mean ± SEM (n=4). *p<0.05, **p<0.01, ***p<0.001 vs control.

156 mRNA expression of estrogen responsive genes in MG-63 cells

To further understand the estrogenic effects of DBT, ERE-dependent luciferase activity and mRNA expression of estrogen responsive genes were determined in

MG-63 cells upon treatment. As shown in Figure 5.14A, similarly to estradiol, DBT at both doses significantly increased ERE-dependent luciferase activity in MG-63 cell and DBT at 1.0mg/ml significantly increased the ERE luciferase by three fold which was far more potent than estradiol (p<0.05 vs control). The result suggested that the direct estrogenic effects of DBT in MG-63 cells might be ERE-dependent. DBT at 1.0 and 2.0mg/ml exerted comparable effects to estradiol on inducing mRNA expression of osteocalcin (OCN) (Figure 5.14B, p<0.001 vs control) and ALP (Figure 4.14C, p<0.05 vs control) in MG-63 cells. Osteoprogerin (OPG) is expressed on mesenchrymal stem cells and suppresses the osteoclastogenesis. DBT at both doses significantly up-regulated mRNA expression of OPG (Figure 5.14D, p<0.05 vs control) but down-regulated the mRNA expression of RANKL (Figure 5.14E, P<0.01 vs control) in MG-63 cell upon treatment. In addition, a 1.4 fold and 2.1 fold increase in OPG/RANKL ratio were detected in MG-63 cells upon DBT treatment at 1.0 and

2.0mg/ml (Figure 5.14F, p<0.01, p<0.01 vs control), respectively. These results suggested that DBT may protect bone cell by activating osteoblastic activity and suppressing osteoclastic activity via regulating ERE-dependent transcription.

157

A B

4 * 2.0

3 1.5 * *** 2 * *** 1.0

1 0.5

(relatibe to control) (relatibe ERE luciferase activity ERE luciferase 0 0.0 -7 -7

control E2 (10 M) 1.0 2.0 control E2(10 M) 1.0 2.0 OCN/GAPDH (relative to control) (relative OCN/GAPDH DBT (mg/ml) DBT (mg/ml)

C D

E F

Figure 5.14 Effects of DBT on ERE-dependent luciferase activity and mRNA expression of estrogen-responsive genes in MG-63 cells MG-63 cells were cultured and subjected to vehicle, estradiol (10-7M) and DBT (1.0 and 2.0mg/ml) for 48 hours in phenol red-free MEM containing 5% of charcoal-stripped fetal bovine serum (cs-FBS). Upon treatment, A. ERE-dependent luciferase activity was determined by Dual Luciferase Reported Assay System; B. mRNA expression of OCN; C. mRNA expression of ALP; D mRNA expression of OPG; E. mRNA expression of RANKL; F. OPG/RANKL were determined by Real-time PCR. Results were from two independent experiments and expressed as mean ± SEM (n=4). P*<0.05, **p<0.01, ***p<0.001 vs control.

158

Possible mechanisms involved in actions of DBT in MG-63 cells

To reveal the possible pathways mediating actions of DBT, MG-63 cells were co-treated with DBT and estrogen antagonist ICI182,780, MAPK inhibitor U0126 or

PI3K inhibitor. As shown in Figure 5.15A and B, ICI182,780, U0126 and LY294002 completely abolished the stimulatory effects of estradiol on both cell proliferation and

ALP activity in MG-63 cells to the same level as control group (p<0.01, p<0.001 vs

E2 treatment). The stimulatory effects of DBT on cell proliferation were completely blocked by ICI182,780 and partially blocked by co-treatment with U0126 to 80% of the effect of DBT, respectively (Figure 5.15A, p<0.001 vs DBT treatment) while

LY294002 exerted no blocking effects on actions of DBT on cell proliferation (Figure

5.15A, P<0.001 vs control). As shown in Figure 5.15B, ICI182,780, U0126 and

LY294002 completely blocked the effects of DBT (p<0.001 vs DBT treatment) and reduced the ALP activity to levels even lower than that of control (p<0.01, p<0.001 vs control). Our results demonstrated that ICI182,780 completely blocked the stimulatory effects of DBT on cell proliferation and ALP activity whileU0126 exerted partial blocking on the stimulatory effects of DBT on cell proliferation and cell differentiation, but LY294002 only completely blocked action of DBT on ALP activity rather than cell proliferation, suggesting the possible involvement of ER,

MAPK and PI3K pathways in actions of DBT in MG-63 cell.

159

no antagonist ICI pretreatment U0126 pretreatment 600 LY294002 pretreatment ***

400

*** ## *** 200 *** ### ###

### ###

cell proliferation cell (relative to control) to (relative 0 -7 C E2 (10 M) DBT (2mg/ml)

no antagonist ICI pretreatment U0126 pretreatment 250 LY294002 pretreatment *** 200

150 * ## ## ### ### 100 ** ### ** ### ** ***

50 *** *** ***

ALP activity (%) activity ALP (relative to control) to (relative 0 -7 C E2 (10 M) DBT (2mg/ml)

Figure 5.15 Blocking effects of ICI182,780, U0126 and LY294002 on actions of DBT in MG-63 cells MG-63 cells were cultured and subjected to vehicle, estradiol (10-7M) and DBT (2mg/ml) with or without ICI182,780 (10-6M), U0126 (10-6M) or LY294002 (10-6M) for 48 hours in phenol red-free MEM containing 5% of cs-FBS (cs-FBS). Upon treatment, A. Cell proliferation was measured by MTS assay; B. ALP activity was determined by ALP assay. Results were from two independent experiments and expressed as mean ± SEM (n=4). *p<0.05, **p<0.01, ***p<0.001 vs control.

160

5.3 Discussion

Danggui Buxue Tang (DBT), the most popular Chinese Medicine for postmenopausal women, has been reported to contain phytoestrogens by which the estrogenic- actions of DBT are exerted. Clinical studies demonstrated that DBT is effective in the activation of hematogenesis, anti-oxidation activity, prevention against osteoporosis

(Choi et al., 2011; Xie et al., 2012). It has also been shown to be effective in accelerating fracture healing of lower limb, helping functional recovery of hip joint in studies (Liu et al., 2014) and alleviating incidence and intensity of hot flashes in postmenopausal women (Haines et al., 2008). In addition, DBT has been shown to mimic estrogen in receptor phosphorylation in vitro (Zheng et al., 2012b). In particular, DBT significantly increased activities of ERE-dependent luciferase in

MG-63 cells in dose-dependent manner (Choi et al., 2011). The biological effects of

DBT in MG-63 cells could be abolished by co-treatment with a specific estrogen receptor antagonist, ICI182,780. Similarly, DBT was shown to induce phosphorylation of ER and extracellular signal-regulated kinase 1/2 (ERK1/2) in

MCF-7 cells (Gao et al., 2007b). These results suggest that DBT may contain phytoestrogen that exert its effects via ER. In my study, the long-term estrogenic effects of DBT decoction in four estrogen-sensitive tissue including bone, brain, breast and uterus in the mature OVX rats and the direct estrogenic effects of DBT in

161 four estrogen-sensitive cell lines were determined. The results clearly showed that

DBT exerted estrogenic effects in both in vivo and in vitro models and its effects have been proved to be tissue selective (Table 5.5).

Long term oral administration with DBT at 3g/kg.day, a dosage calculated from the dose used in human (Li et al., 2006), significantly prevented OVX rats from disruptions in bone and central nervous system induced by estrogen-deficiency.

Among the three sites of bone assessed in my study, the decreased bone mineral density, the deteriorated bone microarchitecture and bone properties were all dramatically attenuated in OVX rats by treatment with DBT, and the effects were similar to these of rats treated with 17ß-estradiol. The effect of estradiol against estrogen deficiency-induced osteoporosis has been reported in both human (Black et al., 2016; Ju et al., 2015) and animal studies (Jing et al., 2014; Niu et al., 2012; Song et al., 2015) as well as our own studies in both OVX rat and mice models (Mok et al.,

2010; Wong et al., 2013). It is exciting to note that the effects of DBT at lumbar spine appeared to be more potent than the effects of estradiol in OVX rats. Besides, the

OVX-induced increase in bone turnover biomarker, osteocalcin and urinary deoxypyridinoline, were also suppressed in DBT-treated OVX rats. These results indicated that DBT protected bone from estrogen deficiency-induced osteoporosis in rats possibly via suppression of bone turnover in a way similar to estradiol.

162

Epidemiological studies have reported that postmenopausal women are at high risk of

Parkinson’s disease (Ascherio et al., 2003; Ragonese et al., 2004) while the supplement with exogenous estrogen significantly reduces the risk of PD in postmenopausal women (Currie et al., 2004), suggesting a potential regulation of estrogen in PD. Moreover, observations from epidemiological studies suggest that moderate exposure to exogenous estrogen reduces the risk of memory impairment, irreversible damage of neurons (Erickson et al., 2007) and improves cognitive performance of postmenopausal women (Sundermann et al., 2006). Furthermore, according to results of the retrospective studies, the earlier the women receive HRT, the later dementia happens, indicating that HRT may delay the onset of AD (Bagger et al., 2005). These results suggest that estrogen protects women from damages in the central nervous system associated with estrogen deficiency.

Striatum, a main part of substantia nigra-striatum dopaminergic system, was collected in the present study to evaluate the estrogenic effects of DBT in central nervous system in OVX rats. mRNA expression level of tyrosine hydroxylase (TH) and dopamine transporter (DAT) in striatum, two specific factors regulated by estrogen in the biosynthesis and release of dopamine, were determined to evaluate the neuroprotective effects of DBT in OVX rats. mRNA expression of TH dramatically decreased while DAT mRNA expression significantly decreased in striatum of OVX

163 rats, indicating the estrogen deficiency induced damages in striatum that might lead to the decrease in dopamine biosynthesis and enhancement of dopamine release.

Consistently with the results presented in chapter 4, treatment with 17ß-estradiol significantly restored mRNA expression of TH and DAT in striatum of OVX rats to levels comparable to these in sham rats. Interestingly, the potency of DBT for restoring TH and DAT mRNA expression in striatum of OVX rats was shown to be much higher than estradiol. These results demonstrated that DBT exerted estrogen-like action in protecting central nervous system of OVX rats from damages caused by estrogen deficiency.

The protective effects of DBT in bone and central nervous system were also confirmed in human osteoblast MG-63 and human neuroblastoma SH-SY5Y cells in the present study, respectively. The stimulatory effects of DBT on increasing cell viability and ERE-dependent luciferase activities in SH-SY5Y cells suggested that

DBT exerted direct ERE-dependent estrogenic effects in neurons. Similarly, DBT significantly stimulated cell proliferation, ALP activities and the ERE-luciferase activities in MG-63 cells. Such effects in MG-63 cells were consistent with previously published results by Choi (Choi et al., 2011). Moreover, the expression level of genes for bone formation including osteocalcin (OCN), alkaline phosphoatase (ALP) and osteoprotegerin OPG were significantly up-regulated while the bone resorption genes

164 including OPG and RANKL were markedly down-regulated in MG-63 cells. These results suggest that DBT directly regulates bone remodeling via enhancing bone formation and suppressing bone resorption.

Uterus and breast are main target tissues of estrogen and also the target tissue where side effects of HRT or estrogen are most frequently reported (Hinds et al., 2010). It is of special importance to investigate the potential side effects of DBT in uterus and breast since it might contain phytoestrogens and exerted estrogen-like effects at least in bone and brain. Our results showed that estradiol rather than DBT treatment significantly reversed the estrogen deficiency-induced decrease in uterine weight of

OVX rats. Similarly to estradiol, DBT also showed the trend to restore the estrogen deficiency-induced decrease in mRNA expression of the estrogen responsive genes in uterus including complement component 3 (C3), progesterone receptor (PR) and estrogen receptor (ER). However, no statistical significance was detected between rats treated with DBT and OVX rats. In the endometrium, hyperplasia was observed in rats treated with estradiol not DBT. These results indicated that DBT acted differently to estradiol in uterus. As for the effect in breast, no significant change in both number and morphology of mammary gland was observed in OVX rats treated with DBT when compared with the OVX rats. The results from the in vitro experiment clearly showed that DBT directly induced estrogenic responses in human breast cancer

165

MCF-7 cells and human endometrial cancer Ishikawa cells, two estrogen receptor (ER) positive cell lines. DBT significantly stimulated cell proliferation and ERE-dependent reporter activities in MCF-7 cells as well as ALP activities in Ishikawa cell at wide concentrations. Additionally, DBT significantly up-regulated the mRNA expression of estrogen responsive genes including ER and pS2 in MCF-7 cell and ALP in Ishikawa cells in estrogen like manner, suggesting that DBT. These results clearly indicated that

DBT exerted direct estrogenic effects in ER positive cells in vitro possibly via regulation of ERE-dependent transcription.

The stimulatory effects of DBT in Ishikawa cells seemed to be inconsistent with its actions in uterus of OVX rats. The discrepancies might be caused by the differences between the concentration of DBT used in these studies, the bioavailabilities of the phytoestrogens presence in DBT, as well as the fact that DBT used in the in vitro studies have not been metabolized. The concentrations of DBT used in the in vitro study were much higher than the circulating levels of DBT achieved by treatment of

OVX rats with 3g/kg.day of DBT. In addition, some of the bioactive phytoestrogen in

DBT might not be absorbed efficiently or be metabolized into other metabolites, thereby reducing its direct estrogenic responses in the uterus of OVX rats.

Based on our results, we are confident to summarize that DBT selective protected bone and brain from estrogen deficiency-induced damages in OVX rats without

166 causing undesirable side effects in uterus and breast. Moreover, DBT directly exerted estrogenic effects in estrogen sensitive cell lines and the estrogenic effects may be mediated by regulations of ERE-dependent transcription. However, how DBT exerts the estrogenic effects in tissue-selective manner is still unclear.

Estrogen deficiency has been believed to be the main cause of menopause and result in series of symptoms. To address how DBT selectively acted in estrogen sensitive tissues, the circulating level of estradiol as well as the two gonadotropins, follicle-stimulating hormone (FSH) and luteizing hormone (LH), were measured.

Estrogens negatively regulate the secretion of FSH and LH from pituitary gland via the feedback control system of the hypothalamic-pituitary-ovarian axis (Marieb et al.,

2007). Sharp decreased estradiol level accompanied by the dramatic increased FSH and LH level in serum are typical hormonal pattern found in post-menopausal women

(Dal Maso et al., 2003; Navarro et al., 2012a). The results of the present study were in agreement with previous findings (Ma et al., 2013; Xu et al., 2014) that the serum level of estradiol dramatically declined in rats after OVX which was accompanied by the increase in FSH and LH levels. Exposure to 17ß-estradiol (2mg/kg.day) significantly suppressed the increase in FSH and LH in OVX rats via the feedback regulation. According to our results, the serum level of estradiol significantly increased while estrogen deficiency-induced increases in FSH and LH were inhibited

167 in OVX rats in response to treatment with DBT. Interestingly, the magnitude of suppression by DBT on FSH and LH in OVX rats was comparable to that by

17ß-estradiol. This finding was in agreement with the results of an earlier study using mature rats with one of the ovary being removed (Tan et al., 2010) which suggested that DBT may contain phytoestrogen and regulate the functions of hypothalamus-pituitary-ovary axis. The effect of DBT on increasing circulating level of estradiol and suppressing serum level of gonadotropins might be one possible reason for its protective actions in bone and brain in OVX rats. The suppressing effect of DBT on FSH level should be of particular importance for prevention and treatment of postmenopausal osteoporosis since recent studies indicate that it is the dramatically increased FSH level rather than the decreased estradiol level that directly causes bone loss (Sun et al., 2006).

Soy flavonoids have been reported to inhibit the effects of estrogen in breast of postmenopausal women and facilitate clearance of estradiol from uterus and breast, resulting in estrogen antagonistic effects (Wood et al., 2007; Wood et al., 2006).

Based on the previous findings and our own results, it was inferred that DBT might act in similar way as flavonoids as it significantly increased circulating estradiol level but cause no estrogenic effects in uterus and mild hyperplasia in breast in OVX rats.

Aromatase is the key enzyme for the bioconversion from androgen to estrogen in

168 several extragonadal tissues including adipose tissues, placenta and adrenal gland, which become the main sources of estrogen in postmenopausal women (Simpson,

2002). In my study, the mRNA expression level of aromatase in adipose tissue in

OVX rats was significantly up-regulated by DBT by 54%, suggesting the adipose tissues might become one main source of estradiol, which might explain the increase in estradiol level in OVX rats treated with DBT.

Furthermore, the possible mechanisms mediating the actions of DBT were also investigated in MG-63 cells. Co-treatment of MG-63 cells with specific ER antagonist

ICI182,780, MAPK inhibitor U0126 and PI3K inhibitor LY294002 completely or partially blocked the stimulatory effects of DBT on cell proliferation and (or) ALP activities. These results further suggested that ER, MAPK and PI3K pathways might be involved in mediating the stimulatory actions of DBT in MG-63 cells. These results suggested the actions of DBT in MG-63 cells were mediated not only by the genomic pathway but also by the rapid signaling pathways. However, the roles of rapid signaling in mediating the actions of DBT require further investigation.

Taken together, DBT acts in estrogen-like manner on majority parameters analyzed in bone and brain in OVX rats while acts in different manner to that of estrogen in uterus and breast, suggesting that the estrogenic effects of DBT in estrogen sensitive tissues are selective. Possible mechanisms for the tissue-selective effects of DBT in vivo

169 might be that DBT significantly promoted circulating estradiol while facilitate the clearance of estradiol and catalyze estradiol into more benign metabolites in local tissues like breast and uterus at the same time. Our study in vitro study confirmed the tissue selective effects of DBT. Moreover, results from in vitro demonstrated that actions of DBT in human osteoblastic MG-63 cells might be mediated by ER, MAPK and PI3K pathways. Based on results of present study, DBT appears to be a new safe and practical alternative approach to HRT in management of post-menopausal symptoms.

170

Table 5.5 Summary of the tissue-selective estrogenic effects of DBT in four estrogen sensitive tissues in comparison to 17ß-estradiol

17ß-estradiol DBT Uterus Uterus Weight Increased No change Gene expression Increased without significance No change Morphology Increased without significance No change Ishikawa ALP activity Stimulatory effect Stimulatory effect at 0.05 to 2.0mg/ml cell ALP ER gene expression ALP: increased (+++) ALP: increased (+) ER: increased (++) ER: increased without significance ERE luciferase activity Increased (+) Increased (+) Breast Breast Morphology Hyperplasia Mild hyperplasia MCF-7 cell Cell proliferation Stimulatory effect Stimulatory effect at 0.05 to 2.0mg/ml Estrogen responsive genes IGF-IR, pS2, ER: increased (+); ER pS2: increased (+)IGF-IR: no change ERE luciferase activity Increased (+) Increased (+) Bone Bone BMD Increased (+++) Increased (++) Bone property Improved (+++) Improved (++) OCN, DPD OCN decreased (-);DPD decreased (---) OCN decreased (-); DPD decreased (--) MG-63 cell Cell proliferation Stimulatory effect (++) Stimulatory effect (++) ALP activity Stimulatory effect (+) Stimulatory effect at 1.0 and 2.0mg/ml (++) Estrogen responsive genes OCN, ALP, OPG, OPG/RANKL: OCN, ALP, OPG, OPG/RANKL: increased increased (+);RANKL: decreased (-) (+);RANKL: decreased (-) ERE luciferase activity Increased (+) Increased (+) CNS Striatum Estrogen responsive genes TH increased (++);DAT decreased (++) TH increased (++); DAT decreased (++) SH-SY5Y Cell proliferation Stimulatory effect (+) Stimulatory effect (+++) cell Estrogen responsive genes ER: increased (+) ER: increased (+++) ERE luciferase activity Increased (+) Increased (+) Increase (+++) > (++) > (+); Decrease (---) > (--) > (-)

171

Chapter 6

Characterization of the potential interactions

between DBT and SERMs (Tamoxifen and

Raloxifene) in mature ovariectomized (OVX)

rats and estrogen-sensitive cell lines

172

6.1 Introduction

Selective estrogen receptor modulators (SERMs) are compounds with both estrogenic and anti-estrogenic activities at different sites of the body that are clinically used for treatment for menopausal symptoms (Dahlman-Wright et al., 2006). They are ER ligands that are full or partial agonist in some tissue, like bone, but antagonist in other tissue such as breast, which makes it possible for SERMs to exert actions like estrogen in target tissues with lower side effects than estrogen (Komm et al., 2014).

SERMs represent a group of molecules with varying levels of estrogenic agonist and antagonist in different tissues. According to their chemical structure, SERMs can be classified as trihenylethylene, benzothiophene, or benzopyran compounds (Dutertre et al., 2000). Tamoxifen, the first generation of SERMs, belongs to the trihenylethylene

SERMs and has been clinically used for treatment of breast cancer and prevention against osteoporosis (Gambacciani, 2013). Raloxifene is the second generation of

SERMs that is prescribed for osteoporosis and breast cancer for postmenopausal women (Abdelhamid et al., 2011). However, treatment-related adverse symptoms, such as fatigue, nausea, vomiting and other menopausal symptoms, make it impossible for the patients to tolerate treatment and lead them to seek for additional or alternative help (Shapiro et al., 2001).

Many patients use complementary and alternative medicine for either treatment of

173 their menopausal symptoms or alleviating treatment-related adverse symptoms. A previous study in Shanghai showed that 78.6% of breast cancer patient concurrently used TCM after diagnosis and during the treatment of breast cancer (Chen et al.,

2008a). This rate in Taiwan has been found to be 76.8% during the 10-year visiting period (Lai et al., 2012). These TCMs have been proved to contain phytoestrogens.

With the increasing popularity of Chinese Medicine in postmenopausal women who are concurrently prescribed with the western drugs, concerns have been raised about the potential problems of drug-herb interactions since they exert estrogenic effects via the same estrogen receptors. As many of the actions of phytoestrogen are mediated by

ERs, it is important to characterize if herbal medicine containing phytoestrogens will interact with prescribed drugs that target ER, such as tamoxifen and raloxifene to either increase or decrease the pharmacological or toxicological effects of either compounds.

Indeed, potential herb-drug interaction has posed significant challenges to the medical communities due to the increasing interest of using herbal medicines among patients prescribed with conventional drugs. For example, orally feeding with Biochanin A, an isoflavone, reduced bioavailability of tamoxifen and its metabolite in female rats

(Singh et al., 2012). Supplement with either genistein or 8-prenylnaringenin significantly neutralized the effects of tamoxifen in MCF-7 cells, which was

174 suggested to be better avoided during breast cancer treatment (van Duursen et al.,

2013). The in vitro observation was further confirmed in a study performed in athymic nude mice (Du et al., 2012b). On the contrary, a clinical study conducted in Taiwan demonstrated that consumption of Chinese herbal products containing coumestrol, genistein, or daidzein was negatively correlated with the incidence of endometrial cancer risk among the tamoxifen-treated female breast cancer survivors (Hu et al.,

2015). Thus, there is a strong demand to characterize the potential herb-drug interactions that might occur.

Based on our results presented in chapter 5, DBT, a popular TCM formula prescribed for women to improve their health, has been demonstrated to selectively exert estrogenic effects in estrogen sensitive tissues as it protected bone and brain from estrogen deficiency-induced damages without causing side effects in uterus or breast.

In the present study, DBT was chosen as one representative to address the potential interactions between TCMs and tamoxifen or raloxifene. Two preclinical models, the mature ovariectomized rats and four estrogen-sensitive cell lines, were employed to determine whether DBT interacts with SERMs in bone, brain, breast and uterus. The experimental design of the animal study was described in Chapter 3.

175

6.2 Results

6.2.1 Characterization of interactive effects of DBT and SERMs in OVX rats

6.2.1.1 Interactive effects between DBT and tamoxifen, raloxifene on body weight gain in OVX rats

Tamoxifen and raloxifene are ER compounds and they behave as estrogen antagonist in breast tissue while mimic effects of estrogen in other tissues like uterus and bone.

Treatment with 17ß-estradiol for 12 weeks significantly suppressed the body weight gain in OVX rats (Figure 6.1, p<0.001 vs OVX rats). Both tamoxifen and raloxifene showed the trend to suppress the increase in body weight in OVX rats while the changes did not reach statistical significance. The results were consistent with the findings reported by others (Lopez et al., 2006; Meli et al., 2004) and demonstrated that SERMs act as estrogen agonist on food intake and energy expenditure in OVX rats. Co-treatment of OVX rats with DBT and tamoxifen reduced the body weight gain to even lower level (Figure 6.1, p<0.05 vs OVX rats) while combination of DBT and raloxifene did not suppress the increase in body weight in OVX rats. However, no significant differences were determined in body weight gain between tamoxifen alone and DBT plus tamoxifen or raloxifene and DBT plus raloxifene. The results showed that DBT did not alter the effects of either tamoxifen or raloxifene on body weight gain in OVX rats, indicating no interactive effects between DBT and either tamoxifen or raloxifene on body weight gain in OVX rats.

176

5 ^

-5

-10 ^^^ E2 Body weight gain (% gain weight Body the of initial) OVX Ralo DBT Sham Tamo

DBT+Ralo DBT+Tamo

Figure 6.1 Effects of DBT, tamoxifen, raloxifene and their combinations on body weight gain of OVX rats OVX rats were orally administrated with vehicle, 17ß-estradiol (2mg/kg.day), tamoxifen (1mg/kg.day), raloxifene (3mg/kg.day), DBT (3g/kg.day) as well as combination of DBT and tamoxifen or raloxifene for 12 consecutive weeks. Body weight of rats was measured every two weeks during the whole experiment. Percentages of the changes in body weight from baseline to the end of treatment over their baseline body weight were regarded as the body weight gain of rats. Sham: sham operated + vehicle treatment; OVX: OVX operated + vehicle treatment; E2: OVX operated + 17ß-estradiol treatment; Tamo: OVX operated + tamoxifen treatment; Ralo: OVX operated + raloxifene treatment; DBT: OVX operated + DBT treatment; DBT+Tamo: OVX operated + DBT treatment + tamoxifen treatment; DBT+Ralo: OVX operated + DBT treatment + raloxifene treatment. Data was expressed as mean ± SEM. ***p<0.001 vs sham; ^^^p<0.001 vs OVX. n=8 or 9.

177

6.2.1.2 Interactive effects between DBT and tamoxifen, raloxifene on serum reproductive hormones in OVX rats

In the present study, circulating estradiol level dramatically reduced after removal of bilateral ovaries in OVX rats and such change was accompanied by the increase in serum levels of FSH and LH (Figure 6.2A, B and C, p<0.01, p<0.001 vs OVX).

Supplementation with exogenous estradiol reversed these changes in reproductive hormones induced by ovariectomy (Figure 6.2 A, B and C, P<0.001). As shown in

Figure 5.2A, B and C, tamoxifen showed the trend to increase circulating estradiol level and significantly suppressed the increased FSH (p<0.001 vs OVX rats) and LH level (p<0.05 vs OVX rats) in OVX rats. Similarly, raloxifene significantly increased estradiol level (Figure 6.2A, p<0.01 vs OVX rats) and suppressed the FSH (Figure

6.2B, p<0.001 vs OVX rats) while attenuated the increase in LH level (Figure 6.2C) without statistical significance. The results indicated that SERMs could alter the regulation of hypothalamus-pituitary-gonadal axis in OVX rats. As mentioned in chapter 4, DBT significantly increased circulating level of estradiol via up-regulating the biosynthesis of estradiol in adipose tissues in OVX rats via the enzymatic activities of aromatase and reduced FSH and LH level via negative feedback.

Compared with the treatment with tamoxifen or raloxifene alone, the effects of co-treatment with DBT appeared to be just combined effect rather than interactive effect as no statistical difference has been found.

178

A B

400 20 350 ^^^ *** 300 15 250 200 10 100 ^^^ ^^^ ^^^ ^^^ 80 ^^^ ^^ ^^ ^^^ 60 ^^^ ^^^ 5 40 20 **

0 FSH (ng/ml) of level serum 0 serum level of estradiol (pg/ml) estradiol of level serum E2 E2 OVX Ralo DBT OVX Ralo DBT sham Tamo sham Tamo

DBT+Ralo DBT+TamoDBT+Ralo DBT+Tamo C

6000 ***

4000 ^ ^ ^^^ ^^^

2000

serum level of LH (pg/ml) of level serum 0

E2 OVX Ralo DBT sham Tamo

DBT+Ralo DBT+Tamo

Figure 6.2 Effects of DBT, tamoxifen, raloxifene and their combinations on serum reproductive hormones in OVX rats Blood was obtained from the Sham and OVX rats treated with vehicle, 17ß-estradiol (2mg/kg.day), tamoxifen (1mg/kg.day), raloxifene (3mg/kg.day), DBT (3g/kg.day) and combination of DBT with tamoxifen or raloxifene for 12 weeks. Serum level of estradiol, FSH and LH were measured by commercial kits by following manufacturers’ instruction. A. Serum level of estradiol; B. Serum level of FSH; C. Serum level of LH. Data was expressed as mean ± SEM. **p<0.01, ***p<0.001 vs sham; ^^^p<0.001 vs OVX. n=8 or 9.

179

6.2.1.3 Interactive effects between DBT and tamoxifen, raloxifene on uterus in

OVX rats

The biggest drawback in the treatment of breast cancer with tamoxifen is its agonistic effect in the uterus while such agonistic effect in uterus has been reduced when treatment with raloxifene (Zujewski, 2002). However, both of them still exert strong estrogenic effect in uterus. In the present study, both tamoxifen and raloxifene significantly increased the weight of uterus in OVX rats by more than 80% (Figure

6.3A, p<0.001, p<0.001 vs OVX rats). These results were in agreement with previously published studies (Seidlova-Wuttke et al., 2009; Stygar et al., 2003), confirming the estrogen agonistic effects of SERMs in uterus. As mentioned in chapter 5, treatment with DBT did not alter the weight of uterus in OVX rats while co-treatment with DBT and SERMs significantly increased the weight of uterus in

OVX rats (Figure 6.3A, p<0.05, p<0.05 vs OVX rats). However, there were no statistical differences in the uterus weight between rats treated with SERMs alone and those co-treated with DBT and SERMs (Figure 6.3A).

The mRNA expression of three estrogen-responsive genes, complement component 3

(C3), progesterone receptor (PR) and estrogen receptor (ER) in rat uterus in response to treatment with SERMs or DBT alone as well as their combinations were determined by real-time PCR to further evaluate the potential interactive effects between DBT and SERMs on uterus in OVX rats. Results of chapter 5 demonstrated

180 that mRNA expression of C3, ER and PR decreased in rat uterus by more than 50% in response to OVX. In particular, the mRNA expression of C3 reduced to be extremely low after OVX. Both DBT and estradiol appeared to up-regulate the mRNA expression of C3, ER and PR in OVX rats but the changes were not statistically significant. As shown in Figure 6.3B-6.3D, treatment with tamoxifen or raloxifene and their combination with DBT also showed the similar trend to restore estrogen deficiency-induced changes in the mRNA expression of C3, ER and PR in uterus of

OVX rats, confirming that SERMs act as estrogen agonists in uterus. However, these effects were not statistically significant. In terms of restoring estrogen deficiency-induced changes in mRNA expression of estrogen-responsive genes in uterus, DBT neither enhance nor weaken the effects of SERMs, indicating no interactive effects between DBT.

Furthermore, the morphology of endometrium in OVX rats was visualized by H&E staining to evaluate the potential interactive effects between DBT and SERMs. As shown in Figure 5.3E, compared to sham rats, OVX rats experienced atrophy in endometrium in response to OVX. 17ß-estradiol dramatically induced hyperplasia in the endometrium and treatment of OVX rats with tamoxifen and raloxifene, but not

DBT, exert estrogen-like but weaker hyperplasia effects in the endometrium of OVX rats. Moreover, DBT did not alter the effects of either tamoxifen or raloxifene in

181 inducing hyperplasia in endometrium of OVX rats. These results suggested that DBT did not alter the estrogen agonistic effects of tamoxifen and raloxifene on uterus in

OVX rats.

182

A B

15 2.5

2.0 ^^^ 10 1.5

1.0 5 ^^^ ^^^ ^ ^ ER/GAPDH 0.5 ***

0.0 0

Uterus index (mg/g body weight) (mg/g body Uterus index E2 E2 OVX Ralo DBT OVX Ralo DBT sham Tamo sham Tamo

DBT+Ralo DBT+Ralo DBT+Tamo DBT+Tamo

C D

5 PR/GAPDH

10 C3/GAPDH

0 E2 OVX Ralo DBT sham Tamo E2 OVX Ralo DBT sham Tamo DBT+TamoDBT+Ralo

DBT+TamoDBT+Ralo E

Figure 6.3 Effects of DBT, tamoxifen, raloxifene and their combinations on

183 uterus index, endometrial morphology and mRNA expression of estrogen-responsive genes in uterus of OVX rats Uterus was freshly collected and weighed from Sham and OVX rats treated with vehicle, 17ß-estradiol (2mg/kg.day), tamoxifen (1mg/kg.day), raloxifene (3mg/kg.day), DBT (3g/kg.day) and combination of DBT with tamoxifen or raloxifene for 12 weeks. A. Uterus index: The weight (Giunta et al.) of uterus was divided by the body weight (g) as uterus index to evaluate the estrogenic effects of DBT in uterus. B. mRNA expression of component complement 3(C3); C. mRNA expression of progesterone receptor (PR); D. mRNA expression of estrogen receptor (ER) mRNA expression of C3, PR and ER were determined by Real time-PCR; E. Morphology of endometrium was visualized by H&E staining. Data was expressed as mean ± SEM. ***p<0.001 vs sham; ^p<0.05, ^^^p<0.001 vs OVX. n=8 or 9.

184

6.2.1.4 Interactive effects between DBT, tamoxifen, raloxifene on breast of OVX rats

As shown in chapter 4, atrophy was observed in the mammary gland of OVX rats which was significantly attenuated by treatment with 17ß-estradiol and slightly attenuated by DBT, respectively. Tamoxifen and raloxifene are developed for treatment or prevention of breast cancer because of their estrogen antagonist activity.

In the present study, the number of both mammary gland and breast milk duct decreased in OVX rats upon treatment with tamoxifen and raloxifene (Figure 6.4).

Co-treatment of OVX rats with DBT did not alter the effects of either tamoxifen or raloxifene in breast (Figure 6.4). The results suggested that DBT did not affect the estrogen antagonistic effects of either tamoxifen or raloxifene in the breast tissues of

OVX rats, indicating DBT at the current dose might be safe even for breast cancer patients prescribed with SERMs.

185

Figure 6.4 Effects of DBT, tamoxifen, raloxifene and their combinations on histology of breast (400X) in OVX rats The second breast together with its surrounding skin was collected from the Sham and OVX rats treated with vehicle, 17ß-estradiol (2mg/kg.day), tamoxifen (1mg/kg.day), raloxifene (3mg/kg.day), DBT (3g/kg.day) and combination of DBT with tamoxifen or raloxifene for 12 weeks. H&E staining was performed to visualize the morphology of mammary gland of rats using Methods described in Chapter 3. Sham: sham operated + vehicle treatment; OVX: OVX operated + vehicle treatment; E2: OVX operated + 17ß-estradiol treatment; Tamo: OVX operated + tamoxifen treatment; Ralo: OVX operated + raloxifene treatment; DBT: OVX operated + DBT treatment; DBT+Tamo: OVX operated + DBT treatment + tamoxifen treatment; DBT+Ralo: OVX operated + DBT treatment + raloxifene treatment. n=8 or 9.

186

6.2.1.5 Interactive effects between DBT and tamoxifen, raloxifene on brain tissues of OVX rats

Studies so far have demonstrated that SERMs decrease neuronal damages caused by different forms of neuronal injuries in animal models of several neural dysfunctions including Parkinson’s disease and Alzheimer’s disease (Bourque et al., 2012;

Morissette et al., 2008; Tian et al., 2009) (Breuer et al., 2000; Du et al., 2004). In the present study, treatment of OVX rats with both tamoxifen and raloxifene alone significantly up-regulated the mRNA expression of TH in striatum (Figure 5.5A, p<0.01, p<0.001 vs OVX). Moreover, such protective effect of tamoxifen and raloxifene was much stronger than that of estradiol (Figure 6.5A). Especially the protective potency in raloxifene treated rats appeared to be two fold of that in estradiol treated rats (Figure 6.5A). The down-regulation of DAT mRNA expression in striatum exerted by tamoxifen and raloxifene alone was comparable to that of estradiol (Figure 6.5B, P<0.001 vs OVX). Again, the neuroprotective potency of raloxifene on decreasing DAT mRNA expression in striatum was higher than it of estradiol. These results indicated that SERMs may exert neuroprotective effects in central nervous system of OVX rats in estrogen-like manner.

As presented in chapter 5, DBT also exerted comparable neuroprotective effects to estradiol against estrogen deficiency-induced damage in OVX rats as it significantly up-regulated TH mRNA expression and suppressed mRNA expression of DAT in

187 striatum (Figure 5.7A and B). Co-treatment of OVX rats with DBT neither enhance nor weaken the neuroprotective effects of SERMs in striatum (Figure 6.5A and B), indicating that DBT and SERMs did not interact with each other in central nervous system of OVX rats.

188

A B

4 1.5 ^^^ *** 3 1.0 ^^^ ^^^ ^^ 2 * ^^ ^^^ 0.5 ^^^ ^^^

TH/GAPDH ^^^ DAT/GAPDH ^^^ 1 ^^^

0.0 0

E2 E2 OVX Ralo DBT OVX Ralo DBT sham Tamo sham Tamo

DBT+Ralo DBT+Ralo DBT+Tamo DBT+Tamo Figure 6.5 Effects of DBT, tamoxifen, raloxifene and their combinations on mRNA expression of estrogen-responsive genes in striatum of OVX rats Striatum was freshly collected from the Sham and OVX rats treated with vehicle, 17ß-estradiol (2mg/kg.day), tamoxifen (1mg/kg.day), raloxifene (3mg/kg.day), DBT (3g/kg.day) and combination of DBT with tamoxifen or raloxifene for 12 weeks. A. mRNA expression of TH; B. mRNA expression of DAT were determined by Real time-PCR. Data was expressed as mean ± SEM. ***p<0.001 vs sham; ^p<0.05, ^^p<0.01, ^^^p<0.001 vs OVX. n=8 or 9.

189

6.2.1.6 Interactive effects between DBT and tamoxifen, raloxifene on bone of

OVX rats

BMD and bone properties

Raloxifene is one of the SERMs approved by FDA for prevention and treatment of osteoporosis and significantly reduces the risk of bone fracture in postmenopausal women as an anti-resorptive agent (Ensrud et al., 2006; Hegde et al., 2016).

Tamoxifen has been shown to have effects on bone by inhibiting bone resorption in postmenopausal women (Powles et al., 1996). In the present study, both raloxifene and tamoxifen significantly increased bone mineral density (BMD) (Figure 6.6A, B and C, p<0.001 vs OVX) and dramatically improved the trabecular bone which could be obviously observed from the microarchitecture (Figure 6.6 D, E and F) at distal femur, proximal tibia and lumbar spine in OVX rats. Moreover, both tamoxifen and raloxifene significantly improved the trabecular bone properties at distal femur, proximal tibia and lumbar spine as the BS, BV/TV, Tb.N and Tb.Th significantly increased while Tb.Sp significantly decreased upon treatment (Table 6.1, 6.2 and 6.3, p<0.05, p<0.01, p<0.001 vs OVX). Such findings were in agreement with the clinical results (Hadji et al., 2009; Haghverdi et al., 2014).

As shown in chapter 5, DBT protected bone from estrogen deficiency-induced osteoporosis in OVX rats as it significantly increased the BMD (Figure 5.8A, B and

C), improved microarchitecture of trabecular bone (Figure 5.8D, E and F, Table 5.2,

190

5.3 and 5.4) at all the three sites tested. Co-treatment of tamoxifen and raloxifene with

DBT also significantly increased BMD (Figure 6.6 A, B and C, p<0.001 vs OVX) and markedly improved trabecular bone microarchitecture (Figure 6.6 D, E and F) as well as trabecular bone properties (Table 6.1, 6.2 and 6.3, p<0.05, p<0.01, p<0.001 vs

OVX) at the three sites tested in the present study. However, no significant difference between tamoxifen alone and DBT plus tamoxifen or raloxifene alone and DBT plus raloxifene has been found, indicating DBT did not alter the effects of SERMs on

BMD, bone properties and bone microarchitecture in OVX rats.

191

A B

500 500 400 400 ^^^ ^^^ ^^^ 300 ^^^ ^^^ 300 ^^^ ^^^ ^^^ ^^^ ^^^ ^^ ^^ 200 200 *** 100 ***

100 HA/ccm) (mg BMD BMD (mg HA/ccm) (mg BMD

0 0

2 E 2 E ralo DBT OVX ralo DBT sham OVX tamo sham tamo DBT+ralo DBT+ralo DBT+tamo DBT+tamo

500

400 ^^^ ^^^ ^^^ ^^^ ^^^ ^^^ 300 ***

200

100 BMD (mg HA/ccm) (mg BMD

E 2 ralo DBT sham OVX tamo DBT+ralo DBT+tamo

192

Figure 6.6 Effects of DBT, tamoxifen, raloxifene and their combinations on micro-architecture and bone mineral density at distal femur, proximal tibia and lumbar spine as well as the bone turnover biomarkers in OVX rats Upon treatment, whole left leg and spine of the Sham and OVX rats treated with vehicle, 17ß-estradiol (2mg/kg.day), tamoxifen (1mg/kg.day), raloxifene (3mg/kg.day), DBT (3g/kg.day) and combination of DBT with tamoxifen or raloxifene for 12 weeks were collected for micro-CT scanning. Bone mineral density

193 and trabecular bone microarchitecture at distal femur, proximal tibia and lumbar spine were determined by Micro-CT. A. BMD of distal femur; B. BMD of proximal tibia; C. BMD of lumbar spine; D. Microarchitecture of distal femur; E. Microarchitecture of proximal tibia; F. Microarchitecture of lumbar spine. Sham: sham operated + vehicle treatment; OVX: OVX operated + vehicle treatment; E2: OVX operated + 17ß-estradiol treatment; Tamo: OVX operated + tamoxifen treatment; Ralo: OVX operated + raloxifene treatment; DBT: OVX operated + DBT treatment; DBT+Tamo: OVX operated + DBT treatment + tamoxifen treatment; DBT+Ralo: OVX operated + DBT treatment + raloxifene treatment. Data was expressed as mean ± SEM. n=8 or 9.

194

Table 6.1 Effects of DBT, tamoxifen, raloxifene and their combinations on trabecular bone properties at distal femur of OVX rats BS (mm2) BV/TV Tb.N (1/mm) Tb.TH (mm) Tb.Sp (mm) Sham 239.3±4.49 0.484±0.017 4.73±0.07 0.102±0.003 0.109±0.005

OVX 86.7±8.85*** 0.121±0.014*** 1.66±0.19*** 0.067±0.002*** 0.565±0.081***

E2 182.0±7.73^^^ 0.305±0.018^^^ 3.38±0.15^^^ 0.086±0.003^ 0.214±0.015^^^ Tamo 162.2±8.66^^^ 0.244±0.019^^^ 2.96±0.151^^^ 0.0800.003 0.268±0.019^^^ Ralo 144.9±7.81^^^ 0.232±0.013^^^ 2.92±0.140^^^ 0.0800.002 0.353±0.040^^^ DBT 141.1±13.25^^ 0.207±0.032^ 2.64±0.30^^^ 0.076±0.002 0.353±0.041^^^ DBT+Tamo 163.1±5.84^^^ 0.255±0.021^^^ 3.06±0.176^^^ 0.0820.002 0.244±0.022^^^ DBT+Ralo 166.7±7.22^^^ 0.265±0.010^^^ 3.13±0.128^^^ 0.0840.002^ 0.232±0.011^^^

Table 6.2 Effects of DBT, tamoxifen, raloxifene and their combinations on trabecular bone properties at proximal tibia of OVX rats BS (mm2) BV/TV Tb.N (1/mm) Tb.TH (mm) Tb.Sp (mm)

Sham 228.8±9.43 0..517±0.023 4.97±0.05 0.104±0.004 0.097±0.005

OVX 55.0±7.67** 0.076±0.010*** 1.23±0.16*** 0.062±0.001*** 0.856±0.144***

E2 155.1±10.26^^ 0.255±0.023^^^ 3.38±0.19^^^ 0.076±0.002^ 0.214±0.014^^^

Tamo 128.5±10.67^^ 0.207±0.021^^ 2.93±0.21^^^ 0.072±0.003 0.287±0.026^^^

Ralo 117.3±6.13^^ 0.199±0.015^ 2.87±0.14^^^ 0.070±0.004 0.286±0.017^^^

DBT 110.4±18.62 0.172±0.047 2.46±0.31^^^ 0.070±0.004 0.420±0.061^^^

DBT+Tamo 129.4±9.69^^ 0.214±0.025^^ 3.12±0.20^^^ 0.073±0.003 0.257±0.023^^^

DBT+Ralo 131.8±10.06 0.224±0.018^^ 2.93±0.16^^^ 0.077±0.003^ 0.274±0.021^^^

195

Table 6.3 Effects of DBT, tamoxifen, raloxifene and their combinations on trabecular bone properties at lumbar spine of OVX rats BS (mm2) BV/TV Tb.N (1/mm) Tb.Th (mm) Tb.Sp (mm) Sham 89.9±2.73 0.442±0.012 4.12±0.09 0.107±0.001 0.136±0.006 OVX 62.6±5.18** 0.209±0.016*** 2.62±0.17*** 0.079±0.001*** 0.313±0.026***

E2 90.4±3.33^^ 0.350±0.010^^^ 3.71±0.06^^^ 0.094±0.002^^ 0.176±0.006^^^

Tamo 89.7±5.93^ 0.366±0.014^^^ 3.65±0.11^^^ 0.100±0.002^^^ 0.176±0.008^^^ Ralo 93.0±4.72^ 0.375±0.011^^^ 3.67±0.09^^^ 0.102±0.002^^^ 0.171±0.007^^^ DBT 81.3±4.78^ 0.307±0.019^^ 3.40±0.09^^^ 0.090±0.004 0.205±0.010^^^

DBT+Tamo 90.8±5.64^^ 0.361±0.021^^^ 3.60±0.11^^^ 0.010±0.003^^^ 0.180±0.011^^^ DBT+Ralo 81.4±2.99^ 0.371±0.010^^^ 3.54±0.05^^^ 0.105±0.002^^^ 0.178±0.005^^^

Bone properties of trabecular bone at proximal tibia and distal femur as well as lumbar vertebra were also determined by Micro-CT. Data was expressed as mean ± SEM. *p<0.05, **p<0.01, ***p<0.001 vs sham; ^p<0.05, ^^p<0.01, ^^^p<0.001 vs OVX. n=8 or 9. BS: Bone surface; BV/TV: Bone volume/total volume; Tb.N: Trabecular bone number; Tb.Th: Trabecular bone thickness; Tb.Sp: Trabecular separation.

196

Biochemical markers for bone turnover

The effects of tamoxifen, raloxifene alone and their combinations with DBT on biochemical markers were also performed to evaluate the potential interactions between DBT and SERMs. As shown in Figure 6.7A and B, both serum osteocalcin and urinary DPD significantly increased in rats in response to OVX (p<0.01, p<0.001 vs sham) while 17ß-estradiol appeared to suppress the increase in osteocalcin and significantly reduced the urinary level of DPD (p<0.001 vs OVX). Both tamoxifen and raloxifene exerted comparable effects to estrogen as they suppressed the increase in serum osteocalcin without statistical significance (Figure 6.7A) while they exerted estrogen-like but weaker effects on urinary DPD as they significantly reduced urinary level of DPD in OVX rats (Figure 6.7B, p<0.001 vs OVX), indicating a suppressing effect of SERMs on bone turnover. As presented in chapter 5, DBT also significantly reduced the serum osteocalcin and urinary DPD in OVX rats (Figure 5.9A and B).

Similar inhibitory effects on serum osteocalcin and urinary DPD were also determined in OVX rats co-treated with DBT and tamoxifen (Figure 6.7 A and B,

P<0.001 vs OVX). Co-treatment of raloxifene with DBT also significantly decreased the urinary DPD level (Figure 6.7B, p<0.001 vs OVX) while the serum osteocalcin appeared to decrease without statistical significance (Figure 6.7A). However, no statistical difference between tamoxifen alone and DBT plus tamoxifen or raloxifene

197 alone and DBT plus raloxifene has been found, indicating DBT did not alter the inhibitory effects of SERMs on bone turnover in OVX rats.

Our results suggested that both DBT and SERMs protected bone from osteoporosis caused by estrogen deficiency in OVX rats but they did not interact with each other in bone.

198

A B

150 20 ** *** ^ ^^^ 15 100 ^^^ 10 ^^^ ^^^ ^^^ ^^^ 50

5 ^^^ osteocalcin (ng/ml) osteocalcin 0

0 (nmol/mmol) DPD/Creatinine

E2 E2 OVX Ralo DBT OVX Ralo DBT sham Tamo sham Tamo

DBT+Ralo DBT+Ralo DBT+Tamo DBT+Tamo

Figure 6.7 Effects of DBT, tamoxifen, raloxifene and their combinations on the bone turnover biomarkers in OVX rats Serum level of osteocalcin and urinary DPD were measured by commercial kits by following manufacturers’ instruction. A. Serum level of osteocalcin; B. Urinary deoxypyridinoline. Data was expressed as mean ± SEM. *p<0.05, **p<0.01, ***p<0.001 vs sham; ^p<0.05, ^^p<0.01, ^^^p<0.001 vs OVX. n=8 or 9.

199

6.2.2 Characterization of interactive effects between DBT and SERMs in estrogen sensitive cells

6.2.2.1 Interactive effects between DBT and tamoxifen, raloxifene in MCF-7 cells

Tamoxifen is clinically approved for treatment of breast cancer and raloxifene is approved for prevention of breast cancer since they both act as estrogen antagonist in breast (Ronghe et al., 2016). In the present study, both tamoxifen and raloxifene decreased MCF-7 cell proliferation at a wide range of concentrations as shown in figure 5.8 (p<0.05, p<0.01, p<0.001 vs control). The higher the concentrations of

SERMs, the more potent the inhibitory effects. Upon co-treatment for 48 hours, DBT at 0.1mg/ml did not affect the inhibitory effect of both tamoxifen and raloxifene while

DBT at 0.5mg/ml significantly reversed the decreasing effects of tamoxifen and raloxifene in MCF-7 cells as the cell proliferation significantly increased (Figure 6.8

A, P<0.01, P<0.001 vs tamoxifen alone; Figure 6.8B, p<0.01 vs ralxofene alone).

Moreover, DBT appeared to reverse the effects of tamoxifen at lower concentrations, i.e. 10-10 and 10-12M, to higher level than that of estradiol (Figure 6.8A). The results indicated the potential interaction between DBT and SERMs in breast tissues might be dosage dependent.

200

C E2 Tamo ^^ ^^^ ^^^ Tamo+DBT0.1 1.5 Tamo+DBT0.5 *** *** ** *** *** *** *** 1.0 ** *** **

0.5

0.0

Cell viablity (Relative to control) (Relative viablity Cell Tamo-12 Tamo-10 Tamo-8 Tamo-6

C E2 Ralo Ralo+DBT0.1 2.0 ^^ ^^ Ralo+DBT0.5 *** 1.5 *** *** *** *** *** * 1.0 * *** ** **

0.5

0.0

Cell viablity (Relative to control) (Relative viablity Cell Ralo-12 Ralo-10 Ralo-8 Ralo-6

Figure 6.8 Interactive effects of DBT and tamoxifen or raloxifene on cell proliferation in MCF-7 cells MCF-7 cells were cultured and subjected to vehicle, estradiol (10-8M), tamoxifen (10-12, 10-10, 10-8, 10-6M), raloxifene (10-12, 10-10, 10-8, 10-6M) alone and co-treatment of DBT (0.1, 0.5mg/ml) plus tamoxifen or raloxifene for 48 hours as described in Chapter 3. After treatment, cell proliferation was measured by MTS assay. A. Interactive effects between DBT and tamoxifen; B. Interactive effects between DBT and raloxifene. Results were from two independent experiments and expressed as mean ± SEM (n=4). *p<0.05, **p<0.01, ***p<0.001 vs control; ^^p<0.01, ^^^p<0.001 vs tamoxifen or raloxifene alone.

201

6.2.2.2 Interactive effects between DBT and tamoxifen, raloxifene in Ishikawa cells

As shown in figure 6.9, neither tamoxifen nor raloxifene at all concentrations showed any effects on ALP activity in Ishikawa cells. Upon co-treatment for 48 hours, DBT at

0.1mg/ml did not alter the effect of either tamoxifen or raloxifene in Ishikawa cells

(Figure 6.9A and B). However, DBT at 0.5mg/ml significantly enhanced effects of tamoxifen and raloxfiene on ALP activity in Ishikawa cells (Figure 6.9A, p<0.01, p<0.05 vs tamoxifen alone; Figure 6.9B, p<0.05, p<0.01 vs raloxifene alone). The results indicated that DBT might alter the effects of SERMs on ALP activities in

Ishikawa cells.

202

C E2 Tamo Tamo+DBT0.1 ^ ^^ ^^ Tamo+DBT0.5 *** 1.5 *** *** *** *** *** **

1.0

0.5

0.0

ALP activity (relative to control) (relative activity ALP Tamo-12 Tamo-10 Tamo-8 Tamo-6

C E2 Ralo Ralo+DBT0.1 Ralo+DBT0.5 ^ 1.5 *** ^^ *** *** *** * ** ** *** *** 1.0

0.5

0.0

ALP activity (relative to control) (relative activity ALP Ralo-12 Ralo-10 Ralo-8 Ralo-6

Figure 6.9 Interactive effects of DBT and tamoxifen or raloxifene on ALP activity in Ishikawa cells Ishikawa cells were cultured and subjected to vehicle, estradiol (10-8M), tamoxifen (10-12, 10-10, 10-8, 10-6M), raloxifene (10-12, 10-10, 10-8, 10-6M) alone and co-treatment of DBT (0.1, 0.5mg/ml) plus tamoxifen or raloxifene for 48 hours as described in Chapter 3. ALP activity was measured by ALP kit. A. Interactive effects between DBT and tamoxifen; B. Interactive effects between DBT and raloxifene. Results were from two independent experiments and expressed as mean ± SEM (n=4). *p<0.05, **p<0.01, ***p<0.001 vs control; ^p<0.05, ^^p<0.01 vs tamoxifen or raloxifene alone.

203

6.2.2.3 Interactive effects between DBT and tamoxifen, raloxifene in SH-SY5Y cells

SERMs have been reported to protect neurons of central nervous system from neuronal injuries in animal models including Parkinson’s disease and Alzheimer’s disease (Bourque et al., 2012; Breuer et al., 2000; Du et al., 2004; Morissette et al.,

2008; Tian et al., 2009). In the present study, both tamoxifen and raloxifene significantly stimulated SH-SY5Y cell proliferation (Figure 6.10A and B, p<0.01, p<0.001 vs control) at majority of the concentrations tested except 10-6M, confirming the neuroprotective effects of SERMs reported by other researchers and our own results from the in vivo study. As presented in chapter 5, DBT also promoted

SH-SY5Y cell proliferation (Figure 5.12A). To investigate the potential interactions between DBT and SERMs, cells were co-treated with DBT plus tamoxifen or raloxifene for 48 hours. The results demonstrated that DBT at 0.25mg/ml appeared to enhance the effects of tamoxifen and raloxifene but the results did not reach statistical significance (Figure 6.10A and B). However, DBT at 0.5mg/ml significantly enhanced the effects of both tamoxifen and raloxifene on SH-SY5Y cell proliferation (Figure

6.10A, p<0.05, p<0.01 vs tamoxifen alone; Figure 6.10B, p<0.05 vs raloxifene alone).

The results indicated the possible interactions between the effects of DBT and SERMs in SH-SY5Y cells and the interaction appears to be dose-dependent.

204

C E2 Tamo Tamo+D0.25 Tamo+D0.5

2.0 ^^ ^ ^^ ^ 1.5 * ** ** * * *** ** *** *** *** ** *** * *** 1.0

0.5

0.0

Cell viablity (Relative to control) (Relative viablity Cell Tamo-12 Tamo-10 Tamo-8 Tamo-6

C E2 Ralo Ralo+D0.25 ^ ^ ^ Ralo+D0.5 ^ 1.5 ** ** * * * ** * *** *** *** ** *** *** *** 1.0

0.5

0.0

Cell viablity (Relative to control) (Relative viablity Cell Ralo-12 Ralo-10 Ralo-8 Ralo-6

Figure 6.10 Interactive effects of DBT and tamoxifen or raloxifene on cell proliferation in SH-SY5Y cells SH-SY5Y cells were cultured and subjected to vehicle, estradiol (10-8M), tamoxifen (10-12, 10-10, 10-8, 10-6M), raloxifene (10-12, 10-10, 10-8, 10-6M) alone and co-treatment with DBT (0.25, 0.5mg/ml) for 48 hours as described in Methods (Chapter 3). Cell proliferation was measured by MTS assay. A. Interactive effects between DBT and tamoxifen; B. Interactive effects between DBT and raloxifene. Results were from two independent experiments and expressed as mean ± SEM (n=4). *p<0.05, **p<0.01, ***p<0.001 vs control; ^p<0.05, ^^p<0.01 vs tamoxifen or raloxifene alone.

205

6.2.2.4 Interactive effects between DBT and tamoxifen, raloxifene in MG-63 cells

As presented in chapter 5, DBT significantly increased ALP activities of MG-63 cells and DBT at 1.0 and 2.0mg/ml exerted the most potent effects which increased ALP activities by more than one fold (Figure 5.13B). In the present study, raloxifene, but not tamoxifen, significantly promoted ALP activities in MG-63 cells (Figure 6.11B, p<0.01, p<0.05 vs control). To investigate the interactive effects between DBT and tamoxifen or raloxifene, MG-63 cells were subjected to co-treatment with their combination. Upon co-treatment for 48 hours, DBT at 1.0 and 2.0mg/ml significantly enhanced the effects of both tamoxifen and raloxifene at lower concentrations (10-12,

10-10, 10-8 M) on ALP activities (Figure 6.11A, p<0.05, p<0.01 vs tamoxifen alone;

Figure 5.11B, p<0.01, p<0.001 vs raloxifene alone).

206

C E2 Tamo Tamo+D1.0 2.5 ^^ ^^ ^ Tamo+D2.0 ^^ ^^ ^^ 2.0 * ** ** ** ** * 1.5 *** *** *** *** 1.0

0.5

0.0

ALP activity (relative to control) (relative activity ALP Tamo-12 Tamo-10 Tamo-8 Tamo-6

C E2 ^^^ Ralo ^^^ Ralo+D1.0 ^^ Ralo+D2.0 2.0 ^^^ ^^^ ** *** *** *** *** *** 1.5 *** *** *** * *** ** ** *** 1.0

0.5

0.0

ALP activity (relative to control) (relative activity ALP Ralo-12 Ralo-10 Ralo-8 Ralo-6

Figure 6.11 Interactive effects of DBT and tamoxifen or raloxifene on ALP activities in MG-63 cells MG-63 cells were cultured and subjected to vehicle, estradiol (10-8M), tamoxifen (10-12, 10-10, 10-8, 10-6M), raloxifene (10-12, 10-10, 10-8, 10-6M) alone and co-treatment of DBT (1.0, 2.0mg/ml) plus tamoxifen or raloxifene for 48 hours as described in Methods (Chapter 3). ALP activities were measured by ALP kit. A. Interactive effects between DBT and tamoxifen; B. Interactive effects between DBT and raloxifene. Results were from two independent experiments and expressed as mean ± SEM (n=4). *p<0.05, **p<0.01, ***p<0.001 vs control; ^p<0.05, ^^p<0.01, ^^^p<0.001 vs tamoxifen or raloxifene alone.

207

6.3 Discussion

Tamoxifen has been widely used for treatment of breast cancer at all stages (Fisher et al., 1989) and it indeed reduces the incidence of breast cancer in pre- and post-menopausal women at high risk (Powles et al., 2007). Raloxifene is clinically prescribed for treatment of osteoporosis and prevention of breast cancer and it significantly decreases the incidence of bone fracture (Peterson et al., 2011). They belong to selective estrogen receptor modulator (SERMs) that act as estrogen agonist or antagonist depending on the target tissues (Pinkerton et al., 2014). Both tamoxifen and raloxifene behave as estrogen antagonist in mammary gland (Smith et al., 2004) while they mimics estrogen in uterus and skeleton in both human (Love et al., 1992) and rodents (Spiro et al., 2013; Turner et al., 1987). However, the activities of tamoxifen and raloxifene in the central nervous system, another target tissue of estrogen, have not yet been understood.

Treatment-related adverse symptoms of SERMs, such as fatigue, nausea, vomiting and other menopausal symptoms, make it impossible for the patients to tolerate treatment and lead them to seek for additional or alternative help (Shapiro et al., 2001).

Many patients use complementary and alternative medicine for either treatment of their menopausal symptoms or release treatment-related adverse symptoms. In China, the Chinese herbal medicine is a major alternative approach. The rate of patient

208 concurrently use prescribed western drugs and herbal medicine was high, amount to

78.6% (Chen et al., 2008b) in Shanghai and 76.8% in Taiwan (Lai et al., 2012). It means that combination of western drugs and TCM has been becoming more and more popular. With the increasing popularity of combination of western drugs together with TCM, the concerns have been raised whether it is safe in terms of the potential interactions between them since they exert via the same estrogen receptors.

DBT is a commonly prescribed Traditional Chinese Medicine formula for postmenopausal women who are prescribed with western drugs to either alleviate their menopausal symptoms and or prevent the recurrence of original disease. In chapter 5, DBT was demonstrated to selectively exert protective effects on bone and central nervous system against ovariectomy-induced disruptions in OVX rats without causing undesirable adverse effects in both breast and uterus. Moreover, DBT directly exerted estrogenic effects in four estrogen receptor-positive cell lines as it stimulated cell proliferation, up-regulated mRNA expression level of estrogen-regulated genes and increased estrogen response element (ERE)-dependent luciferase activities in these four cell lines. The promoting effects of DBT on cell proliferation and differentiation in human osteosarcoma MG-63 cells could be completely or partially abolished by ICI182,780, U0126 and LY 294,002, indicating the possible involvement of ER, MAPK and PI3K pathways in the actions of DBT. Our results established that

209

DBT contain phytoestrogens and selectively exerted estrogenic effects in estrogen sensitive tissues.

In the present study, DBT has been chosen as representative of phytoestrogens exerting tissue-selective effects in estrogen sensitive tissues to investigate the potential herb-drug interactions. By using both the mature OVX rats and four estrogen receptor-positive cell lines, the potential interactions between DBT and tamoxifen or raloxifene were determined as summarized in Table 6.4. No undesirable interaction between DBT and tamoxifen or raloxifene has been found on majority of the parameters analyzed in OVX rat model. However, DBT at high dose, but not low dose, appeared to enhance or alleviate the actions of SERMs on cell proliferation or ALP activity in four cell lines.

The bone mineral density (BDM), bone surface (BS), ration of bone volume /total volume (BV/TV), trabecular bone number (Tb.N) and trabecular bone thickness

(Tb.Th) dramatically decreased while Tb.Sp markedly increased at distal femur, proximal tibia and spine of OVX rats in response to OVX. Exposure to estradiol significantly restored these changes in bone by increasing BMD, BS, BV/TV, Tb.N,

Tb.Th and reducing Tb.Sp in OVX rats. Both tamoxifen and raloxifene exerted estrogen-like bone protective effects at the three sites scanned in the present study.

The results were consistent with previous findings in literatures in both

210 postmenopausal women (Ensrud et al., 2006; Hegde et al., 2016; Powles et al., 1996) and rat or mice model (Kangas et al., 2014; Ramalho-Ferreira et al., 2015). In particular, tamoxifen and raloxifene were more potent than estradiol in increasing

BMD at the lumbar spine of OVX rats. As presented in chapter 5, DBT also protected bone from the OVX-induced osteoporosis at distal femur, proximal tibia and the lumbar spine in OVX rats in a way similar to estrogen. However, combinations of

DBT and tamoxifen or raloxifene did not alter the effects of either treatment on BMD and trabecular bone properties, indicating that there is no interaction between DBT and tamoxifen or raloxifene. Moreover, similar to estradiol, tamoxifen and raloxifene significantly inhibited the OVX-induced increase in urinary deoxypyridinoline, a specific biomarker for bone resorption, confirming the anti-resorption activities of

SERMs (Matheny et al., 2013). Serum osteocalcin and urinary DPD levels were not different between rats treated with tamoxifen or raloxifene alone and those treated with combination of DBT plus tamoxifen or raloxifene. These results indicate that

DBT did not alter the effects of SERMs on bone turnover markers in OVX rats.

Results from the in vitro studies showed that both raloxifene not tamoxifen stimulated

ALP activities in MG-63 cells at a wide range of concentrations, which was in agreement with a study performed in human bone marrow-derived stem cells

(Matsumori et al., 2009). In chapter 4, DBT was found to increase ALP activities in

211

MG-63 cells by more than 70% in which DBT at 1.0 and 2.0mg/ml were shown to exert the most potent effects and was chosen as the dosages used in co-treatment experiments. Our results showed that DBT at both 1.0 and 2.0mg/ml significantly enhanced the actions of raloxifene on ALP activities in MG-63 cells. However, the effects appear to be additive rather than interactive since DBT alone already stimulated the ALP activities to an extremely high level. Based on our in vivo experiments using OVX rats, there was no interaction between DBT and tamoxifen or raloxifene on bone.

Our study was the first to report the protective effects of tamoxifen and raloxifene in the substantia nigra-striatum dopaminergic system. Our study showed that tamoxifen and raloxifene significantly reversed the changes in mRNA expression level of TH and DAT in rats induced by OVX. Moreover, the neuroprotective effects of both tamoxifen and raloxifene on TH and DAT mRNA expressions in striatum of OVX rat were much more potent than that of estradiol. Similarly, DBT was also demonstrated to exert strong neuroprotective effects in OVX rats. However, the effects of tamoxifen or raloxifene alone on TH and DAT mRNA expression in OVX rats were not different from those co-treated with DBT and tamoxifen or raloxifene. Our in vitro studies confirmed the neuroprotective effects of tamoxifen and raloxifene as they significantly promoted the viability of SH-SY5Y cells at a wide range of

212 concentrations. In addition, DBT at 0.5mg/ml enhanced the neuroprotective of

SERMs while DBT at 0.25mg/ml did not alter the effects of SERMs in SH-SY5Y cells.

The positive activities of tamoxifen and raloxifene have been established in bone and brain in OVX rats and no interaction between them and DBT has been found.

However, their actions on uterus and breast should be of much more importance in assessing the safety of their use in combinations with DBT. Both tamoxifen and raloxifene significantly increased the weight of uterus in OVX rats by more than 80% and up-regulated C3 mRNA expression, confirming their estrogen agonist activities in uterus. These results are in agreement with findings in previously published studies

(Seidlova-Wuttke et al., 2009; Stygar et al., 2003). Upon co-treatment with DBT and

SERMs, the uterine weight significantly increased compared with OVX rats. However, no differences between co-treatment and SERMs alone have been found. Neither

SERMs alone nor combination of them with DBT altered the mRNA expression level of the other two estrogen-regulated genes. Therefore, the effects of combinations of

DBT and SERMs might be just the additive rather than interactive effects. Our in vitro studies showed that neither tamoxifen nor raloxifene alone exerted stimulatory effects on ALP activities in Ishikawa cells while co-treatment with DBT at a higher dose, i.e.

0.5mg/ml, significantly promoted the ALP activity and enhanced the effects of

213

SERMs in Ishikawa cell.

Similarly, neither SERMs alone nor their combination with DBT showed any effects on mammary glands in OVX rats as no difference in the number and morphology of mammary gland was detected. Furthermore, in vitro studies showed that DBT at

0.5mg/ml reversed the inhibitory effects of tamoxifen and raloxifene on viability of

MCF-7 cells, indicating that DBT might affect the actions of SERMs against breast cancer. However, this discrepancy in results between in vivo and in vitro might be resulted from the much higher dosages of DBT used in the in vitro study than the dose detected in animal and the DBT used in in vitro study was not metabolically activated.

Our results demonstrated both tamoxifen, raloxifene alone and their combinations with DBT suppressed the increase in body weight, confirming their estrogen agonist activities on body weight in OVX rats (Lopez et al., 2006; Wade et al., 1993). At the same time, these treatments significantly increased the serum level of estradiol in

OVX rats and suppressed the FSH and LH levels. However, DBT did not alter the effects of either tamoxifen or raloxifene on body weight gain and serum level of reproductive hormone, indicating that DBT did not alter effects of SERMs on body weight gain and serum levels of reproductive hormones in OVX rats.

Both tamoxifen and raloxifene exerted estrogen like protective in bone and brain against estrogen deficiency-induced damages in OVX rats without causing abnormal

214 proliferation even hyperplasia in breast. However, both tamoxifen and raloxifene induced hyperplasia in endometrium in OVX rats. Co-treatment with DBT did not alter the actions of SERMs in bone, brain, uterus and breast in OVX rats. Results from in vitro study showed that DBT at higher dosages enhanced the effects of SERM in

MG-63, SH-SY5Y cell and Ishikawa cells while weaken the anti-breast cancer effects of SERMs in MCF-7 cell. However, the results of in vivo study are more relevant than those of in vitro study since the dosages used in cells are much higher than the dose used in animal. Therefore, dosage is the key factor for combination of DBT and

SERMs that must be taken into consideration when making clinical decision. In conclusion, our study demonstrated that DBT did not interact with tamoxifen and raloxifene in estrogen sensitive tissues and cell lines.

Our study provides evidences for the safety in the combined use of DBT and SERMs for treatment of menopause-related symptoms.

215

Table 6.4 Summary of potential interactions between DBT and tamoxifen or raloxifene in four estrogen sensitive tissues

tamoxifen DBT + tamoxifen raloxifene DBT + raloxifene Uterus Uteru Weight Increased(+) Increased(+) Increased(+) Increased(+) s Gene ER: no change ER: no change ER, C3, PR: increased ER, C3, PR: increased expression C3, PR: increased C3, PR: increased without without significance without significance without significance significance Morphology Mild hyperplasia Mild hyperplasia Mild hyperplasia Mild hyperplasia ALP activity of No significant change Stimulatory co-treated No significant change Stimulatory co-treated with Ishikawa cell with DBT (0.5mg/ml) at DBT (0.5mg/ml) at 10-12, 10-12, 10-10 and 10-8M 10-10M Breast Morphology of breast Atrophy Atrophy Atrophy Atrophy Cell proliferation of Inhibitory effects Stimulatory co-treated No significant change Stimulatory co-treated with MCF-7 cell with DBT (0.5mg/ml) DBT (0.5mg/ml) Bone Bone BMD Increased (++) Increased (++) Increased (++) Increased (++) Bone property Improved (++) Improved (++) Improved (++) Improved (++) OCN, DPD OCN: no change OCN decreased (-) OCN: no change OCN: no change DPD: decreased (-) DPD decreased (-) DPD decreased (-) DPD decreased (-) ALP activity of MG-63 No significant effect Stimulatory effects Stimulatory effects at Enhanced by co-treatment cell co-treated with DBT at 1.0 wide concentrations (+) with DBT at 1.0 and and 2.0mg/ml (+++) 2.0mg/ml (+++) CNS Striat Estrogen TH increased (++) TH increased (++) TH increased (+++) TH increased (++) um responsive DAT decreased (++) DAT decreased (++) DAT decreased (++) DAT decreased (++) genes SH-SY5Y cell Stimulatory effects at Enhanced by co-treatment Stimulatory effects at Enhanced by co-treatment wide concentrations with DBT at 0.5mg/ml wide concentrations (+) with DBT at 0.5mg/ml (++) (+) (++) Increase (+++) > (++) > (+); Decrease (---) > (--) > (-)

216

Chapter 7

Discussion and Conclusion

217

6.1 Discussion

Phytoestrogens are groups of non-steroidal compounds that are derived from natural plants (Brzezinski et al., 1999) and have been reported to be able to attenuate post-menopause related symptoms (Borrelli et al., 2010; Carmignani et al., 2010;

Drews et al., 2007; Manonai et al., 2006). They have been found to exert both agonistic and antagonistic effects, depending on the endogenous level of estrogen and tissue types (Kim et al., 2012). These properties make use of phytoestrogen a potential alternative approach to HRT that attenuates postmenopausal symptoms in target tissues, such as bone and central nervous system, with lower or even no adverse effects in other estrogen sensitive tissues like uterus and breast.

Tissue-selective effects of icariin and DBT

The present study increases our understanding of the tissue selective estrogenic effects of icariin and DBT, two representatives of phytoestrogens, in four estrogen sensitive tissues. In chapter 4, we demonstrated that icariin exerted tissue-selective estrogenic effects in ways that are different from estradiol. The potent estrogenic effects of icariin were found in bone and brain, but not in reproductive tissues, of OVX rats.

Such effects of icariin were similar to those of estradiol which also significantly protected bone and brain from estrogen deficiency in OVX rats. Moreover, icariin at

500 and 3000ppm appeared to be more potent than estradiol in increasing bone

218 mineral density (BMD) in OVX rats. In contrast, icariin acted differently from estradiol in both uterus and breast in OVX rats. Our study showed that estradiol, but not icariin, induced proliferation and even hyperplasia as well as altered gene expression of uterus and breast tissues in OVX rats. Our in vitro studies showed that icariin mimicked estrogen in dramatically promoting cell proliferation and (or) ALP activities as well as the mRNA expression of estrogen responsive genes in human breast cancer MCF-7, human endometrial cancer Ishikawa, human neuroblastoma

SH-SY5Y and human osteoblastic MG-63 cells. However, one important but unexpected finding is that the regulation of estrogen-responsive genes by icariin was distinct from that of estradiol in which icariin did not alter the estrogen response element (ERE)-dependent luciferase activities in the four cell lines. These results suggested that the estrogenic effects of icariin were independent of ERE-mediated transcription in these four types of cells.

In chapter 5, our results provided experimental evidences for the effectiveness of DBT in management of postmenopausal syndrome. Our study supported that the estrogenic actions of DBT were tissue-selective in OVX rats. DBT acted similarly to estradiol in protecting bone and brain from estrogen deficiency-induced manifestations. However, the actions of DBT in uterus were shown to be different from those of estradiol.

Estradiol, but not DBT, was found to stimulate proliferation and even hyperplasia,

219 induce estrogen-responsive gene expression in uterus and increase uterus weight in

OVX rats. Furthermore, DBT was found to induce weak estrogenic effects in mammary gland in OVX rats. Such stimulatory effects of DBT in mammary gland might be due to the increase in circulating estradiol level in OVX rats treated with

DBT. In addition, DBT was showed to up-regulate the mRNA expression of aromatase in adipose tissue, which is the key enzyme that catalyzes the bioconversion from androgen to estradiol (Simpson, 2002). Besides estradiol, the increase in FSH and LH level in OVX rats were also reversed by treatment with DBT. These results suggested that DBT may regulate the functions of hypothalamus-pituitary-gonadal axis (Figure 7.1). Our in vitro studies indicated that DBT showed weak estrogen-like effects as both DBT and estradiol significantly promoted cell proliferation and (or)

ALP activities, mRNA expression level of estrogen-responsive genes and

ERE-dependent luciferase activities in these estrogen-sensitive cell lines. These results suggested that DBT directly exerted estrogenic effects in estrogen sensitive cell line and activated ERE-dependent transcription of estrogen responsive genes.

Moreover, the stimulatory effects of DBT on cell proliferation and (or) ALP activity in human osteoblastic MG-63 cells could be completely or partially blocked in the presence of estrogen receptor (ER) antagonist ICI182,780, MAPK inhibitor MAPK and PI3K inhibitor LY294002. Such results indicated that the actions of DBT might

220 be ER-dependent and mediated at least in part via non-genomic signaling pathways besides the classical ER pathway.

Underlying mechanisms for the tissue-selectivity of phytoestrogens

Our results have confirmed that TCM-derived phytoestrogens including icariin and

DBT exerted protective effects in bone and brain in tissue-selective manner without causing undesirable adverse effects in uterus and breast (Figure 7.2). To determine if these positive effects were mediated via modulation of circulating estradiol level, we also measured the serum level of estradiol in OVX rats upon treatment. Our results showed that DBT significantly increased the circulating estradiol level by more than two fold in OVX rats. Icariin has also been reported to increase circulating estradiol level via the up-regulation of gene and protein expression of aromatase in human ovarian granulosa-like KGN cells (Yang et al., 2013). Such effects were shown to be dose and time-dependent. Icariin at 50uM showed the most prominent effects in KGN cells upon treatment for three hours, at which the biosynthesis of estradiol and the mRNA and protein expression level of aromatase were significantly increased. Based on our results and previous findings in literature, the modulation effects of phytoestrogens on sex hormone and hypothalamus-pituitary-gonadal axis (Low et al.,

2007) might at least in part account for their positive effects in bone and brain (Figure

7.2). The stimulatory effect of DBT on circulating estradiol level has been reported by

221 others in rats with one ovary removed (Tan et al., 2010). Our study was the first to report that DBT increased the estradiol level in rat with bilateral ovaries removed.

Furthermore, our study showed that the stimulatory effect of DBT on circulating estradiol might be in part mediated by the induction of mRNA expression of aromatase in adipose tissue, one of the major extragonadal sources of estradiol. These results indicated that adipose tissue has become one source of estradiol in OVX rats treated with DBT via the induction of activity of aromatase. This finding indicates that

DBT might be useful for postmenopausal women with extremely low circulating estradiol level for increasing endogenous estradiol synthesis.

Another important finding is that DBT could suppress the increase in follicle-stimulating hormone (FSH) level in OVX rats as a feedback to estrogen deficiency. Recent studies indicated that the increase in FSH level, rather than the decrease in estradiol, contributed to the increase in bone loss in postmenopausal women (Sun et al., 2006). Thus, the suppressive effect of DBT on FSH level in OVX rats might be more important than its inductive effects on estradiol for treatment of postmenopausal osteoporosis.

It is of interest to note that neither icariin nor DBT exert estrogenic effects in uterus and breast tissue of OVX rats. This seems to be in conflict with our observations that demonstrate the increase of circulating estradiol level by DBT as well as its positive

222 effects in bone and brain of OVX rats. Indeed, recent studies indicated that the estrogenic effects of estradiol was actively regulated by local estradiol metabolism in the tissue rather than the circulating level of estradiol (Huhtinen et al., 2012).

Phytoestrogens have been found to facilitate the clearance of estrogens from local tissues like uterus and breast and catabolize the estrogens to more benign

2-hydroxylated metabolites (Wood et al., 2007). The decreased tissue levels of E2 and increased concentrations of 2-methoxy-E2 were found upon treatment with daidzein, an isoflavone. Such profile of E2 and its metabolite in local tissue has been reported to be related to the anti-cancer effects of daidzein in breast and uterus (Pfeiffer et al.,

2006). These results indicate that phytoestrogens modulate the estradiol level in local tissues in way that is different from that in the circulation and that the estrogenic effects are dependent on the concentration of estradiol in local tissue rather than the circulating estradiol. Thus, it is possible that DBT and icariin might act like other phytoestrogen and facilitate the clearance of estradiol from the local tissues, thereby reducing estrogenic responses in uterus and breast tissues in OVX rats. Moreover, the agonistic or antagonistic effects of phytoestrogens have been demonstrated to be dependent on its own concentrations as well as the estradiol concentrations in local environments (Kim et al., 2012; Ososki et al., 2003). Genistein has been shown to be estrogen agonist at a low concentration as it stimulated the proliferation of MCF-7 cell,

223 whereas high concentrations exerted inhibitory effects (Martin et al., 1978; Wang et al., 1998). Isoflavones were shown to act as estrogen agonist in low-estradiol concentrations to stimulate cell proliferation (Hwang et al., 2006). Thus, the fact that

DBT and icariin behave as agonists in bone and brain tissues but as antagonists in uterus and breast tissues might also depend on the ratio of phytoestrogen to estradiol concentrations in local environments of each tissue. Future study will be needed to determine the effects of DBT and icariin on the local estradiol levels at different tissues in OVX rats.

Our results also indicated that the actions of icariin and DBT were ER dependent since specific ER antagonist ICI182,780 could abolish their stimulatory effects in either MCF-7 or MG-63 cells. The two subtypes of ERs, ERα and ERß, mediate majority of actions of estrogen. They are co-expressed in majority of the ER-positive tissues including brain, breast, uterus and bone while only ERα or ERß is found in liver or gastrointestinal tract, respectively. However, even in the same tissue, the expression level of ERα and ERß is different (Kuiper et al., 1997). The two subtypes of ERs belong to the nuclear receptor and mediate different biological actions

(Enmark et al., 1997; Johansson et al., 1999) (Kuiper et al., 1996). ERα promotes cell proliferation and survival as well as E2-mediated gene expression while ERß exerts inhibitory effects on actions of ERα (Nelson et al., 2014; Williams et al., 2008).

224

Moreover, ERα and ERß are shown to interact with each other. For example, isoforms of ERß could dimerize with ERα and silence the signaling mediated by ERα (Chen et al., 2005). Such wide distribution and different expression level in certain tissue of

ERs as well as the different role of mediating estrogenic effects and interactions between them might result in the distinct responses to estrogen from tissue to tissue.

At the transcriptional level, both DBT and icariin regulated the specific estrogen-responsive genes in each estrogen sensitive cell line in estrogen-like manner.

Transcriptional regulations by DBT in these estrogen sensitive cell lines were demonstrated to be estrogen response element (ERE)-dependent. However, one unexpected result from my study is that regulation of icariin on estrogen-regulated genes was ERE-independent. ER signaling is regulated in the presence of a series of transcriptional co-factors that either enhance or repress the estrogen-regulated genes

(McDonnell et al., 2002), such as steroid receptor coactivator-2 (SRC-2), the activator protein (AP)-1, nuclear factor-kB (NF-kB) and stimulating protein-1 (Sp-1) (Marino et al., 2006) in both direct or indirect genomic pathways. Moreover, the relative expression of these co-activators and co-repressor varies from cell type to cell type

(Ciana et al., 2001; Krum et al., 2008; Marino et al., 2006; Penttinen et al., 2007). In responses to stimulation of estradiol, different subsets of co-factors will be activated or inactivated, resulting in distinct transcriptional events. Besides, the non-genomic

225 pathways also induce transcriptional events via numerous interactions with several other pathways like insulin growth factor-I (IGF-I) pathway (Kahlert et al., 2000;

Song et al., 2007), endothelial growth factor (EGF) pathway (Zhang et al., 2012b).

Thus, the final responses of a particular tissue/cell type could be dependent on a number of conditions including the subset of co-factors, the relative levels of co-activators and co-repressors as well as the signaling pathways by which the estrogenic effects are exerted (Marino et al., 2006), resulting in a cell type specific transcriptional responses. Thus, it is possible that the tissue-selective effects of icariin and DBT as reported in the present study might be achieved in part by these transcriptional conditions. Future study will be needed to explore the differential mechanism involved in the activation of transcriptional events by icariin and DBT in different estrogen sensitive tissues and cells.

Selective estrogen receptor modulators (SERMs) are ER ligands that are either agonist in some tissue like bone while antagonist in other tissue such as uterus and breast.

Phytoestrogens have been proved to be novel ER modulators (Ahn et al., 2014; van de

Schans et al., 2016). In chapter 5, our results demonstrated DBT is a novel selective estrogen receptor modulator since it exerted tissue-selective effects in target tissues of estrogen. Moreover, the actions of DBT have been proved to be ER-dependent. Since they might exert their effects via similar ERs, the interactions between DBT and

226

SERMs may possibly occur. In chapter 6, we demonstrated that co-treatment of DBT did not affect the effects of either tamoxifen or raloxifene in four estrogen sensitive tissues, including bone, brain, uterus and breast tissues, in OVX rats. The slight enhancement or suppressive effects of SERMs by DBT at high dose in vitro appeared to be additive, rather than interactive. Our results indicated that DBT did not interact with SERMs and the combinations the use of DBT and SERMs are therefore recommended. However, special attentions must be paid to the selection of dosages of

DBT when making clinical decisions.

Chinese Medicine has been used for improving women’s health and treatment of menopause or treatment-related symptoms. Such applications of TCM could be tracked back to ancient times. Moreover, interests in the use of TCM for management of menopause have been on the rise outside of China. However, due to the complex compositions and actions of TCM, the prescriptions of TCM have been primarily based on experiences rather than modern scientific evidences. The present study not only confirmed the effectiveness of TCM for treatment of estrogen deficiency-induced disorders but also provided scientific evidences for the possible mechanisms of TCM in achieving tissue-specific estrogen responses. These results clearly indicated that use of TCM-derived phytoestrogens might be safe alternative to HRT for management of menopausal symptoms by which novel mechanisms are involved in achieving

227 selective responses in estrogen sensitive tissues.

Taken together, both icariin and DBT selectively protected bone and brain from estrogen deficiency-induced changes without inducing undesirable estrogenic effects in uterus and breast in OVX rats. Besides, DBT appeared to act on hypothalamus-pituitary-gonadal axis to restore the abnormal profile of sex hormone in

OVX rats. Compared with effects of estrogen in targeting tissues, DBT and icariin showed good tissue-selectivity. DBT did not interact with SERMs in either OVX rats or estrogen-sensitive cell lines. Our results clearly suggest that that DBT could be safely used in combination with SERMs in the postmenopausal women.

228

Figure 7.1 Modulating effects of DBT on sex hormone profile in OVX rats

DBT significantly up-regulated the mRNA expression in adipose tissue by which DBT increased circulating estrogen level, indicating that the extragonadal tissues such as adipose tissues might become one major source of estradiol in OVX rats treated with DBT. In addition, the increase in FSH and LH level were significantly suppressed upon treatment with DBT. These results suggested that DBT might modulate he hypothalamus-pituitary-gonadal axis in OVX rats.

229

Figure 7.2 Possible mechanisms for the tissue-selective effects of phytoestrogens

(1) Systematically, phytoestrogens increase the circulating estradiol level in OVX rats via up-regulation of mRNA expression of aromatase in extragonadal tissue. (2) Simultaneously, they facilitate the clearance of estradiol from local tissues and catalyze the metabolism of estradiol into more benign metabolites in local tissues, resulting in the different levels of estradiol in circulation and local tissues (Pfeiffer et al., 2006; Wood et al., 2007). (Rocca et al.) At the molecular level, phytoestrogens act as agonist or antagonist in local tissues via activation of different ER signaling pathways (Kim et al., 2012; Marino et al., 2006; Ososki et al., 2003). (4) Therefore, local tissues like bone, brain, uterus and breast specifically response to phytoestrogens, resulting in the tissue selective effects of phytoestrogens.

230

7.2 Conclusion

In conclusion, our results provide strong evidence for the use of icariin and DBT as alternative approaches to HRT for management of postmenopausal syndrome. Results from our study strongly support that TCM-derived phytoestrogens, such as icariin and

DBT, could exert tissue-selective effects in estrogen sensitive tissues. Moreover, evidence from the present study suggests that phytoestrogen or phytoestrogen-containing TCM, such as DBT, does not interact with the effects of western drugs (tamoxifen and raloxifene) in estrogen-sensitive tissues. Special attentions should be paid by medicinal communities to the selections of dosages of herbal medicine for simultaneous use of SERMs when making clinical decisions on holistic management of menopause symptoms, osteoporosis and breast cancer in women.

231

7.3 Limitation and recommendations for further research

The present study is a valuable study and provides scientific evidences for the clinical community. However, there are still some limitations.

1. First and foremost is about the dose of the drugs. A single dose of DBT was

chosen in the present study due to the limitation of the scale of animal experiment.

The administration of icariin was not accurate enough to ensure the dosages while

it is already the best choice due to the extremely low solubility of icariin in water

even in organic solvents.

2. Ovariectomized model is good for assessing estrogen-like effects but might not be

sufficient to address the pathological conditions such Alzheimer’s disease or

breast cancer. Cell study might have lower relevance as the systemic effect of

herbal medicine cannot be imitated in the in vitro conditions.

3. The present study was conducted in animal and cell experiments rather than

clinical trial. Clinical trials are necessary to investigate the potential interactions

between Herbal medicine and prescribed drugs.

Based on the results obtained in the present study, future research work is recommended to be conducted in the following aspects:

1. The effects of icariin and DBT on estrogen level (circulating vs local tissue),

estrogen metabolism (aromatase activity and expression in local tissues),

232

expression of ERs (ERα and ERß) as well as interactions between them and ERs

should be investigated for the possible mechanism for their tissue-selectivity;

2. The main active component(s) in the DBT extract that predominantly exerts the

estrogenic effects should be determined;

3. Estrogenic effects of DBT need to be investigated at various dosages;

4. Clinical trials are recommended for the investigation of the potential interactions

between Herbal medicine and prescribed drugs.

233

References

Abdelhamid R, Luo J, Vandevrede L, Kundu I, Michalsen B, Litosh VA, et al. Benzothiophene Selective Estrogen Receptor Modulators Provide Neuroprotection by a novel GPR30-dependent Mechanism. ACS Chem Neurosci. 2011. 2:256-268.

Acevedo-Rodriguez A, Mani SK, Handa RJ. Oxytocin and Estrogen Receptor beta in the Brain: An Overview. Front Endocrinol (Lausanne). 2015. 6:160.

Adlercreutz H. Phyto-oestrogens and cancer. The lancet oncology. 2002. 3:364-373.

Ahn HN, Jeong SY, Bae GU, Chang M, Zhang D, Liu X, et al. Selective Estrogen Receptor Modulation by Larrea nitida on MCF-7 Cell Proliferation and Immature Rat Uterus. Biomol Ther (Seoul). 2014. 22:347-354.

Al-Sweidi S, Morissette M, Bourque M, Di Paolo T. Estrogen receptors and gonadal steroids in vulnerability and protection of dopamine neurons in a mouse model of Parkinson's disease. Neuropharmacology. 2011. 61:583-591.

Almey A, Milner TA, Brake WG. Estrogen receptors in the central nervous system and their implication for dopamine-dependent cognition in females. Horm Behav. 2015. 74:125-138.

Almey A, Milner TA, Brake WG. Estrogen receptor alpha and G-protein coupled estrogen receptor 1 are localized to GABAergic neurons in the dorsal striatum. Neurosci Lett. 2016. 622:118-123.

An BH, Jeong H, Zhou W, Liu X, Kim S, Jang CY, et al. Evaluation of the Biological Activity of Opuntia ficus indica as a Tissue- and Estrogen Receptor Subtype-Selective Modulator. Phytother Res. 2016. 30:971-980.

Ascherio A, Chen H, Schwarzschild MA, Zhang SM, Colditz GA, Speizer FE. Caffeine, postmenopausal estrogen, and risk of Parkinson's disease. Neurology. 2003. 60:790-795.

Avis NE, Brockwell S, Colvin A. A universal menopausal syndrome? Am J Med. 2005. 118 Suppl 12B:37-46.

234

Bagger YZ, Tanko LB, Alexandersen P, Qin G, Christiansen C. Early postmenopausal hormone therapy may prevent cognitive impairment later in life. Menopause. 2005. 12:12-17.

Baum LW. Sex, hormones, and Alzheimer's disease. J Gerontol A Biol Sci Med Sci. 2005. 60:736-743.

Bean LA, Ianov L, Foster TC. Estrogen receptors, the hippocampus, and memory. Neuroscientist. 2014. 20:534-545.

Benassayag C, Perrot-Applanat M, Ferre F. Phytoestrogens as modulators of steroid action in target cells. Journal of Chromatography B. 2002. 777:233-248.

Berthois Y, Katzenellenbogen JA, Katzenellenbogen BS. Phenol red in tissue culture media is a weak estrogen: implications concerning the study of estrogen-responsive cells in culture. Proc Natl Acad Sci U S A. 1986. 83:2496-2500.

Beyene Y. Cultural significance and physiological manifestations of menopause. A biocultural analysis. Cult Med Psychiatry. 1986. 10:47-71.

Biedler JL, Helson L, Spengler BA. Morphology and growth, tumorigenicity, and cytogenetics of human neuroblastoma cells in continuous culture. Cancer Res. 1973. 33:2643-2652.

Black DM, Rosen CJ. Clinical Practice. Postmenopausal Osteoporosis. N Engl J Med. 2016. 374:254-262.

Borrelli F, Ernst E. Alternative and complementary therapies for the menopause. Maturitas. 2010. 66:333-343.

Bourque M, Dluzen DE, Di Paolo T. Signaling pathways mediating the neuroprotective effects of sex steroids and SERMs in Parkinson's disease. Front Neuroendocrinol. 2012. 33:169-178.

Brailoiu E, Dun SL, Brailoiu GC, Mizuo K, Sklar LA, Oprea TI, et al. Distribution and characterization of estrogen receptor G protein-coupled receptor 30 in the rat central nervous system. J Endocrinol. 2007. 193:311-321.

235

Brambilla DJ, McKinlay SM. A prospective study of factors affecting age at menopause. J Clin Epidemiol. 1989. 42:1031-1039.

Breuer B, Anderson R. The relationship of tamoxifen with dementia, depression, and dependence in activities of daily living in elderly nursing home residents. Women Health. 2000. 31:71-85.

Brinton RD, Yao J, Yin F, Mack WJ, Cadenas E. Perimenopause as a neurological transition state. Nat Rev Endocrinol. 2015. 11:393-405.

Brisken C, O'Malley B. Hormone action in the mammary gland. Cold Spring Harb Perspect Biol. 2010. 2:a003178.

Brodowska A, Laszczynska M, Brodowski J, Masiuk M, Starczewski A. Analysis of pituitary gonadotropin concentration in blood serum and immunolocalization and immunoexpression of follicle stimulating hormone and luteinising hormone receptors in ovaries of postmenopausal women. Histol Histopathol. 2012. 27:241-248.

Brooks SC, Locke ER, Soule HD. Estrogen receptor in a human cell line (MCF-7) from breast carcinoma. J Biol Chem. 1973. 248:6251-6253.

Brown LM, Clegg DJ. Central effects of estradiol in the regulation of food intake, body weight, and adiposity. J Steroid Biochem Mol Biol. 2010. 122:65-73.

Brzezinski A, Debi A. Phytoestrogens: the "natural" selective estrogen receptor modulators? Eur J Obstet Gynecol Reprod Biol. 1999. 85:47-51.

Buhling KJ, Daniels BV, Studnitz FS, Eulenburg C, Mueck AO. The use of complementary and alternative medicine by women transitioning through menopause in Germany: results of a survey of women aged 45-60 years. Complement Ther Med. 2014. 22:94-98.

Burns KA, Rodriguez KF, Hewitt SC, Janardhan KS, Young SL, Korach KS. Role of estrogen receptor signaling required for endometriosis-like lesion establishment in a mouse model. Endocrinology. 2012. 153:3960-3971.

Cai DJ, Zhao Y, Glasier J, Cullen D, Barnes S, Turner CH, et al. Comparative effect of soy protein, soy isoflavones, and 17beta-estradiol on bone metabolism in adult

236 ovariectomized rats. J Bone Miner Res. 2005. 20:828-839.

Cao H, Ke Y, Zhang Y, Zhang CJ, Qian W, Zhang GL. Icariin stimulates MC3T3-E1 cell proliferation and differentiation through up-regulation of bone morphogenetic protein-2. Int J Mol Med. 2012. 29:435-439.

Carmignani LO, Pedro AO, Costa-Paiva LH, Pinto-Neto AM. The effect of dietary soy supplementation compared to estrogen and placebo on menopausal symptoms: a randomized controlled trial. Maturitas. 2010. 67:262-269.

Carr MC. The emergence of the metabolic syndrome with menopause. J Clin Endocrinol Metab. 2003. 88:2404-2411.

Chalupka S. Soy isoflavones for the prevention of menopausal symptoms and bone loss--a safe and effective alternative to estrogen? Aaohn J. 2011. 59:504.

Chang J, Wang Z, Tang E, Fan Z, McCauley L, Franceschi R, et al. Inhibition of osteoblastic bone formation by nuclear factor-kappaB. Nat Med. 2009. 15:682-689.

Channing CP, Coudert SP. Contribution of granulosa cells and follicular fluid to ovarian estrogen secretion in rhesus monkey in vivo. Endocrinology. 1976. 98:590-597.

Chaube R, Joy KP. Estradiol-17beta modulates dose-dependently hypothalamic tyrosine hydroxylase activity inhibited by alpha-methylparatyrosine in the catfish Heteropneustes fossilis. Endocrine. 2011. 40:394-399.

Chen B, Gajdos C, Dardes R, Kidwai N, Johnston SR, Dowsett M, et al. Potential of endogenous estrogen receptor beta to influence the selective ER modulator ERbeta complex. Int J Oncol. 2005. 27:327-335.

Chen DF, Li X, Xu Z, Liu X, Du SH, Li H, et al. Hexadecanoic acid from Buzhong Yiqi decoction induced proliferation of bone marrow mesenchymal stem cells. J Med Food. 2010. 13:967-970.

Chen M, Cui YK, Huang WH, Man K, Zhang GJ. Phosphorylation of estrogen receptor alpha at serine 118 is correlated with breast cancer resistance to tamoxifen. Oncol Lett. 2013. 6:118-124.

237

Chen Y, Zhao YH, Jia XB, Hu M. Intestinal absorption mechanisms of prenylated flavonoids present in the heat-processed Epimedium koreanum Nakai (Yin Yanghuo). Pharm Res. 2008a. 25:2190-2199.

Chen Z, Gu K, Zheng Y, Zheng W, Lu W, Shu XO. The use of complementary and alternative medicine among Chinese women with breast cancer. J Altern Complement Med. 2008b. 14:1049-1055.

Chiu PY, Leung HY, Siu AH, Poon MK, Dong TT, Tsim KW, et al. Dang-Gui Buxue Tang protects against oxidant injury by enhancing cellular glutathione in H9c2 cells: role of glutathione synthesis and regeneration. Planta medica. 2007. 73:134-141.

Choi RC, Gao QT, Cheung AW, Zhu JT, Lau FT, Li J, et al. A chinese herbal decoction, danggui buxue tang, stimulates proliferation, differentiation and gene expression of cultured osteosarcoma cells: genomic approach to reveal specific gene activation. Evid Based Complement Alternat Med. 2011. 2011:307548.

Ciana P, Di Luccio G, Belcredito S, Pollio G, Vegeto E, Tatangelo L, et al. Engineering of a mouse for the in vivo profiling of estrogen receptor activity. Mol Endocrinol. 2001. 15:1104-1113.

Clarkson TB, Utian WH, Barnes S, Gold EB, Basaria SS, Aso T, et al. The role of soy isoflavones in menopausal health: report of The North American Menopause Society/Wulf H. Utian Translational Science Symposium in Chicago, IL (October 2010). Menopause. 2011. 18:732-753.

Cohen LS, Soares CN, Poitras JR, Prouty J, Alexander AB, Shifren JL. Short-term use of estradiol for depression in perimenopausal and postmenopausal women: a preliminary report. Am J Psychiatry. 2003. 160:1519-1522.

Coulam CB, Adamson SC, Annegers JF. Incidence of premature ovarian failure. Obstet Gynecol. 1986. 67:604-606.

Crisafulli A, Marini H, Bitto A, Altavilla D, Squadrito G, Romeo A, et al. Effects of genistein on hot flushes in early postmenopausal women: a randomized, double-blind EPT- and placebo-controlled study. Menopause. 2004. 11:400-404.

238

Cui J, Zhu M, Zhu S, Wang G, Xu Y, Geng D. Inhibitory effect of icariin on Ti-induced inflammatory osteoclastogenesis. J Surg Res. 2014. 192:447-453.

Currie LJ, Harrison MB, Trugman JM, Bennett JP, Wooten GF. Postmenopausal estrogen use affects risk for Parkinson disease. Arch Neurol. 2004. 61:886-888.

Dahlman-Wright K, Cavailles V, Fuqua SA, Jordan VC, Katzenellenbogen JA, Korach KS, et al. International Union of Pharmacology. LXIV. Estrogen receptors. Pharmacol Rev. 2006. 58:773-781.

Dahlman-Wright K, Qiao Y, Jonsson P, Gustafsson JA, Williams C, Zhao C. Interplay between AP-1 and estrogen receptor alpha in regulating gene expression and proliferation networks in breast cancer cells. Carcinogenesis. 2012. 33:1684-1691.

Dal Maso RC, Cavagna Neto M, Yu L, Juliano Y, Novo NF, Cury MC, et al. [Sex hormones profile in women on dialysis program in treatment with erythropoietin]. Rev Assoc Med Bras (1992). 2003. 49:418-423.

Daly E, Gray A, Barlow D, McPherson K, Roche M, Vessey M. Measuring the impact of menopausal symptoms on quality of life. Bmj. 1993. 307:836-840.

DeMichele A, Troxel AB, Berlin JA, Weber AL, Bunin GR, Turzo E, et al. Impact of raloxifene or tamoxifen use on endometrial cancer risk: a population-based case-control study. J Clin Oncol. 2008. 26:4151-4159.

Drews K, Seremak-Mrozikiewicz A, Puk E, Kaluba-Skotarczak A, Malec M, Kazikowska A. [Efficacy of standardized isoflavones extract (Soyfem) (52-104 mg/24h) in moderate and medium-severe climacteric syndrome]. Ginekol Pol. 2007. 78:307-311.

Du B, Ohmichi M, Takahashi K, Kawagoe J, Ohshima C, Igarashi H, et al. Both estrogen and raloxifene protect against beta-amyloid-induced neurotoxicity in estrogen receptor alpha-transfected PC12 cells by activation of telomerase activity via Akt cascade. J Endocrinol. 2004. 183:605-615.

Du M, Yang X, Hartman JA, Cooke PS, Doerge DR, Ju YH, et al. Low-dose dietary genistein negates the therapeutic effect of tamoxifen in athymic nude mice. Carcinogenesis. 2012a.bgs017.

239

Du M, Yang X, Hartman JA, Cooke PS, Doerge DR, Ju YH, et al. Low-dose dietary genistein negates the therapeutic effect of tamoxifen in athymic nude mice. Carcinogenesis. 2012b. 33:895-901.

Dutertre M, Smith CL. Molecular mechanisms of selective estrogen receptor modulator (SERM) action. J Pharmacol Exp Ther. 2000. 295:431-437.

Enmark E, Pelto-Huikko M, Grandien K, Lagercrantz S, Lagercrantz J, Fried G, et al. Human estrogen receptor beta-gene structure, chromosomal localization, and expression pattern. J Clin Endocrinol Metab. 1997. 82:4258-4265.

Ensrud K, Genazzani AR, Geiger MJ, McNabb M, Dowsett SA, Cox DA, et al. Effect of raloxifene on cardiovascular adverse events in postmenopausal women with osteoporosis. Am J Cardiol. 2006. 97:520-527.

Erickson KI, Colcombe SJ, Elavsky S, McAuley E, Korol DL, Scalf PE, et al. Interactive effects of fitness and hormone treatment on brain health in postmenopausal women. Neurobiol Aging. 2007. 28:179-185.

Fan C, Yang Y, Liu Y, Jiang S, Di S, Hu W, et al. Icariin displays anticancer activity against human esophageal cancer cells via regulating endoplasmic reticulum stress-mediated apoptotic signaling. Sci Rep. 2016. 6:21145.

Filardo EJ, Quinn JA, Bland KI, Frackelton AR, Jr. Estrogen-induced activation of Erk-1 and Erk-2 requires the G protein-coupled receptor homolog, GPR30, and occurs via trans-activation of the epidermal growth factor receptor through release of HB-EGF. Mol Endocrinol. 2000. 14:1649-1660.

Fisher B, Costantino J, Redmond C, Poisson R, Bowman D, Couture J, et al. A randomized clinical trial evaluating tamoxifen in the treatment of patients with node-negative breast cancer who have estrogen-receptor–positive tumors. New England Journal of Medicine. 1989. 320:479-484.

Fohr B, Schulz A, Battmann A. Sex steroids and bone metabolism: comparison of in vitro effects of 17beta-estradiol and testosterone on human osteosarcoma cell lines of various gender and differentiation. Exp Clin Endocrinol Diabetes. 2000. 108:414-423.

240

Funakoshi T, Yanai A, Shinoda K, Kawano MM, Mizukami Y. G protein-coupled receptor 30 is an estrogen receptor in the plasma membrane. Biochem Biophys Res Commun. 2006. 346:904-910.

Galmiche G, Richard N, Corvaisier S, Kottler ML. The expression of aromatase in gonadotropes is regulated by estradiol and gonadotropin-releasing hormone in a manner that differs from the regulation of luteinizing hormone. Endocrinology. 2006. 147:4234-4244.

Gambacciani M. Selective estrogen modulators in menopause. Minerva Ginecol. 2013. 65:621-630.

Gao G, Liao M, Feng Y. [Determination of flavonoids and quality evaluation of Chinese traditional drug "puhuang"]. Yao Xue Xue Bao. 1998. 33:300-303.

Gao Q, Mezei G, Nie Y, Rao Y, Choi CS, Bechmann I, et al. Anorectic estrogen mimics leptin's effect on the rewiring of melanocortin cells and Stat3 signaling in obese animals. Nat Med. 2007a. 13:89-94.

Gao QT, Cheung JK, Choi RC, Cheung AW, Li J, Jiang ZY, et al. A Chinese herbal decoction prepared from Radix Astragali and Radix Angelicae Sinensis induces the expression of erythropoietin in cultured Hep3B cells. Planta Medica. 2008. 74:392-395.

Gao QT, Choi RC, Cheung AW, Zhu JT, Li J, Chu GK, et al. Danggui buxue tang--a Chinese herbal decoction activates the phosphorylations of extracellular signal-regulated kinase and estrogen receptor alpha in cultured MCF-7 cells. FEBS Lett. 2007b. 581:233-240.

Giunta M, Rigamonti A, Bonomo S, Gagliano M, Müller E, Scarpini E, et al. Estrogens need insulin-like growth factor I cooperation to exert their neuroprotective effects in post-menopausal women. Journal of endocrinological investigation. 2013. 36:97-103.

Gong AG, Lau KM, Zhang LM, Lin H, Dong TT, Tsim KW. Danggui Buxue Tang, Chinese Herbal Decoction Containing Astragali Radix and Angelicae Sinensis Radix, Induces Production of Nitric Oxide in Endothelial Cells: Signaling Mediated by Phosphorylation of Endothelial Nitric Oxide Synthase. Planta medica. 2016.

241

82:418-423.

Gray SA, Mannan MA, O'Shaughnessy PJ. Development of cytochrome P450 aromatase mRNA levels and enzyme activity in ovaries of normal and hypogonadal (hpg) mice. J Mol Endocrinol. 1995. 14:295-301.

Gruber CJ, Tschugguel W, Schneeberger C, Huber JC. Production and actions of estrogens. N Engl J Med. 2002. 346:340-352.

Guo B, Wang C, Xiao P, Chen J. Determination of flavonids and quality evaluation of sagittate epimedium (Epimedium sagittatum). Chinese traditional and herbal drugs. 1995. 27:584-586.

Hadji P, Ziller M, Kieback DG, Dornoff W, Tessen HW, Menschik T, et al. Effects of exemestane and tamoxifen on bone health within the Tamoxifen Exemestane Adjuvant Multicentre (TEAM) trial: results of a German, 12-month, prospective, randomised substudy. Ann Oncol. 2009. 20:1203-1209.

Haghverdi F, Farbodara T, Mortaji S, Soltani P, Saidi N. Effect of raloxifene on parathyroid hormone in osteopenic and osteoporotic postmenopausal women with chronic kidney disease stage 5. Iran J Kidney Dis. 2014. 8:461-466.

Haines C, Lam P, Chung T, Cheng K, Leung P. A randomized, double-blind, placebo-controlled study of the effect of a Chinese herbal medicine preparation (Dang Gui Buxue Tang) on menopausal symptoms in Hong Kong Chinese women. Climacteric. 2008. 11:244-251.

Hall J. The Autonomic Nervous System and the Adrenal Medulla. Hall JE, Guyton AC Guyton and Hall textbook of medical physiology 12th edition Philadelphia, PA: Saunders/Elsevier. 2011.729-740.

Hansen KR, Craig LB, Zavy MT, Klein NA, Soules MR. Ovarian primordial and nongrowing follicle counts according to the Stages of Reproductive Aging Workshop (STRAW) staging system. Menopause. 2012. 19:164-171.

Haslam SZ, Osuch JR, Raafat AM, Hofseth LJ. Postmenopausal hormone replacement therapy: effects on normal mammary gland in humans and in a mouse postmenopausal model. J Mammary Gland Biol Neoplasia. 2002. 7:93-105.

242

Hauschka PV, Lian JB, Cole DE, Gundberg CM. Osteocalcin and matrix Gla protein: vitamin K-dependent proteins in bone. Physiol Rev. 1989. 69:990-1047.

Hegde V, Jo JE, Andreopoulou P, Lane JM. Effect of osteoporosis medications on fracture healing. Osteoporos Int. 2016. 27:861-871.

Heino TJ, Chagin AS, Savendahl L. The novel estrogen receptor G-protein-coupled receptor 30 is expressed in human bone. J Endocrinol. 2008. 197:R1-6.

Hekimoglu A, Bilgin HM, Kurcer Z, Ocak AR. Effects of increasing ratio of progesterone in estrogen/progesterone combination on total oxidant/antioxidant status in rat uterus and plasma. Arch Gynecol Obstet. 2010. 281:23-28.

Hinds L, Price J. Menopause, hormone replacement and gynaecological cancers. Menopause Int. 2010. 16:89-93.

Ho KJ, Liao JK. Nonnuclear actions of estrogen. Arterioscler Thromb Vasc Biol. 2002. 22:1952-1961.

Häggström M. Diagram of the pathways of human steroidogenesis. Medicine. 2014. 1.

Holinka CF, Anzai Y, Hata H, Kimmel N, Kuramoto H, Gurpide E. Proliferation and responsiveness to estrogen of human endometrial cancer cells under serum-free culture conditions. Cancer Res. 1989. 49:3297-3301.

Holinka CF, Hata H, Kuramoto H, Gurpide E. Effects of steroid hormones and antisteroids on alkaline phosphatase activity in human endometrial cancer cells (Ishikawa line). Cancer Res. 1986a. 46:2771-2774.

Holinka CF, Hata H, Kuramoto H, Gurpide E. Responses to estradiol in a human endometrial adenocarcinoma cell line (Ishikawa). J Steroid Biochem. 1986b. 24:85-89.

Holzer G, Einhorn TA, Majeska RJ. Estrogen regulation of growth and alkaline phosphatase expression by cultured human bone marrow stromal cells. J Orthop Res. 2002. 20:281-288.

243

Hsieh TP, Sheu SY, Sun JS, Chen MH, Liu MH. Icariin isolated from Epimedium pubescens regulates osteoblasts anabolism through BMP-2, SMAD4, and Cbfa1 expression. Phytomedicine. 2010. 17:414-423.

Hu Y-C, Wu C-T, Lai J-N, Tsai Y-T. Detection of a negative correlation between prescription of Chinese herbal products containing coumestrol, genistein or daidzein and risk of subsequent endometrial cancer among tamoxifen-treated female breast cancer survivors in Taiwan between 1998 and 2008: A population-based study. Journal of ethnopharmacology. 2015. 169:356-362.

Huang G, Li M. The role of EphB4 and IGF-IR expression in breast cancer cells. Int J Clin Exp Pathol. 2015. 8:5997-6004.

Huhtinen K, Desai R, Stahle M, Salminen A, Handelsman DJ, Perheentupa A, et al. Endometrial and endometriotic concentrations of estrone and estradiol are determined by local metabolism rather than circulating levels. J Clin Endocrinol Metab. 2012. 97:4228-4235.

Hwang CS, Kwak HS, Lim HJ, Lee SH, Kang YS, Choe TB, et al. Isoflavone metabolites and their in vitro dual functions: they can act as an estrogenic agonist or antagonist depending on the estrogen concentration. J Steroid Biochem Mol Biol. 2006. 101:246-253.

Ikeda H, Pastuszko A, Ikegaki N, Kennett RH, Wilson DF. 3,4-dihydroxyphenylalanine (dopa) metabolism and retinoic acid induced differentiation in human neuroblastoma. Neurochem Res. 1994. 19:1487-1494.

Ishida Y, Killinger DW, Khalil MW, Yang K, Strutt B, Heersche JN. Expression of steroid-converting enzymes in osteoblasts derived from rat vertebrae. Osteoporos Int. 2002. 13:235-240.

Jensen EV, DeSombre ER. Estrogen-receptor interaction. Science. 1973. 182:126-134.

Jing D, Cai J, Wu Y, Shen G, Li F, Xu Q, et al. Pulsed electromagnetic fields partially preserve bone mass, microarchitecture, and strength by promoting bone formation in hindlimb-suspended rats. J Bone Miner Res. 2014. 29:2250-2261.

244

Jing YP, Liu W, Wang JX, Zhao XF. The steroid hormone 20-hydroxyecdysone via nongenomic pathway activates Ca2+/calmodulin-dependent protein kinase II to regulate gene expression. J Biol Chem. 2015. 290:8469-8481.

Johansson L, Thomsen JS, Damdimopoulos AE, Spyrou G, Gustafsson JA, Treuter E. The orphan nuclear receptor SHP inhibits agonist-dependent transcriptional activity of estrogen receptors ERalpha and ERbeta. J Biol Chem. 1999. 274:345-353.

Joshi S, Guleria R, Pan J, DiPette D, Singh US. Retinoic acid receptors and tissue-transglutaminase mediate short-term effect of retinoic acid on migration and invasion of neuroblastoma SH-SY5Y cells. Oncogene. 2006. 25:240-247.

Jouyandeh Z, Nayebzadeh F, Qorbani M, Asadi M. Metabolic syndrome and menopause. J Diabetes Metab Disord. 2013. 12:1.

Ju YI, Sone T, Ohnaru K, Tanaka K, Fukunaga M. Effect of swimming exercise on three-dimensional trabecular bone microarchitecture in ovariectomized rats. J Appl Physiol (1985). 2015. 119:990-997.

Kahlert S, Nuedling S, van Eickels M, Vetter H, Meyer R, Grohe C. Estrogen receptor alpha rapidly activates the IGF-1 receptor pathway. J Biol Chem. 2000. 275:18447-18453.

Kameda T, Mano H, Yuasa T, Mori Y, Miyazawa K, Shiokawa M, et al. Estrogen inhibits bone resorption by directly inducing apoptosis of the bone-resorbing osteoclasts. J Exp Med. 1997. 186:489-495.

Kang HK, Lee S-B, Kwon H, Sung CK, Park YI, Dong M-S. Peripubertal administration of icariin and icaritin advances pubertal development in female rats. Biomolecules & therapeutics. 2012. 20:189.

Kangas L, Harkonen P, Vaananen K, Peng Z. Effects of the selective estrogen receptor modulator ospemifene on bone in rats. Horm Metab Res. 2014. 46:27-35.

Kanis JA, Melton LJ, 3rd, Christiansen C, Johnston CC, Khaltaev N. The diagnosis of osteoporosis. J Bone Miner Res. 1994. 9:1137-1141.

Kaufman S. Tyrosine hydroxylase. Adv Enzymol Relat Areas Mol Biol. 1995.

245

70:103-220.

Khosla S, Oursler MJ, Monroe DG. Estrogen and the skeleton. Trends Endocrinol Metab. 2012. 23:576-581.

Kim SH, Park MJ. Effects of phytoestrogen on sexual development. Korean J Pediatr. 2012. 55:265-271.

Komm BS, Mirkin S. An overview of current and emerging SERMs. J Steroid Biochem Mol Biol. 2014. 143:207-222.

Kong L, Tang M, Zhang T, Wang D, Hu K, Lu W, et al. Nickel nanoparticles exposure and reproductive toxicity in healthy adult rats. Int J Mol Sci. 2014. 15:21253-21269.

Kousteni S, Bellido T, Plotkin LI, O'Brien CA, Bodenner DL, Han L, et al. Nongenotropic, sex-nonspecific signaling through the estrogen or androgen receptors: dissociation from transcriptional activity. Cell. 2001. 104:719-730.

Kousteni S, Chen JR, Bellido T, Han L, Ali AA, O'Brien CA, et al. Reversal of bone loss in mice by nongenotropic signaling of sex steroids. Science. 2002. 298:843-846.

Krum SA, Miranda-Carboni GA, Lupien M, Eeckhoute J, Carroll JS, Brown M. Unique ERalpha cistromes control cell type-specific gene regulation. Mol Endocrinol. 2008. 22:2393-2406.

Kuiper GG, Carlsson B, Grandien K, Enmark E, Haggblad J, Nilsson S, et al. Comparison of the ligand binding specificity and transcript tissue distribution of estrogen receptors alpha and beta. Endocrinology. 1997. 138:863-870.

Kuiper GG, Enmark E, Pelto-Huikko M, Nilsson S, Gustafsson JA. Cloning of a novel receptor expressed in rat prostate and ovary. Proc Natl Acad Sci U S A. 1996. 93:5925-5930.

Labos G, Trakakis E, Pliatsika P, Augoulea A, Vaggopoulos V, Basios G, et al. Efficacy and safety of DT56a compared to hormone therapy in Greek post-menopausal women. J Endocrinol Invest. 2013. 36:521-526.

Lacey DL, Timms E, Tan HL, Kelley MJ, Dunstan CR, Burgess T, et al.

246

Osteoprotegerin ligand is a cytokine that regulates osteoclast differentiation and activation. Cell. 1998. 93:165-176.

Lai JN, Wu CT, Wang JD. Prescription pattern of chinese herbal products for breast cancer in taiwan: a population-based study. Evid Based Complement Alternat Med. 2012. 2012:891893.

Lannigan DA. Estrogen receptor phosphorylation. Steroids. 2003. 68:1-9.

Lau WS, Chen WF, Chan RY, Guo DA, Wong MS. Mitogen-activated protein kinase (MAPK) pathway mediates the oestrogen-like activities of ginsenoside Rg1 in human breast cancer (MCF-7) cells. Br J Pharmacol. 2009. 156:1136-1146.

Lei Y, Gao Q, Li Y. [Study on effects of Astragalus, Angelica and their combination on vascular endothelial cell proliferation in vitro]. Zhongguo Zhong xi yi jie he za zhi Zhongguo Zhongxiyi jiehe zazhi= Chinese journal of integrated traditional and Western medicine/Zhongguo Zhong xi yi jie he xue hui, Zhongguo Zhong yi yan jiu yuan zhu ban. 2003. 23:753-756.

Leung PC, Siu WS. Herbal treatment for osteoporosis: a current review. J Tradit Complement Med. 2013. 3:82-87.

Li C, Li Q, Mei Q, Lu T. Pharmacological effects and pharmacokinetic properties of icariin, the major bioactive component in Herba Epimedii. Life sciences. 2015. 126:57-68.

LI Q, ZHOU PY, Li H, Xie JH, XUE SQ, SHANG XH. Discriminatory Analysis of Climacteric Syndrome Patients of Shen Deficiency Syndrome. Chinese Journal of Integrated Traditional and Western Medicine. 2013a. 33:1064-1068.

Li W, Frierman D, Luna B, Flaws B Diseases of the kidney & bladder: diagnosis & treatment with Chinese medicine.2006 Blue Poppy Enterprises, Inc.

Li XF, Xu H, Zhao YJ, Tang DZ, Xu GH, Holz J, et al. Icariin Augments Bone Formation and Reverses the Phenotypes of Osteoprotegerin-Deficient Mice through the Activation of Wnt/ beta -Catenin-BMP Signaling. Evid Based Complement Alternat Med. 2013b. 2013:652317.

247

Littlefield BA, Gurpide E, Markiewicz L, McKinley B, Hochberg RB. A simple and sensitive microtiter plate estrogen bioassay based on stimulation of alkaline phosphatase in Ishikawa cells: estrogenic action of delta 5 adrenal steroids. Endocrinology. 1990. 127:2757-2762.

Liu B, Geng G, Lin R, Ren C, Wu JH. Learning from estrogen receptor antagonism: structure-based identification of novel antiandrogens effective against multiple clinically relevant androgen receptor mutants. Chem Biol Drug Des. 2012. 79:300-312.

Liu WJ, Xin ZC, Xin H, Yuan YM, Tian L, Guo YL. Effects of icariin on erectile function and expression of nitric oxide synthase isoforms in castrated rats. Asian journal of andrology. 2005. 7:381-388.

Liu ZQ, Ding ZQ, Guo CY, Liang BW, Fu XJ, Chen ZB, et al. Effects of Stress Stimulation combined with Danggui Buxue Decoction for the Treatment of Comminuted Shaft Fracture of Lower Extremity: A Randomized Controlled Clinical Study. Chinese J Trad Feb Traum & Orthop. 2014. 22:21-24.

Lock M. Symptom reporting at menopause: a review of cross-cultural findings. J Br Menopause Soc. 2002. 8:132-136.

Lodish H Molecular cell biology.2008 Macmillan.

Lopez M, Lelliott CJ, Tovar S, Kimber W, Gallego R, Virtue S, et al. Tamoxifen-induced anorexia is associated with fatty acid synthase inhibition in the ventromedial nucleus of the hypothalamus and accumulation of malonyl-CoA. Diabetes. 2006. 55:1327-1336.

Love RR, Mazess RB, Barden HS, Epstein S, Newcomb PA, Jordan VC, et al. Effects of tamoxifen on bone mineral density in postmenopausal women with breast cancer. N Engl J Med. 1992. 326:852-856.

Low YL, Dunning AM, Dowsett M, Folkerd E, Doody D, Taylor J, et al. Phytoestrogen exposure is associated with circulating sex hormone levels in postmenopausal women and interact with ESR1 and NR1I2 gene variants. Cancer Epidemiol Biomarkers Prev. 2007. 16:1009-1016.

248

Ma HP, Ming LG, Ge BF, Zhai YK, Song P, Xian CJ, et al. Icariin is more potent than genistein in promoting osteoblast differentiation and mineralization in vitro. J Cell Biochem. 2011. 112:916-923.

Ma HR, Wang J, Qi HX, Gao YH, Pang LJ, Yang Y, et al. Assessment of the estrogenic activities of chickpea (Cicer arietinum L) sprout isoflavone extract in ovariectomized rats. Acta Pharmacol Sin. 2013. 34:380-386.

Ma P, Zhang S, Su X, Qiu G, Wu Z. Protective effects of icariin on cisplatin-induced acute renal injury in mice. Am J Transl Res. 2015. 7:2105-2114.

Maclennan AH, Broadbent JL, Lester S, Moore V. Oral oestrogen and combined oestrogen/progestogen therapy versus placebo for hot flushes. Cochrane Database Syst Rev. 2004. 18:CD002978.

Makarova MN, Pozharitskaya ON, Shikov AN, Tesakova SV, Makarov VG, Tikhonov VP. Effect of lipid-based suspension of Epimedium koreanum Nakai extract on sexual behavior in rats. Journal of Ethnopharmacology. 2007. 114:412-416.

Manonai J, Songchitsomboon S, Chanda K, Hong JH, Komindr S. The effect of a soy-rich diet on urogenital atrophy: a randomized, cross-over trial. Maturitas. 2006. 54:135-140.

Marieb EN, Hoehn K Human anatomy & physiology.2007 Pearson Education.

Marino M, Galluzzo P, Ascenzi P. Estrogen signaling multiple pathways to impact gene transcription. Curr Genomics. 2006. 7:497-508.

Martin PM, Horwitz KB, Ryan DS, McGuire WL. Phytoestrogen interaction with estrogen receptors in human breast cancer cells. Endocrinology. 1978. 103:1860-1867.

Matheny JB, Slyfield CR, Tkachenko EV, Lin I, Ehlert KM, Tomlinson RE, et al. Anti-resorptive agents reduce the size of resorption cavities: a three-dimensional dynamic bone histomorphometry study. Bone. 2013. 57:277-283.

Matsumori H, Hattori K, Ohgushi H, Dohi Y, Ueda Y, Shigematsu H, et al. Raloxifene: its ossification-promoting effect on female mesenchymal stem cells. J Orthop Sci. 2009. 14:640-645.

249

McDonnell DP, Norris JD. Connections and regulation of the human estrogen receptor. Science. 2002. 296:1642-1644.

Meli R, Pacilio M, Raso GM, Esposito E, Coppola A, Nasti A, et al. Estrogen and raloxifene modulate leptin and its receptor in hypothalamus and adipose tissue from ovariectomized rats. Endocrinology. 2004. 145:3115-3121.

Meng X, Pei H, Lan C. Icariin Exerts Protective Effect Against Myocardial Ischemia/Reperfusion Injury in Rats. Cell Biochem Biophys. 2015. 73:229-235.

Miksicek RJ. Estrogenic flavonoids: structural requirements for biological activity. Proc Soc Exp Biol Med. 1995. 208:44-50.

Mirkin S, Archer DF, Taylor HS, Pickar JH, Komm BS. Differential effects of menopausal therapies on the endometrium. Menopause. 2014. 21:899-908.

Mo Z, Liu M, Yang F, Luo H, Li Z, Tu G, et al. GPR30 as an initiator of tamoxifen resistance in hormone-dependent breast cancer. Breast Cancer Res. 2013. 15:R114.

Mok SK, Chen WF, Lai WP, Leung PC, Wang XL, Yao XS, et al. Icariin protects against bone loss induced by oestrogen deficiency and activates oestrogen receptor-dependent osteoblastic functions in UMR 106 cells. Br J Pharmacol. 2010. 159:939-949.

Morissette M, Al Sweidi S, Callier S, Di Paolo T. Estrogen and SERM neuroprotection in animal models of Parkinson's disease. Mol Cell Endocrinol. 2008. 290:60-69.

Murphy LC, Weitsman GE, Skliris GP, Teh EM, Li L, Peng B, et al. Potential role of estrogen receptor alpha (ERalpha) phosphorylated at Serine118 in human breast cancer in vivo. J Steroid Biochem Mol Biol. 2006. 102:139-146.

Muyan M, Callahan LM, Huang Y, Lee AJ. The ligand-mediated nuclear mobility and interaction with estrogen-responsive elements of estrogen receptors are subtype specific. J Mol Endocrinol. 2012. 49:249-266.

Nakajima K. Sex-related differences in response of plasma lipids to simvastatin: the

250

Saitama Postmenopausal Lipid Intervention Study. S-POLIS Group. Clin Ther. 1999. 21:2047-2057.

Nasu M, Sugimoto T, Kaji H, Chihara K. Estrogen modulates osteoblast proliferation and function regulated by parathyroid hormone in osteoblastic SaOS-2 cells: role of insulin-like growth factor (IGF)-I and IGF-binding protein-5. J Endocrinol. 2000. 167:305-313.

Navarro D, Acosta A, Robles E, Diaz C. Hormone profile of menopausal women in Havana. MEDICC Rev. 2012a. 14:13-15.

Navarro D, Acosta A, Robles E, Díaz C. Hormone profile of menopausal women in Havana. MEDICC review. 2012b. 14:13-15.

Nelson AW, Tilley WD, Neal DE, Carroll JS. Estrogen receptor beta in prostate cancer: friend or foe? Endocr Relat Cancer. 2014. 21:T219-234.

Nelson LR, Bulun SE. Estrogen production and action. J Am Acad Dermatol. 2001. 45:S116-124.

Nie J, Luo Y, Huang X-N, Gong Q-H, Wu Q, Shi J-S. Icariin inhibits beta-amyloid peptide segment 25–35 induced expression of β-secretase in rat hippocampus. European journal of pharmacology. 2010. 626:213-218.

Nilsson S, Makela S, Treuter E, Tujague M, Thomsen J, Andersson G, et al. Mechanisms of estrogen action. Physiol Rev. 2001. 81:1535-1565.

Nishida M, Kasahara K, Kaneko M, Iwasaki H, Hayashi K. [Establishment of a new human endometrial adenocarcinoma cell line, Ishikawa cells, containing estrogen and progesterone receptors]. Nihon Sanka Fujinka Gakkai Zasshi. 1985. 37:1103-1111.

Niu YB, Li YH, Kong XH, Zhang R, Sun Y, Li Q, et al. The beneficial effect of Radix Dipsaci total saponins on bone metabolism in vitro and in vivo and the possible mechanisms of action. Osteoporos Int. 2012. 23:2649-2660.

Nowacka-Cieciura E, Sadowska A, Pacholczyk M, Chmura A, Tronina O, Durlik M. Bone Mineral Density and Bone Turnover Markers Under Bisphosphonate Therapy Used in the First Year After Liver Transplantation. Ann Transplant. 2016. 21:241-249.

251

O'Neill K, Chen S, Brinton RD. Impact of the selective estrogen receptor modulator, raloxifene, on neuronal survival and outgrowth following toxic insults associated with aging and Alzheimer's disease. Experimental neurology. 2004a. 185:63-80.

O'Neill K, Chen S, Diaz Brinton R. Impact of the selective estrogen receptor modulator, tamoxifen, on neuronal outgrowth and survival following toxic insults associated with aging and Alzheimer's disease. Exp Neurol. 2004b. 188:268-278.

Obermeyer CM, Reher D, Saliba M. Symptoms, menopause status, and country differences: a comparative analysis from DAMES. Menopause. 2007. 14:788-797.

Ohta H, Makita K, Kawashima T, Kinoshita S, Takenouchi M, Nozawa S. Relationship between dermato-physiological changes and hormonal status in pre-, peri-, and postmenopausal women. Maturitas. 1998. 30:55-62.

Olde B, Leeb-Lundberg LM. GPR30/GPER1: searching for a role in estrogen physiology. Trends Endocrinol Metab. 2009. 20:409-416.

Organization WH. Research on the menopause in the 1990s: report of a WHO scientific group. 1996.

Ososki AL, Kennelly EJ. Phytoestrogens: a review of the present state of research. Phytother Res. 2003. 17:845-869.

Paech K, Webb P, Kuiper GG, Nilsson S, Gustafsson J, Kushner PJ, et al. Differential ligand activation of estrogen receptors ERalpha and ERbeta at AP1 sites. Science. 1997. 277:1508-1510.

Pan Y, Kong L, Xia X, Zhang W, Xia Z, Jiang F. Antidepressant-like effect of icariin and its possible mechanism in mice. Pharmacology Biochemistry and Behavior. 2005. 82:686-694.

Pang WY, Wang XL, Mok SK, Lai WP, Chow HK, Leung PC, et al. Naringin improves bone properties in ovariectomized mice and exerts oestrogen-like activities in rat osteoblast-like (UMR-106) cells. Br J Pharmacol. 2010. 159:1693-1703.

Parazzini F. Determinants of age at menopause in women attending menopause clinics

252 in Italy. Maturitas. 2007. 56:280-287.

Peng W, Adams J, Hickman L, Sibbritt DW. Complementary/alternative and conventional medicine use amongst menopausal women: results from the Australian Longitudinal Study on Women's Health. Maturitas. 2014. 79:340-342.

Penttinen P, Jaehrling J, Damdimopoulos AE, Inzunza J, Lemmen JG, van der Saag P, et al. Diet-derived polyphenol metabolite enterolactone is a tissue-specific estrogen receptor activator. Endocrinology. 2007. 148:4875-4886.

Pesonen S. [On the effect of gonadotropins and corticotropin on the secretion of estrogens and androgens by the adrenal glands and gonads]. Geburtshilfe Frauenheilkd. 1960. 20:764-769.

Peterson GM, Naunton M, Tichelaar LK, Gennari L. Lasofoxifene: selective estrogen receptor modulator for the prevention and treatment of postmenopausal osteoporosis. Ann Pharmacother. 2011. 45:499-509.

Pfeiffer E, Graf E, Gerstner S, Metzler M. Stimulation of estradiol glucuronidation: a protective mechanism against estradiol-mediated carcinogenesis? Mol Nutr Food Res. 2006. 50:385-389.

Picherit C, Coxam V, Bennetau-Pelissero C, Kati-Coulibaly S, Davicco MJ, Lebecque P, et al. Daidzein is more efficient than genistein in preventing ovariectomy-induced bone loss in rats. J Nutr. 2000. 130:1675-1681.

Pinkerton JV, Thomas S. Use of SERMs for treatment in postmenopausal women. J Steroid Biochem Mol Biol. 2014. 142:142-154.

Powles TJ, Ashley S, Tidy A, Smith IE, Dowsett M. Twenty-year follow-up of the Royal Marsden randomized, double-blinded tamoxifen breast cancer prevention trial. J Natl Cancer Inst. 2007. 99:283-290.

Powles TJ, Hickish T, Kanis JA, Tidy A, Ashley S. Effect of tamoxifen on bone mineral density measured by dual-energy x-ray absorptiometry in healthy premenopausal and postmenopausal women. J Clin Oncol. 1996. 14:78-84.

Presgraves SP, Ahmed T, Borwege S, Joyce JN. Terminally differentiated SH-SY5Y

253 cells provide a model system for studying neuroprotective effects of dopamine agonists. Neurotox Res. 2004. 5:579-598.

Ragonese P, D'Amelio M, Salemi G, Aridon P, Gammino M, Epifanio A, et al. Risk of Parkinson disease in women: effect of reproductive characteristics. Neurology. 2004. 62:2010-2014.

Rahmani P, Morin S. Prevention of osteoporosis-related fractures among postmenopausal women and older men. Cmaj. 2009. 181:815-820.

Ramalho-Ferreira G, Faverani LP, Prado FB, Garcia IR, Jr., Okamoto R. Raloxifene enhances peri-implant bone healing in osteoporotic rats. Int J Oral Maxillofac Surg. 2015. 44:798-805.

Rankinen T, Zuberi A, Chagnon YC, Weisnagel SJ, Argyropoulos G, Walts B, et al. The human obesity gene map: the 2005 update. Obesity. 2006. 14:529-644.

Register TC, Jayo MJ, Anthony MS. Soy phytoestrogens do not prevent bone loss in postmenopausal monkeys. J Clin Endocrinol Metab. 2003. 88:4362-4370.

Revankar CM, Cimino DF, Sklar LA, Arterburn JB, Prossnitz ER. A transmembrane intracellular estrogen receptor mediates rapid cell signaling. Science. 2005. 307:1625-1630.

Ribeiro JR, Freiman RN. Estrogen signaling crosstalk: Implications for endocrine resistance in ovarian cancer. J Steroid Biochem Mol Biol. 2014. 143:160-173.

Rocca WA, Bower JH, Maraganore DM, Ahlskog JE, Grossardt BR, de Andrade M, et al. Increased risk of parkinsonism in women who underwent oophorectomy before menopause. Neurology. 2008. 70:200-209.

Rocca WA, Grossardt BR, Shuster LT. Oophorectomy, estrogen, and dementia: a 2014 update. Mol Cell Endocrinol. 2014. 389:7-12.

Ronghe A, Chatterjee A, Bhat NK, Padhye S, Bhat HK. Tamoxifen synergizes with 4-(E)-{(4-hydroxyphenylimino)-methylbenzene, 1,2-diol} and 4-(E)-{(p-tolylimino)-methylbenzene-1,2-diol}, novel azaresveratrol analogs, in inhibiting the proliferation of breast cancer cells. Oncotarget. 2016. 16:10106.

254

Ruan W, Kleinberg DL. Insulin-like growth factor I is essential for terminal end bud formation and ductal morphogenesis during mammary development. Endocrinology. 1999. 140:5075-5081.

Schwartz AG, Ray RM, Cote ML, Abrams J, Sokol RJ, Hendrix SL, et al. Hormone Use, Reproductive History, and Risk of Lung Cancer: The Women's Health Initiative Studies. J Thorac Oncol. 2015. 10:1004-1013.

Seidlova-Wuttke D, Jarry H, Wuttke W. Effects of estradiol benzoate, raloxifen and an ethanolic extract of Cimicifuga racemosa in nonclassical estrogen regulated organs of ovariectomized rats. Planta Med. 2009. 75:1279-1285.

Shanle EK, Xu W. Selectively targeting estrogen receptors for cancer treatment. Adv Drug Deliv Rev. 2010. 62:1265-1276.

Shapiro CL, Recht A. Side effects of adjuvant treatment of breast cancer. N Engl J Med. 2001. 344:1997-2008.

Shevde NK, Bendixen AC, Dienger KM, Pike JW. Estrogens suppress RANK ligand-induced osteoclast differentiation via a stromal cell independent mechanism involving c-Jun repression. Proc Natl Acad Sci U S A. 2000. 97:7829-7834.

Sievert LL, Obermeyer CM, Saliba M. Symptom groupings at midlife: cross-cultural variation and association with job, home, and life change. Menopause. 2007. 14:798-807.

Simpson ER. Aromatization of androgens in women: current concepts and findings. Fertil Steril. 2002. 77 Suppl 4:S6-10.

Singh SP, Raju K, Ali MM, Kohli K, Jain GK. Reduced Bioavailability of Tamoxifen and its Metabolite 4‐Hydroxytamoxifen After Oral Administration with Biochanin A

(an Isoflavone) in Rats. Phytotherapy Research. 2012. 26:303-307.

Smith CL, O'Malley BW. Coregulator function: a key to understanding tissue specificity of selective receptor modulators. Endocr Rev. 2004. 25:45-71.

255

Smith L, Brannan RA, Hanby AM, Shaaban AM, Verghese ET, Peter MB, et al. Differential regulation of oestrogen receptor beta isoforms by 5' untranslated regions in cancer. J Cell Mol Med. 2010. 14:2172-2184.

Song RX, Zhang Z, Chen Y, Bao Y, Santen RJ. Estrogen signaling via a linear pathway involving insulin-like growth factor I receptor, matrix metalloproteinases, and epidermal growth factor receptor to activate mitogen-activated protein kinase in MCF-7 breast cancer cells. Endocrinology. 2007. 148:4091-4101.

Song SH, Wang D, Mo YY, Ding C, Shang P. [Antiosteoporotic effects of naringenin on ovariectomy-induced osteoporosis in rat]. Yao Xue Xue Bao. 2015. 50:154-161.

Soules MR, Sherman S, Parrott E, Rebar R, Santoro N, Utian W, et al. Executive summary: Stages of Reproductive Aging Workshop (STRAW) Park City, Utah, July, 2001. Menopause. 2001. 8:402-407.

Spiro AS, Khadem S, Jeschke A, Marshall RP, Pogoda P, Ignatius A, et al. The SERM raloxifene improves diaphyseal fracture healing in mice. J Bone Miner Metab. 2013. 31:629-636.

Stournaras C, Gravanis A, Margioris AN, Lang F. The actin cytoskeleton in rapid steroid hormone actions. Cytoskeleton (Hoboken). 2014. 71:285-293.

Stygar D, Muravitskaya N, Eriksson B, Eriksson H, Sahlin L. Effects of SERM (selective estrogen receptor modulator) treatment on growth and proliferation in the rat uterus. Reprod Biol Endocrinol. 2003. 1:40.

Sun L, Peng Y, Sharrow AC, Iqbal J, Zhang Z, Papachristou DJ, et al. FSH directly regulates bone mass. Cell. 2006. 125:247-260.

Sundermann E, Gilbert PE, Murphy C. Estrogen and performance in recognition memory for olfactory and visual stimuli in females diagnosed with Alzheimer's disease. J Int Neuropsychol Soc. 2006. 12:400-404.

Sundstrom SA, Komm B, Ponce-de-Leon H, Yi Z, Teuscher C, Lyttle C. Estrogen regulation of tissue-specific expression of complement C3. Journal of Biological Chemistry. 1989. 264:16941-16947.

256

Tan X, S., Hong L. Influence of Danggui Buxue Tang on serum level of estradiol and FSH as well as histology of uterus of OVX rats. Chinese Journal of Basic in Traditional Chinese Medicine. 2010. 16:1008-1009.

Teng F, Zeng YY, Huang XY, Yang Z, Yao ML, Song B, et al. [Effect of icariin on intermediate and advanced activation of murine T lymphocytes in vitro]. Xi Bao Yu Fen Zi Mian Yi Xue Za Zhi. 2008. 24:1059-1061.

Tham DM, Gardner CD, Haskell WL. Clinical review 97: Potential health benefits of dietary phytoestrogens: a review of the clinical, epidemiological, and mechanistic evidence. J Clin Endocrinol Metab. 1998. 83:2223-2235.

Thomas RS, Sarwar N, Phoenix F, Coombes RC, Ali S. Phosphorylation at serines 104 and 106 by Erk1/2 MAPK is important for estrogen receptor-alpha activity. J Mol Endocrinol. 2008. 40:173-184.

Tian DS, Liu JL, Xie MJ, Zhan Y, Qu WS, Yu ZY, et al. Tamoxifen attenuates inflammatory-mediated damage and improves functional outcome after spinal cord injury in rats. J Neurochem. 2009. 109:1658-1667.

Tsai C-L, Wu H-M, Lin C-Y, Lin Y-J, Chao A, Wang T-H, et al. Estradiol and tamoxifen induce cell migration through GPR30 and activation of focal adhesion kinase (FAK) in endometrial cancers with low or without nuclear estrogen receptor α (ERα). PLoS One. 2013a. 8:e72999.

Tsai CL, Wu HM, Lin CY, Lin YJ, Chao A, Wang TH, et al. Estradiol and tamoxifen induce cell migration through GPR30 and activation of focal adhesion kinase (FAK) in endometrial cancers with low or without nuclear estrogen receptor alpha (ERalpha). PLoS One. 2013b. 8:e72999.

Turner RT, Wakley GK, Hannon KS, Bell NH. Tamoxifen prevents the skeletal effects of ovarian hormone deficiency in rats. J Bone Miner Res. 1987. 2:449-456.

Umland EM, Cauffield JS, Kirk JK, Thomason TE. Phytoestrogens as therapeutic alternatives to traditional hormone replacement in postmenopausal women. Pharmacotherapy. 2000. 20:981-990.

Urano T, Tohda C. Icariin improves memory impairment in Alzheimer's disease model

257 mice (5xFAD) and attenuates amyloid beta-induced neurite atrophy. Phytother Res. 2010. 24:1658-1663. van de Schans MG, Vincken JP, de Waard P, Hamers AR, Bovee TF, Gruppen H. Glyceollins and dehydroglyceollins isolated from soybean act as SERMs and ER subtype-selective phytoestrogens. J Steroid Biochem Mol Biol. 2016. 156:53-63. van Duursen MB, Smeets EE, Rijk JC, Nijmeijer SM, van den Berg M. Phytoestrogens in menopausal supplements induce ER-dependent cell proliferation and overcome breast cancer treatment in an in vitro breast cancer model. Toxicol Appl Pharmacol. 2013. 269:132-140.

Villaseca P. Non-estrogen conventional and phytochemical treatments for vasomotor symptoms: what needs to be known for practice. Climacteric. 2012. 15:115-124.

Vrtacnik P, Ostanek B, Mencej-Bedrac S, Marc J. The many faces of estrogen signaling. Biochem Med (Zagreb). 2014. 24:329-342.

Wade GN, Heller HW. Tamoxifen mimics the effects of estradiol on food intake, body weight, and body composition in rats. Am J Physiol. 1993. 264:R1219-1223.

Wang C, Kurzer MS. Effects of phytoestrogens on DNA synthesis in MCF-7 cells in the presence of estradiol or growth factors. Nutr Cancer. 1998. 31:90-100.

Wang C, Prossnitz ER, Roy SK. Expression of G protein-coupled receptor 30 in the hamster ovary: differential regulation by gonadotropins and steroid hormones. Endocrinology. 2007. 148:4853-4864.

Wang CC, Cheng KF, Lo WM, Law C, Li L, Leung PC, et al. A randomized, double-blind, multiple-dose escalation study of a Chinese herbal medicine preparation (Dang Gui Buxue Tang) for moderate to severe menopausal symptoms and quality of life in postmenopausal women. Menopause. 2013. 20:223-231.

Wang H, Dwyer-Lindgren L, Lofgren KT, Rajaratnam JK, Marcus JR, Levin-Rector A, et al. Age-specific and sex-specific mortality in 187 countries, 1970-2010: a systematic analysis for the Global Burden of Disease Study 2010. Lancet. 2012. 380:2071-2094.

258

Weitzmann MN, Roggia C, Toraldo G, Weitzmann L, Pacifici R. Increased production of IL-7 uncouples bone formation from bone resorption during estrogen deficiency. J Clin Invest. 2002. 110:1643-1650.

Wickerham DL, Fisher B, Wolmark N, Bryant J, Costantino J, Bernstein L, et al. Association of tamoxifen and uterine sarcoma. J Clin Oncol. 2002. 20:2758-2760.

WIDE L, WIDE M. Higher Plasma Disappearance Rate in the Mouse for Pituitary Follicle-Stimulating Hormone of Young Women Compared to that of Men and Elderly Women*. The Journal of Clinical Endocrinology & Metabolism. 1984. 58:426-429.

Williams C, Edvardsson K, Lewandowski SA, Strom A, Gustafsson JA. A genome-wide study of the repressive effects of estrogen receptor beta on estrogen receptor alpha signaling in breast cancer cells. Oncogene. 2008. 27:1019-1032.

Won Jeong K, Chodankar R, Purcell DJ, Bittencourt D, Stallcup MR. Gene-specific patterns of coregulator requirements by estrogen receptor-alpha in breast cancer cells. Mol Endocrinol. 2012. 26:955-966.

Wong KC, Pang WY, Wang XL, Mok SK, Lai WP, Chow HK, et al. Drynaria fortunei-derived total flavonoid fraction and isolated compounds exert oestrogen-like protective effects in bone. Br J Nutr. 2013. 110:475-485.

Wood CE, Register TC, Cline JM. Soy isoflavonoid effects on endogenous estrogen metabolism in postmenopausal female monkeys. Carcinogenesis. 2007. 28:801-808.

Wood CE, Register TC, Franke AA, Anthony MS, Cline JM. Dietary soy isoflavones inhibit estrogen effects in the postmenopausal breast. Cancer Res. 2006. 66:1241-1249.

Wu J, Du J, Xu C, Le J, Xu Y, Liu B, et al. Icariin attenuates social defeat-induced down-regulation of glucocorticoid receptor in mice. Pharmacology Biochemistry and Behavior. 2011. 98:273-278.

Wu ZX, Wang XH, Deng HZ. [Effects of Yi Fu Ning soft gelatin capsules on reproductive endocrine-immune function of ovariectomized rats]. Zhong Yao Cai. 2009. 32:1083-1086.

259

Xiao HH, Fung CY, Mok SK, Wong KC, Ho MX, Wang XL, et al. Flavonoids from Herba epimedii selectively activate estrogen receptor alpha (ERalpha) and stimulate ER-dependent osteoblastic functions in UMR-106 cells. J Steroid Biochem Mol Biol. 2014. 143:141-151.

Xie QF, Xie JH, Dong TT, Su JY, Cai DK, Chen JP, et al. Effect of a derived herbal recipe from an ancient Chinese formula, Danggui Buxue Tang, on ovariectomized rats. J Ethnopharmacol. 2012. 144:567-575.

Xu P, Cao X, He Y, Zhu L, Yang Y, Saito K, et al. Estrogen receptor-alpha in medial amygdala neurons regulates body weight. J Clin Invest. 2015. 125:2861-2876.

Xu Y, Ma XP, Ding J, Liu ZL, Song ZQ, Liu HN, et al. Treatment with qibaomeiran, a kidney-invigorating Chinese herbal formula, antagonizes estrogen decline in ovariectomized rats. Rejuvenation Res. 2014. 17:372-381.

Yager JD. Mechanisms of estrogen carcinogenesis: The role of E2/E1-quinone metabolites suggests new approaches to preventive intervention--A review. Steroids. 2015. 99:56-60.

Yan FF, Liu Y, Liu YF, Zhao YX. Herba Epimedii water extract elevates estrogen level and improves lipid metabolism in postmenopausal women. Phytother Res. 2008. 22:1224-1228.

Yang L, Lu D, Guo J, Meng X, Zhang G, Wang F. Icariin from Epimedium brevicornum Maxim promotes the biosynthesis of estrogen by aromatase (CYP19). J Ethnopharmacol. 2013. 145:715-721.

Yang P, Guan YQ, Li YL, Zhang L, Zhang L, Li L. Icariin promotes cell proliferation and regulates gene expression in human neural stem cells in vitro. Mol Med Rep. 2016. 14:1316-1322.

YoungLai EV. The bioconversion of androstenedione and testosterone by rabbit granulosa cells and the effect of ovine LH on thecal aromatization. Steroids Lipids Res. 1973. 4:11-16.

Zhang CZ, Wang SX, Zhang Y, Chen JP, Liang XM. In vitro estrogenic activities of Chinese medicinal plants traditionally used for the management of menopausal

260 symptoms. J Ethnopharmacol. 2005. 98:295-300.

Zhang D, Zhang J, Fong C, Yao X, Yang M. Herba epimedii flavonoids suppress osteoclastic differentiation and bone resorption by inducing G2/M arrest and apoptosis. Biochimie. 2012a. 94:2514-2522.

Zhang G, Qin L, Shi Y. Epimedium-derived phytoestrogen flavonoids exert beneficial effect on preventing bone loss in late postmenopausal women: a 24-month randomized, double-blind and placebo-controlled trial. J Bone Miner Res. 2007. 22:1072-1079.

Zhang X, Meng J, Wang ZY. A switch role of Src in the biphasic EGF signaling of ER-negative breast cancer cells. PLoS One. 2012b. 7:e41613.

Zhang Y-Z, Xu F, Yi T, Zhang J-Y, Xu J, Tang Y-N, et al. Chemical profile analysis and comparison of two versions of the classic TCM formula Danggui Buxue Tang by HPLC-DAD-ESI-IT-TOF-MSn. Molecules. 2014. 19:5650-5673.

Zheng KY-z, Choi RC-y, Guo AJ-y, Bi CW-c, Zhu KY, Du CY-q, et al. The membrane permeability of Astragali Radix-derived formononetin and calycosin is increased by Angelicae Sinensis Radix in Caco-2 cells: a synergistic action of an ancient herbal decoction Danggui Buxue Tang. Journal of pharmaceutical and biomedical analysis. 2012a. 70:671-679.

Zheng KY, Choi RC, Guo AJ, Bi CW, Zhu KY, Du CY, et al. The membrane permeability of Astragali Radix-derived formononetin and calycosin is increased by Angelicae Sinensis Radix in Caco-2 cells: a synergistic action of an ancient herbal decoction Danggui Buxue Tang. J Pharm Biomed Anal. 2012b. 70:671-679.

Zheng KY, Choi RC, Xie HQ, Cheung AW, Guo AJ, Leung K-w, et al. The expression of erythropoietin triggered by Danggui Buxue Tang, a Chinese herbal decoction prepared from Radix Astragali and Radix Angelicae Sinensis, is mediated by the hypoxia-inducible factor in cultured HEK293T cells. Journal of Ethnopharmacology. 2010. 132:259-267.

Zierau O, Zheng KY, Papke A, Dong TT, Tsim KW, Vollmer G. Functions of danggui buxue tang, a chinese herbal decoction containing astragali radix and angelicae sinensis radix, in uterus and liver are both estrogen receptor-dependent

261 and-independent. Evidence-Based Complementary and Alternative Medicine. 2014. 2014.

Zujewski J. Selective estrogen receptor modulators (SERMs) and retinoids in breast cancer chemoprevention. Environ Mol Mutagen. 2002. 39:264-270.

262