Bioactivity of Naringenin in Metabolic Dysregulation and Obesity-Associated Breast

Cancer in a Mouse Model of Postmenopause

DISSERTATION

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

Jia-Yu Ke

Ohio State University Nutrition Graduate Program

The Ohio State University

2015

Dissertation Committee:

Dr. Martha A. Belury, Advisor

Dr. Earl Harrison

Dr. Kichoon Lee

Dr. Lisa D Yee

! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! Copyright by

Jia-Yu Ke

2015

! ! ! ! ! Abstract

Loss of ovarian function after menopause is associated with accumulation of body fat in the waist region. These changes in body composition increase the risk of developing central obesity, metabolic syndrome, and other chronic diseases.

Additionally, breast cancer is the second leading cause of cancer deaths in women, with the majority occurring in women past the age of menopause. Obesity is associated with increased risk of postmenopausal breast cancer and the underlying mechanism(s) likely involves obesity-related metabolic dysregulation. Therefore, improvement of metabolic status may be a useful approach to decrease the risk of breast cancer.

Naringenin is a flavonone found in citrus fruits and tomatoes. It possesses antitumor properties as well as ameliorates obesity-associated metabolic dysregulation.

Therefore, we hypothesized that dietary naringenin not only improves metabolic disturbances resulting from loss of ovarian function but also ameliorates metabolic disturbances associated with postmenopausal obesity. Furthermore, we hypothesized that naringenin inhibits mammary tumor growth induced by postmenopausal obesity.

The first objective of this research was to evaluate the effect of naringenin on metabolic changes resulting from loss of ovarian function. Ovariectomized C57BL/6J female mice were fed a control diet (10% calories from fat) for 11 weeks. Mice were either continued on the control diet or switched to the control diet supplemented with 3% naringenin for the next 11 weeks. Ovariectomized mice, even fed a control diet, exhibited

ii elevated fasting glucose levels and increased adiposity. Plasma leptin and leptin mRNA in adipose depots as well as adipose tissue inflammation were decreased in mice supplemented with naringenin. We also observed that mice fed a naringenin diet had less hepatic lipid accumulation with corresponding alterations of hepatic gene expression associated with de novo lipogenesis, fatty acid oxidation, and gluconeogenesis. In summary, dietary naringenin attenuated many of the metabolic disturbances associated with ovariectomy in female mice.

The second objective of this research was to examine the effect of naringenin on metabolic disturbances in a mouse model of postmenopausal obesity. First, we measured naringenin concentrations in plasma, liver, perigonadal and subcutaneous adipose tissues, and muscle of ovariectomized C57BL/6J female mice after 11 weeks of naringenin supplementation. Naringenin accumulated 5-12 times more in mice fed a 3% naringenin diet than in mice fed a 1% naringenin diet. Then we fed ovariectomized mice a high-fat diet (60% kcal fat) for 11 weeks to induce obesity and half of the mice were then supplemented with 3% naringenin for another 11 weeks. Supplementation with dietary naringenin decreased weight gain, hyperglycemia, and intra-abdominal adiposity.

Naringenin-fed mice exhibited elevated locomotor activity, maintained muscle mass, and reduced muscle diacylglycerol content. Real-time PCR analysis in skeletal muscle revealed decreased mRNA level for genes involved in de novo lipogenesis, lipolysis, and triglyceride synthesis/storage. In conclusion, naringenin supplementation attenuated metabolic dysregulation in obese ovariectomized mice.

iii

The third objective of this research was to evaluate the effects of naringenin on the growth of breast tumors in a mouse model of postmenopausal obesity.

Ovariectomized mice were fed a high fat diet for 3 weeks to induce obesity and then supplemented with 1% or 3 % naringenin, or metformin as a positive control. After 2 weeks of experimental diets, E0771 murine breast cancer cells were inoculated into one mammary fat pad and the tumor size was monitored for 3 weeks. Naringenin supplementation significantly decreased body weight, adipose depot mass, as well as mRNA expression of inflammatory cytokines in both mammary and perigonadal adipose tissues. Naringenin supplementation suppressed tumor growth in the early stage but final tumor weight was not significantly different from the high fat group. Metformin reduced tumor growth and weight, without affecting body weight, tissue weights, and adipose tissue inflammation. Collectively, our data demonstrated that naringenin and metformin alter mammary tumorigenesis via different mechanisms.

Overall, these experiments suggest that naringenin supplementation may correct metabolic dysregulation induced by excess caloric consumption and/or loss of ovarian function.

! ! ! ! ! Acknowledgments

I would like to express my sincerest gratitude to my advisor, Dr. Martha Belury, for providing me the opportunity to conduct this research and numerous opportunities to learn new skills and collaborate with other researchers, for encouraging me to challenge myself academically, and for helping me with many fellowship/grant applications, abstracts, posters, and papers throughout the past 5 years of my Ph.D. study. I would like to extend my gratitude to Dr. Lisa Yee for teaching me everything about the breast cancer and for helping me with fellowship applications, abstracts, posters, and papers. I also would like to sincerely thank to other members of my committee, Dr. Earl Harrison and

Dr. Kichoon Lee for their time and advice from start to finish. This research could not have been done without all the collaborators, Dr. Kimerly Powell, Dr. Rebecca Andridge,

Dr. Steven Schwartz, Dr. Ken Riedl, Dr. Santosh K. Maurya, Dr. Muthu Periasamy, and

Shana Straka. I would like to thank them for sharing their time and extensive knowledge and expertise.

I would like to thank my colleagues Min, Mike, Sarah, Queenie, Josephine,

Taylor, Sharon, and Brad for helping me with experiments. Without them, I would not have accomplished all the studies in my 5 years. I would like to especially thank Kara,

Essam, and Rachel for the unwavering support, and their suggestions, help with experiments, and friendship have been invaluable throughout my Ph.D. study. v

Additionally, I am grateful to have join OSUN and would like to thank to OSUN faculty and students, especially Dr. Irene Hatsu, Kom Kamonpatana, Rumana Yasmeen, Han-Yi

Lin, Shirley Tan, Yi Guo, Eunice Mah, Jinhui Li, Yi-Hsuan Liu, Leanna Perez, and

Kellie Weinhold, for their support and suggestions.

Thank you to the organizations that contributed financially to support my education and research: The Graduate School University Fellowship, College of

Education and Human Ecology (EHE) Dissertation Research Fellowships, Leta Gigax

Duhamel Scholarship, Helen Sells CLARKSON Memorial Scholarship, Food Innovation

Center (FIC) Doctoral Research Grant, SEEDS: The OARDC Research Enhancement

Competitive Grants Program, Russell Klein Memorial Award, and travel awards from

OSUN, FIC, EHE, and Department of Human Sciences.

Finally, I would like to thank my Taiwanese friends in Columbus for bringing me joy and laughter. I am extremely grateful to Fabi and Kara’s family, my husband, my sister and my parents for countless encouragement and unconditional support.

Vita

2006 ...... B.S. Biochemical Science & Technology,

National Taiwan University, Taipei, Taiwan

2008 ...... M.S. Molecular Medicine, National Taiwan

University, Taipei, Taiwan

2010 to present ...... The Ohio State University Ph.D. Program in

Nutrition, The Ohio State University,

Columbus, OH, USA

! ! Publications

Abstracts

Ke JY, Hsiao YH, Straka SR, Yee LD, Belury MA. Comparison of the citrus flavonoid naringenin and metformin for effects on breast cancer in obese ovariectomized mice. Proceedings: AACR Annual Meeting 2015; April 18-22, 2015; Philadelphia, PA

Cotton B, Ke JY, Hsiao YH, Banh T, Belury MA. Naringenin attenuates muscle loss in ovariectomized mice fed a high-fat diet. Poster Presentation, EHE Student Research Forum, The Ohio State University, Columbus, OH, Feb 2015. Peer-reviewed conference.

vii

Ke JY, Tian M, Kliewer KL, Schwartz SJ, Reidl KM, Tsai SY, Yee LD, Belury MA Evaluation of a citrus flavonoid as a chemopreventive agent against breast cancer. Proceedings: AACR Annual Meeting 2014; April 5-9, 2014; San Diego, CA

Ke JY, Schwartz SJ, Riedl KM, Yee LD, Kliewer K, and Belury MA Accumulation of dietary naringenin and metabolites in mice. FASEB J April 9, 2013 27:636.2

Ke JY, Tian M, Kliewer KL, Belury MA The effect of naringenin on the phosphorylation of AMPK in diet-induced obese mice. FASEB J March 29, 2012 26:818.2 Research Publications

Research Publications

Ke JY, Kliewer KL, Hamad E, Cole RM, Powell KA, Andridge RR, Straka SR, Yee LD, Belury MA. The flavonoid, naringenin, decreases adipose tissue mass and attenuates ovariectomy-associated metabolic disturbances in mice. Nutrition & Metabolism 2015, 12:1

Kliewer KL, Ke JY, Lee HY, Stout MB, Cole RM, Samuel VT, Shulman GI, Belury MA Short-term food restriction followed by controlled refeeding promotes gorging behavior, enhances fat deposition, and diminishes insulin sensitivity in mice J. Nutr. Biochem. 2015 Mar

Kliewer KL, Ke JY, Tian M, Cole RM, Andridge RR, Belury MA Adipose tissue lipolysis and energy metabolism in early cancer cachexia in mice. Cancer Biol Ther. 2014 Dec 2:0.

Han DS, Huang HP, Wang TG, Hung MY, Ke JY, Chang KT, Chang HY, Ho YP, Hsieh WY, Yang WS. Transcription activation of myostatin by trichostatin A in differentiated C2C12 myocytes via ASK1-MKK3/4/6-JNK and p38 mitogen-activated protein kinase pathways. J Cell Biochem. 2010 Oct 15;111(3):564-73

Fields of Study

Major Field: Nutrition

viii

Table of Contents

ABSTRACT ...... ii ACKNOWLEDGMENTS ...... v VITA ...... vii TABLE OF CONTENTS ...... ix LIST OF TABLES ...... xi LIST OF FIGURES ...... xii CHAPTER 1. INTRODUCTION ...... 1 1.1 Introduction ...... 1 1.2 Aims ...... 3 CHAPTER 2. LITERATURE REVIEW ...... 5 2.1 Menopause and metabolic changes ...... 5 2.1.1 Definitions of menopause ...... 5 2.1.2 Menopause and metabolic changes: Findings from human studies ...... 6 2.1.3 The metabolic effect of hormone replacement therapy in postmenopausal women .... 10 2.1.4 Menopause and metabolic changes: Findings from animal models ...... 12 2.2 Obesity and postmenopausal breast cancer ...... 22 2.2.1 Breast cancer ...... 22 2.2.2 Breast cancer subtypes ...... 23 2.2.3 The prevalence, incidence, and mortality of breast cancer ...... 27 2.2.4 Obesity and postmenopausal breast cancer ...... 27 2.2.5 Metformin and breast cancer ...... 31 2.3 Naringenin and its bioactivities against diet-induced metabolic dysregulation and breast cancer ...... 33 2.3.1 Naringenin ...... 33 2.3.2 Absorption, bioavailability, and metabolism of naringenin ...... 34 2.3.3 The effects of naringenin on diet-induced metabolic dysregulation ...... 36 2.3.4 The effects of naringenin on breast cancer ...... 42 CHAPTER 3. THE FLAVONOID, NARINGENIN, DECREASES ADIPOSE TISSUE MASS AND ATTENUATES OVARIECTOMY-ASSOCIATED METABOLIC DISTURBANCES IN MICE ...... 43 3.1 Abstract ...... 44 3.2 Introduction ...... 46 3.3 Materials and Methods ...... 48 3.4 Results ...... 54 ix

3.5 Discussion ...... 65 3.6 Acknowledgements ...... 71 CHAPTER 4. CITRUS FLAVONOID, NARINGENIN, INCREASES LOCOMOTOR ACTIVITY AND REDUCES DIACYLGLYCEROL ACCUMULATION IN SKELETAL MUSCLE OF OBESE OVARIECTOMIZED MICE ...... 72 4.1 Abstract ...... 73 4.2 Introduction ...... 74 4.3 Materials and Methods ...... 76 4.4 Results ...... 83 4.5 Discussion ...... 99 4.6 Acknowledgements ...... 104 CHAPTER 5. THE EFFECT OF CITRUS FLAVONOID NARINGENIN AND METFORMIN ON THE GROWTH OF BREAST CANCER CELLS IN OBESE OVARIECTOMIZED MICE ...... 105 5.1 Abstract ...... 106 5.2 Introduction ...... 107 5.3 Materials and methods ...... 109 5.4 Results ...... 116 5.5 Discussion ...... 133 5.6 Acknowledgments ...... 139 CHAPTER 6 ...... 140 EPILOGUE ...... 140 LIST OF REFERENCES ...... 143 APPENDIX A. ILLUSTRATION OF DIFFERENT FAT DEPOTS ANALYZED BY MAGNETIC RESONANCE IMAGING (MRI) ...... 170

List of Tables

Table 1. Effect of loss of ovarian hormones on body composition and metabolism (Rat)16! Table 2. Effect of loss of ovarian hormones on body composition and metabolism (Mouse) ...... 19! Table 3. Effects of naringin and naringenin on diet-induced metabolic dysregulation .... 39! Table 4. Composition of experimental diets ...... 49! Table 5. Real-time PCR primers and probes of Taqman gene expression assay ...... 53! Table 6. Effect of dietary naringenin on tissue mass in OVX C57BL/6J female mice .... 58! Table 7. Effect of dietary naringenin on feed efficiency and tissue mass in obese ovariectomized C57BL/6J female mice ...... 87! Table 8. Plasma insulin levels and tissue weights ...... 123!

List of Figures

Figure 1. Female breast anatomy (86)...... 23! Figure 2.Chemical structure of naringenin aglycone. (162) ...... 34! Figure 3. Effects of dietary naringenin on caloric intake, body weight, and metabolic measurements...... 55! Figure 4. Effects of dietary naringenin on adiposity, plasma adipokines, and adipose tissue gene expression...... 59! Figure 5. Effects of dietary naringenin on plasma lipids and lipid accumulation in liver and muscle...... 61! Figure 6. Effects of dietary naringenin on gene expression in liver and muscle...... 63! Figure 7. Naringenin accumulation in plasma and tissues after 11 weeks of supplementation...... 84! Figure 8. Effect of naringenin supplementation on body weight and caloric intake...... 86! Figure 9. Effect of naringenin supplementation on indirect calorimetry and locomotor activity...... 89! Figure 10. Effect of naringenin supplementation on adiposity, fat distribution, and MCP/Ccl2 mRNA expression in perigonadal adipose tissue...... 91! Figure 11. Effect of naringenin supplementation on plasma analysis...... 92! Figure 12. Effect of naringenin supplementation on hepatic and muscle lipid profiles. .. 93! Figure 13. Effect of naringenin supplementation on mRNA expression in muscle...... 96! Figure 14. Effect of naringenin supplementation on AMPK and Akt activity in muscle. 98! Figure 15. Naringenin decreases cell viability of E0771 breast cancer cells...... 117! Figure 16. Naringenin increased the accumulation of sub-G1 cells...... 118! Figure 17. The effect of naringenin on AMPK and Akt activity, and CyclinD1 protein expression...... 119! Figure 18. The effect of naringenin and metformin on body weight cumulative caloric intake and fasting glucose...... 122! Figure 19. Naringenin accumulation in plasma and tissues after 5 weeks of supplementation...... 125! Figure 20. The effect of naringenin and metformin on mRNA expression in perigonadal and mammary adipose tissues...... 127! Figure 21. The effect of naringenin and metformin on tumor growth and tumor weight...... 129! Figure 22. Effects of naringenin and metformin on AMPK activity downstream signaling...... 131! xii

Figure 23. Illustration of different fat depots analyzed by MRI...... 170! ! !

xiii

CHAPTER 1

INTRODUCTION

1.1 Introduction

Menopause is associated with increased central obesity (1), metabolic syndrome

(2), and other chronic diseases (3-6). Given the high prevalence of obesity (38%) and metabolic syndrome (50%) among postmenopausal women in the United States (7, 8), weight loss and management of metabolic syndrome in postmenopausal women is a growing field in medical practice and scientific investigation for prevention of chronic diseases.

Breast cancer is the second leading cause of cancer deaths in women (9). About

67% of incidence and 80% of deaths were occurred in women past the age of menopause

(10). Obesity is an established risk factor in postmenopausal breast cancer, suggesting that obesity-associated metabolic dysregulation may have critical roles in the pathogenesis of postmenopausal breast cancer. Reduction of obesity by calorie restriction and/or weight loss has been shown to reduce the development of mammary tumors (11,

12). Finding approaches that limit weight gain, modify the pattern of central obesity, and

1 correct metabolic derangements associated with menopause, may reduce the risk of postmenopausal obesity-induced breast cancer.

A naturally-derived flavonoid, naringenin, mimics some of the metabolic and cellular mechanisms of metformin (13, 14), a glucose lowering drug that has shown to reduce incidence of breast cancer . Naringenin is abundant in citrus fruits and tomatoes and was recently evaluated for its effects on metabolic disturbances. Male mice fed a high-fat diet supplemented with dietary naringenin showed suppression of diet-induced weight gain, adipose mass, blood glucose and insulin levels, and increased insulin sensitivity (15). Naringenin also inhibits proliferation and induces apoptosis in various cancer cells, including breast cancer cells (13, 16-22). Altogether, it appears that naringenin mitigates obesity-related metabolic disturbances and exhibits anti-breast cancer activity. Furthermore, naringenin readily accumulates in plasma after ingestion of orange juice, grapefruit juice (23), and tomato paste or sauce (24), suggesting that naringenin is bioavailable from the diet.

Based on previous knowledge in the field, I hypothesized that naringenin ameliorates metabolic dysregulation and inhibits growth of mammary tumors in postmenopausal obesity. Although the effects of naringenin on diet-induced metabolic disturbances have been extensively studied in male rodents. It is unclear whether naringenin exerts similar effects in the postmenopausal endocrine milieu. This research utilized ovariectomized mice to model postmenopausal status to test our hypotheses.

Many studies have been shown that ovariectomized mice gain weight and develop metabolic disturbances even when fed a normal diet, a pattern much like metabolic

2 disturbances seen in postmenopausal women. In Chapter 3, we characterized metabolic disturbances induced by ovariectomy and examined whether naringenin reverses these metabolic changes in ovariectomized mice. Furthermore, to address the interaction of postmenopausal obesity and breast cancer, we investigated the extent that naringenin ameliorates metabolic dysregulation in a setting of postmenopausal obesity. These results are presented in Chapter 4. Finally, we compared the effect of dietary naringenin with metformin on mammary tumorigenesis in a mouse model of postmenopausal obesity. We utilized a syngeneic orthotopic mouse model of breast cancer that mammary tumor grows faster in mice fed with a high-fat diet compared to which fed with a low-fat diet (25).

Additionally, this model is an immunocompetent breast cancer mouse model that maintains dynamic interaction between the immune system and tumor in the mammary tumor microenvironment. The results regarding the effects of naringenin and metformin on mammary tumorigenesis are presented in Chapter 5.

1.2 Aims

Aim 1: To evaluate the effect of naringenin on metabolic changes resulting from loss of ovarian function. Although naringenin ameliorates metabolic disturbances in male rodents with diet-induced insulin resistance, the effects of naringenin on adipose tissue accumulation and metabolic abnormalities associated with loss of ovarian function have not been studied. The working hypothesis for this aim is that loss of ovarian function by ovariectomy results in adipose tissue accumulation and insulin resistance, while naringenin supplementation attenuates these metabolic disturbances.

Aim 2: To examine the effect of naringenin on metabolic disturbances in a mouse model of postmenopausal obesity. Most studies supplemented dietary naringenin to high-fat diets when the mice were lean and healthy, and found beneficial effects of naringenin on metabolism. However, whether naringenin reverses diet-induced metabolic disturbances when mice are already obese is not well studied. The working hypothesis for this aim is that naringenin suppresses weight gain and reverses insulin resistance induced by a combination of high-fat diet and loss of ovarian function.

Aim 3: To compare the effect of naringenin and metformin on the growth of breast tumors in a mouse model of postmenopausal obesity. Obesity is an established risk factor of postmenopausal breast cancer. Studies have shown that improvement of metabolic status by metformin reduces the risk of breast cancer. We previously found that naringenin acts like metformin to induce AMP-activated protein kinase (AMPK) and reduce glucose production in rat hepatoma cells. However, the effects of naringenin or metformin on mammary tumor growth in a setting of postmenopausal obesity remain undefined. The working hypothesis of this aim is that both naringenin and metformin induces AMPK signaling, improves insulin sensitivity, and inhibits the growth of breast tumor in a mouse model of postmenopausal obesity.

CHAPTER 2

LITERATURE REVIEW

2.1 Menopause and metabolic changes

2.1.1 Definitions of menopause

Menopause is a natural biological process that usually occurs in women 45-55 years of age. Menopause is associated with depletion of the ovarian pool of non-growing follicles and cease of ovulation (26), resulting in reduced secretion of the ovarian hormones estrogen and progesterone. The complete process of menopause usually has three parts: perimenopause, menopause, and postmenopause. Due to erratic fluctuations of estrogen and progesterone and disruption of hormonal feedback loops during perimenopause, women may start to notice signs and symptoms including changes in menstrual period, hot flashes, and night sweats (27). Menopause is defined as the time when a woman has not had a period for 12 months, indicating that the function of the ovaries totally ceases. At this point, a woman advances to postmenopausal status for the rest of her life. However, not all the women undergo natural menopause. Some surgeries

(e.g., oophorectomy) or medical treatments (e.g., chemotherapy that damages ovaries)

5 that remove or damage ovaries also cause menopause. In these cases, women may experience more severe symptoms because of the rapid loss of estrogen.

2.1.2 Menopause and metabolic changes: Findings from human studies

The prevalence of obesity is higher (38% vs. 32%) in postmenopausal women compared to men in the same age range (28). Although several cross-sectional studies showed that postmenopausal women are heavier than premenopausal women (8, 29), other cross-sectional studies observed that menopause-associated weight changes was associated with aging (30, 31). Longitudinal data that followed premenopausal women found that women gained similar amount of weight regardless of their menopausal status at the end of the study (32, 33), suggesting that body weight gain in middle-aged women is a consequence of aging rather than hormonal changes at menopause.

Muscle loss with aging may explain weight gain in middle-aged women. Muscle mass normally diminishes with age, a process called sarcopenia. It has been shown that skeletal muscle mass decreased over time in middle-age women (34). Concomitantly, a study showed that 24 hours energy expenditure and physical activity energy expenditure declined over time in both pre- and post-menopausal women (35). Because muscle mass significantly correlates with resting energy expenditure (36), muscle loss in middle-aged women may result in decreased energy expenditure, potentially leading to weight gain.

In contrast to changes in body weight, changes of body composition and distribution of adipose depots might be associated with both aging and hormonal changes resulting from menopause. Extensive studies indicate that menopause is associated with a

6 significant shift from gynoid (fat storage around the hips) to android (fat storage around the waist) fat distribution (37). Increases in waist circumference and fat mass in middle- age women are associated with both aging and menopause (38). Pooled results from randomized-controlled trials showed significant reduction in abdominal obesity (waist circumference or % abdominal fat) in women with hormone-replacement therapy compared to placebo or no treatment (35), suggesting that estrogens are regulators of fat distribution. However, several studies suggested that other hormones, including bioavailable testosterone, sex hormone-binding globulin (SHBG), and follicle-stimulating hormone (FSH), may play a more important role in changes in body fat distribution during menopause transition. Sex hormone-binding globulin (SHBG) is a circulating protein that binds to testosterone, dihydrotestosterone, and estradiol. Bioavailable testosterone decreases when SHBG increases. Studies have shown that bioavailable testosterone (free testosterone) and SHBG are stronger predictors of visceral fat and risk of developing metabolic syndrome compared to total estradiol (39, 40). FSH is a hormone secreted by the pituitary gland. The blood levels of FSH are negatively regulated by estradiol and thus FSH levels increase during menopause when the blood estradiol levels are low. Levels of FSH are also positively correlated with fat mass and waist circumference while inversely associated with lean and skeletal muscle mass (41).

Future studies are needed to identify the key hormonal regulator that modulates body composition changes during menopausal transition.

Due to changes in hormones and body composition at menopause, postmenopausal women have higher risk of developing metabolic syndrome (35).

Metabolic syndrome is diagnosed when a person has three or more of the following risk factors: waist circumference more than 40 inches for men and 35 inches for women, triglyceride level higher than 150 mg/dL, high-density lipoprotein (HDL) cholesterol lower than 40 mg/dL for men and 50 mg/dL for women, blood pressure higher than

130/85 mm Hg, and fasting blood sugar higher than 100 mg/dL (42). Over 50% of postmenopausal women in the United States have metabolic syndrome (43). A longitudinal study reported a steady increase in the risk of developing metabolic syndrome during the course of the menopausal transition and 13.7% of the cohort had developed metabolic syndrome by the time of the final menstrual period (7).

Dyslipidemia (defined as total/HDL cholesterol ratio >6 in men and >5 in women in the study) was strongly associated with waist circumference in women (41). In the same study, men 55-64 and 35-44 years of age had the same risk of dyslipidemia, while women

55-64 years of age (mostly postmenopausal population) had more than doubled risk of dyslipidemia compared to women 35-44 years of age (44). A longitudinal study also found significant reductions in HDL cholesterol levels and increases in low-density lipoprotein (LDL) cholesterol levels in women who became postmenopausal compared to their counterparts who were still premenopausal (44). In addition to dyslipidemia, a study that followed women from premenopausal to postmenopausal status found that increases in inflammatory markers, including serum amyloid A (SAA), C-reactive protein (CRP), tissue plasminogen activator antigen (tPA) (45). The changes in CRP and tPA were positively correlated with increased intra-abdominal fat. The same study also showed

8 effects on leptin and adiponectin, adipokines secreted from adipocyte, with blood leptin positively and adiponectin negatively correlating with increases in intra-abdominal fat.

Epidemiological studies indicate that nonalcoholic fatty liver disease (NAFLD) is more prevalent in men than in women. However, the prevalence increases in postmenopausal women and exceeds that of men in the same age range (46). A cross- sectional study showed that the prevalence of NAFLD increased in women in the late menopausal transition stage and post-menopausal women compared with premenopausal women (4). Women (both pre- and post-menopausal) with NAFLD have significantly lower levels of serum estrogen compared to women without NAFLD, suggesting that estrogens may have a protective effect against NAFLD in women (47).

Metabolic syndrome is associated with increased incidence of cardiovascular disease and diabetes (48). It is well-established that the risk of developing type 2 diabetes increases with age. However, whether menopausal status affects the risk for diabetes independent of age is less clear. A study comparing diabetes risks between pre- and postmenopausal women did not find a strong association (49). Another study comparing diabetes risk in pre- and post-menopausal women with glucose intolerance also showed that postmenopausal women had a similar risk of progressing to diabetes compared with premenopausal women after adjustment for age (50). On the other hand, premenopausal women generally have lower risk of developing cardiovascular disease and stroke (2, 51), while postmenopausal women have the same level of risk for cardiovascular-disease- related morbidity and mortality as men in the same age range (52), suggesting that menopausal status may be a risk factor for cardiovascular disease. However, results from

9 a meta-analysis showed that postmenopausal status did not increase the risk of cardiovascular disease (52). Therefore, even though menopause may increase risk factors of cardiovascular diseases and diabetes, whether menopausal status independent of age influences risks of diabetes and cardiovascular diseases remains controversial (53, 54).

2.1.3 The metabolic effect of hormone replacement therapy in postmenopausal women

Hormone replacement therapy (HRT) has been used in perimenopausal and postmenopausal women to prevent/treat menopausal symptoms and menopause-related metabolic dysregulation. The main types of hormones involved are estrogens, progesterone or progestins, and sometimes testosterone. Estrogen or combinations of estrogen/progesterone treatments are the most common types of hormone replacement therapy administered to postmenopausal women.

In a meta-analysis of 107 randomized-controlled trials (5), HRT users exhibited increased lean body mass (3.3%, 95% CI = 0.02–6.6%) and reduced waist circumference

(-0.8%, CI = -1.2 to -0.4%) and abdominal fat (-6.8%, 95% CI= -11.8 to -1.9%) when compared to women using placebo or no treatment. HRT reduced both fasting glucose and insulin levels and improved insulin sensitivity estimated by homeostasis model assessment (HOMA-IR) (-12.9%, CI= -8.6 to -17.1%). The incidence of diabetes mellitus is 30% lower in HRT users compared to placebo or no treatment groups (HR = 0.70, 95%

CI = 0.6–0.9). HRT also increased HDL cholesterol (5.1%, 95% CI = 3.6 to 6.7%) and reduced LDL cholesterol (-11.0%, 95% CI= -12.3 to -9.6%) but had no effect on

10 triglycerides. Mean blood pressure and procoagulant factors, fibrinogen and plasminogen activator inhibitor-1 (PAI-1), were also reduced in HRT users. However, HRT use correlated with increased C-reactive protein (CRP), a proinflammatory marker. Further analysis showed that oral HRT treatment had more significant effects than transdermal

HRT treatment. Transdermal HRT only significantly decreased LDL/HDL ratio and fibrinogen levels but did not increase CRP levels.

Several studies with small sample sizes have shown that the prevalence of

NAFLD is lower in women using hormone replacement therapy (HRT) compared to women not using HRT. However, it is still unclear whether HRT protects against

NAFLD in postmenopausal women (39, 55, 56).

Of note, increases in cardiovascular events and breast cancer have been observed in HRT users in two large, randomized, double-blind, placebo-controlled trials, Women’s

Health Initiative (WHI) (57) and the Heart Estrogen–Progestin Replacement Study

(HERS) (58). HRT is no longer recommended to prevent chronic diseases, although considered for management of menopausal symptoms in some women (59). However,

Hodis et al argued that HRT treatment might be beneficial for peri-menopausal women or younger postmenopausal women, because the populations studied in WHI and HERS were older postmenopausal women with more than 10 years of postmenopausal status.

Further investigation on the dosage, regime, and timing of HRT treatment is warranted to maximum the beneficial effects of HRT treatment and minimum side effects (60).

2.1.4 Menopause and metabolic changes: Findings from animal models

The ovariectomized (OVX) animal model is a widely used model of menopause

(61). In this model, both ovaries of the female animal are removed to study the effect of ovarian hormones on organs in the body. Ovariectomized rodents share multiple characteristics and similar endocrine changes found in human (62). In rodents, the plasma levels of 17β-estradiol normally reduce to low or undetectable after 1-2 weeks of ovariectomy. Extensive studies have investigated metabolic changes induced by estrogen depletion in ovariectomized rodents. Although some studies fed ovariectomized rodents high fat diets in order to observe more significant changes, the present review only addresses the metabolic differences between ovariectomized and non-ovariectomized

(ovary-intact) animals. Furthermore, the effects of estradiol treatment and exercise are discussed. To assess whether estradiol treatment can recover the detrimental effect of loss of ovarian function, the rodents are surgically implanted with slow-releasing subcutaneous estradiol pellets.

Table 1 and Table 2 show metabolic characteristics of ovariectomized rats (Table

1.) and mice (Table 2.) compared to non-ovariectomized rodents. Ovariectomy induces body weight gain and increases adiposity in both rats and mice. Hyperphagia has been reported in ovariectomized rats but not in ovariectomized mice (62-67). Instead of altering food intake, several studies demonstrate that ovariectomized mice have lower energy expenditure and activity levels, especially in the dark phase (when mice are most active and consume most food) of the light/dark cycle (66, 68-70). A study comparing ovariectomy-induced metabolic and behavioral changes in rats and mice reported that

12 body weight gain in ovariectomized rats was due to hyperphagia and reduced locomotor activity, while the weight gain in ovariectomized mice was more related to reduced energy expenditure and locomotor activity (71). Estradiol treatment has been shown to decrease body weight gain and food intake and to increase energy expenditure in ovariectomized rats. Exercise training also decreases body weight gain in ovariectomized rats but does not affect food intake (66).

Unlike findings from human studies, no difference in muscle mass was observed between ovariectomized and ovary-intact rodents (72). However, accumulation of lipids or triacylglycerides in skeletal muscle of ovariectomized rodents has been reported (72,

73). A significant increase in intramyocellular lipids is observed in ovariectomized mice, associated with up-regulated fatty acid transporters and impaired mitochondrial function

(73, 74). Exercise training has been shown to decrease lipid content in different muscles in both ovariectomized and ovary-intact rats (73, 74).

An increase in adiposity has been consistently observed in both ovariectomized rats and mice. Ovariectomy induces whole body adiposity and does not exhibit depot- specific effect (65, 74). Increases in adipose tissue mass and adipocyte size are observed in both visceral and subcutaneous depots of ovariectomized mice compared to ovary- intact mice (72). Leptin is highly correlated with fat mass (68) and higher blood leptin levels are consistently observed in ovariectomized animals compared to ovary-intact animals (63, 67, 75-77). Additionally, adipose inflammation is observed in both depots but is more severe in visceral than subcutaneous adipose tissue. Estradiol treatment and exercise training have been shown to effectively decrease adipose tissue accumulation

13 and blood leptin levels in ovariectomized animals (70, 72, 78). Protein and mRNA expression of adipose tissue inflammatory markers were also decreased by estradiol treatment in ovariectomized animals (72).

Loss of ovarian hormones from ovariectomy in animals is associated with impaired glucose homeostasis. Elevated fasting glucose and insulin levels have been observed in ovariectomized rodents (65, 67, 68, 72, 76, 79). Several studies performed glucose tolerance test (80), insulin tolerance test (77), and HOMA-IR calculation (69, 72) and found impaired glucose utilization and insulin resistance in ovariectomized rodents.

Both estradiol treatment and exercise training exhibited positive effects on lowering blood glucose levels, while the effect on lowering insulin levels was less prominent (65-

67, 69, 70, 72, 73, 78, 81, 82). Estradiol treatment also improved insulin sensitivity, measured by insulin tolerance test, in ovariectomized mice (80).

Dyslipidemia occurs in ovariectomized mice. Several studies consistently reported elevated blood cholesterol levels in ovariectomized mice (64, 67, 74, 82).

However, the effects of ovariectomy on blood triacylglycerides and non-esterified fatty acids are mixed. The blood levels of triglycerides have been shown to be the same or even decreased in ovariectomized rodents compared to the ovary-intact rodents (64, 65,

68, 69, 74, 80, 81). Ovariectomized rats exhibited higher total cholesterol levels, higher

LDL levels, lower HDL levels and higher atherogenic index even though the blood triglyceride levels were lower compared with the ovary-intact rats (72). Only a few studies have reported elevated non-esterified fatty acids (NEFA) levels in ovariectomized rodents (65, 67, 74) while most of studies failed to find a difference (64, 68, 69, 72, 78,

80-82). The effect of estradiol treatment on plasma lipids is also controversial. Estradiol increased blood triglyceride and NEFA levels in some studies. Other studies showed estradiol had no effect or decreased blood NEFA levels. In contrast, exercise training has been shown to decrease both blood triglyceride and NEFA levels (80).

Loss of ovarian function in rodents is associated with lipid/triglycerides accumulation in liver (64, 65, 68, 69, 72, 78, 81, 82). The ectopic lipid accumulation in liver may be due to an up-regulation of de novo lipogenesis and a reduction in lipid oxidation (64, 65, 68, 69, 74, 80, 81). Furthermore, gene expression of proinflammatory cytokines was higher and evidence of hepatic steatosis was observed in ovariectomized animals (64, 65, 68, 69, 72, 74, 81). Estradiol treatment and exercise training exerted similar effects on reducing liver lipid accumulation and modulating expression of hepatic genes involved in de novo lipogenesis, lipid oxidation, and inflammation in ovariectomized rats (64, 65, 68, 69, 72, 74, 81).

In summary, loss of ovarian hormones in rodents has been linked to increases in body weight, adiposity, and fasting glucose and cholesterol levels. Adipose tissue inflammation and ectopic lipid accumulation in liver and muscle are also observed in ovariectomized animals.

Table 1. Effect of loss of ovarian hormones on body composition and metabolism (Rat) OVX vs. Non-OVX OVX + estradiol vs. OVX OVX + Exercise vs. OVX

Effect Reference Effect Reference Effect Reference

Body Weigh Gain Increase (63, 64, 72, 79) Decrease (65, 72, 79) Decrease (78)

Food Intake Increase (63-65, 72) Decrease (65, 66) No Effect (72)

Decrease (72) Energy Expenditure Increase (83) N/A No Effect (83)

Adiposity Increase (65, 66, 74, 79) Decrease (65, 72, 78, 79) Decrease (72, 78)

16 Plasma Analytes (Fasting)

Glucose Increase (65, 72) Decrease (65, 72, 78) No Effect (81)

Insulin Increase (65, 72, 81) Decrease (65, 72, 81) Decrease (72, 78)

No Effect (64, 72) Triglycerides Increase (78, 81) Decrease (65) Decrease (65, 74, 78) Continued

Table 1. Continued OVX vs. Non-OVX OVX + estradiol vs. OVX OVX + Exercise vs. OVX

Effect Reference Effect Reference Effect Reference

Plasma Analytes (Fasting)

Cholesterol Increase (72, 74) N/A N/A

Increase (64) Decrease (65) NEFA Decrease (65) No Effect (64, 78, 81) No Effect (72)

Leptin Increase (63, 78) Decrease (72) Decrease (72, 78) Liver

17 Lipid Increase (72) N/A Decrease (74)

Triglycerides Increase (64, 65, 74, 81) Decrease (72, 78, 81) Decrease (72, 78, 81)

De Novo Lipogenesis Increase (64, 65, 72, 81) Decrease (72) Decrease (72)

Inflammatory markers Increase (72) Decrease (72) Decrease (72) Continued

Table 1. Continued OVX vs. Non-OVX OVX + estradiol vs. OVX OVX + Exercise vs. OVX

Effect Reference Effect Reference Effect Reference Adipose

Inflammatory markers Increase (72) Decrease (79) N/A

Skeletal Muscle

Lipid Increase (79) N/A Decrease (74)

NEFA, non-esterified fatty acids; OVX, ovariectomized; Ref, reference Rat strain: Sprague-Dawley, Wistar, Long-Evans

Table 2. Effect of loss of ovarian hormones on body composition and metabolism (Mouse) OVX vs. Non-OVX OVX + estradiol vs. OVX OVX + Exercise vs. OVX

Effect Reference Effect Reference Effect Reference Body Weigh Gain Increase (68, 74, 80, 84) No Effect (77) No Effect (80)

Food Intake No Effect (67, 68, 80) N/A N/A

Energy Expenditure Decrease (66-68) N/A N/A

Adiposity Increase (66-68, 84) Decrease (70, 80) Decrease (73, 80)

Plasma Analytes (Fasting)

Glucose Increase (68, 76, 80) Decrease (70, 80) Decrease (80, 82)

No Effect (69, 70) Insulin Increase (67, 76, 80) No Effect (73, 80) Decrease (67, 80) Triglyceride No Effect (66, 68) N/A N/A

Cholesterol Increase (69, 82) N/A N/A

Continued

Table 2. Continued OVX vs. Non-OVX OVX + estradiol vs. OVX OVX + Exercise vs. OVX

Effect Reference Effect Reference Effect Reference Plasma Analytes (Fasting)

Increase (67, 80) Increase (70, 80) NEFA N/A No Effect (67-69) Decrease (73, 82)

Leptin Increase (67, 76, 80) Decrease (70, 77) N/A

Liver

Triacylglyceride N/A N/A Decrease (67, 73)

De Novo Lipogenesis Increase (68, 69, 80) Decrease (70, 82) No Effect (67, 73)

Inflammatory markers Increase (68, 69, 80) N/A N/A

Continued

Table 2. Continued OVX vs. Non-OVX OVX + estradiol vs. OVX OVX + Exercise vs. OVX

Effect Reference Effect Reference Effect Reference

Adipose

Adipocyte Size Increase (68, 69, 82) Decrease (70, 82) N/A

Inflammatory markers Increase (67-69) N/A N/A

Skeletal Muscle

Intramyocellular lipids Increase (73, 82) N/A N/A

NEFA, non-esterified fatty acids; OVX, ovariectomized; Ref, reference Mouse strain: C57BL/6

2.2 Obesity and postmenopausal breast cancer

2.2.1 Breast cancer

A malignant tumor is a group of cancer cells that spread to surrounding tissues or even distant areas of the body. Breast cancer arises from breast epithelial and occurs mostly in women. As shown in Figure 1, a women’s breast is comprised of lobules (milk- producing glands), ducts (tubes that carry the milk from the lobules to the nipple), and stroma (adipose tissue, connective tissue, blood vessels, and lymphatic vessels) (80).

Breast cancer that starts in the epithelial cells lining the breast ducts is called ductal carcinoma while breast cancer that starts in the epithelial cells lining the milk glands is called lobular carcinoma. Tissues samples obtained from breast biopsy or tumors removed from surgery are examined under the microscope to determine if the breast cancer is in situ or invasive, ductal or lobular, how different the cancer cells resemble normal cells (feature of grade), and whether the cancer has spread to the lymph nodes

(85). Ductal carcinoma in situ (DCIS) is the most common type of in situ breast cancer.

In situ stems from the Latin phrase meaning “it’s original place.” DCIS indicates that the cancer cells is noninvasive and have not grown beyond the ductal basement membrane.

Although DCIS is considered non-invasive, studies have suggested that DCIS increases the risk of developing an invasive breast cancer (86).

Compared to DCIS, most of other types of breast cancers are invasive (87). These cancer cells break through the ductal or glandular basement membrane and infiltrate into surrounding breast tissues (e.g., adipose tissue). Cells that are able to metastasize may travel to other parts of the body via the bloodstream or lymphatic system. Invasive duct

22 carcinoma (IDC) is the most common type, accounting for approximately 80% of invasive breast cancers diagnoses.

Figure 1. Female breast anatomy (86).

2.2.2 Breast cancer subtypes

Breast cancer now is considered as a heterogeneous disease composed by a spectrum of tumor subtypes with distinct molecular, pathological, and clinical features.

In 2000, Perou et al identified 496 genes (intrinsic gene subset) that showed higher variation in expression between different tumors than between repeated tumor samples

(before and after neoadjuvant chemotherapy) (88). By hierarchical clustering, breast cancer tumors were separated into two main branches, estrogen receptors (ER) positive or

ER negative. The tumors in the ER positive group had high expression of estrogen

23 receptor and were characterized by the relatively high expression of many genes expressed by breast luminal cells. Therefore, they were also called luminal-like tumors.

Tumors with low to absent expression of ER (ER negative) were separated into three groups, basal-like, ERBB2 positive (HER2-enriched), and normal breast-like subtypes.

The basal-like tumor was characterized by low expression of the luminal genes, low expression of the HER2 gene cluster, and high expression of basal epithelial genes. The tumors were named as “basal-like” was because they showed the expression of genes that were typically expressed within basal epithelial cells of the skin and airways (89).

ERBB2 positive (HER2-enriched) subtype showed high expression of several genes in the ERBB2 amplicon. Normal breast-like group was characterized by highest expression of many genes characterized of basal epithelial cells, adipose tissue and other non- epithelial cell types but low expression of genes characteristic of luminal epithelial cells.

Later, Sorlie et al separated the luminal/ER+ group to luminal A, B and/or C and showed that these 5 and/or 6 subtypes are highly correlated with overall and relapse-free survival

(90). However, due to the heterogeneity of breast tumors, the definition of the subtypes has changed over time and the development of the methods that can robustly assign subtype to individual tumors is still ongoing (91).

The classifier that identifies the four tumor subtypes, luminal A, luminal B,

HER2-enriched, and basal-like, and the normal breast-like group is consistently idendified and commonly accepted (92). Although some studies have identified new subtypes, for example, claudin-low or mesenchymal-like subtype (90), these subtypes have not been fully validated for clinical routine use. Therefore, we will discuss the 4 main subtypes in the present review. 24 1. Luminal A: Luminal A tumors have higher expression of ER and/or progesterone receptor (PR) and ER-regulated genes but low expression of HER2 and Ki-67 (cellular marker for proliferation). Luminal A breast cancer is the most common subtype, accounting for about 70 % of breast cancers (93). These tumors grow slower and are less aggressive compared to other subtypes. Patients who have luminal A tumors generally have better prognosis, in part because women who have ER+ tumors can be given hormone therapy to lower estrogen levels or to block the effects of estrogen on the growth of breast cancer cells (94).

Hormonal treatments, including selective estrogen receptor modulators (SERMs), estrogen receptor down regulators, and aromatase inhibitors, are common adjuvants administered to estrogen receptor positive breast cancer patients until drug resistance develops (86).

2. Luminal B: Luminal B tumors, like luminal A tumors, have higher expression of ER and/or PR, but the expression of HER2 and proliferation-associated genes

(eg. Ki67) are also higher in luminal B tumors. Luminal B tumors also tend to be

TP53 mutant and show lower expression of ER and ER-regulated genes (85). It is not always easy to distinguish a luminal B tumor from a luminal A tumor since the expression of genes defining these two groups is a continuum. About 10% of breast cancers are luminal B (90). Compared to luminal A tumors, luminal B tumors are associated with poorer prognosis in the first 5 years after diagnosis, but then the difference decreases and there is no difference between these 2 subtypes by 8 years after diagnosis (94). 25

3. HER2-enriched (ERBB2+): HER2-enriched tumors do not express hormone receptors but have elevated expression of HER2 and many other genes that reside near HER2 in the genome (95). HER2-enriched tumors comprise about 5-10% of breast cancers (90) and tend to grow and spread more aggressively. This subtype of breast cancers used to have poorer prognosis compared to luminal breast cancers. However, the recurrence and death rate decrease significantly after the use of anti-HER2 targeted therapies (94).

4. Basal-like: The basal-like tumors have low expression of the luminal genes, low expression of the HER2 gene cluster, and high expression of basal epithelial genes. The tumors in this group have been known as “triple-negative” in clinical term because they typically do not express or have low expression of ER, PR, or

HER2. Although the majority of triple-negative cancers are of basal-like phenotype, basal-like and triple-negative cancers are not synonymous. It has been found that only 77% of molecular basal-like tumors were triple-negative and only

71% of triple-negative cancers were basal-like subtype by gene expression profiling (96). The basal-like breast cancer is the second most-common subtype of breast cancers, accounting for about 15% of all breast cancers (97). The incidence is almost double in African-American women compared to all other racial/ethnic groups (98). Premenopausal women and women with a BRCA1 gene mutation also have higher risk of developing basal-like breast cancers (94). Because the heterogeneity of basal-like breast tumor, it remains a big challenge for physicians 26 and patients, and women diagnosed with basal-like breast cancer generally have a

poorer prognosis than those diagnosed with other types of breast cancers (90).

2.2.3 The prevalence, incidence, and mortality of breast cancer

According to the Surveillance, Epidemiology, and End Results (SEER), there were an estimated 2,975,314 women living with breast cancer in the United States in

2012 and about 12% of U.S. women will develop breast cancer at some point during their lifetime. The rate of breast cancer incidence has been stable with an estimated 231,840 new cases of breast cancer in 2015, accounting for 29% of new cancer cases in women in the United States (99). Although the rate of breast cancer mortality has declined, breast cancer is still the second leading cause of death in women with 40,290 women estimated to die of breast cancer in 2015 (15% of all cancer death) (9).

2.2.4 Obesity and postmenopausal breast cancer

Incidence and death rates for breast cancer increase with age. In addition, 67% of new cases and 80% of breast cancer deaths occurred in women 55 years of age and older based on the data from 2008-2012 (9). Therefore, the majority of women diagnosed with breast cancer are postmenopausal. A direct association between adult weight gain/obesity and postmenopausal breast cancer has been reported in many studies and now obesity is regarded as an established risk factor for postmenopausal breast cancer (10). The risk of postmenopausal breast cancer was about 2.5 times higher in postmenopausal obese women (BMI >31.1) compared to postmenopausal lean women (BMI ≤ 22.6) (100).

Results from a meta-analysis of 50 studies showed that every 5 kg increase in adult 27 weight gain (from age 18-25 years to study enrollment) was associated with an 11% increased risk of postmenopausal breast cancer (101), and in a longitudinal study, 10 kg weight gain or more after menopause was associated with an 18% increased risk of breast cancer in postmenopausal women (102). Furthermore, the risk of postmenopausal breast cancer was 1.9 times higher in women in the highest quintile of waist circumference compared to women in the lowest quintile (103). Body mass index (BMI) and central obesity were also associated with increased risk of death for breast cancer (104, 105).

Altogether, postmenopausal body weight, body mass index (BMI), weight gain, and central obesity are associated with increased risk and mortality of postmenopausal breast cancer (103, 104, 106, 107).

Although many lines of evidence reveal a direct association between obesity and breast cancer incidence and prognosis, the underlying mechanism linking obesity to postmenopausal breast cancer is not clearly understood. It may be related to the fact that postmenopausal women are more susceptible to central adiposity (1, 101) and metabolic syndrome (2, 41, 108). At least 4 main pathways have been proposed that connect obesity and postmenopausal breast cancer: estrogen metabolism, adipokine (leptin and adiponectin) signaling, adipose inflammation, and insulin signaling.

1. Estrogen metabolism

Cytochrome P450 aromatase is an enzyme that converts androgens to estrogens

(42). Before menopause, the ovarian follicle is a major site for aromatase

expression and thus estrogen production. However, because ovarian function

ceases in postmenopausal women, adipose tissue becomes the major peripheral 28 site for aromatase expression and estrogen production (109). Obese postmenopausal women have higher level of circulating levels of estrogen than normal-weight women (110). Postmenopausal women with the highest levels of endogenous estrogen have shown twice the risk of developing breast cancer compared to women with the lowest levels (111). Furthermore, obesity in postmenopausal women strongly associates with the risk of ER+/PR+ breast cancer but not other breast cancer subtypes (112, 113). Therefore, increased availability of estrogen in the obese postmenopausal women promotes the development of estrogen-responsive breast tumors.

2. Adipokine signaling

Adipokines leptin and adiponectin, synthesized and secreted from adipocytes, have been shown to actively participate in the pathogenesis of breast cancer (114,

115). Circulating leptin levels are positively correlated with adiposity (116) and predict breast cancer risk independently of BMI and waist circumference (117).

Leptin levels also positively correlate with total body aromatic activity (118) and tumor size and stage (119) in postmenopausal breast cancer patients. In vitro and preclinical animal studies have shown that leptin crosstalks with various signaling pathways, including estrogen/aromatase, insulin/insulin-like growth factor 1(IGF-

1), angiogenic network, and inflammatory cytokines (120). Furthermore, obese mice lacking leptin or leptin receptor have failed to develop mammary tumor or exhibited slower tumor growth, suggesting a critical role of leptin in mammary cancer initiation and progression (121-123). On the contrary, circulating 29 adiponectin levels are inversely associated with breast cancer risk (124). In particular, lower serum adiponectin concentrations are associated with larger tumor size and higher histologic grades (125). The anti-carcinogenic effect of adiponectin is likely mediated through effects to promote apoptosis and/or inhibit signaling pathways involved in proliferation, cell cycle, and survival (115, 126).

3. Adipose tissue inflammation

Obesity is associated with a low-grade systemic inflammation characterized by elevated levels of circulating proinflammatory mediators secreted by liver, white adipose tissue, and immune cells (116, 127, 128). Systemic levels of inflammatory markers, including C-reactive protein (CRP), serum amyloid A

(SAA), and interleukin 6 (IL-6), have been linked to poor prognosis of breast cancer (129, 130). Adipocytes and macrophages recruited to the adipose tissue secrete inflammatory mediators, such as tumor necrosis factor-α (TNF-α), IL-6, and prostaglandin E2 (PGE2), and induce aromatase expression (131).

Additionally, TNF-α and IL-6 modulate the expression of genes involved in proliferation, survival, and angiogenesis and may directly affect tumor growth

(132, 133). Local adipose tissue inflammation is also observed in the breast tissue of overweight and obese postmenopausal women (134) and is associated with increased aromatase expression in obese breast cancer patients (135). Adipocytes located close to breast cancer cells have been shown to provide free fatty acids as a fuel source for cancer cell growth as well as secreting angiogenic and inflammatory mediators that promote cancer cell invasion (136). 30

4. Activation of insulin signaling pathway

Insulin is a polypeptide hormone secreted from pancreatic β cells in response to

elevated glucose concentration. Insulin is a powerful mitogenic factor. It binds to

insulin receptor and stimulates downstream PI3K/Akt/mTOR (Phosphoinositide

3-kinase/ Protein Kinase B/ mammalian target of rapamycin) pathway signaling,

which promotes cell proliferation and blocks apoptosis in vitro and in vivo (137-

140). Increased BMI in postmenopausal women is correlated with higher

circulating insulin (141, 142). A prospective case-control study has shown that

postmenopausal women with higher insulin levels have a greater risk of breast

cancer (143). Furthermore, the expression of insulin receptor in breast cancer cells

and tissues is more than 6-fold higher compared to normal breast tissues (143)

and breast cancer cells are incapable of down-regulating insulin receptors in

response to excess insulin (144). Therefore, a higher concentration of insulin

continually activates insulin receptors, initiates downstream signaling pathway,

and leads to proliferation and anti-apoptosis of cancer cells.

2.2.5 Metformin and breast cancer

Metformin is an oral hypoglycemic drug used to treat type 2 diabetes mellitus and metformin usage is associated with reduced risk of breast cancer in women with diabetes

(145, 146). Many cell culture and animal studies have demonstrated that metformin inhibits mitochondrial respiratory-chain complex 1, induces a decrease in cellular ATP concentrations, and activates AMP-activated protein kinase α (AMPKα) (147). AMPKα 31 acts as an energy sensor and regulates energy metabolism in tissues, such as liver, muscle, and adipose tissue (148). It has been shown that metformin, via AMPKα signaling, inhibits transcription of key gluconeogenic genes and lipid synthesis in the liver, reduces fatty acid synthesis and lipolysis in adipose tissue, increases glucose and fatty acid uptake, mitochondrial oxidation, and glycolysis in muscle, and decreases insulin secretion in pancreas (149). In addition to reversing hyperglycemia and hyperinsulinemia, metformin exerts antitumor activity through AMPKα. Metformin has been shown to inhibit the proliferation of breast cancer cells through AMPKα-mediated suppression of protein translation (mTOR pathway inhibition) (147) and inhibition of cyclin D1 expression (150). Since metformin has been widely used in patients with type 2 diabetes mellitus, it is currently regarded as a promising target for breast cancer prevention and treatment.

32 2.3 Naringenin and its bioactivities against diet-induced metabolic dysregulation and breast cancer

2.3.1 Naringenin

Flavonoids comprise a large group of polyphenolic plant metabolites that are found in our daily food, such as vegetables, fruits, grains, nuts, wine, and tea. It is estimated that U.S. adults consume an average of 189.7 ± 11.2 mg/ day of flavonoids

(151). All naturally occurring flavonoids have a 15-carbon backbone ring structure (C6-

C3-C6) consisting of two aromatic rings and a heterocyclic ring. Based on their molecular structures, flavonoids are divided into 6 main classes: flavones, flavanones, flavonols, isoflavones, anthocyanidins and catechins. Epidemiological studies have shown an inverse relationship between dietary flavonoid intakes and chronic diseases

(152-154). A meta-analysis of epidemiologic studies reported that the risk of breast cancer significantly decreased in women with high intake of flavonols and flavones compared to women with low intake (155). It also showed that flavonols, flavones or flavan-3-ols intake is associated with a significant reduced risk of breast cancer in postmenopausal while not in pre-menopausal women.

Naringenin (Figure 2) is a flavanone abundant in citrus fruits and tomatoes (155,

156). It is present in 3 main forms, aglycone (naringenin), neohesperidoside (naringin), and rutinoside/rhamnoglucoside (narirutin) (157). The glycosidic forms, naringin and narirutin, are the major forms of naringenin in citrus fruits (158-160) and the aglyconic form (naringenin) mainly found in tomato skin (156, 161). Naringenin bioavailability and bioactivities have been studied in recent years because of its potential beneficial effects on human health (162, 163). 33

CHEMOPREVENTIVE /THERAPEUTIC EFFECTS OF NARINGENIN 31

Skin Cancer NGN significantly increased long-term cell survival after UVB irradiation in immortalized p53-mutant human keratino- cyte Ha CaT cells, detected by colony forming assay. It enhanced the removal of cyclobutane pyrimidine dimers from the genome observed by both direct quantization of cyclobutane pyrimidine dimers in genomic DNA and immuno-locali- zation of the damage within the nuclei. Modulation of the UVB-induced poly (ADP-ribose) polymerase-1 (PARP-1) Figure FIG.2.Chemical 2. Chemical structure structure ofof naringenin. naringenin aglycone. (164) cleavage, caspase activation, and Bax Bcl2 ratio by NGN ⁄ treatment, indicated an antiapoptotic effect. This antiapoptotic effect in UVB-damaged cells occurred at least in part via cas- 2.3.2 Absorption, bioavailability, and metabolism of naringenin NGN AS A CHEMOPREVENTIVE AND THERAPEUTIC pase cascade pathway and protected the cells from UVB- AGENT AGAINST CANCER induced apoptosis, detected by the DNA ladders, agarose gel The glycosidic moiety is a major determinant of the absorption site and electrophoresis, and flow cytometric analysis, respectively. Epidemiological Studies Demonstrating the Cancer- This provides a molecular basis for the action of NGN as a Preventive Effects bioavailability of naringenin. The glycosidic moietiespromising need tonatural be cleav flavonoided prior in preventing to skin aging and carci- In a study conducted in a Hawaiian population, a decreased nogenesis (83). However NGN was found to be completely rate of lung cancer was associated with increased intake of absorption of naringenin aglycone (163, 165, 166). Studiesineffective in rodents in decreasing have shown UV-B-induced that erythema alone. In onions and apples (rich in the flavone quercetin) and white combination with lecithin, the photoprotective activity of grapefruit (a rich source of the flavanone NGN) (76). In naringenin monoglucosides are absorbed as efficientlyNGN as naringenin, was enhanced while to a significant naringenin extent. The topical activity another study, conducted on 10,054 Finnish people, lower of NGN must be optimized by using a suitable penetration incidence of cerebrovascular disease and asthma was associ- neohesperidoside and rutinosides exhibit a delay in intestinalenhancer absorption so that the NGN(167,can 168 be). employedIn as a topical photo- ated with higher intakes of hesperetin and NGN (2). A direct protective agent (84). correlationthe was small also intestine, reported in monoglu people,cos havingides high are content hydrolyzed of by lactase phloridzin hydrolase NGN and hesperetin in diet and a lower risk of stomach cancer (77). These(LPH), studies an epithelial indicated a beta protective-glucosidase effect of on NGN the on brushGliomas border membrane (166, 169). both cancer as well as noncancer diseases resulting in an over- Treatment of C6 cells-implanted rats, with NGN increased all positiveSubsequently, health benefit. naringenin aglycone is passively absorbed.the expressionAdditionally, of glialmonoglucosides fibrillary acidic protein. Furthermore, NGN has also been found to limit glial activation and downre- may be actively absorbed by the sodium-dependent gulate glucose the transporter NF-kB expression (SGLT1) level and and its target genes. These Protective Effect Against Chemical Carcinogens observations suggest that NGN may inhibit glial cell tumori- Biotransformationhydrolyzed by of cytosolic some toxicants β-glucosidases such as ( polycyclic170-172). Ingenesis contrast, (85). naringenin The observation with more that NGN abrogates the ische- aromatic hydrocarbons by the cytochrome p450 (CYP) 1B1 mic brain injury by suppressing NF-kB mediated enzymecomplex make them glycosidic genotoxic moieties, agents. such NGN as (at neohesperidoside or above neuroinflammation and rutinosides, in middlerequire cerebral gut artery occlusion male 5 mM) inhibits this enzyme. NGN (1 mM) can reduce Wistar rats suggests that NGN can act as a potential neuropro- CYP1B1microorganisms mRNA expression to inducedhydrolyze by 7,12-dimethylbenz(a)the glycosidic bonds (tectant166, 168 in patients, 173). withBoth high the risk sm ofall ischemic stroke (86). anthracene (DMBA) (78). Further study illustrated that the

Downloaded by [University of Montana] at 09:49 14 January 2015 suppressionintestinal was and at the colonic transcriptional epithelium level. can Usingmetabolize reporter naringenin via phase 2 conjugation by gene assays as well as the electromobility shift assay, it was Breast Cancer veriﬁedadding that NGN glucuronic counteracted acid DMBA-induced or sulfate groups XRE (165 binding, 167). In Qin addition et al. to (2011) the gastrointestinal established a breast cancer resection at 21675 and protected against DMBA-induced oral carcino- model by injecting the fourth mammary fat-pad of female genesis (79). NGN has also shown protection against azoxy BALB/c mice with 4T1 cells. 4T1 is a mouse mammary carci- methane induced colon carcinogenesis (80). NGN administra-34 noma cell line with many similarities to human breast cancer. tion (200 mg/kg body weight) largely upregulated the redox 4T1 cells or 4T1 breast cancer resection models are highly status to decrease the risk of cancer in N-methyl-N’-nitro-N- malignant and quickly develop spontaneous metastases to vari- nitrosoguanidine-induced gastric carcinoma in animal models ous organs with pulmonary metastasis as the most common of stomach cancer. This up-regulation of antioxidants by NGN and predominant cause of death (87). NGN administration pro- treatment might be responsible for the anticancer effects in longed the survival period in breast cancer 4T1 resection gastric carcinoma (81). More recently, Subramanian and Arul model (88). This happens very rarely in this breast cancer (82) showed that NGN can attenuate the NDEA induced hepa- model system as it is highly malignant and poorly immuno- tocarcinogenesis in rats through the downregulation of NF-kB, genic. This property of NGN was attributed to the inhibition of vascular endothelial growth factor, and matrix metallo protei- lung metastasis by activation of the immune system as NGN nases (MMPs). administration increased the number of interferon-g and tract, the liver is the main organ involved in flavonoid metabolism and mediates phase 2 conjugation.

Naringenin accumulates in blood and tissues mostly as its conjugated metabolites

(25, 174-177). Naringenin glucuronides are the main metabolites in the plasma and urine while naringenin sulfates are more abundant in liver, spleen, heart and brain (177, 178).

Bioavailability is defined as the fraction of an orally ingested substance that is absorbed and available for physiologic activity or storage (178). Bioavailability of naringenin from different food sources has been studied in healthy men and women. After consumption of cooked tomato paste, the time needed to reach peak concentration (Tmax) of naringenin in plasma is about 2 hours (179) compared to 4.8 hours and 5.5 hours after consuming orange and grapefruit juices, respectively (25). The differences in Tmax could be due to the different forms of naringenin ingested (glycosidic forms in orange and grapefruit juices and aglyconic form in tomato skin) that may be absorbed via different routes in the digestive tract as described earlier. The urinary excretion of naringenin from orange and grapefruit juices was nearly complete in 24 hours (24, 177). Comparing the relative bioavailability of naringenin with other flavonoids, quercetin and hesperetin shows that naringenin has slightly higher accumulation in the plasma in a study of 10 male participants consuming a single dose of juice mix (30 mg/L quercetin, 28 mg/L naringenin and 32 mg/L hesperetin) for 48 hours (178). Studies in animals have shown that naringenin can accumulate in tissues (eg. liver, gut, kidney, spleen, lung, ovary, uterus, mammary tissue, and brain) after repeated doses (175, 177, 180). The accumulation of naringenin metabolites in the serum reaches a steady state after 5 days of repeated oral administration (213mg/kg twice a day) (181). Furthermore, repeated oral 35 administration results in a higher and faster accumulation of naringenin in the serum

(176), suggesting that daily consumption may increase bioavailability of naringenin in humans.

2.3.3 The effects of naringenin on diet-induced metabolic dysregulation

Extensive studies have used animal models to investigate the effect of naringenin or naringin on metabolic dysregulation induced by high-fat, high-cholesterol, or high fructose diets (Table 3). Suppression of body weight gain by naringenin or naringin supplementation has been observed in mice (13, 176, 182-184). Supplementation with

0.02% of naringin or 1% naringenin suppressed a high-fat diet-induced body weight gain in C57BL/6 mice (184, 185). In some studies, naringenin or naringin reduced adiposity without observing significant changes in body weight (185, 186). Obesity-induced inflammation in adipose tissue is linked to the development of insulin resistance (187).

Naringenin has been shown to inhibit Toll-like receptor 2 (TLR2) expression, a membrane receptor on adipocytes involved in obesity-induced adipose tissue inflammation, and suppress expression of inflammatory markers in adipose tissue (188).

Furthermore, monocyte chemoattractant protein-1 (MCP-1/CCL2) is a key chemokine in obesity-related macrophage infiltration into adipose tissue (185). A study demonstrated that macrophage infiltration and up-regulation of MCP-1 occurred in the early phase of high-fat diet-induced obesity before significant increases in body weight gain and adiposity were observed. Naringenin supplementation of a high-fat diet fed to male

C57BL/6 mice suppressed macrophage infiltration and MCP-1 mRNA expression in epididymal fat without affecting body weight and adiposity (189). 36 Naringenin or naringin also exhibits lipid-lowering effects in blood and liver of rodents (13, 182, 183, 186, 190-195). Studies have shown that naringenin or naringin supplementation lowers blood LDL, VLDL and total cholesterol levels as well as hepatic cholesterol levels (13, 182-184, 186, 191-195). 3-hydroxy-3-nethylglutaryl coenzyme A

(HMG-CoA) reductase is a rate-limiting enzyme for cholesterol synthesis. Acyl coenzyme A: cholesterol acyltransferase (ACAT) is a key regulatory enzyme controlling cellular cholesterol homeostasis in liver. The activities of HMG-CoA reductase and

ACAT reduced decreased by more than 40% in rodents consuming diets supplemented with naringenin or naringin (184, 191, 194), which may explain the cholesterol-lowering effects of naringenin or naringin. Furthermore, feeding high cholesterol or high-fat/high cholesterol diets resulted in elevated circulating VLDL particles, a lipoprotein that transports endogenous lipids from the liver to the rest of the body, while naringenin supplementation significantly suppressed VLDL secretion and ameliorated dyslipidemia in a mouse model of insulin resistance (182, 195). Naringenin or naringin supplementation also promotes fatty acid oxidation and suppresses fatty acid synthesis in liver (13, 192).

In addition to lowering blood and hepatic lipids, naringenin or naringin decreases blood glucose and insulin levels as well as improves glucose utilization and insulin sensitivity in animal models of insulin resistance (13, 182-184, 186). Naringenin ameliorated high-fructose-induced hyperlipidemia and insulin resistance by increasing activity of glycolytic enzymes and decreasing activity of gluconeogenic enzymes. Several studies have proposed that the effects of naringenin or naringin on metabolic dysregulation were mediated by AMPK signaling (184). However, more studies are 37 needed to identify the specific molecular targets of naringenin in order to better understand naringenin-activated signaling mechanism(s).

38 Table 3. Effects of naringin and naringenin on diet-induced metabolic dysregulation Dose Model Outcome Ref Naringin 0.05% HF/HC-fed Reduced plasma and hepatic cholesterol levels. (164) Sprague-Dawley Decreased HMG-CoA reductase and ACAT rats activities in liver. 0.02% HF/HC-fed Reduced plasma TG and cholesterol levels. (191) Sprague-Dawley Decreased hepatic cholesterol levels and HMG- rats CoA reductase and ACAT activities 0.02% HF-fed Reduced serum levels of total, HDL, LDL (194) C57BL/6 mice cholesterol. Decreased hepatic cholesterol levels and FA synthesis and increased FA oxidation. Improved glucose utilization and insulin sensitivity. Reduced adiposity. Increased AMPK activity in liver. 0.16% HF/high- Decreased plasma total cholesterol and FFA. (184) (~100 fructose-fed Reduced adipose tissue mass. Improved mg/kg/day) Wistar rats glucose utilization. Naringenin 0.10% HC-fed Reduced plasma and hepatic cholesterol levels. (186) Sprague-Dawley Decreased HMG-CoA reductase and ACAT rats activities in liver. 1% Palm oil-fed Reduced plasma TG and cholesterol levels. (195) ICR mice Increased expression and activities of hepatic enzymes involved in fatty acid oxidation. Continued

39 Table 3. Continued Dose Model Outcome Ref 1% or 3% HF-fed LDLR- Suppressed body weight gain. Reduced plasma (192) null mice levels of TG, and total, VLDL, and LDL cholesterol. Decreased hepatic TG and cholesterol levels, FA synthesis and TG synthesis, and increased FA oxidation. Improved glucose utilization and insulin sensitivity. Reduced adipose tissue mass. 3% HF-fed Suppressed body weight gain. Reduced plasma (13) C57BL/6J mice levels of TG, and total and LDL cholesterol. Decreased hepatic TG and cholesterol levels. Improved glucose utilization and insulin sensitivity. Reduced adiposity 3% HF-fed LDLR- Suppressed body weight gain. Reduced plasma (13) null mice levels of TG, and total, VLDL, and LDL cholesterol. Decreased hepatic TG and cholesterol levels. Reduced plasma glucose and insulin levels, and adiposity. 50 mg/kg High-fructose- Reduced plasma TG and FFA. Improved (183) BW fed Wistar rats glucose utilization and insulin sensitivity. Increased activities of enzymes involved in insulin signaling. 0.003%, Long-Evans rats Did not affect body weight. Reduced plasma (193) 0.006%, and hepatic TG and cholesterol levels. 0.012% Decreased adipose tissue mass. Continued

40 Table 3. Continued Dose Model Outcome Ref 1% HF-fed Decreased body weight and fat pad weight. (187) C57BL/6J mice Reduced blood glucose levels. Decreased adipose tissue inflammation 3% HF/HC-fed Decreased body weight gain. Reduced plasma (185) LDLR-null mice levels of TG, and total, VLDL, and LDL cholesterol. Decreased hepatic TG and cholesterol levels and inflammation. Improved glucose utilization and insulin sensitivity. Decreased adiposity and adipose inflammation. 3% HC-fed LDLR- Decreased body weight gain. Reduced plasma (182) null mice levels of TG, and total, VLDL, and LDL cholesterol. Decreased hepatic TG and cholesterol levels and inflammation. Improved glucose utilization and insulin sensitivity. Decreased adiposity and adipose inflammation. 100 mg/kg HF-fed Decreased macrophage infiltration and MCP-1 (182) BW C57BL/6J mice mRNA and protein expression. ACAT, acyl-CoA: cholesterol acyltransferase; AMPK, AMP-activated protein kinase; FA, fatty acid; HDL, high density lipoprotein; HC, high-cholesterol; HF, high-fat; HMG- CoA, 3-hydroxy-3-methylglutaryl-coenzyme A; LDLR, low density lipoprotein receptor; MCP-1/CCL2: monocyte chemoattractant protein-1; Ref, reference; TG, triglyceride; VLDL, very low-density lipoprotein.

41 2.3.4 The effects of naringenin on breast cancer

Several in vitro studies have found that naringenin modulates estrogen receptors

(ER) extra-nuclear signaling pathways. Naringenin induces cell cycle arrest and apoptosis in cancer cells through ERα or ERβ signaling (190). Although naringenin is a low affinity and low potency ligand to the ERα receptor, it antagonizes 17β-estradiol and impairs ERK (extracellular signal-regulated kinase) and AKT (serine/threonine-protein kinases) signaling pathways for survival and proliferation while inducing p-38 pro- apoptotic signaling pathway (22). Tumors use glucose as the primary fuel for energy and elevated level blood glucose provides unlimited supply to the tumor for growth (196).

Naringenin impairs insulin stimulated AKT and MAPK (mitogen-activated protein kinase) activities and suppresses insulin-mediated glucose uptake in MCF-7 human breast cancer cells (197). Few studies have investigated the effect of naringenin on breast cancer in animals. A study using an orthotopic 4T1 breast cancer resection model showed that naringenin did not inhibit cell proliferation and tumor growth of 4T1 mouse mammary carcinoma (23). Nonetheless, naringenin inhibited lung metastasis and extended survival of those mice. Of note, 4T1 breast cancer cells are triple-negative breast cancer cells (ER-

, PR-, and HER2-). Given that naringenin induces cell cycle arrest and apoptosis through

ER signaling and that ER-positive breast cancer is the most common subtype of breast cancer, future studies are needed to examine whether naringenin may be a chemopreventive or therapeutic agent against ER-positive breast cancer.

CHAPTER 3

The flavonoid, naringenin, decreases adipose tissue mass and attenuates

ovariectomy-associated metabolic disturbances in mice

* CITATION: Ke JY, Kliewer KL, Hamad E, Cole RM, Powell KA, Andridge RR,

Straka SR, Yee LD, Belury MA. The flavonoid, naringenin, decreases adipose tissue mass and attenuates ovariectomy-associated metabolic disturbances in mice. Nutrition &

Metabolism 2015, 12:1. doi:10.1186/1743-7075-12-1

43 3.1 Abstract

Adverse metabolic changes associated with loss of ovarian function increase the risk of developing metabolic syndrome and non-alcoholic fatty liver disease (NAFLD) in postmenopausal women. Naringenin improves metabolic disturbances in vitro and in vivo. In the present study, we tested the effects of naringenin on metabolic disturbances resulting from estrogen deficiency in ovariectomized mice. Ovariectomized C57BL/6 J female mice were fed a control diet (10% calories from fat) for 11 weeks. Mice either continued on the control diet (n = 9) or were switched to the control diet supplemented with 3% naringenin (n = 10) for the next 11 weeks. Energy expenditure was measured by indirect calorimetry and activity was monitored by infrared beam breaks. Intra-abdominal and subcutaneous adiposity was evaluated by magnetic resonance imaging (MRI). Blood biochemical measures of metabolic response included glucose, insulin, adipokines, and lipids. Lipid content in liver and muscle and expression of relevant genes in adipose tissue, liver, and muscle were quantified. Ovariectomized mice fed naringenin exhibited lower fasting glucose and insulin levels compared to controls, with over 50% reduction of intra-abdominal and subcutaneous adiposity. Plasma leptin and leptin mRNA in adipose depots were also decreased in mice fed a naringenin diet. Monocyte chemoattractant protein-1 (MCP1/Ccl2) and interleukin 6 (IL-6/Il6) mRNA expression levels were significantly lower in perigonadal adipose tissue of naringenin-supplemented mice. We also observed that mice fed a naringenin diet had less hepatic lipid accumulation with corresponding alterations of hepatic gene expression associated with de novo lipogenesis, fatty acid oxidation, and gluconeogenesis. Our results suggest that dietary naringenin

44 attenuates many of the metabolic disturbances associated with ovariectomy in female mice.

45 3.2 Introduction

During menopause, many women experience weight gain and accumulation of body fat in the waist region (198). These changes of body composition increase the risk of developing metabolic syndrome (38), non-alcoholic fatty liver disease (42), and heart disease (4). Although exogenous estrogen has been shown to be protective against many menopause-related metabolic abnormalities, long-term usage of hormone replacement therapy may increase the risk of breast cancer in addition to negative implications for cardiovascular diseases in postmenopausal women (54). Lifestyle changes to reduce body weight, including healthy diet and regular exercise, are the initial strategies recommended for prevention and treatment of menopause-related metabolic disturbances.

Biologically active phytochemicals have attracted considerable attention for their potential health-promoting benefits (58, 199). Naringenin is a flavonoid that is abundant in citrus fruits and tomatoes (156, 200). Previously, we found that naringenin acts in a manner similar to metformin, a medicine used for treating type 2 diabetes, to reduce hepatic glucose production in hepatocytes (157). In addition, naringenin improved some aspects of glucose and lipid homeostasis and mitigated adipose tissue inflammation in vivo (13, 14, 187, 192, 201). However, the effects of naringenin on adipose depot mass and metabolic abnormalities associated with estrogen deficiency have not been studied.

Metabolic changes induced by estrogen depletion from ovariectomy share many similar characteristics with changes in menopausal women that are independent of energy intake, e.g., weight gain, increased adiposity, adipose tissue inflammation, and the development of fatty liver with inflammation (68, 164). These similarities make ovariectomized mice a good model to study physiological changes after menopause. 46 Therefore, in the present study we investigated the effect of 3% wt/wt naringenin supplementation in female ovariectomized (OVX) mice. The aim of our study is to determine whether naringenin ameliorates weight gain and attenuates accumulation of subcutaneous and abdominal adipose tissues, with resultant decreases in fasting glucose and ectopic lipid accumulation in muscle and liver of OVX mice. Our findings suggest that many of the effects of naringenin on dysregulated metabolism are related to effects on decreasing adipose mass and ectopic lipid deposition in muscle and liver of OVX mice fed diet with naringenin.

47 3.3 Materials and Methods

3.3.1 Animals and diets

Twenty-week old C57BL/6J female mice were ovariectomized at 19 weeks old (Jackson

Laboratory; Bar Harbor, ME, USA) and housed 4–5 per cage at 22 ± 0.5°C on a 12:12-h light-dark cycle. After 2 weeks of acclimation, mice (n=19) were fed ad libitum a semi- purified diet for 11 weeks to enhance weight gain (D12450J, Research Diets Inc. New

Brunswick, NJ, USA, formula is shown in Table 4). Then mice were randomized by weight into either CON (n=9) or NAR (n=10) group (The randomization was based on 19 mice as one mouse died prior to assignment to diet groups). For the next 11 weeks, the

CON group continued on the control diet and the NAR group received the control diet supplemented with 3% wt/wt naringenin (Sigma, St. Louis, MO, USA), custom prepared by Research Diets Inc. This dose of naringenin is based on a previously published study showing amelioration of metabolic disturbances in C57BL/6J male mice fed a high-fat diet with 3% naringenin (69). Body weight and food intake were measured daily. At week 22, mice were fasted for 5 h, anesthetized with isoflurane for blood collection via cardiac puncture, and then euthanized by cervical dislocation. Blood was collected into

EDTA-coated blood collection tubes and plasma was obtained after centrifugation.

Tissues, including subcutaneous adipose tissue (SCAT, thoracic and abdominal mammary fat pads), liver, perigonadal adipose tissue (PGAT), and quadriceps skeletal muscle, were excised, weighed, snap frozen in liquid nitrogen, and stored at -80 ºC until further analysis. All procedures were in accordance with institution guidelines and approved by the Institutional Animal Care and Use Committee at The Ohio State

University. 48

Table 4. Composition of experimental diets CON NAR (D12450J) Nutrient gm% kcal% gm% kcal% Protein 19.2 20 18.7 20 Carbohydrate 67.3 70 65.3 70 Fat 4.3 10 4.1 10 kcal/g 3.85 3.73

Ingredient gm kcal gm kcal Casein 200 800 200 800 L-Cystine 3 12 3 12 Corn Starch 506.2 2024.8 506.2 2024.8 Maltodextrin 10 125 500 125 500 Sucrose 68.8 275.2 68.8 275.2 Cellulose, BW200 50 0 50 0 Soybean Oil 25 225 25 225 Lard 20 180 20 180 Mineral Mix S10026 10 0 10 0 DiCalcium Phosphate 13 0 13 0 Calcium Carbonate 5.5 0 5.5 0 Potassium Citrate, H2O 16.5 0 16.5 0 Vitamin Mix V10001 10 40 10 40 Choline Bitartrate 2 0 2 0 Naringenin 0 0 32.7 0

Naringenin (%) 0 3.01

3.3.2 Fasting glucose analysis

At weeks 5 and 21, glucose was measured from tail vein blood samples after a 5h period of fasting (OneTouch Ultra blood glucose meter, LifeScan Inc., Milpitas, CA, USA).

3.3.3 Plasma analysis

49 Plasma insulin, leptin, and adiponectin levels were measured by ELISA (Millipore,

Billerica, MA, USA) according to the manufacturer’s instructions. Plasma triglycerides were examined using a Cholestech LDX analyzer (Cholestech Corporation, Hayward,

CA, USA). Plasma free fatty acids (NEFA C, Wako Chemicals, Richmond, VA, USA) and total cholesterol (Pointe Scientific Inc., Canton, MI, USA) were determined by enzymatic colorimetric assays.

3.3.4 Indirect Calorimetry

During week 17, six mice in each group were housed individually in metabolic chambers at 22°C, allowed free access to food and water, and acclimated 24 hours prior to metabolic assessments. Measurements were taken for a 24-h period, including a 12-h light cycle and a 12-h dark cycle. Oxygen consumption (VO2), carbon dioxide production

(VCO2), and physical activity (by infra-red beam breaks) were measured every 20 minutes using a computer-controlled, open-circuit Oxymax /CLAMS System (Columbus

Instruments, Columbus, OH, USA). Respiratory exchanging ratio (RER) was calculated as the ratio of VCO2 to VO2. Heat, the standard measure of energy expenditure, was calculated with the formula (13),

Heat (kcal/hr) = (3.815 + 1.232 x RER) x (VO2) x (1L/1000 ml x 1kg/1000g) x (body weight)

3.3.5 Magnetic resonance imaging (MRI)

During week 19, total, intra-abdominal, and subcutaneous adiposity were analyzed by

MRI using a Bruker Biospin 94/30 magnet (Billerica, MA, USA) and a 70 mm diameter 50 linear volume coil. T1-weighted coronal images of the whole mouse torso were collected using a respiratory-gated RARE sequence (TR/TE=1570/7.5ms, RARE factor=4,

FOV=70x45 mm2, matrix size=256x192, slice thickness=1 mm, navg=2). The details and definition of fat areas are shown in Appendix A.

3.3.6 Histology

Liver samples were collected at necropsy and fixed in 10% neutral buffered formalin overnight and transferred to 70% ethanol for storage. Tissues were then embedded in paraffin, sectioned, and stained with hematoxylin and eosin (H&E).

3.3.7 Total lipids, triacylglyceride and diacylglyceride analysis of liver and muscle

Total lipids were extracted from liver or muscle samples with 2:1 (v/v) chloroform: methanol and washed with 0.88% KCL (202). The chloroform phase was transferred to a weighed test tube and dried under nitrogen gas at room temperature. The dried sample and test tube was weighed again to calculate total extracted lipid. Following the lipid extraction, triacylglycerol and diacylglycerol were obtained using solid-phase extraction

(203) and solubilized in tert-butanol, methanol, Triton X-100 (204). Analysis was performed using enzymatic colorimetric assay (205).

3.3.8 RNA extraction and quantitative real-time polymerase chain reaction (qRT-

PCR) analysis of gene expression

Total RNA was extracted from liver and muscle samples with QIAzol lysis reagent

(Qiagen, Valencia, CA, USA) and from perigonadal adipose tissue and subcutaneous 51 adipose tissue using RNeasy Lipid Tissue Mini Kit (Qiagen) following manufacturer's instructions. Total RNA was then reversed transcribed to cDNA using High Capacity cDNA Reverse Transcription Kit (Applied Biosystems, Foster City, CA, USA). qRT-

PCR analysis was performed with ABI Prism 7300 sequence detection system (Applied

Biosystems) using TaqMan Gene Expression Assays (Applied Biosystems, Table 5).

Target gene expression was normalized to 18S rRNA for perigonadal adipose tissue or glyceraldehyde 3-phosphate dehydrogenase (GAPDH) for liver, subcutaneous adipose tissue, and muscle. Endogenous Controls (VIC probes) were amplified in the same reaction and expressed as 2-ΔΔCt compared to the CON group (206).

52 Table 5. Real-time PCR primers and probes of Taqman gene expression assay Gene No. Acox1 Mm00443579_m1 Adipoq (Adiponectin) Mm00456425_m1 Ccl2 (MCP1) Mm00441242_m1 Cpt1a Mm00550438_m1 Cpt1b Mm00487200_m1 Dgat2 Mm01273905_m1 Emr1 (F4/80) Mm00802529_m1 Fas Mm00662319_m1 G6pc Mm00839363_m1 Il6 Mm00446190_m1 Lep (Leptin) Mm00434759_m1 Pck2 (PEPCK) Mm00440636_m1 Ppara Mm00440939_m1 Ppargc1a (PGC1a) Mm00447183_m1 Pparg Mm00440945_m1 Scd1 Mm00772290_m1 Srebf1 (SREBP1) Mm00550338_m1 Tnf (TNFa) Mm00443258_m1

3.3.9 Statistical analysis

All data are presented as mean ± standard error (SEM) with p<0.05 considered significantly different. Statistical analysis was performed using SPSS version 20.0 software (SPSS, Inc., Chicago, IL, USA). Significance was determined using two-tailed unpaired Student's t test. Caloric intake, heat, and adipose mass were also analyzed by

ANCOVA with body weight as a covariate (207). Pearson correlation coefficient was used for the correlation analyses.

53 3.4 Results

3.4.1 Effect of naringenin on caloric intake, body weight, and metabolic measurements

Caloric intake decreased significantly when mice were switched to a naringenin- containing diet at week 12 (Figure 3A) without complete recovery to baseline in the following weeks. However, excluding week 12, daily caloric intake from week 0-11 to week 13-22 did not differ significantly between the two treatment groups (p=0.075).

Body weights were significantly reduced in the NAR group from week 12 until the end of the study (Figure 3B). There were no differences in caloric intake between groups after adjusting for body weight (ANCOVA, p=0.508). There were also no differences in ambulatory activity (Figure 3C) and energy expenditure after controlling for body weight differences between groups (Figure 3D).

Fasting glucose levels were significantly elevated in the CON mice from 121.8 ±

5.8 mg/dl at week 5 to 163.0 ± 5.2 mg/dl at week 18 (Figure 3E), while unchanged in

NAR mice (128.6 ± 4.8 at week 5 and 127.1 ± 10.2 at week 18). Additionally, diets enriched with naringenin resulted in decreased fasting insulin levels and HOMA-IR values (Figure 3F & G), surrogate markers of insulin sensitivity. Using an insulin tolerance test, we observed that glucose levels were significantly lower in the NAR mice compared to the CON mice at 45 min (48.0 ± 3.4 mg/dl vs. 64.8 ± 8.1 mg/dl) and 60 min

(51.4 ± 8.2 mg/dl vs. 77.0 ± 6.1 mg/dl) after injection of insulin (data not shown).

However, no difference was observed in areas under the curve (AUC) between groups.

Figure 3. Effects of dietary naringenin on caloric intake, body weight, and metabolic measurements. OVX C57BL/6J mice were fed the control diet for 11 weeks then randomized to continuation of the control diet (n=9) or switched to 3% wt/wt naringenin supplementation of the control diet (n=10) for the weeks 12-22. Average daily caloric intake (A) of two groups (n=2/group) and weekly body weight (B) of the CON (n=9) and NAR (n=10) group were determined. At week 17, CLAMS chambers were used to measure locomotor activity (ambulation) in the horizontal plane by infrared beam breaks (C) and estimated energy expenditure (heat) after controlling for body weight (D) in the dark and light phase for 24-h following a 24-h acclimation (n=6/group). Fasting glucose levels (E) were measured at week 5 and week 18, (CON, n=9; NAR, n=10). Fasting insulin levels (F) were determined after 22 weeks of experimental period (n=6/group). HOMA-IR values (G) were derived from fasting plasma glucose and insulin (n=4/group). Values are presented as mean ± SEM. Significance between groups was determined by Student’s t test, except metabolic data of heat was analyzed by ANCOVA with body weight as a covariate. #P < 0.05 compared CON with NAR.

55 3.4.2 Effect of naringenin on adiposity, plasma adipokines, and adipose tissue gene expression

MRI analysis was performed in a subset of mice from both groups (n=6), which revealed a reduction in total, intra-abdominal and subcutaneous adiposity by 54, 59, and

50%, respectively, in mice fed naringenin versus control diets (Figure 4A & C).

Additionally, naringenin was significantly associated with decreased perigonadal (PGAT) and subcutaneous (SCAT) adipose tissues (Table 6). Because body weights differed significantly between groups and body weight had a significant effect on perigonadal and subcutaneous adipose tissue mass (p < 0.001), we used ANCOVA to adjust for this difference when comparing adipose tissue mass. Mice fed naringenin had significantly decreased perigonadal adipose tissue mass (p = 0.006) after controlling for body weight in comparison to control mice. Fasting plasma leptin levels decreased by 80% in the

NAR mice (Figure 4B) and strongly correlated with total, intra-abdominal, and subcutaneous adiposity determined by MRI (Total, r = 0.85, p = 0.008; Intra-abdominal, r

= 0.86, p = 0.006; subcutaneous, r = 0.80, p = 0.016), as well as both perigonadal and subcutaneous adipose tissue mass (r = 0.95 and 0.97 respectively, both p < 0.001).

Interestingly, plasma leptin levels correlated positively with insulin levels (r = 0.76, p =

0.004), as previously shown in women (208). Dietary naringenin was not associated with changes in levels of adiponectin in plasma and mRNA expression (Adipoq) in perigonadal adipose tissue between groups (Figure 4B & D). Mice fed naringenin also had significantly lower leptin mRNA expression (Lep) in adipose depots, with a 60% reduction in perigonadal adipose tissue and a 55% reduction in subcutaneous adipose tissue (Figure 4D & E). 56 We measured mRNA levels of genes encoding for several markers of inflammation related to obesity, including the chemokine MCP1 (Ccl2), proinflammatory cytokine IL6 (Il6) and TNFα (Tnf), and macrophage-specific marker F4/80 (Emr1) in adipose depots (Figure 4D & E). The diet with naringenin significantly down-regulated mRNA levels of MCP1 (56% reduction) and IL6 (40% reduction) in perigonadal adipose tissue, but had no significant effect on MCP1 and IL6 levels in subcutaneous adipose tissue. TNFα and F4/80 mRNA was not affected by the naringenin supplementation in perigonadal adipose tissue.

57 Table 6. Effect of dietary naringenin on tissue mass in OVX C57BL/6J female mice CON NAR

Tissue Weights (g)

# Perigonadal Adipose Tissue 1.44 ± 0.15 0.38 ± 0.05 # Subcutaneous Adipose Tissue 1.21 ± 0.14 0.52 ± 0.03 # Liver 0.94 ± 0.05 0.80 ± 0.03 Muscle (Quadriceps) 0.36 ± 0.01 0.32 ± 0.02

Tissue Weight Percentages (%)

# Perigonadal Adipose Tissue 5.23 ± 0.35 1.81 ± 0.19 # Subcutaneous Adipose Tissue 4.34 ± 0.34 2.55 ± 0.15 # Liver 3.45 ± 0.09 3.89 ± 0.12 # Muscle (Quadriceps) 1.33 ± 0.07 1.55 ± 0.07 Significance between groups was determined by Student’s t-test (n= 9- # 10). Data represent the mean ± SEM. P < 0.05 compared NAR with CON

Figure 4. Effects of dietary naringenin on adiposity, plasma adipokines, and adipose tissue gene expression. At week 19, MRI analysis was performed in a subset of mice (n=6/group) to measure percentage of total body fat and percentage of two adipose tissue depots, intra-abdominal and subcutaneous adipose tissues (A). Fasting plasma leptin and adiponectin in the CON (n=6) and NAR (n=5-6) groups were measured at the end of study at 22 weeks (B). Panel (C) shows representative coronal MRI views from each group. Effects of dietary naringenin on the mRNA levels of genes in perigonadal adipose tissue (D) and in subcutaneous adipose tissue (E) are shown (CON, n=6-8; NAR, n=10). Values are presented as mean ± SEM. Significance between groups was determined by Student’s t test. # P < 0.05 compared CON with NAR.

59 3.4.3 Effect of naringenin on plasma, hepatic, and muscle lipid profile

Dietary naringenin reduced total cholesterol levels (Figure 5A), but did not change the levels of triglyceride and non-esterified fatty acids in plasma (Figure 5B & C).

As ectopic lipid accumulation in liver has been observed in OVX animals (209), we tested hepatic lipid contents in both groups. Dietary naringenin decreased hepatic total lipids and triacylglyceride levels (Figure 5D & E) but did not affect diacylglyceride levels (data not shown). Small lipid droplets were scattered throughout H&E stained liver sections from the CON group, while no lipid droplets were detected in the NAR group

(Figure 5F). OVX has been shown to increase muscle lipid contents (210). Therefore, we examined muscle lipid levels and found naringenin reduced total lipids levels (Figure 5G) but had no effect on either triacylglyceride (Figure 5H) or diacylglyceride levels in skeletal muscle (data not shown).

Figure 5. Effects of dietary naringenin on plasma lipids and lipid accumulation in liver and muscle. Plasma cholesterol (A; CON, n=6; NAR, n=10), triglyceride (B; n=4/group) and NEFA (C; n=6/group) after 5-h fasting obtained prior to necropsy were measured by enzymatic colorimetric method. Total lipids (D; CON, n=9; NAR n=9) and triglyceride (E; CON, n=8; NAR n=9) per gram of liver section were determined. Panel (F) shows representative H&E staining of livers (10x) from each group. Total lipids (G) and triglyceride (H) per gram of muscle section were also determined (CON, n=9; NAR n=9). Values are presented as mean ± SEM. Significance between groups was determined by Student’s t test. #P < 0.05 compared CON with NAR.

61 3.4.4 Effect of naringenin on hepatic and muscle mRNA expression

Srebf1, Fasn, and Scd1 encode proteins involved in de novo lipogenesis.

Increased hepatic expression of Srebf1 was observed in the NAR mice (Figure 6A).

However, naringenin down-regulated Scd1 mRNA by 38% but had no effect on another lipogenic enzyme, Fasn. No differences in mRNA levels of genes related to steatotic liver

(74, 211) were detected, i.e. Pparg and Dgat2 (Figure 6A). Expression of genes involved in fatty acid oxidation, Cpt1α (mitochondrial) was significantly higher in the mice fed a naringenin diet, but Acox1 (peroxisomal) was decreased (Figure 6B). PGC1α (Ppargc1a) is a transcriptional coactivator regulating genes involved in fatty acid oxidation and gluconeogenesis. Dietary naringenin induced PGC1α mRNA (4-fold) as well as PEPCK

(Pck2) mRNA (3.5-fold), a rate-limiting enzyme in hepatic gluconeogenesis (Figure 6B), but not G6Pase (G6pc) mRNA, another enzyme involved in gluconeogenesis. Having observed reduced lipid content in skeletal muscle, we measured the mRNA levels of genes involved in fatty acid metabolism (Figure 6C). However, naringenin had no effect on the expression of Cpt1β and PGC1α in muscle. Interestingly, Fasn mRNA expression was higher in muscle tissue of NAR relative to CON mice but did not reach significance

(p=0.07).

Figure 6. Effects of dietary naringenin on gene expression in liver and muscle. mRNA expression of hepatic genes related to de novo lipogenesis and hepatic steatosis (A), mRNA expression of hepatic genes related to beta-oxidation and gluconeogenesis (B), and mRNA expression of genes related to de novo lipogenesis and beta-oxidation in muscle (C) were quantified using qRT- PCR (n=6-10/group). Values are presented as 63 mean ± SEM. Significance between groups was determined by Student’s t test. #P < 0.05 compared CON with NAR.

64 3.5 Discussion

Estrogen deficiency leads to metabolic changes and increased risk of metabolic syndrome (4, 38, 42, 212). To our knowledge, this is the first paper to examine the effects of naringenin on energy metabolism and adiposity in OVX female mice, a preclinical model of menopause. We report here that OVX mice exhibited metabolic disturbances expected with estrogen deficiency including elevated fasting glucose and obesity (2, 210,

213). The addition of naringenin to the diet 1) decreased body weight, fasting glucose and insulin, 2) reduced body fat and perigonadal adipose tissue inflammation, 3) decreased plasma cholesterol and ectopic lipid accumulation in liver and muscle, and 4) affected expression of hepatic genes involved in de novo lipogenesis and fatty acid oxidation.

Therefore, our study results suggest that a 3% wt/wt dietary naringenin supplementation ameliorates many metabolic derangements associated with estrogen deficiency.

We observed that naringenin supplementation caused an initial reduction of caloric intake, followed by decrease in caloric intake and body weight over the subsequent ten weeks of feeding. Reduced caloric intake after naringenin supplementation in the present study was unexpected because previous studies had reported no differences in food intake in male rodent models supplemented with naringenin (4, 13, 182, 187, 192, 193). No changes in food intake were observed even in the same strain, wild-type C57BL/6J males (183, 185). Discrepancies between our data from studies in C57BL/6J males may be attributed to the fat content in the diet, age of the mice, and different effect of naringenin on appetite between male and female OVX mice.

Estrogen interacts with neuropeptides, decreases food intake, and reduces weight gain

(13). Hyperphagia has been reported in some models of menopause (214) but not all (66, 65 69, 70, 214). Instead of altering food intake, several studies demonstrated that OVX mice have lower energy expenditure and activity levels, especially in the dark phase of the light/dark cycle (66, 69, 70). These findings are consistent with human data demonstrating decreased free-living and 24 h energy expenditure, and reduced physical activity during the menopausal transition (71). The totality of evidence instead suggests that reduced energy expenditure and increased energy efficiency, rather than overeating, results in the menopause-associated metabolic disturbances. We did not observe differences in spontaneous physical activity and in energy expenditure between groups after controlling for body weight. It is unclear whether the effects of naringenin on body weight and other OVX-related metabolic disturbances are attributed to reduced caloric intake and/or other mechanisms. In future studies, a pair-fed group will be included to control for possible differences in energy intake as a potential confounding factor.

Dietary naringenin attenuated hyperglycemia and hyperinsulinemia induced by a high-fat or a fructose-enriched diet (13, 36). We observed development of hyperglycemia in OVX mice from week 5 to week 18, consistent with Roger et al who showed elevated fasting glucose in mice after 12 weeks of ovariectomy compared to sham-operated mice

(193), while naringenin supplementation prevented the development of hyperglycemia and lowered fasting insulin concentration and HOMA-IR value. Potter et al suggest that a significant difference in insulin resistance (determined by insulin tolerance test) may not develop until as late as 26 weeks post-ovariectomy (68), which may explain why insulin reduced blood glucose levels more effectively in mice supplemented with naringenin but the glucose AUC did not reach significance in the present study.

66 Estrogen deficiency increases the susceptibility to weight gain and central obesity in both humans and mice (68, 69, 192). Hong et al. demonstrated that adiposity in OVX female mice, examined by DEXA, were comparable with those of male mice (35).

Additionally, despite consuming a low-fat diet, the OVX mice accumulated about 40% of body fat and about 25% of reduction in percent body fat after undergoing a 30% calorie restriction (84). Similarly, we found that OVX mice accumulated about 40% of body fat as assessed by MRI. Although naringenin reduced caloric intake by ~14%, total adiposity decreased by approximately 50%. Additionally, naringenin significantly reduced perigonadal adipose tissue mass, even after controlling for body weight. Dietary naringenin has linked to decreased adiposity in several male mouse studies independent of caloric intake (13, 84, 183). These data further suggest that the effect of naringenin on adiposity is not simply attributable to lower caloric intake.

Increased visceral fat and pro-inflammatory activity were observed in postmenopausal women (185, 215). Previous studies indicate adipose tissues inflammation occurred early in OVX mice (e.g., 12 weeks after ovariectomy) and progressively worsens as indicated by increased infiltration and activation of immune cells and decreased insulin sensitivity (46, 68). Genetic deletion of MCP1 reduces body fat, increases glucose tolerance, and ameliorates adipose inflammation in visceral fat pad in OVX mice, but has no effect on sham-operated mice. These findings suggest that

MCP1 may be a mediator of OVX-induced metabolic disturbances attributed to adipose tissue inflammation (69). Yoshida et al demonstrated that naringenin inhibited high-fat- diet induced toll-like receptor 2 mRNA expression and suppressed mRNA levels of proinflammatory mediators, TNFα and MCP1, in epididymal/perigonadal adipose tissue 67 (216). We found that naringenin down-regulated mRNA levels of MCP1 but not TNFα in perigonadal fat. Future studies will investigate the mechanisms of naringenin action on adipose tissue inflammation of OVX mice.

Interestingly, we found that naringenin significantly reduced MCP1 and IL6 mRNA in perigonadal adipose tissue, but had no significant effect on these markers of inflammation in subcutaneous adipose tissue. Rogers et al (185) suggested that adipose tissue inflammation associated with OVX is more severe in perigonadal adipose tissue than in subcutaneous adipose tissue, which may explain why we found a more significant effect of naringenin on perigonadal adipose tissue inflammation. Additionally, by

LC/MS-MS analyses, naringenin accumulated 1.7 fold more in perigonadal adipose than in subcutaneous adipose tissue after 11 weeks of 3% naringenin supplementation (3.03 ±

2.00 µmole/kg vs. 5.11 ± 1.26 µmole/kg, data unpublished), suggesting that naringenin may have less influence on subcutaneous adipose tissue.

Estrogen deficiency has been connected to hepatic fat accumulation in women and rodents (4, 68, 210). In the present study, we demonstrated that naringenin supplementation led to decreased hepatic lipid accumulation and changed mRNA levels of some genes involved in de novo lipogenesis and fatty acid oxidation. Several studies have shown that naringenin induces PPARα activity and downstream enzymes involved in fatty acid oxidation, such as CPT1α, UCP2, and ACOX1 (4, 192). Interestingly, we observed increased Cpt1α mRNA levels but decreased Acox1 levels, implicating discordance in mitochondrial and peroxisomal β-oxidation in response to OVX and naringenin treatment. Goldwasser et al. (187) also showed that naringenin induced a

68 fasted-like state in hepatocytes, inhibiting fatty acid and cholesterol synthesis and increasing fatty acid oxidation. Consistent with these in vitro results, we also observed increased Srebf1 and PGC1α mRNA in the liver the mice supplemented with naringenin.

However, we did not observe increased of microsomal triglyceride transfer protein (Mttp) expression and were unable to detect low-density lipoprotein (LDL) receptor mRNA expression (data not shown), which are downstream genes of Srebf1 and inducible by naringenin in vitro (201). Additionally, we found increased hepatic expression of genes controlling gluconeogenesis in the mice fed diets with naringenin, a normal physiological reaction in response to fasting. Lack of induction of these two genes in the CON mice may be related to their high levels of fasting glucose.

We demonstrated that naringenin decreased plasma cholesterol but did not change the levels of plasma triacylglyceride and non-esterified fatty acids in OVX mice.

However, the effects of OVX on blood triacylglyceride and non-esterified fatty acids are mixed. Several studies have shown no difference in blood triacylglyceride and non- esterified fatty acids between sham-operated and OVX rodents (65, 68, 69, 71, 81, 217), while others have observed increased levels in OVX rodents (74, 82). It is possible that

OVX had no effect on triacylglyceride and non-esterified fatty acids in the current study.

Therefore we did not observe changes in plasma triacylglyceride and non-esterified fatty acids in naringenin-supplemented mice.

Estrogen deficiency-associated metabolic disturbances have been widely demonstrated in OVX mice (36, 66-68, 71, 80, 84). There were several limitations of the present study. Due to the atrophy of the uterus observed in our OVX mice, we could not measure the weight of uterus to determine whether there is an estrogenic or anti- 69 estrogenic effect of naringenin. Additionally, we did not have sham-operated mice as a control to evaluate the degrees of metabolic disturbances in OVX mice and thus the extent of naringenin effects. However, the elevated fasting glucose, adiposity, and hepatic lipid accumulation observed in OVX mice in the present study were markedly reduced in female mice supplemented with naringenin. In addition, as mentioned earlier, we did not design this study to include a pair-fed group since previous studies evaluating metabolic effects of diet with naringenin did not observe differences in food intake in male rodents.

Naringenin readily accumulates in plasma after ingestion of orange juice, grapefruit juice (69), and tomato paste or sauce (24), suggesting that it is bioavailable in individuals who consume naringenin food sources regularly. In the present study, we found that mice developed higher fasting glucose, adiposity, and hepatic steatosis after loss of ovarian function, similar to what has been observed in postmenopausal women (2,

25). Naringenin supplementation attenuated these estrogen-deficiency-associated metabolic disturbances in OVX female mice, suggesting the potential influence of dietary naringenin on metabolic syndrome in postmenopausal women. Further work in preclinical and human intervention studies will help to determine if naringenin is able to protect against menopausal-associated metabolic syndrome in humans.

70 3.6 Acknowledgements

The authors thank Josephine Fouts for assistance with mouse experimentation, and Dr. Santosh K. Maurya and Dr. Muthu Periasamy for support and assistance with indirect calorimetry measurements. The authors would like to acknowledge the OSU

Small Animal Imaging Core for MR imaging. This research was supported by the Carol S.

Kennedy Endowment (M.A.B.), NCI 1R03CA162551-01A1 (M.A.B.) and the Food

Innovation Center Doctoral Research Grant (J-Y. K).

CHAPTER 4

Citrus flavonoid, naringenin, increases locomotor activity and reduces

diacylglycerol accumulation in skeletal muscle of obese ovariectomized mice

Jia-Yu Ke, Rachel M Cole, Essam M Hamad, Yung-Hsuan Hsiao, Bradley M Cotten,

Kimerly A Powell, Martha A Belury

72 4.1 Abstract

Estrogen deficiency has been associated with central obesity, muscle loss, and metabolic syndrome in postmenopausal women. This study assessed naringenin accumulation in tissues and investigated the hypothesis that naringenin reverses diet- induced metabolic disturbances in obese ovariectomized mice. In study 1, we measured naringenin concentrations in plasma, liver, perigonadal and subcutaneous adipose tissues, and muscle of ovariectomized C57BL/6J female mice after 11 weeks of naringenin supplementation. Naringenin accumulated 5-12 times more in mice fed a 3% naringenin diet than in mice fed a 1% naringenin diet. In study 2, ovariectomized mice were fed a high-fat diet (60 kcal% fat) for 11 weeks and half of the mice were then supplemented with 3% naringenin for another 11 weeks. Dietary naringenin suppressed weight gain, lowered hyperglycemia, and decreased intra-abdominal adiposity evaluated by magnetic resonance imaging. Naringenin-fed mice exhibited elevated locomotor activity monitored by infrared beam breaks, maintained muscle mass, and reduced muscle diacylglycerol content. Real-time PCR analysis in muscle revealed decreased mRNA level for genes involved in de novo lipogenesis, lipolysis, and triglyceride synthesis/storage. Our data suggest that long-term 3% naringenin supplementation results in significant naringenin accumulation in plasma and tissues, associated with attenuated metabolic dysregulation while maintaining muscle mass in obese ovariectomized mice.

73 4.2 Introduction

Naringenin, a dietary flavonoid, displays a diverse range of biological functions, such as anti-carcinogenic, anti-inflammatory, and anti-atherogenic activities (4, 163).

Naringenin is abundant in citrus fruits and tomatoes (157, 164) and its bioavailability has been studied in human and murine models. Naringenin accumulated significantly in plasma (156, 168) and its metabolites were detected in liver, kidney, spleen, heart, and brain of rats fed naringenin (175). In human studies, naringenin readily accumulates in plasma after consumption of orange juice, grapefruit juice (177), and tomato paste (24).

Therefore, naringenin appears to be readily bioavailable from the diet, suggesting that beneficial health effects could be attributed to this flavonoid in individuals who consume naringenin food sources regularly.

Although naringenin accumulation in tissues was measured in rodents after short- term oral administration of naringenin, to our knowledge, few studies have examined naringenin concentrations after long-term feeding in tissues that heavily participate in the regulation of metabolic homeostasis, including liver, skeletal muscle, and adipose tissues.

Thus, we assessed the accumulation of naringenin in mice supplemented with 1% or 3% dietary naringenin for 11 weeks.

Additionally, reports from the Centers for Disease Control and Prevention showed that the prevalence of obesity in women aged 60 years and older increased from 31% in

2003-2004 to 38% in 2011-2012 (25). Eighty-five percent of women undergo menopause by age 55 and the loss of ovarian function at menopause has been associated with weight gain, central obesity, muscle loss and metabolic syndrome (8, 38, 42). Therefore, weight management and health maintenance in postmenopausal women is a growing field in 74 medical practice and scientific investigation. Naringenin appears to be a promising candidate for mitigating adverse metabolic changes in postmenopausal women. We recently showed that naringenin attenuated estrogen deficiency related to metabolic disturbances, including high blood glucose, body fat accumulation, and liver steatosis, in mice fed a normal-fat diet (10% calories from fat) (218). Additionally, several studies demonstrated that dietary supplementation of naringenin to high-fat diets suppressed diet- induced obesity and attenuated metabolic disturbances in male rodents (13, 187, 219).

However, most of the studies supplemented high-fat diets with naringenin in the beginning of the study, when the mice were lean and healthy. Few studies have examined whether naringenin reverses diet-induced metabolic disturbances when mice are already obese, a scenario more reflective of postmenopausal status. Therefore, in the present study, we fed ovariectomized mice, a mouse model mimicking postmenopausal women, a very high-fat diet (60% calories from fat) to induce obesity followed by naringenin supplementation in order to examine if naringenin exerts similar beneficial effects on metabolism in obese ovariectomized mice.

75 4.3 Materials and Methods

4.3.1 Animals and diets

C57BL/6J mice were obtained from Jackson Laboratory (Bar Harbor, ME, USA) and acclimated to the new environment for 1-2 weeks. Mice were housed 5 per cage at 22 ±

0.5°C on a 12:12-hour light-dark cycle. Body weight and food intake were measured daily. All procedures were in accordance with institution guidelines and approved by the

Institutional Animal Care and Use Committee at The Ohio State University.

4.3.2 Study 1: Naringenin accumulation

Five-week old female C57BL/6J mice, ovariectomized at 4 weeks of age, were fed a semi-purified high-fat diet (HFD, D12451, 45 % calories from fat, Research Diets Inc.

New Brunswick, NJ, USA) for 6 weeks to generate high-fat diet induced obesity. Then mice were randomized by weight to one of three diet groups: CON group, continued on the high-fat diet, 1% NAR group, switched to the high-fat diet supplemented with 1 wt/wt

% naringenin, and 3% NAR group, switched to the high-fat diet supplemented with 3 wt/wt % naringenin. Naringenin powder used in the diets was purchased through Sigma

(W530098, St. Louis, MO, USA) and was custom prepared by Research Diets Inc. The dose of naringenin is based on a previously published study in C57BL/6J male mice showing amelioration of metabolic disturbances in mice fed a high-fat diet with 1% and

3% naringenin (185). After 11 weeks of experimental diets, mice were fasted for 5 hours, anesthetized with isoflurane for blood collection via cardiac puncture, and then euthanized by cervical dislocation. Blood was collected in EDTA-coated collection tubes and plasma was obtained after centrifugation at 1,500 g for 20 min at 4 ºC. Plasma (~100 76 µl) for total naringenin accumulation was transferred to another tube with 10% vol/vol of glacial acetic acid and stored at -80 ºC. Tissues (~100 mg) for total naringenin accumulation, including subcutaneous adipose tissues (SCAT, thoracic and abdominal mammary fat pads), liver, perigonadal fat (PGAT), and quadriceps skeletal muscle, were excised, weighed, placed in a tube with 10% vol/wt of glacial acetic acid, frozen in liquid nitrogen, and stored at -80 ºC.

4.3.3 Tissue sample extraction for total naringenin analysis

About 100 mg of tissue samples (4-5 mice / dietary group) were probe sonicated with

400µL water and 800µL of acetonitrile to precipitate proteins before centrifugation.

Pellets were re-suspended in 1.2 mL 2:1 volume/ volume acetonitrile: water and probe sonicated again and then centrifuged. For plasma, 2 volume of acetonitrile were added to plasma for the first extraction and repeated as described above. The supernatants were pooled and dried by SpeedVac. Residues were resuspended in 1mL of 2M acetate buffer, pH 5.5, and glucuronidase/sulfatase was added to deconjugate naringenin metabolites for

2hr. Naringenin aglycone was extracted twice with 3 volumes diethyl ether and dried under nitrogen. Residues were redissolved in 300µL methanol for HPLC-MS/MS analysis.

4.3.4 HPLC-MS/MS

Samples were injected on an Agilent 1200SL HPLC and separated with a 4.6 x 75mm,

3µm ACE C18AR column (MacMod Analytical Inc., Chadds Ford, PA) using a 0.1%

77 formic acid in water and 0.1% formic acid in acetonitrile mobile phase gradient. HPLC eluent was interfaced with a QTrap 5500 mass spectrometer (ABSciex, Concord, Canada) via an electrospray probe in both negative and positive polarity mode.

4.3.5 Study 2: The effects of naringenin on ovariectomized mice with diet-induced obesity

Because we did not observe developed mammary glands in the tissue sections stained with hematoxylin and eosin (H&E) and the CON mice stopped gaining weight after 10 weeks of the high-fat diet in study 1, we used older mice to better mimic the postmenopausal state and fed the mice with higher fat diet to induce obesity in study 2.

Twenty-week old female C57BL/6J mice (n=20) mice, ovariectomized at 19 weeks of age, were fed a semi-purified very high-fat diet (VHF diet, D12492, 60% calories from fat, Research Diets Inc.) for 11 weeks to induce obesity. The mice were then randomized by weight to one of two diet groups: VHF group, continued on the high-fat diet, and

VHFN group, switched to the very high-fat diet supplemented with 3% naringenin. After

11 weeks on the randomized diets, mice were fasted for 5 hours, anesthetized with isoflurane for blood collection via cardiac puncture, and then euthanized by cervical dislocation. Blood was collected into EDTA-coated tubes and plasma was obtained after centrifugation at 1,500 g for 20 min at 4 ºC. Tissues, as mentioned in Study 1, were excised, weighed, snap frozen in liquid nitrogen, and stored at -80 ºC until further analysis.

4.3.6 Fasting glucose analysis 78 After a 5 hour-fast, glucose was measured via tail vein puncture (OneTouch Ultra blood glucose meter, LifeScan Inc., Milpitas, CA, USA) at week 0 and 5.

4.3.7 Plasma analysis

Plasma insulin levels were measured by ELISA (Millipore, Billerica, MA, USA) according to the manufacturer’s instructions. Plasma glucose and triglycerides were assayed using a Cholestech LDX analyzer (Cholestech Corporation, Hayward, CA,

USA). Total plasma cholesterol (Pointe Scientific Inc., Canton, MI, USA) was determined by enzymatic colorimetric assays.

4.3.8 Indirect calorimetry and locomotor activity

During week 17, 6-8 mice in each group were housed individually in metabolic chambers at 22°C, allowed free access to food and water, and acclimated for 24-hour prior to metabolic assessments. Measurements were taken for a 24-hour period, including a 12- hour light cycle and a 12-hour dark cycle. Oxygen consumption (VO2), carbon dioxide production (VCO2), and ambulatory activity (by infrared beam breaks) were measured every 20 minutes using a computer-controlled, open-circuit Oxymax /CLAMS System

(Columbus Instruments, Columbus, OH, USA). Respiratory exchange ratio (RER) was calculated as the ratio of VCO2 to VO2. Heat, representing the standard measure of energy expenditure, was calculated by the formula,

Heat (kcal/hr) = (3.815 + 1.232 x RER) x (VO2) x (1L/1000 ml x 1kg/1000g) x (body weight)

79 4.3.9 Magnetic resonance imaging (MRI)

During week 19, total, intra-abdominal, and subcutaneous adiposity were analyzed by

MRI using a Bruker Biospin 94/30 magnet (Billerica, MA, USA) and a 70 mm diameter linear volume coil, as previously described (13). T1-weighted coronal images of the whole mouse torso were collected using a respiratory-gated RARE sequence

(TR/TE=1570/7.5ms, RARE factor=4, FOV=70x45 mm2, matrix size=256x192, slice thickness=1 mm, navg=2). The whole body and abdominal masks were then used to calculate the percentage of segmented voxels in the whole body and abdomen, respectively. Subcutaneous fat was calculated by subtracting intra-abdominal fat from total fat.

4.3.10 Triacylglycerol and diacylglycerol analysis

Total lipids were extracted from liver and muscle samples with 2:1 (v/v) chloroform: methanol and washed with 0.88% KCL (219). The chloroform phase was transferred to a weighed test tube and dried under nitrogen gas at room temperature. The dried sample and test tube was weighed again to calculate total extracted lipid. Following the lipid extraction, triacylglycerol and diacylglycerol were obtained using solid-phase extraction

(203) and solubilized in tert-butanol, methanol, Triton X-100 (204). Analysis was performed using enzymatic colorimetric assay (205).

4.3.11 RNA extraction and RT-PCR analysis of gene expression

Total RNA was extracted from perigonadal adipose tissue using RNeasy Lipid Tissue

Mini Kit (Qiagen, Valencia, CA, USA) following manufacturer’s instructions. Muscle 80 total RNA was extracted with TriZol reagent (Invitrogen, Carlsbad, CA, USA) and precipitated with the addition of a high salt solution (1.2M NaCl, 0.8 M sodium citrate).

RNA was then reversed transcribed to cDNA using a High Capacity cDNA Reverse

Transcription Kit (Applied Biosystems, Foster City, CA, USA). Quantitative RT-PCR analysis of individual cDNA was performed with ABI Prism 7300 sequence detection system (Applied Biosystems) using TaqMan Gene Expression. Target gene expression was normalized to 18S rRNA for perigonadal adipose tissue or glyceraldehyde 3- phosphate dehydrogenase (GAPDH) for muscle tissue. Endogenous Controls (VIC probes) were amplified in the same reaction and expressed as 2^(-ddCt) compared to the

VHF group (206).

4.3.12 Western blotting

Muscle was homogenized in lysis buffer (20 mM Trizma base, 1% Triton-X100, 50 mM

NaCl, 250 mM sucrose, 50 mM NaF, and 5 mM Na4P2O7 ! 10H2O) with the Complete

Mini Protease Inhibitor Cocktail (Roche Diagnostics, Indianapolis, IN, USA) and

PhosSTOP Phosphatase Inhibitor Cocktail Tablets (Roche Diagnostics). Homogenates were incubated for 1 hour and centrifuged at 16,000 g for 15 minutes at 4°C. Supernatant was collected and protein concentration was determined by BCA Protein Assay Kit

(Pierce Biotechnology, Rockford, IL, USA). Protein (40 µg/sample) was separated on polyacrylamide gels and transferred to nitrocellulose membranes using transfer buffer containing 25 mM Tris,192 mM glycine, and 10% methanol. Membranes were blocked with 5% non-fat dry milk in Tris-buffered saline with 0.1% Tween-20 (TBST) and incubated overnight at 4°C with primary antibodies (Cell Signaling Technology, 81 Danvers, MA, USA). After washing, membranes were incubated for 1 hour at room temperature with HRP-linked anti-rabbit IgG, (Cell Signaling Technology). Bands were visualized with chemiluminescence (SuperSignal West Femto Maximum Sensitivity

Substrate, Thermo Fisher Scientific, Rockford, IL, USA) using Carestream Image Station

4000GL PRO (Carestream Health, Inc., Rochester, NY, USA). Densitometric analysis was conducted using Carestream Molecular Imaging Software. GAPDH was used as a loading control.

4.3.13 Statistical analysis

All data are presented as mean ± standard error (SEM) with P<0.05 considered significantly different. Statistical analysis was performed using SPSS version 20.0 software (SPSS, Inc., Chicago, IL, USA). Significance was determined using two-tailed unpaired Student’s t test, and naringenin accumulation results were analyzed by one-way analysis of variance (ANOVA) followed by Tukey's multiple comparison tests. Heat

(energy expenditure) was analyzed by ANCOVA with body weight as a covariate (207).

Pearson correlation coefficient was used for correlation analyses.

82 4.4 Results

4.4.1 Naringenin accumulation

To examine whether increasing doses of dietary naringenin results in increased naringenin accumulation in plasma and tissues, ovariectomized mice were fed high-fat diets containing 0%, 1%, and 3% of naringenin for 11 weeks. No adverse clinical signs, sudden death, and differences in spleen and kidney weights were observed (data not shown), suggesting that 3% naringenin was well tolerated in these mice. Body weights and cumulative food intake were not different between diet groups (data not shown).

Total naringenin concentrations in plasma and tissues were 5-12 times higher in mice fed a diet with 3% naringenin compared to mice fed 1% naringenin (Fig. 7). Mice fed 3% naringenin accumulated 18.03 ± 4.72 µM total naringenin in plasma compared to 1.50 ±

0.46 µM in mice fed a diet with 1% naringenin. Relative accumulation of naringenin in tissues was greatest in liver (1% NAR, 7.61 ± 1.14 µmol/kg tissue and 3% NAR, 37.04 ±

8.33 µmol/kg tissue) followed by perigonadal adipose tissue (PGAT, 1% NAR, 0.85 ±

0.30 µmol/kg tissue and 3% NAR, 5.11 ± 1.26 µmol/kg tissue), and lastly subcutaneous adipose tissues (SCAT) and muscle, which had similar magnitudes of naringenin accumulation (SCAT, 1% NAR, 0.47 ± 0.19 µmol/kg tissue and 3% NAR, 3.02 ± 1.99

µmol/kg tissue; muscle, 1% NAR, 0.30 ± 0.06 µmol/kg tissue and 3% NAR, 2.17 ± 1.04

µmol/kg tissue).

Figure 7. Naringenin accumulation in plasma and tissues after 11 weeks of supplementation. OVX C57BL/6J mice were fed a high-fat diet for 6 weeks and switched to the high-fat diet with 0%, 1%, and 3% wt/wt naringenin. After 11 weeks of the diets, total naringenin concentrations were 5-12 times higher in plasma (n=4/group) or tissue samples (n=5/group) of mice fed 3% naringenin than the levels in mice fed 1% naringenin. Values are presented as mean ± SEM. Data were analyzed using one-way ANOVA with post hoc Tukey’s test. Significant difference (P < 0.05) between groups with different mean superscripts.

84 4.4.2 Body weight and caloric intake

In the second study, mice were fed a very high-fat diet (VHF) to induce an obese phenotype. After 11 weeks on a very high-fat diet, mice gained 19 g (Fig. 8A) and exhibited dysregulated glucose metabolism as manifested by elevated blood glucose levels (178.4 ± 4.3 mg/dL compared to 135.3 ± 4.5 mg/dL at week 0). Then, half of the mice were switched to the VHF supplemented with 3% naringenin (VHFN). Mice fed naringenin gained only 0.89 ± 1.52 g in 11 weeks of treatment, compared to VHF mice, which gained 4.09 ± 0.83 g (Fig. 8A). Cumulative caloric intake was not different between groups (Fig. 8B). Feed efficiency was calculated to measure the ability of the mice to transform ingested calories into body weight. A decrease in the feed efficiency ratio was observed in VHF group from weeks 0 - 11 to weeks 12 - 22 as body weight of the VHF group reached a plateau at week 16. Nonetheless, naringenin tended to lower feed efficiency (p=0.06, Table 7).

Figure 8. Effect of naringenin supplementation on body weight and caloric intake. OVX C57BL/6J mice were fed a very high-fat diet for 11 weeks. Then half of the mice (VHFN) were given a very high-fat diet supplemented with 3% naringenin for another 11 weeks. (A) Body weight (n=10/group). (B) Cumulative caloric intake of mice (n=2/group). Values are presented as mean ± SEM. Significance between groups was determined by Student’s t test. *P < 0.05 compared VHFN with VHF.

86 Table 7. Effect of dietary naringenin on feed efficiency and tissue mass in obese ovariectomized C57BL/6J female mice VHF VHFN Feed Efficiency (mg/kcal) Before Supplementation (0-11 week) 0.022 ± 0.002 0.021 ± 0.001 After Supplementation (12-22 week) 0.005 ± 0.001 0.001 ± 0.002

Tissue Weights (g)

Subcutaneous Adipose Tissue 3.78 ± 0.23 3.43 ± 0.24

Liver 1.42 ± 0.07 1.35 ± 0.04

Perigonadal Adipose Tissue 3.79 ± 0.09 3.20 ± 0.18

Quadriceps Muscle 0.36 ± 0.01 0.39 ± 0.01

Tissue Weight Percentages (%)

Subcutaneous Adipose Tissue 8.26 ± 0.28 8.06 ± 0.33

Liver 3.12 ± 0.09 3.22 ± 0.08

Perigonadal Adipose Tissue 7.81 ± 0.25 7.55 ± 0.23

Quadriceps Muscle 0.79 ± 0.02 0.93 ± 0.03* Feed efficiency ratio = body weight gain (mg) /cumulative caloric intake (kcal) Significance between groups was determined by Student's T-test (n=10/group). Data represent the mean ± SEM. *P < 0.05 compared VHFN with VHF

87 4.4.3 Locomotor activity, energy expenditure, and respiratory exchanging ratio

Indirect calorimetry was performed to assess metabolic changes after naringenin supplementation. Compared to ambulatory activity levels of ovariectomized mice fed a normal-fat diet from our previous study (208), the activity levels were reduced by 50% in

VHF mice (156 ± 13 counts vs. 76 ± 11 counts in the light phase and 428 ± 52 counts vs.192 ± 42 counts in the dark phase). In the current study, mice with naringenin supplementation (VHFN) increased activity counts about 2 fold in both light and dark phases compared to VHF mice (VHFN: 134 ± 15 vs. VHF: 76 ± 11 in the light phase and

VHFN: 455 ± 40 vs. VHF: 192 ± 42 in the dark phase, Fig. 9A & B). There were no differences in energy expenditure between groups after adjusting for body weight (Fig.

9C). RER was around 0.7 in both groups, indicating that fat was the predominant fuel source for energy need (Fig. 9D). Interestingly, mice supplemented with naringenin had higher RER at specific time points in the light phase, suggesting a decrease in reliance on fatty acid oxidation for energy needs.

88 (A) (B)

2000 600 Light Dark VHF VHF * * VHFN VHFN * 1500 400 * * 1000

200 Activity Count Activity Count 500 *

0 0 6 AM 9 AM 12 PM 3 PM 6 PM 9 PM 12 AM 3 AM 6 AM Dark Light Time (C) (D)

0.85 1.0 VHF Light Dark VHF VHFN VHFN 0.8 0.80

0.6 * * 0.75 * * 0.4

Heat (kcal/hr) 0.70 0.2 RER (Arbitrary Units)

0.0 0.65 Dark Light 6 AM 9 AM 12 PM 3 PM 6 PM 9 PM 12 AM 3 AM 6 AM Time

Figure 9. Effect of naringenin supplementation on indirect calorimetry and locomotor activity. At week 17, metabolic parameters were measured in the CLAMS chambers for 24 hours following a 24-hour acclimation (n=6-8/group). (A) Ambulatory locomotor activity in the horizontal plane measured by infrared beam breaks. (B) Ambulatory counts in the dark and light phase. (C) Estimated energy expenditure (heat) after controlling for body weight in the dark and light phase. (D) Respiratory exchanging ratio (RER). Values are presented as mean ± SEM. Significance between groups was determined by Student’s t test, and metabolic data of heat was analyzed by ANCOVA with body weight as a covariate. *P < 0.05 compared VHFN with VHF.

89 4.4.4 Adiposity and tissue weights

After 19 weeks of VHF diet (at the time when the MRI analysis was performed), the mice accumulated 64% of body fat and 30% of the body fat accumulated in intra- abdominal area. Dietary naringenin significantly decreased total body fat by 11% and intra-abdominal adiposity by 20% compared to VHF mice (Fig. 10A & B). Subcutaneous adiposity was unaffected. At the end of the study, neither subcutaneous nor perigonadal adipose tissue mass were significantly different between the groups (Table 7). Monocyte chemoattractant protein-1 (MCP1/Ccl2) is a key inflammatory mediator in adipose tissues. In our previous study with ovariectomized mice fed a normal-fat diet (219), naringenin significantly down-regulated mRNA levels of MCP-1 mRNA levels in perigonadal adipose tissue. When compared with our previous ovariectomized mice fed a normal-fat diet, MCP-1 expression was 4-fold higher in perigonadal adipose tissue of

VHF mice (data not shown). In the current study, naringenin significantly reduced MCP-

1 mRNA expression by 40% in perigonadal adipose tissue of obese ovariectomized mice.

Weight loss is often associated with a decrease in lean mass (219). Here we found maintenance of quadriceps muscle mass in naringenin-fed mice (Table 7) and negatively correlated with total and intra-abdominal adiposity (Total, r = -0.808, p = 0.001; intra- abdominal, r= -0.924, p<0.001). The muscle mass was significantly higher in VHFN mice compared to VHF mice after normalized to body weight muscle mass (Table 7).

90 (A) (B) (C)

100 1.5

1.0 60 * * 40 0.5 Adiposity (%) * 20 normalized to 18s Ccl2 mRNA expression

VHF VHFN 0 0.0 VHF VHFN VHF VHFN VHF VHFN VHFVHFN Total Intra-abdominal Subcutaneous

Figure 10. Effect of naringenin supplementation on adiposity, fat distribution, and MCP/Ccl2 mRNA expression in perigonadal adipose tissue. (A) Representative coronal views of MRI from each group. (B) Percentage of total body fat and percentage of two adipose tissue depots, intra-abdominal and subcutaneous adipose tissues were determined by MRI (n=6/group). (C) mRNA expression of MCP1/Ccl2 in perigonadal adipose tissue (n=8/group). Values are presented as mean ± SEM. Significance between groups was determined by Student’s t test. *P < 0.05 compared VHFN with VHF.

4.4.5 Plasma analyses and hepatic and muscle lipid content

Naringenin supplementation resulted in decreased fasting plasma glucose levels

(Fig. 11A), but did not affect fasting plasma insulin levels (Fig. 11B). Naringenin supplementation also failed to attenuate plasma cholesterol and triglyceride in obese ovariectomized mice (Fig. 11C & D). We quantified the levels of triacylglycerol and diacylglycerol in liver and muscle. No significant difference was observed in extracted triacylglycerol and diacylglycerol levels in liver between groups (Fig. 12A & B). On the contrary, extracted diacylglycerol levels in muscle were significantly lower in naringenin-fed mice without changes in extracted triacylglycerol levels (Fig. 12C & D).

Figure 11. Effect of naringenin supplementation on plasma analysis. (A) Fasting plasma glucose levels, (B) Fasting plasma insulin levels, (C) Fasting plasma cholesterol levels, and (D) Fasting plasma triglyceride levels. Values are presented as mean ± SEM (n=5-8/group). Significance between groups was determined by Student’s t test. *P < 0.05 compared VHFN with VHF.

Figure 12. Effect of naringenin supplementation on hepatic and muscle lipid profiles. (A) Extracted triacylglycerol content per gram of liver section. (B) Extracted diacylglycerol content per gram of liver section. (C) Extracted triacylglycerol content per gram of muscle section. (D) Extracted diacylglycerol content per gram of muscle section. Values are presented as mean ± SEM (n=9-10/group). Significance between groups was determined by Student’s t test. *P < 0.05 compared VHFN with VHF.

93 4.4.6 Mitochondrial biogenesis in muscle

Exercise activates signaling cascades, induces changes in gene expression, and results in mitochondrial biogenesis (220). Since VHFN mice showed elevated locomotor activity, we measured expression of mRNA for mitochondrial biogenesis (Fig. 13A).

Peroxisome proliferator-activated receptor-γ coactivator 1 alpha (ppargc1a, PGC1α), a transcriptional coactivator, acts as a central regulator of mitochondrial biogenesis and its expression has been shown to be increased in human muscle after exercise (221). When compared with ovariectomized mice fed a normal-fat diet (222), the RNA levels of

PGC1α were 6-fold lower in the muscle of VHF mice. The levels of PGC1α transcript were about 1.4 fold greater in the muscle of VHFN mice compared to VHF mice, but did not reach significance (p=0.17). Nuclear respiratory factor 1 (NRF1) interacts with

PGC1α and induces expression of transcription factors essential for proper mitochondrial

DNA transcription, including mitochondrial transcription factor A (Tfam), mitochondrial transcription factor B1 (Tfb1m), and mitochondrial transcription factor B2 (Tfb2m)

(219). No difference was observed in mRNA levels of these genes between the groups. A previous study showed that PGC1α was up-regulated 2 hours after a single bout of exercise in human muscle but down-regulated after 24 hours, while mRNA levels of

COX IV (COX4I1), a nuclear encoded protein of the mitochondrial respiratory chain, and mitochondrial fusion proteins Mfn1 and Mfn2 were induced until 24-hour post-exercise

(221). Therefore, we measured the transcript levels of these late-responsive genes but still found no difference between two groups of mice. We also measured mitochondrial DNA

(mtDNA) copy number, a surrogate marker of mitochondrial function (223), yet found no

94 difference. Collectively, our data suggested that increased activity did not induce mitochondrial biogenesis in postmenopausal obese mice.

4.4.7 Markers of lipid metabolism in muscle

Since we observed a reduction in muscle diacylglycerol in the naringenin-fed mice, we measured muscle mRNA levels of genes involved in lipid metabolism. The expression of lipoprotein lipase (Lpl), an enzyme that releases fatty acids from circulating triglycerides and facilitates uptake of fatty acids by muscle cells, and the expression of carnitine palmitoyltransferase I beta (Cpt1b), a key regulator of mitochondrial fatty acid transport, were not different between groups (Fig. 13B). However, the expression of fatty acid synthase (Fasn) and stearoyl-CoA desaturase (Scd1), two genes encoding proteins involved in de novo lipogenesis, was significantly lower (by 70%) in the mice fed a naringenin diet. Transcript levels of diacylglycerol acyltransferase-2 (Dgat2), an enzyme that incorporates diacylglyceride to triacylglyceride, were down-regulated in naringenin- fed mice. We also measured the levels of genes involved in intramuscular lipolysis, e.g., adipose triglyceride lipase (ATGL, Pnpla2) and hormone-sensitive lipase (HSL, Lipe).

The expression of ATGL was unaffected, while, mRNA levels of HSL were decreased by

40% in naringenin-fed mice.

Figure 13. Effect of naringenin supplementation on mRNA expression in muscle. mRNA expression of genes for mitochondrial biogenesis (A) and genes involved in lipid metabolism (B). Values are presented as mean ± SEM (n=9-10/group). Significance between groups was determined by Student’s t test. *P < 0.05 compared VHFN with VHF.

96 4.4.8 AMPK and Akt signaling in muscle

Previous studies have shown that naringenin stimulates glucose uptake in L6 myotubes by activating 5’-AMP-activated protein kinase (AMPK) (222). However, we did not detect a significant difference between groups in the levels of AMPK Thr172 phosphorylation, a measure of AMPK activation (Fig. 14A). It has been suspected that accumulation of intracellular fatty acid metabolites, diacylglycerol and ceramide, are the key intermediates that affect insulin signaling, leading to insulin resistance in skeletal muscle (224). Since we observed significant decreases in both plasma glucose levels and muscle diacylglycerol content in the mice fed naringenin, we measured the activity of protein kinase Akt/PKB, a central mediator of insulin signaling. However, we did not observed increased phosphorylation levels of Akt at both phosphorylation sites, Ser 473 and Thr 308, in muscle of naringenin-fed mice (Fig. 14B).

Figure 14. Effect of naringenin supplementation on AMPK and Akt activity in muscle. Ratio of phosphorylated AMPK Thr172 to total AMPK protein with representative blots (A) and ratio of phosphorylated Akt Ser473 and Thr308 to total Akt protein with representative blots (B) in muscle. Values are presented as mean ± SEM (n=5-10/group). Significance between groups was determined by Student’s t test. *P < 0.05 compared VHFN with VHF.

98 4.5 Discussion

The present study reveals novel information on naringenin accumulation in tissues after long-term supplementation. Eleven weeks of 3% naringenin feeding resulted in naringenin accumulation in plasma, liver, perigonadal and subcutaneous adipose tissues, and muscle more than 3 times of the amount accumulated in mice fed 1% naringenin.

Furthermore, the present study evaluated the effects of dietary naringenin on body weight and energy metabolism in a mouse model of postmenopausal obesity. Dietary naringenin suppressed body weight gain, decreased intra-abdominal fat, and reduced plasma glucose levels. Ovariectomized mice fed naringenin also exhibited a higher locomotor activity, the maintenance of muscle mass, a reduction in muscle diacylglycerol content, and changes in muscle lipid metabolism.

Naringenin bioavailability has gained much interest due to its potential beneficial effects on human health. Studies in humans and rodents suggest that bioavailability of naringenin could be affected by the form of naringenin consumed. In humans, after consumption of cooked tomato paste, the time needed to reach peak concentration

(Tmax) of naringenin in plasma was 2 hours (225) compared to 4.8 hours and 5.5 hours after consuming orange and grapefruit juices, respectively (25). The differences in Tmax could be due to the different forms of naringenin ingested (glycosidic forms such as naringenin 7-neohesperidoside and -rhamnoglucoside in orange and grapefruit juices and the aglyconic form, naringenin, in tomato skin) that may be absorbed via different routes in the digestive tract. Rodent studies have also demonstrated that naringenin, aglycone form, and naringenin-7-glucoside have faster absorption kinetics and higher 24-hour urinary excretion compared to naringenin-7-rhamnoglucoside (narirutin), one of the 99 predominant forms in grapefruit (24, 168). Therefore, in the present study, we supplemented aglyconic form of naringenin into the diets to reduce inter-individual variances on hydrolysis of glycosidic linkage. At the end of the study, we estimated that the mice fed a 3% naringenin diet consumed about 1g/kg body weight of naringenin, which equates to 9 g of naringenin or 17L of grapefruit juice for a 60 kg person based on surface area (175).

Most studies examined bioavailability of naringenin by using plasma kinetics and urinary excretion as indicators and only gave a single-dose administration. However, the response to naringenin in blood and urine could simply reflect the release of naringenin from the food matrix and absorption and not accurately indicate the degree of tissue uptake (226). Therefore, the bioavailability of naringenin and its metabolites in target tissues is much more important than the knowledge of its concentration in plasma and urine. In the present study, we demonstrated that the concentrations of total naringenin in tissues of mice fed a diet with 3% naringenin were more than 3 times higher (5-12 times) than the accumulation in tissues of mice fed a 1% naringenin-containing diet, suggesting that higher naringenin levels in tissues from 3% naringenin supplementation may have resulted in the beneficial effects presently observed. Our pilot studies (227) and other studies (228) showed that the major forms of naringenin in blood and organs are sulfates and glucuronides. In the present study, we only measured total naringenin concentration and did not identify and quantify of naringenin and its metabolites in tissues. The release, expression, and function of enzymes responsible for glucuronide or sulfate conjugation have shown to be altered during inflammation or progression of nonalcoholic fatty liver disease (177, 228). Therefore, it is intriguing to investigate the changes of naringenin 100 metabolite profile and identify the potential biological activity of the metabolites during progression of the diseases.

Recently, we reported that dietary naringenin decreased body weight, reduced adiposity, and improved glucose metabolism in ovariectomized mice (229). In the present study, we also observed reduction of weight gain and abdominal adiposity in the obese ovariectomized mice fed a naringenin-enriched diet. In accordance with our previous data, dietary naringenin reduced intra-abdominal adiposity as well as expression of MCP-

1/ccl2 in perigonadal adipose tissue of the obese mice. Although plasma glucose levels were lower in obese mice fed naringenin, we did not observe changes in plasma insulin.

However, the mice began naringenin supplementation when they had already gained

179% of their baseline body weight. It is possible that naringenin supplementation may result in better outcomes if naringenin is given before the onset of severe obesity.

Bjursell et al (2008) demonstrated that reduced locomotor activity occurred in mice

3-5 hour after switching to a high-fat diet and the reduction was sustained after 21 days of feeding, when energy intake, energy absorption, and energy expenditure were normalized

(219). In the present study, dietary naringenin did not affect caloric intake but increased locomotor activity levels in obese ovariectomized mice to the levels shown in our previous study with ovariectomized mice fed a normal-fat diet (230). It is likely that recovery of locomotor activity by naringenin results in the suppression of body weight in obese mice. Additionally, Janssen et al (219) showed that a high prevalence of women over age 60 in the Third National Health and Nutrition Examination Survey (NHANES

III) exhibited sarcopenia (loss of skeletal muscle mass with aging). Sarcopenic obesity

(low muscle mass with high body fat) increases with aging and is associated with 101 functional impairment, disabilities, and falls (218). We found that naringenin maintained quadriceps muscle mass in obese ovariectomized mice. Further investigation is required to determine the influence of naringenin in the maintenance of muscle health and prevention of sarcopenia.

Mulvihill et al (2009) demonstrated that dietary naringenin decreased fatty acid synthesis in muscle but did not affect fatty acid oxidation in muscle of low-density lipoprotein receptor (LDLR) knockout mice fed a high-fat diet (231). In line with their findings, dietary naringenin did not affect fatty acid oxidation in muscle of obese ovariectomized mice but down-regulated mRNA expression of genes involved in lipid metabolism, including de novo lipogenesis, lipolysis, and triglyceride synthesis/storage.

Although fatty acid synthase expression in skeletal muscle is not abundant and the role of de novo lipogenesis was assumed negligible (13), skeletal muscle specific knockdown of fatty acid synthase increased whole body glucose tolerance and insulin sensitivity in mice fed a high-fat diet (232). Concurrently, the levels of diacylglycerol but not triacylglycerol were lower in the obese mice fed a naringenin-enriched diet, suggesting little contribution of de novo lipogenesis on triacylglyceride content in muscle. Expression of two genes that directly regulate diacylglycerol metabolism, DGAT2 and HSL, were significantly decreased in naringenin-fed obese mice. Interestingly, HSL knockout mice exhibited impaired insulin signaling and diacylglycerol accumulation in muscle (233). Furthermore,

DGAT2 overexpression in muscle resulted in accumulation of triacylglycerol and ceramide but a reduction in diacylglycerol levels (234). Thus, the reduction of mRNA expression of HSL and DGAT2 may be the result, rather than the cause of lower diacylglycerol content in muscle. Several lines of evidence suggest that diacylglycerol 102 acts like a signaling molecule to activate protein kinase C (PKC) isoforms, resulting in impaired insulin signaling in liver and muscle (235). Although we observed lower glucose levels in naringenin-fed mice, we did not observe changes in Akt activity in muscle, or changes in total body insulin sensitivity by performing an insulin tolerance test

(data not shown). Future studies should use hyperinsulinemic-euglycemic clamp to further investigate the effect of naringenin on muscle insulin sensitivity.

As life expectancy increases, the postmenopausal period for women will likely increase as well. Considering the prevalence of obesity in women and the increasing rate of obesity after menopause (236), the need for approaches that reduce the rate of postmenopausal obesity is crucial. This present study demonstrates that naringenin supplementation maintains body weight, increases activity, reduces intra-abdominal adiposity, and decreases plasma glucose levels in obese ovariectomized mice.

Additionally, in muscle, dietary naringenin maintains tissue mass, reduces diacylglycerol content, and alters lipid metabolism. The results suggest that naringenin may modulate energy metabolism in obese menopausal women and is a potential supplement to treat postmenopausal obesity.

103 4.6 Acknowledgements

The authors thank Josephine Fouts for assistance with mouse experimentation, Dr.

Rebecca R. Andridge for statistical consulting, Dr. Kenneth M Riedl for insights on naringenin accumulation, and Dr. Santosh K. Maurya and Dr. Muthu Periasamy for assistance with indirect calorimetry measurements. The authors would also like to acknowledge the Nutrient & Phytochemical Analytic Shared Resource (NPASR) for naringenin accumulation analysis and OSU Small Animal Imaging Core for MR imaging.

This research was supported by the NCI 1R03CA162551 (M.A.B.) and the Food

Innovation Center Doctoral Research Grant (J-Y. K).

104

CHAPTER 5

The effect of citrus flavonoid naringenin and metformin on the growth of breast

cancer cells in obese ovariectomized mice

Jia-Yu Ke, Yung-Hsuan Hsiao, Rachel M Cole, Shana R Straka, Lisa D Yee, Martha A

Belury

105 5.1 Abstract

Naringenin is a flavonoid abundant in grapefruits that may display beneficial effects on metabolic health and tumorigenesis. We presented here that naringenin inhibited cell proliferation and induced cells death in E0771 murine breast cancer cells.

We also compared the effect of naringenin and metformin, a glucose-lowering drug shown to decrease breast cancer incidence in diabetic women, on suppression of mammary tumor growth in a mouse model of postmenopausal obesity. Ovariectomized mice were fed a high-fat (HF) diet for 3 weeks and then mice were randomized to HF alone, LN (HF + 1% wt/wt naringenin), HN (HF + 3% wt/wt naringenin), and Met (HF + metformin) for 5 weeks. After 2 weeks of experimental diets, E0771 cells were inoculated into one mammary fat pad. HN significantly decreased body weight, adipose depot mass, as well as mRNA expression of inflammatory cytokines in both mammary and perigonadal adipose tissues. HN suppressed tumor growth on day 14 and day 17 but final tumor weight was not significantly different from HF group. Metformin reduced tumor growth and weight, without affecting body weight, tissue weights, and adipose tissue inflammation. Collectively, the data suggests different mechanisms of action for naringenin and metformin to alter mammary tumorigenesis.

106 5.2 Introduction

Breast cancer is the second leading cause of cancer deaths in women (8). About

67% of incidence and 80% of deaths were occurred in women past the age of menopause

(9). Obesity is a well-established risk factor for postmenopausal breast cancer. Results from NIH-AARP Diet and Health Study showed that the risk of breast cancer was twice as high in obese postmenopausal women compared to their lean postmenopausal peers

(50). In the Nurses’ Health Study, women who gained more than 22 pounds after menopause had an 18% greater risk of breast cancer (237). The link between obesity and postmenopausal breast cancer is likely associated with obesity-related hormonal and metabolic dysregulation. Metformin, a common oral medication used for diabetes treatment, may be useful as an approach to decrease the risk of obesity related breast cancer. Associated with decreased incidence of breast cancer in women with diabetes

(103, 146), metformin is now under investigation in clinical trials as adjuvant therapy and chemoprevention for breast cancer (238). One drawback to long-term preventive therapy with metformin may be undesirable side effects, such as, diarrhea, nausea, and vitamin

B12 deficiency (239). Of interest are natural compounds derived from foods that exert similar effects as metformin and could offer a safe, non-toxic alternative approach to breast cancer prevention.

The glucose-lowering and anti-tumorigenic activities of metformin are mediated partially through AMP-activated protein kinase (AMPK) (240), an energy sensor that regulates whole-body energy homeostasis. We previously found that a naturally-derived flavonoid, naringenin, acts like metformin to induce AMP-activated protein kinase

(AMPK) and reduce glucose production in Fao hepatoma cells (148). Naringenin, 107 abundant in citrus fruits and tomatoes, ameliorates metabolic dysregulation associated with diet-induced obesity in animals (14). Male mice fed a high-fat diet supplemented with naringenin showed suppression of diet-induced weight gain, adipose mass, blood glucose and insulin levels, and increased insulin sensitivity (164). We also reported that naringenin exhibited similar effects on blood glucose levels and adipose mass in lean and obese ovariectomized (OVX) mice, a mouse model for human postmenopause (13, 241).

Additionally, studies revealed that naringenin possesses anti-tumorigenic activity in various cancer cells and animal models (18, 20, 163, 196, 219, 242).

Based on these data, we hypothesize that naringenin improves metabolic disturbances and inhibits mammary tumor growth in postmenopausal obesity. In the present study, we examined the inhibitory effect of naringenin in breast cancer cells.

Then we compared the effect of dietary naringenin and metformin, respectively, on mammary tumorigenesis in a female mouse model for postmenopausal obesity.

108 5.3 Materials and methods

In vitro experiments

5.3.1 Cell lines and cell Culture

Murine E0771 breast cancer cell line was purchased from CH3 Biosystems (#940001,

Buffalo, NY, USA) and cultured with RPMI 1640 medium (#31800-022, Life

Technologies, Baltimore, MD, USA) supplemented with 5% fetal bovine serum (#26140-

079, Life Technologies) and 0.25% penicillin and streptomycin (#15140-148, Invitrogen,

Grand Island, New York, USA) at 37°C in a 5% CO2 incubator under standard conditions.

5.3.2 Cell viability assay

E0771 cells were plated at 8x104 cells/well of 12-well plate and were treated the next day

(50-60% confluence) with naringenin (W530098, St. Louis, MO, USA) dissolved in dimethyl sulfoxide (DMSO) at the indicated concentrations or the same volume of

DMSO. Cells were rinsed with phosphate buffered saline (PBS) to remove floating cells after incubation with different doses of naringenin for 24 or 48 hours. The cells were harvested by trypsinization and stained with 0.2% trypan blue. Cell numbers were counted by using a hemocytometer.

5.3.3 Flow cytometric analysis

Cell cycle phase distribution with cellular DNA contents were carried out using flow cytometry. E0771 cells were seeded into 6-cm dishes at density 4x105 cells/dish and treated with DMSO, 50 µM, and 100 µM of naringenin for 24 and 48 hours at 37°C in a 109 5% CO2 incubator. After 24- or 48-hour incubation, the cultured cells were harvested, washed with cold PBS, fixed in 70% ethanol and treated with RNase A (10 mg/ml).

Fixed cells were stained with propidium iodide (PI) dye followed by incubation for 30 min at room temperature in dark. The PI fluorescence of individual nuclei was measured using flow cytometer (FACS Calibur, Becton Dickinson, San Jose, USA). Data were analyzed with the Cell Quest Pro V 3.2.1 software (Becton Dickinson).

In vivo experiments

We have chosen ovariectomized C57BL/6 mice injected with a syngeneic breast carcinoma cell line (E0771) as our study model because 1) this model showed increased tumor weights in mice fed a high-fat diet compared to those fed a low-fat diet (22) and is thus a suitable model for our study design; 2) this model is a immunocompetent breast cancer mouse model that more fully mimics the human tumor microenvironment compared to human xenograft models in immunocompromised mice.

5.3.4 Animals and experimental procedures

Twenty-week old ovariectomized C57BL/6J mice were obtained from Jackson

Laboratory (Bar Harbor, ME, USA) and allowed to acclimate to the new environment for

2 to 3 weeks. Mice were individually housed at 22 ± 0.5°C on a 12:12-hour light-dark cycle. Mice (n=35) were fed a semi-purified high-fat diet (HFD, D12492, 60% calories from fat, Research Diets Inc. New Brunswick, NJ, USA) for 3 weeks and randomized by weight to one of four diet groups: HF group (n=10), continued on the high-fat diet, LN group (n=10), changed to the high-fat diet supplemented with 1 wt/wt % naringenin, HN 110 group (n=10), changed to the high-fat diet supplemented with 3 wt/wt % naringenin and

Met group (n=5), change to the high fat with 170 mg/kg metformin in water. Naringenin powder used for the diets was purchased through Sigma (W530098, St. Louis, MO, USA) and custom prepared by Research Diets Inc. The dose of naringenin is based on previous study showing amelioration of metabolic disturbances in mice fed a high-fat diet with 1% and 3% naringenin (26). After 13 days of treatment, mice were anesthetized with isofluorane and a 0.5 cm incision in the skin above the chest was performed to inject 5 x

105 E0771 cells in 50 µL PBS into one thoracic mammary fat pad. The incision was closed by skin clip and mice received a 5 mg/kg subcutaneous injection of Carprofen in sterile saline solution (0.9%) for pain relief immediately after surgery. Mice were maintained on study diets for another 20 days. Body weight and food intake were measured daily. Two perpendicular dimensions of tumors were measured using a dial caliper. Mice were fasted for 5 hours, anesthetized with isoflurane for blood collection via cardiac puncture, and then euthanized by cervical dislocation. Blood was collected in

EDTA-coated collection tubes and plasma was obtained after centrifugation at 1,500 g for 20 min at 4 ºC. 100 µl of plasma was added with 10% vol/vol of glacial acetic acid and stored at -80 ºC. Thoracic mammary fat pad and tumor (~100 mg) were excised, placed in a tube with 10% vol/wt of glacial acetic acid, frozen in liquid nitrogen, and stored at -80 ºC. for total naringenin accumulation analysis. Tissues, including liver, abdominal mammary fat pads, perigonadal fat (PGAT), and quadriceps skeletal muscle, were excised, weighed, and frozen in liquid nitrogen, and stored at -80 ºC for further analysis. Tumor injection to one mouse in HF group and two mice in LN group was unsuccessful and these three mice were excluded from the analysis. The procedures were 111 in accordance with institution guidelines and approved by the Institutional Animal Care and Use Committee at The Ohio State University.

5.3.5 Metformin

Metformin (#D150959, Sigma Aldrich, St. Louis, MO, USA) was administrated 1.122 mg/ml to the drinking water. Administration of metformin through water instead of a daily intraperitoneal injection is to ensure continuous exposure of mice to the treatment.

Continuous exposure more closely resembles human exposure with extended release tablets of metformin now prescribed for patients with type 2 diabetes. Since mice usually drink 3-5 ml a day (1.5 ml/10 g body weight/day), the dose given through water was about 170 mg/kg body weight a day. The dose administrated here is equivalent to 14 mg/kg/day for human (980 mg for a 70 kg person) based on surface area (13), which is a dose lower than doses used in clinical trials for overweight or obese patients with higher risk for breast cancer (eg. NCT01793948, 850mg orally twice a day).

5.3.6 Tissue sample extraction for total naringenin analysis

About 100 mg of tissue samples (4-5 mice / dietary group) were probe sonicated with

400µL water and 800µL of acetonitrile was added to precipitate proteins before centrifugation. Pellets were re-suspended in 1.2 mL 2:1vol/vol acetonitrile: water and probe sonicated again and then centrifuged. For plasma, 2 vol of acetonitrile were added to plasma for the first extraction and repeated as described above. The supernatants were pooled and dried by speedvac. Residues were resuspended in 1mL of 2M acetate buffer, pH 5.5, and glucuronidase/sulfatase was added to deconjugate naringenin metabolites for 112 2 hours. Naringenin aglycone was extracted twice with 3 volumes diethyl ether and dried under nitrogen. Residues were redissolved in 300uL methanol for HPLC-MS/MS analysis.

5.3.7 HPLC-MS/MS

Samples were injected on an Agilent 1200SL HPLC and separated with a 4.6x75mm,

3µm ACE C18AR column (MacMod Analytical Inc., Chadds Ford, PA) using a 0.1% formic acid in water vs. 0.1% formic acid in acetonitrile mobile phase gradient. HPLC eluent was interfaced with a QTrap 5500 mass spectrometer (ABSciex, Concord, Canada) via an electrospray probe in both negative and positive polarity mode.

5.3.8 Fasting glucose analysis

At week 0 and week 5, glucose was measured via tail vein puncture after a 5-hour fasting

(OneTouch Ultra blood glucose meter, LifeScan Inc., Milpitas, CA, USA).

5.3.9 Plasma insulin analysis

Plasma insulin (80-INSMSU-E01, ALPCO Diagnostics, Salem, NH, USA) level was determined by enzymatic colorimetric assays.

5.3.10 RNA extraction and RT-PCR analysis of gene expression

Total RNA was extracted from adipose tissue using RNeasy Lipid Tissue Mini Kit

(Qiagen, Valencia, CA, USA) following manufacturer’s instructions. Total RNA was extracted from tumor samples with TriZol reagent (Invitrogen, Carlsbad, CA, USA). 113 Then RNA was reversed transcribed to cDNA using a High Capacity cDNA Reverse

Transcription Kit (Applied Biosystems, Foster City, CA, USA). Quantitative RT-PCR analysis of individual cDNA was performed with ABI Prism 7300 sequence detection system (Applied Biosystems) using TaqMan Gene Expression. Target gene expression was normalized to 18S rRNA for perigonadal adipose tissue and tumor or hypoxanthine guanine phosphoribosyl transferase (HPRT1) for mammary fat pad. Endogenous Controls

(VIC probes) were amplified in the same reaction and expressed as 2-ddCt compared to the HF group (226).

5.3.11 Western blotting

Tumor was homogenized in ice-cold lysis buffer (20 mM Trizma base, 1% Triton-X100,

50 mM NaCl, 250 mM sucrose, 50 mM NaF, and 5 mM Na4P2O7 ! 10H2O) with the

Complete Mini Protease Inhibitor Cocktail (Roche Diagnostics, Indianapolis, IN, USA) and PhosSTOP Phosphatase Inhibitor Cocktail Tablets (Roche Diagnostics) and then homogenates were incubated at 4°C for 1 hour. For cell lysate, cells were collected, washed with PBS, and spun down. Cell pellet was added with ice-cold lysis buffer, resuspended, and kept on ice for 30 minutes. Homogenates and cell lysates were centrifuged at 16,000 g for 15 minutes at 4°C. Supernatant was collected and protein concentration was determined by BCA Protein Assay Kit (Pierce Biotechnology,

Rockford, IL, USA). Protein (40 µg/sample) was separated on polyacrylamide gels and transferred to nitrocellulose membranes using transfer buffer (25 mM Tris,192 mM glycine, and 10% methanol). Membranes were blocked with 5% non-fat dry milk in Tris- buffered saline with 0.1% Tween-20 (TBST) and incubated overnight at 4°C with 114 primary antibodies (Cell Signaling Technology, Danvers, MA, USA). After washing, membranes were incubated for 1 hour at room temperature with HRP-linked anti-rabbit

IgG, (Cell Signaling Technology). Bands were visualized with chemiluminescence

(SuperSignal West Femto Maximum Sensitivity Substrate, Thermo Fisher Scientific,

Rockford, IL, USA) using Carestream Image Station 4000GL PRO (Carestream Health,

Inc., Rochester, NY). Densitometric analysis was conducted using Carestream Molecular

Imaging Software. GAPDH was used as a loading control. Samples from the HF, LN, and

HN groups were separated on the same polyacrylamide gels and analyzed together, while samples from the HF and Met groups were separated on other gels and analyzed together.

5.3.12 Statistical analysis

All data are presented as mean ± standard error (SEM). Statistical analysis was performed using SPSS version 20.0 software (SPSS, Inc., Chicago, IL, USA). Differences were considered significant at P <0.05. The cell proliferation and flow cytometric data was analyzed using a two-way analysis of variance (ANOVA) (Time × Dose). To measure if the effects of metformin or low and high dose of naringenin were significantly different from the high-fat diet, results were analyzed by student’s t-test (HF vs. Met) or one-way analysis of variance (ANOVA) followed by the Tukey multiple comparison test (HF vs.

LN vs. HN). To compare if the effects of metformin and high dose of naringenin were significantly different, results were analyzed by one-way analysis of variance (ANOVA) followed by the Tukey multiple comparison test (HF vs. Met vs. HN). Fasting glucose levels at week 0 and 5 were analyzed using two-tailed unpaired Student's t-test.

115 5.4 Results

5.4.1 Naringenin inhibits E0771 murine breast cancer cell proliferation and cell cycle progression

To determine whether naringenin decreases cell viability of breast cancer cell in vitro, murine mammary breast cancer E0771 cells were treated with 50, 100, and 200 µM of naringenin or vehicle (DMSO, Dose 0) for 24 and 48 hours. Naringenin treatment decreased E0771 cell number in a dose- and time-dependent manner (Fig 15, p <0.001).

Cell viability after 24 hours was reduced to about 75%, 50%, and 25% at 50µM, 100µM, and 200 µM of naringenin concentration, respectively, compared to controls. Cell viability further decreased to about 40%, 10%, and close to 0% at each concentration after 48-hour naringenin treatment.

The flow cytometric analysis revealed a dose- and time-dependent increase in sub-G1 cell population (cell death) after naringenin treatment (Fig 16, p<0.001). After

48-hour exposure to 50 and 100 µM naringenin, the accumulation of E0771 cells at the sub-G1 phase increased to 8% and 19%, which further increased to 21% and 41% after

72-hour naringenin treatment.

Naringenin increased the expression of activated AMPK and downstream signaling intermediates. AMPK Thr172 phosphorylation, a measure of AMPK activity, increased after 24-hour of naringenin treatment (Fig 17). Phosphorylation of Akt, a central regulator of cell proliferation and survival, at both Ser 473 and Thr 308 decreased with naringenin treatment (Fig 17). We also observed down-regulation of cyclin D1 protein expression. Altogether, data from flow cytometry and western blot showed that naringenin activated AMPK, inhibited cell cycle progression, and promoted cell death. 116

Figure 15. Naringenin decreases cell viability of E0771 breast cancer cells. E0771 cells were treated with naringenin at 0, 50, 100, or 200 µM for 24 hours or 48 hours. Viable cells were identified by Trypan blue dye exclusion and counted in triplicate samples by using a hemocytometer. Representative data from 3 independent experiments are shown. Values are presented as mean ± SEM.

117

Figure 16. Naringenin increased the accumulation of sub-G1 cells. (A) E0771 cells were treated with naringenin at 0, 50 and 100 µM for 24, 48, or 72 hours. Cell cycle progression / stage was analyzed by flow cytometry using propidium iodide DNA staining. The results are from one representative experiment of two or three replicates performed that showed similar patterns. (B) Bar charts showed the percentage of cells in the indicated phases of the cell cycle. Values are presented as mean ± SEM.

118

Figure 17. The effect of naringenin on AMPK and Akt activity, and CyclinD1 protein expression. E0771 cells were treated with DMSO (Dose 0) or naringenin at 50 and 100 µM for 24, or 48 hours. AMPK and AKT phosphorylation levels and CyclinD1 protein expression were determined by Western blotting

119 5.4.2 The effect of naringenin and metformin on body weight, caloric intake, and tissue weights in obese ovariectomized mice

In order to model postmenopausal obesity, mice ovariectomized at 19 weeks of age were fed a high fat diet to generate diet-induced obesity. Our previous study showed that ovariectomized mice exhibited higher glucose levels after 5 weeks of the high-fat diet compared to mice fed a standard diet (207). Therefore, naringenin and metformin interventions were started after mice received 3 weeks of a high-fat diet (week 0) prior to development of hyperglycemia. After 3 weeks of the high-fat diet, mice gained about 8 g, equal to 1/3 of their body weight (Figure 18A, p<0.001). The mice were then randomized to HF group, which continued on the high-fat diet, LN group, which switched to the high- fat diet supplemented with 1% wt/wt naringenin, HN group, which switched to the high- fat diet supplemented with 3% wt/wt naringenin, and Met group, which switched to the high-fat diet with metformin in water (170 mg/kg body weight). E0771 cells were inoculated after 2 weeks of the experimental diets (week 2). HN mice started to lose weight after being placed on the experimental diet and the body weight was significantly lower compared to the HF group at week 4, while the body weight in other 3 groups remained stable till the end of the study (Figure 18A). The HN group also tended to have a lower body weight compared to the Met group (p=0.07). There was no difference in body weight of the LN and Met groups when compared to the HF groups at the end of the study. Although HN group had an 8% decrease in cumulative caloric intake compared to the HF group at the end of the study, it did not reach significance (p=0.15, Figure 18B).

LN (1% naringenin) and Metformin did not affect cumulative caloric intake in ovariectomized mice. 120 We measured fasting glucose at week 0 and prior to necropsy at week 5 (Figure

18C). Although glucose levels among groups did not differ at week 5, fasting glucose levels significantly increased in the HF and LN groups compared to baseline (p=0.02 and p<0.01, respectively), whereas fasting glucose levels for the HN and Met mice remained stable from week 0 and week 5 (p=0.43 and p=0.23, respectively). We did not observe differences in fasting insulin levels among HF, LN, and HN groups or between HF and

Met groups (Table 8). Interestingly, the Met group exhibited higher fasting insulin levels compared to the HN group. HN resulted in significant reductions in mammary, perigonadal, and mesenteric fat pad mass compared to the HF and Met groups, respectively (Table 8). Consistent with what we observed previously in obese ovariectomized mice (241), the muscle mass was maintained in the HN group.

121

Figure 18. The effect of naringenin and metformin on body weight cumulative caloric intake and fasting glucose. (A) Body weight (B) Cumulative food intake (C) Fasting glucose at week 0 and week 5. Values are presented as mean ± SEM (HF, n=9; LN, n=8; HN, n=10, Met, n=5). Significance was determined by Student’s t-test (HF vs. Met) or one-way analysis of variance (ANOVA) followed by the Tukey multiple comparison test (HF vs. LN vs. HN or HF vs. HN vs. Met). Significant differences among HF, LN, and HN are indicated with b in panel (A). Student’s t-test was performed to analyze difference in fasting glucose levels between week 0 and week 5. Significantly different compared fasting glucose levels of individual group at week 0 and week 5 are denoted with * in panel (C).

122 Table 8. Plasma insulin levels and tissue weights HF LN HN Met

Plasma Fasting 0.39 ± 0.04a 0.51 ± 0.09a 0.34 ± 0.04a 0.58 ± 0.14* Insulin (ng/mL)

Tissue Weights (g)

Liver 0.95 ± 0.05a 1.01 ± 0.06a 0.88 ± 0.03a 0.92 ± 0.06

Quadriceps 0.36 ± 0.01a 0.35 ± 0.01a 0.36 ± 0.01a* 0.34 ± 0.01

Mammary Fat Pad 1.21 ± 0.14a 1.03 ± 0.11ab 0.73 ± 0.91b 1.31 ± 0.25

Perigonadal Fat Pad 2.35 ± 0.20a 2.59 ± 0.19a 1.60 ± 0.15b* 2.70 ± 0.33

Perirenal Fat Pad 0.79 ± 0.05a 0.68 ± 0.11ab 0.50 ± 0.08b 0.76 ± 0.11

Mesenteric fat pad 0.34 ± 0.04a 0.31 ± 0.06a 0.13 ± 0.01b* 0.30 ± 0.07

Tissue Weight Percentages (%)

Liver 3.06 ± 0.13a 3.28 ± 0.11a 3.35 ± 0.06a* 2.75 ± 0.12

Quadriceps 1.13 ± 0.03a 1.14 ± 0.06a 1.38 ± 0.06b* 1.07 ± 0.08

Mammary Fat Pad 3.79 ± 0.31a 3.26 ± 0.24ab 2.65 ± 0.25b 3.94 ± 0.50

Perigonadal Fat Pad 7.70 ± 0.28a 8.34 ± 0.32a 5.92 ± 0.39b* 8.23 ± 0.54

Perirenal Fat Pad 2.48 ± 0.13a 2.16 ± 0.16a 1.80 ± 0.24a 2.30 ± 0.17

Mesenteric fat pad 1.06 ± 0.10a 0.95 ± 0.16a 0.49 ± 0.04b* 0.90 ± 0.15 Tissue weight percentages derived by weight of organ/body weight. Data represent the mean ± SEM (HF, n=9; LN, n=8; HN, n=10, Met, n=5). Significance was determined by Student t-test (HF vs. Met) or one-way analysis of variance (ANOVA) followed by the Tukey multiple comparison test (HF vs. LN vs. HN or HF vs. HN vs. Met). # Significantly different between Met and HF. * Significantly different between HN and Met. Significant differences among HF, LN, and HN are labeled with different letters (a, b).

123 5.4.3 Naringenin accumulation in tissues of obese ovariectomized mice

Plasma, thoracic mammary glands, and tumor were collected to assess naringenin accumulation. Total naringenin concentrations in plasma and tissues at week 5 were about 3 times higher in mice fed a diet with 3% naringenin compared to mice fed the 1% naringenin diet (Figure 19). Mice fed 3% naringenin had 35.46 ± 10.15 µM total naringenin in plasma compared to 11.44 ± 8.83 µM in mice fed a diet with 1% naringenin. The tumor had higher accumulation of naringenin (LN, 7.61 ± 1.14 µmol/kg tissue and HN, 37.04 ± 8.33 µmol/kg tissue) compared to mammary fat pad (LN, 0.85 ±

0.30 µmol/kg tissue and HN, 5.11 ± 1.26 µmol/kg tissue). These results showed that dietary naringenin led to detectable levels in various tissue sites including mammary fat pad and tumor, with significantly higher accumulation in the tumor and mammary adipose tissues when compared 3% to 1% dietary naringenin supplementation.

124 Figure 19. Naringenin accumulation in plasma and tissues after 5 weeks of supplementation. Plasma (n=5/group), thoracic mammary adipose tissue (n=5-10/group), and tumor (n=5- 9/group) samples were collected at necropsy to determine total naringenin concentrations after 5 weeks of 1% (LN) and 3% (HN) naringenin supplementation. Values are presented as mean ± SEM. Data were analyzed using one-way ANOVA with post hoc Tukey’s test. Significant differences among HF, LN, and HN are labeled with different letters (a, b).

125 5.4.4 The effect of naringenin and metformin on adipose inflammatory markers

In our previous study, ovariectomized mice fed a 3% naringenin had reduced mRNA expression of inflammatory markers, monocyte chemoattractant protein-1 (MCP-

1/Ccl2) and interleukin 6 (IL-6), and the adipokine leptin (241). Here we also observed a

70% reduction in mRNA expression of MCP-1 in perigonadal adipose tissue (PGAT,

P<0.01) as well as a 40% reduction in the mammary fat pad (p=0.03) of the HN group

(Figure 20A & B). The expression of IL-6 was significantly decreased in perigonadal

(p=0.03) but not in mammary adipose tissue (Figure 20A & B). Leptin is an adipokine highly correlated with adipose mass and associates with obesity-induced breast carcinogenesis (219), and we also found that diets with 3% naringenin reduced the expression of leptin by 40% in mammary adipose depots (PGAT, p=0.07 and Mammary, p=0.02; Figure 20A & B). However, metformin did not change the expression of MCP-1,

IL-6, and leptin in perigonadal or mammary adipose tissues.

126

Figure 20. The effect of naringenin and metformin on mRNA expression in perigonadal and mammary adipose tissues. mRNA expression of (A) MCP1, IL-6, and leptin in perigonadal adipose tissue and mRNA expression of (B) MCP1, IL-6, and leptin in mammary adipose tissue. Values are presented as mean ± SEM (n=5-10/group). Significance was determined by Student’s t- test (HF vs. Met) or one-way analysis of variance (ANOVA) followed by the Tukey multiple comparison test (HF vs. LN vs. HN or HF vs. HN vs. Met). Significant differences among HF, LN, and HN are labeled with different letters (a, b).

127 5.4.5 The effect of naringenin and metformin on tumor growth and final tumor weight The tumors were palpable 8 days after inoculation. Two perpendicular dimensions of tumors were measured using a dial caliper to track tumor growth. Tumors in mice of the HN and Met groups were significantly smaller on days 14 and 17 compared to the HF group (HN, p=0.05 on day 14 and p=0.04 on day 17; Met, p=0.03 on day 14 and p<0.01 on day 17, Figure 21). However, at the end of the study, there was no difference in mean tumor size at day 20 and mean final tumor weight was not significantly different between the HF and HN groups. Tumor growth and final tumor weight in the LN group were not different from the HF group. Metformin significantly suppressed the tumor growth to week 5, and final tumor weights were significantly reduced compared to the HF group

(p<0.01). The HN group and Met group did not have significant difference in tumor size on day 14, 17, and 20, and final tumor weight.

128

Figure 21. The effect of naringenin and metformin on tumor growth and tumor weight. (A) Tumor size was determined as area (length x width) by using a dial caliper to measure two perpendicular dimensions. (B) Tumor weight was measured at necropsy. Values are presented as mean ± SEM (HF, n=9; LN, n=8; HN, n=10, Met, n=5). Significance was determined by Student’s t-test (HF vs. Met) or one-way analysis of variance (ANOVA) followed by the Tukey multiple comparison test (HF vs. LN vs. HN or HF vs. HN vs. Met). # P < 0.05 compared Met with HF. Significant differences among HF, LN, and HN are labeled with different letters (a, b).

129 5.4.6 The effect of naringenin and metformin on AMPK expression and signal transduction in mammary tumors

We evaluated mammary tumor samples for effects on AMPK and downstream gene expression. Although levels of AMPK phosphorylation increased in 5 of 8 tumor samples of the HN group, the overall AMPK phosphorylation levels varied widely and did not significantly increase (Figure 22). Although naringenin inhibited Akt phosphorylation levels in vitro, Akt phosphorylation levels were increased in the HN group compared to the HF groups (Figure 22A & B). The serine-threonine kinase mammalian target of rapamycin (mTOR) plays a critical role in the regulation of protein translation, cell proliferation, and metabolism (142). Its downstream target, p70 S6 kinase, is a key modulator of cell-cycle progression and protein synthesis (243). Phosphorylation levels of mTOR were increased in the HN group compared to the HF groups while we did not observe increased activity of p70 S6 kinase and protein expression of CyclinD1 in the HN groups compared to the HF group (Figure 22A & B). Metformin also did not change phosphorylation levels of AMPK, Akt, and mTOR, and Cyclin D1 protein expression when compared to the HF group and HN group, respectively (Figure 22A &

B).

130

Continued

Figure 22. Effects of naringenin and metformin on AMPK activity downstream signaling. Tumor samples of the HF group were analyzed with naringenin groups (LN & HN) or the Met groups, respectively. (A) Representative immunoblots of AMPK, Akt, mTOR, p70S6K phosphorylation and Cyclin D1 protein expression are shown. (B) Quantitative analysis with densitometry (HF, LN, HN, n=8; Met, n=5). Significance was determined by Student’s t-test (HF vs. Met) or one-way analysis of variance (ANOVA) followed by the Tukey multiple comparison test (HF vs. LN vs. HN or HF vs. HN vs. Met). Significant differences among HF, LN, and HN are labeled with different letters (a, b).

131 Figure 22. Continued

132 5.5 Discussion

About 39.5 % of women aged 40-59 year and 38.1% of women aged over 60 are obese (BMI≥30) (244). Accumulating evidence from clinical and preclinical studies reveals an association between obesity and increased incidence, morbidity, and mortality of postmenopausal breast cancer (8, 245, 246). Obesity-associated metabolic dysregulation (e.g. insulin resistance) and inflammation may have a critical role in the pathogenesis of postmenopausal breast cancer (237). Naringenin has been studied in the last few years because of its potential effects on cancer prevention and treatment (132).

Additionally, animal studies showed that naringenin decreases body weight gain, hyperlipidemia, and adipose tissue accumulation in metabolic syndrome (163). Our previous studies also showed that 3% dietary naringenin supplementation suppressed weight gain, improved fasting glucose levels, and reduced abdominal adiposity in ovariectomized mice fed a normal fat (164) or a high fat diet (219). With the ability to exert both chemopreventive and anti-obesity effects, naringenin is an ideal candidate to study in the prevention and treatment of obesity-associated postmenopausal breast cancer. In the current study, we found that naringenin reduced cell viability and promoted cell death in E0771 breast cancer cells. Naringenin content in plasma (37.04 ± 8.33 µM) from mice fed 3% dietary naringenin was comparable concentrations used in cell culture.

Three percent naringenin supplementation to the high-fat diet was associated with reduced body weight, adiposity, and adipose tissue inflammation in ovariectomized mice.

We also observed early suppression of tumor growth in mice fed diets with 3% naringenin at day 14 and 17. Nonetheless, future studies will need to examine whether the antitumor effect of naringenin was caused by directly targeting proliferation of cancer 133 cells, or indirectly targeting obesity and associated metabolic dysregulation and inflammation, or both.

In the past 10 years, extensive epidemiological evidence has revealed an association of metformin and reduced cancer risk. Recently, metformin was found to be associated with reduced risk of cancer in patients with type 2 diabetes (241). A report from Women’s Health Initiative clinical trials also showed that metformin use lowered incidence of invasive breast cancers in postmenopausal women with type 2 diabetes compared to the incidences in women taking other diabetic medications as well as nondiabetic women (247). A more recent systemic review concluded that metformin is associated with reduced cancer incidence and mortality in patients with diabetes but the association between metformin and breast cancer incidence is less significant (248).

Furthermore, the effect of metformin on cancer prevention/treatment in non-diabetic population is still inconclusive. Several animal studies have reported that metformin has limited effects on tumor growth in mice fed a standard diet (249-254). The reduction in tumor growth was accompanied with increased insulin sensitivity, lower fasting glucose and insulin levels, reduced insulin receptor signaling (insulin receptor, IRS-1, Akt), suggesting that these factors are involved in the molecular mechanisms of metformin’s inhibitory action in high-fat-fed animals. In the current study, we treated obese ovariectomized mice with metformin before the mice exhibited elevated fasting glucose levels. Although fasting glucose levels at week 5 were not different between the HF and

Met groups, mice treated with metformin did not develop elevated fasting glucose from week 0 to week 5 compared to the HF mice. Future studies will perform more precise

134 techniques, e.g. hyperinsulinemic euglycemic glucose clamp, glucose tolerance test, and etc., to assess insulin sensitivity.

This dose of metformin used did not change body weight, fasting insulin levels, adipose tissue mass and adipose tissue inflammation in obese ovariectomized mice, but had a marked inhibitory effect on mammary tumor growth compared to that achieved with naringenin supplementation. However, we were unable to observe differences in

AMPK and Akt activity in the metformin-treated mice. Thompson et al showed that peak plasma concentrations of metformin were achieved 1 hour after an oral dose and the plasma concentrations dropped by half after 4 hours (255). Thus, the lack of differences in phosph-protein levels in the present study may relate to discontinuation of metformin treatment 5 hours prior to necropsy. Gu et al used the same mouse model, diet-induced obesity in ovariectomized mice with allografts of E0771, and showed increased plasma and visceral fat VEGF-A levels in the obese ovariectomized mice compared to the lean mice (255). Furthermore, Chen et al suppressed obesity-induced E0771 mammary tumors by inhibiting IKKβ/mTOR/VEGF pathway (26). However, we measured mTOR activity and VEGF-A expression (data not shown) in tumor and did not find differences between the HF and Met groups. Metformin also did not change cyclin D1 expression, which was shown to be down-regulated by metformin in vitro (256, 257). Therefore, systematic approaches, eg. DNA microarray, are needed in the future to investigate the molecular mechanisms underlying metformin’s inhibitory effect in obesity-related mammary tumorigenesis.

In accordance with our previous findings in both lean and obese ovariectomized mice

(151, 241), mice treated with 3% naringenin exhibited reduced body weight gain and 135 intra-abdominal adiposity. Furthermore, the expression of two inflammatory markers,

MCP-1/CCL2 and IL-6, in perigonadal adipose tissue was suppressed in mice supplemented with 3% naringenin. We also observed significant suppression of both

MCP-1 and IL-6 expression in lean ovariectomized mice treated with 3% naringenin

(219, 241) while we only observed suppression of MCP-1 but not IL-6 expression in obese ovariectomized mice (219, 241). Obese ovariectomized mice in the previous study were fed a high-fat diet for 11 weeks and exhibited overt metabolic dysregulation, which may result in no changes in IL-6 expression. Interestingly, mice supplemented with 1% naringenin did not have reduced body weight and adipose tissue mass compared to the

HF mice, but the expression of MCP-1 in the perigonadal adipose tissue was significantly suppressed in the perigonadal adipose tissue. Yoshida et al also found elevated MCP-1 expression and macrophage infiltration in the perigonadal adipose tissue in mice fed a high-fat diet for 14 days, while naringenin supplementation suppressed MCP-1 expression and macrophage infiltration in perigonadal adipose tissue without affecting body weight and adipose mass (219). Collectively, our results further support that naringenin suppresses MCP-1 expression in perigonadal adipose tissue and may have a potent protective effect against adipose tissue inflammation.

To be noted, 3% naringenin and metformin supplementation had similar effects on tumor growth at day 14 and day 17. However, the inhibitory effect of naringenin was not maintained to the end of the study, suggesting that tumor cell growth was able to overcome the suppressive properties of naringenin treatment at later time points. It has been shown that naringenin induced cell cycle arrest and apoptosis in cancer cells through estrogen receptor signaling (ERα and ERβ) (187). Since E0771 is an estrogen- 136 receptor-positive (ER+) breast cancer, it was possible that naringenin exerts the anti- proliferative activity through estrogen receptor signaling; however, we did not test this hypothesis in the present study. We report here that naringenin inhibited cyclin D1 in breast cancer cells but not in the tumor samples. Cyclin D1 gene amplification and protein expression has been shown to associate with ER-positive status (22) and cyclin

D1 overexpression in ERα positive breast cancer was correlated with resistance to antiestrogen treatment (258). It is possible that the dosage of naringenin tested was unable to inhibit or maintain inhibition of cyclin D1 expression in tumor samples.

Additionally, the tumors at the time of necropsy were of large size and possibly beyond the point at which naringenin could continue to suppress cyclin D1. Clinical observations have shown that ER-positive tumors eventually lose dependence on estrogen receptor resulting in up-regulation of the EGFR/HER2 pathway (epidermal growth factor receptor/human epidermal growth factor receptor 2). Activation of the EGFR/HER2 pathway permits ER positive tumor to escape from the effects of treatments (259), which is associated with elevated phosphorylation levels of Akt and mTOR observed in the naringenin groups (260). However, we did not observe increased phosphorylation levels of Akt-mTOR downstream molecule, p70S6K, supporting our findings that expression of downstream molecules, including cyclin D1 and VEGF-A, was not up-regulated. On the other hand, metformin has been shown to suppress HER2-mediated growth in breast cancer cells (261) and increased the latency and reduced mammary tumor size in mice overexpressed HER2 gene (262). Therefore, metformin may still be able to inhibit beast tumor growth through multiple mechanisms.

137 In conclusion, this is the first study to compare the effect of the citrus flavonoid naringenin with metformin on breast tumor growth in a mouse model of postmenopausal obesity. Additionally, the current study revealed naringenin accumulation in tumor and mammary glands. We showed that naringenin might suppress mammary tumor growth through two non-mutually exclusive mechanisms: 1) direct inhibition of cancer cell proliferation and induces cell death, and 2) indirect modulation of body weight, adiposity, and adipose inflammation. In comparison to naringenin, metformin significantly inhibited tumor growth and final tumor weight without reducing body weight, adiposity, blood insulin, and adipose inflammation. Collectively, naringenin and metformin may decrease mammary tumor growth in a mouse model of postmenopausal obesity through different mechanisms.

138 5.6 Acknowledgments

The authors would like to acknowledge the OSU Analytical Cytometry Shared Resource for flow cytometry analysis and Nutrient & Phytochemical Analytic Shared Resource

(NPASR) for naringenin accumulation analysis. This research was supported by the NCI-

NIH R03 (M.A.B.).

139

Chapter 6

Epilogue

This research was aimed to investigate the effects of naringenin on metabolic dysregulation and mammary growth in a mouse model of postmenopause. We examined the effects of naringenin supplementation on metabolic disturbances induced by loss of ovarian function as well as by a combination of high fat diet-induced obesity and loss of ovarian function (postmenopausal obesity). Then we examined the effects of naringenin on inhibition of mammary tumor growth in the model of postmenopausal obesity. Results from three studies suggest that naringenin suppresses weight gain, decreases adipose mass, lowers fasting glucose levels, and down-regulates inflammatory markers in subcutaneous (mammary) and intra-abdominal (perigonadal and mesenteric) fat pads.

Furthermore, naringenin inhibits proliferation and induces cell death in breast cancer and suppresses the growth of mammary tumor. Of note, naringenin significantly reduces perigonadal fat mass and down-regulates the expression of monocyte chemoattractant protein-1 (MCP-1/CCL2) in perigonadal adipose tissue throughout the three studies.

These results are consistent with published research, indicating that naringenin reduced mRNA expression of MCP-1 and macrophage infiltration in the perigonadal adipose 140 tissue of mice, before changes in body weight, adipose tissue mass, and plasma analytes

(263). In that study, MCP-1mRNA expression is partly mediated by c-Jun NH2-terminal kinase (JNK) signaling pathway. Further studies are needed to investigate if JNK signaling pathway is a molecular target of naringenin in adipose tissues.

We observed that naringenin supplementation decreased cumulative caloric intake in the first and third study. However, the effect of naringenin on cumulative caloric intake was not observed in the second study with obese ovariectomized mice that exhibited metabolic dysregulation. We did observe an initial reduction of caloric intake in the first week but the difference in caloric intake was not different over the subsequent 10 weeks of feeding. To our knowledge, this is the first research to investigate the effects of naringenin in female rodents. Reduced caloric intake after naringenin supplementation in this research was unexpected because previous studies had reported no differences in food intake in male rodent models (13, 182, 187, 190, 192, 193). Since mice in the second and third study were fed the same 3% naringenin diet, the only difference between the two studies were the degree of obesity of the mice. The mice in the second study were fed a high fat diet (60% calories from fat) for 11 weeks and exhibited hyperglycemia at the time of naringenin supplementation, whereas the mice in the third study were fed only 3 weeks of high fat diet and only developed weight gain at the time of supplementation. Previous research has shown that the circulating levels or response of some appetite hormones are different in obese people compared to the normal weight people (183). It is speculated that naringenin may interact with some appetite hormones and thus affect caloric intake. Furthermore, follicle-stimulating hormone (FSH) is a sex hormone that is low in men and premenopausal women but rises significantly during 141 menopause. The increase of FSH during menopause is less pronounced in obese women compared to non-obese women (264). Given that naringenin does not affect caloric intake in male mice and obese ovariectomized female mice, it is likely that naringenin may affect caloric intake via interacting with sex hormones that changes after menopause and in obesity individuals.

This research supports the potential health benefits of naringenin supplementation.

We estimated that the mice fed a 3% naringenin diet consumed about 1g/kg body weight of naringenin, which equates to 9 g of naringenin or 17L of grapefruit juice per day for a

60 kg person based on surface area, which is not an amount that people can consumed from natural food sources on a daily basis. Therefore, it is imperative to develop food products that have more concentrated naringenin or compounds have similar effect but are more bioavailable. Several studies have shown that prenylflavonoids may have higher bioavailability compared to their parent flavonoids (265). Indeed, 8-prenylnaringenin has shown to accumulate in the plasma and muscle tissue at a 6 and 10 fold higher concentration, respectively, compared to naringenin (aglyconic form) after a diet containing 8-prenylnaringenin or naringenin for 22 days (266). Therefore, 8- prenylnaringenin may be a promising candidate for nutrition intervention.

142

List of References

1. Toth MJ, Tchernof A, Sites CK, Poehlman ET. Effect of menopausal status on body composition and abdominal fat distribution. Int J Obes Relat Metab Disord. 2000 Feb;24(2):226-31. PubMed PMID: 10702775. Epub 2000/03/07. eng.

2. Carr MC. The emergence of the metabolic syndrome with menopause. J Clin Endocrinol Metab. 2003 Jun;88(6):2404-11. PubMed PMID: 12788835. Epub 2003/06/06. eng.

3. Agrinier N, Cournot M, Dallongeville J, Arveiler D, Ducimetiere P, Ruidavets JB, et al. Menopause and modifiable coronary heart disease risk factors: a population based study. Maturitas. 2010 Mar;65(3):237-43. PubMed PMID: 20031345. Epub 2009/12/25. eng.

4. Suzuki A, Abdelmalek MF. Nonalcoholic fatty liver disease in women. Womens Health (Lond Engl). 2009 Mar;5(2):191-203. PubMed PMID: 19245356. Epub 2009/02/28. eng.

5. Szmuilowicz ED, Stuenkel CA, Seely EW. Influence of menopause on diabetes and diabetes risk. Nat Rev Endocrinol. 2009 Oct;5(10):553-8. PubMed PMID: 19687788. Epub 2009/08/19. eng.

6. Sugiyama MG, Agellon LB. Sex differences in lipid metabolism and metabolic disease risk. Biochem Cell Biol. 2012 Apr;90(2):124-41. PubMed PMID: 22221155. Epub 2012/01/10. eng.

7. Aguilar M, Bhuket T, Torres S, Liu B, Wong RJ. Prevalence of the metabolic syndrome in the United States, 2003-2012. JAMA. 2015 May 19;313(19):1973-4. PubMed PMID: 25988468. Epub 2015/05/20. eng.

8. Ogden CL, Carroll MD, Kit BK, Flegal KM. Prevalence of childhood and adult obesity in the United States, 2011-2012. JAMA. 2014 Feb 26;311(8):806-14. PubMed PMID: 24570244. Epub 2014/02/27. eng.

9. Siegel RL, Miller KD, Jemal A. Cancer statistics, 2015. CA Cancer J Clin. 2015 Jan-Feb;65(1):5-29. PubMed PMID: 25559415. Epub 2015/01/07. eng.

10. SEER Cancer Statistics Factsheets: Breast Cancer. National Cancer Institute. Bethesda, MD. http://seer.cancer.gov/statfacts/html/breast.html. 143 11. Zhu Z, Jiang W, Zacher JH, Neil ES, McGinley JN, Thompson HJ. Effects of energy restriction and wheel running on mammary carcinogenesis and host systemic factors in a rat model. Cancer Prev Res (Phila). 2012 Mar;5(3):414-22. PubMed PMID: 22246620. Epub 2012/01/17. eng.

12. Dirx MJ, Zeegers MP, Dagnelie PC, van den Bogaard T, van den Brandt PA. Energy restriction and the risk of spontaneous mammary tumors in mice: a meta-analysis. Int J Cancer. 2003 Sep 20;106(5):766-70. PubMed PMID: 12866038. Epub 2003/07/17. eng.

13. Mulvihill EE, Allister EM, Sutherland BG, Telford DE, Sawyez CG, Edwards JY, et al. Naringenin prevents dyslipidemia, apolipoprotein B overproduction, and hyperinsulinemia in LDL receptor-null mice with diet-induced insulin resistance. Diabetes. 2009 Oct;58(10):2198-210. PubMed PMID: 19592617. Epub 2009/07/14. eng.

14. Purushotham A, Tian M, Belury MA. The citrus fruit flavonoid naringenin suppresses hepatic glucose production from Fao hepatoma cells. Mol Nutr Food Res. 2009 Feb;53(2):300-7. PubMed PMID: 19035551. Epub 2008/11/28. eng.

15. .

16. Arul D, Subramanian P. Naringenin (citrus flavonone) induces growth inhibition, cell cycle arrest and apoptosis in human hepatocellular carcinoma cells. Pathol Oncol Res. 2013 Oct;19(4):763-70. PubMed PMID: 23661153. Epub 2013/05/11. eng.

17. Jin CY, Park C, Lee JH, Chung KT, Kwon TK, Kim GY, et al. Naringenin- induced apoptosis is attenuated by Bcl-2 but restored by the small molecule Bcl-2 inhibitor, HA 14-1, in human leukemia U937 cells. Toxicol In Vitro. 2009 Mar;23(2):259-65. PubMed PMID: 19124070. Epub 2009/01/07. eng.

18. Kanno S, Tomizawa A, Hiura T, Osanai Y, Shouji A, Ujibe M, et al. Inhibitory effects of naringenin on tumor growth in human cancer cell lines and sarcoma S-180- implanted mice. Biol Pharm Bull. 2005 Mar;28(3):527-30. PubMed PMID: 15744083. Epub 2005/03/04. eng.

19. Kanno S, Tomizawa A, Ohtake T, Koiwai K, Ujibe M, Ishikawa M. Naringenin- induced apoptosis via activation of NF-kappaB and necrosis involving the loss of ATP in human promyeloleukemia HL-60 cells. Toxicol Lett. 2006 Oct 10;166(2):131-9. PubMed PMID: 16860949. Epub 2006/07/25. eng.

20. Leonardi T, Vanamala J, Taddeo SS, Davidson LA, Murphy ME, Patil BS, et al. Apigenin and naringenin suppress colon carcinogenesis through the aberrant crypt stage in azoxymethane-treated rats. Exp Biol Med (Maywood). 2010 Jun;235(6):710-7. PubMed PMID: 20511675. Epub 2010/06/01. eng.

144 21. So FV, Guthrie N, Chambers AF, Carroll KK. Inhibition of proliferation of estrogen receptor-positive MCF-7 human breast cancer cells by flavonoids in the presence and absence of excess estrogen. Cancer Lett. 1997 Jan 30;112(2):127-33. PubMed PMID: 9066718. Epub 1997/01/30. eng.

22. Totta P, Acconcia F, Leone S, Cardillo I, Marino M. Mechanisms of naringenin- induced apoptotic cascade in cancer cells: involvement of estrogen receptor alpha and beta signalling. IUBMB Life. 2004 Aug;56(8):491-9. PubMed PMID: 15545229. Epub 2004/11/17. eng.

23. Harmon AW, Patel YM. Naringenin inhibits glucose uptake in MCF-7 breast cancer cells: a mechanism for impaired cellular proliferation. Breast Cancer Res Treat. 2004 May;85(2):103-10. PubMed PMID: 15111768. Epub 2004/04/28. eng.

24. Erlund I, Meririnne E, Alfthan G, Aro A. Plasma kinetics and urinary excretion of the flavanones naringenin and hesperetin in humans after ingestion of orange juice and grapefruit juice. J Nutr. 2001 Feb;131(2):235-41. PubMed PMID: 11160539. Epub 2001/02/13. eng.

25. Bugianesi R, Catasta G, Spigno P, D'Uva A, Maiani G. Naringenin from cooked tomato paste is bioavailable in men. J Nutr. 2002 Nov;132(11):3349-52. PubMed PMID: 12421849. Epub 2002/11/08. eng.

26. Gu JW, Young E, Patterson SG, Makey KL, Wells J, Huang M, et al. Postmenopausal obesity promotes tumor angiogenesis and breast cancer progression in mice. Cancer Biol Ther. 2011 May 15;11(10):910-7. PubMed PMID: 21451264. Epub 2011/04/01. eng.

27. Hansen KR, Knowlton NS, Thyer AC, Charleston JS, Soules MR, Klein NA. A new model of reproductive aging: the decline in ovarian non-growing follicle number from birth to menopause. Hum Reprod. 2008 Mar;23(3):699-708. PubMed PMID: 18192670. Epub 2008/01/15. eng.

28. Prior JC, Hitchcock CL. The endocrinology of perimenopause: need for a paradigm shift. Front Biosci (Schol Ed). 2011;3:474-86. PubMed PMID: 21196391. Epub 2011/01/05. eng.

29. Pasquali R, Casimirri F, Labate AM, Tortelli O, Pascal G, Anconetani B, et al. Body weight, fat distribution and the menopausal status in women. The VMH Collaborative Group. Int J Obes Relat Metab Disord. 1994 Sep;18(9):614-21. PubMed PMID: 7812415. Epub 1994/09/01. eng.

30. den Tonkelaar I, Seidell JC, van Noord PA, Baanders-van Halewijn EA, Ouwehand IJ. Fat distribution in relation to age, degree of obesity, smoking habits, parity and estrogen use: a cross-sectional study in 11,825 Dutch women participating in the

145 DOM-project. Int J Obes. 1990 Sep;14(9):753-61. PubMed PMID: 2228408. Epub 1990/09/01. eng.

31. Bjorkelund C, Lissner L, Andersson S, Lapidus L, Bengtsson C. Reproductive history in relation to relative weight and fat distribution. Int J Obes Relat Metab Disord. 1996 Mar;20(3):213-9. PubMed PMID: 8653141. Epub 1996/03/01. eng.

32. Douchi T, Yamamoto S, Nakamura S, Ijuin T, Oki T, Maruta K, et al. The effect of menopause on regional and total body lean mass. Maturitas. 1998 Jun 17;29(3):247- 52. PubMed PMID: 9699196. Epub 1998/08/12. eng.

33. Wing RR, Matthews KA, Kuller LH, Meilahn EN, Plantinga PL. Weight gain at the time of menopause. Arch Intern Med. 1991 Jan;151(1):97-102. PubMed PMID: 1985614. Epub 1991/01/01. eng.

34. Macdonald HM, New SA, Campbell MK, Reid DM. Longitudinal changes in weight in perimenopausal and early postmenopausal women: effects of dietary energy intake, energy expenditure, dietary calcium intake and hormone replacement therapy. Int J Obes Relat Metab Disord. 2003 Jun;27(6):669-76. PubMed PMID: 12833110. Epub 2003/07/02. eng.

35. Sowers M, Zheng H, Tomey K, Karvonen-Gutierrez C, Jannausch M, Li X, et al. Changes in body composition in women over six years at midlife: ovarian and chronological aging. J Clin Endocrinol Metab. 2007 Mar;92(3):895-901. PubMed PMID: 17192296. Epub 2006/12/29. eng.

36. Lovejoy JC, Champagne CM, de Jonge L, Xie H, Smith SR. Increased visceral fat and decreased energy expenditure during the menopausal transition. Int J Obes (Lond). 2008 Jun;32(6):949-58. PubMed PMID: 18332882. Epub 2008/03/12. eng.

37. Illner K, Brinkmann G, Heller M, Bosy-Westphal A, Muller MJ. Metabolically active components of fat free mass and resting energy expenditure in nonobese adults. Am J Physiol Endocrinol Metab. 2000 Feb;278(2):E308-15. PubMed PMID: 10662716. Epub 2000/02/09. eng.

38. Davis SR, Castelo-Branco C, Chedraui P, Lumsden MA, Nappi RE, Shah D, et al. Understanding weight gain at menopause. Climacteric. 2012 Oct;15(5):419-29. PubMed PMID: 22978257. Epub 2012/09/18. eng.

39. Salpeter SR, Walsh JM, Ormiston TM, Greyber E, Buckley NS, Salpeter EE. Meta-analysis: effect of hormone-replacement therapy on components of the metabolic syndrome in postmenopausal women. Diabetes Obes Metab. 2006 Sep;8(5):538-54. PubMed PMID: 16918589. Epub 2006/08/22. eng.

40. Janssen I, Powell LH, Kazlauskaite R, Dugan SA. Testosterone and visceral fat in midlife women: the Study of Women's Health Across the Nation (SWAN) fat patterning

146 study. Obesity (Silver Spring). 2010 Mar;18(3):604-10. PubMed PMID: 19696765. Epub 2009/08/22. eng.

41. Janssen I, Powell LH, Crawford S, Lasley B, Sutton-Tyrrell K. Menopause and the metabolic syndrome: the Study of Women's Health Across the Nation. Arch Intern Med. 2008 Jul 28;168(14):1568-75. PubMed PMID: 18663170. Epub 2008/07/30. eng.

42. Kaaja RJ. Metabolic syndrome and the menopause. Menopause Int. 2008 Mar;14(1):21-5. PubMed PMID: 18380957. Epub 2008/04/03. eng.

43. American Heart A, National Heart L, Blood I, Grundy SM, Cleeman JI, Daniels SR, et al. Diagnosis and management of the metabolic syndrome. An American Heart Association/National Heart, Lung, and Blood Institute Scientific Statement. Executive summary. Cardiol Rev. 2005 Nov-Dec;13(6):322-7. PubMed PMID: 16708441. Epub 2006/05/20. eng.

44. Wietlisbach V, Marques-Vidal P, Kuulasmaa K, Karvanen J, Paccaud F, Project WM. The relation of body mass index and abdominal adiposity with dyslipidemia in 27 general populations of the WHO MONICA Project. Nutr Metab Cardiovasc Dis. 2013 May;23(5):432-42. PubMed PMID: 22209742. Epub 2012/01/03. eng.

45. Matthews KA, Meilahn E, Kuller LH, Kelsey SF, Caggiula AW, Wing RR. Menopause and risk factors for coronary heart disease. N Engl J Med. 1989 Sep 7;321(10):641-6. PubMed PMID: 2488072. Epub 1989/09/07. eng.

46. Lee CG, Carr MC, Murdoch SJ, Mitchell E, Woods NF, Wener MH, et al. Adipokines, inflammation, and visceral adiposity across the menopausal transition: a prospective study. J Clin Endocrinol Metab. 2009 Apr;94(4):1104-10. PubMed PMID: 19126626. Epub 2009/01/08. eng.

47. Ryu S, Suh BS, Chang Y, Kwon MJ, Yun KE, Jung HS, et al. Menopausal stages and non-alcoholic fatty liver disease in middle-aged women. Eur J Obstet Gynecol Reprod Biol. 2015 Jul;190:65-70. PubMed PMID: 25988514. Epub 2015/05/20. eng.

48. Gutierrez-Grobe Y, Ponciano-Rodriguez G, Ramos MH, Uribe M, Mendez- Sanchez N. Prevalence of non alcoholic fatty liver disease in premenopausal, posmenopausal and polycystic ovary syndrome women. The role of estrogens. Ann Hepatol. 2010 Oct-Dec;9(4):402-9. PubMed PMID: 21057159. Epub 2010/11/09. eng.

49. Klein BE, Klein R, Lee KE. Components of the metabolic syndrome and risk of cardiovascular disease and diabetes in Beaver Dam. Diabetes Care. 2002 Oct;25(10):1790-4. PubMed PMID: 12351479. Epub 2002/09/28. eng.

50. Soriguer F, Morcillo S, Hernando V, Valdes S, Ruiz de Adana MS, Olveira G, et al. Type 2 diabetes mellitus and other cardiovascular risk factors are no more common

147 during menopause: longitudinal study. Menopause. 2009 Jul-Aug;16(4):817-21. PubMed PMID: 19387417. Epub 2009/04/24. eng.

51. Kim C, Edelstein SL, Crandall JP, Dabelea D, Kitabchi AE, Hamman RF, et al. Menopause and risk of diabetes in the Diabetes Prevention Program. Menopause. 2011 Aug;18(8):857-68. PubMed PMID: 21709591. Epub 2011/06/29. eng.

52. Lloyd-Jones D, Adams RJ, Brown TM, Carnethon M, Dai S, De Simone G, et al. Executive summary: heart disease and stroke statistics--2010 update: a report from the American Heart Association. Circulation. 2010 Feb 23;121(7):948-54. PubMed PMID: 20177011. Epub 2010/02/24. eng.

53. Atsma F, Bartelink ML, Grobbee DE, van der Schouw YT. Postmenopausal status and early menopause as independent risk factors for cardiovascular disease: a meta- analysis. Menopause. 2006 Mar-Apr;13(2):265-79. PubMed PMID: 16645540. Epub 2006/04/29. eng.

54. Barrett-Connor E. Menopause, atherosclerosis, and coronary artery disease. Curr Opin Pharmacol. 2013 Apr;13(2):186-91. PubMed PMID: 23352690. Epub 2013/01/29. eng.

55. Florentino G, Cotrim HP, Florentino A, Padilha C, Medeiros-Neto M, Bragagnol G, et al. Hormone replacement therapy in menopausal women: risk factor or protection to nonalcoholic fatty liver disease? Ann Hepatol. 2012 Jan-Feb;11(1):147-9. PubMed PMID: 22166577. Epub 2011/12/15. eng.

56. Florentino GS, Cotrim HP, Vilar CP, Florentino AV, Guimaraes GM, Barreto VS. Nonalcoholic fatty liver disease in menopausal women. Arq Gastroenterol. 2013 Jul- Sep;50(3):180-5. PubMed PMID: 24322188. Epub 2013/12/11. eng.

57. Hamaguchi M, Kojima T, Ohbora A, Takeda N, Fukui M, Kato T. Aging is a risk factor of nonalcoholic fatty liver disease in premenopausal women. World J Gastroenterol. 2012 Jan 21;18(3):237-43. PubMed PMID: 22294826. Epub 2012/02/02. eng.

58. Rossouw JE, Anderson GL, Prentice RL, LaCroix AZ, Kooperberg C, Stefanick ML, et al. Risks and benefits of estrogen plus progestin in healthy postmenopausal women: principal results From the Women's Health Initiative randomized controlled trial. JAMA. 2002 Jul 17;288(3):321-33. PubMed PMID: 12117397. Epub 2002/07/19. eng.

59. Grady D, Herrington D, Bittner V, Blumenthal R, Davidson M, Hlatky M, et al. Cardiovascular disease outcomes during 6.8 years of hormone therapy: Heart and Estrogen/progestin Replacement Study follow-up (HERS II). JAMA. 2002 Jul 3;288(1):49-57. PubMed PMID: 12090862. Epub 2002/07/02. eng.

148 60. Manson JE, Chlebowski RT, Stefanick ML, Aragaki AK, Rossouw JE, Prentice RL, et al. Menopausal hormone therapy and health outcomes during the intervention and extended poststopping phases of the Women's Health Initiative randomized trials. JAMA. 2013 Oct 2;310(13):1353-68. PubMed PMID: 24084921. Epub 2013/10/03. eng.

61. Hodis HN, Mack WJ. The timing hypothesis and hormone replacement therapy: a paradigm shift in the primary prevention of coronary heart disease in women. Part 2: comparative risks. J Am Geriatr Soc. 2013 Jun;61(6):1011-8. PubMed PMID: 23692449. Epub 2013/05/23. eng.

62. Diaz Brinton R. Minireview: translational animal models of human menopause: challenges and emerging opportunities. Endocrinology. 2012 Aug;153(8):3571-8. PubMed PMID: 22778227. Epub 2012/07/11. eng.

63. Chen Y, Heiman ML. Increased weight gain after ovariectomy is not a consequence of leptin resistance. Am J Physiol Endocrinol Metab. 2001 Feb;280(2):E315-22. PubMed PMID: 11158936. Epub 2001/02/13. eng.

64. Ngo Sock ET, Cote I, Mentor JS, Prud'homme D, Bergeron R, Lavoie JM. Ovariectomy stimulates hepatic fat and cholesterol accumulation in high-fat diet-fed rats. Horm Metab Res. 2013 Apr;45(4):283-90. PubMed PMID: 23225241. Epub 2012/12/12. eng.

65. Paquette A, Shinoda M, Rabasa Lhoret R, Prud'homme D, Lavoie JM. Time course of liver lipid infiltration in ovariectomized rats: impact of a high-fat diet. Maturitas. 2007 Oct 20;58(2):182-90. PubMed PMID: 17889461. Epub 2007/09/25. eng.

66. Witte MM, Resuehr D, Chandler AR, Mehle AK, Overton JM. Female mice and rats exhibit species-specific metabolic and behavioral responses to ovariectomy. Gen Comp Endocrinol. 2010 May 1;166(3):520-8. PubMed PMID: 20067798. Epub 2010/01/14. eng.

67. Stubbins RE, Najjar K, Holcomb VB, Hong J, Nunez NP. Oestrogen alters adipocyte biology and protects female mice from adipocyte inflammation and insulin resistance. Diabetes Obes Metab. 2012 Jan;14(1):58-66. PubMed PMID: 21834845. Epub 2011/08/13. eng.

68. Rogers NH, Perfield JW, 2nd, Strissel KJ, Obin MS, Greenberg AS. Reduced energy expenditure and increased inflammation are early events in the development of ovariectomy-induced obesity. Endocrinology. 2009 May;150(5):2161-8. PubMed PMID: 19179442. Epub 2009/01/31. eng.

69. Vieira Potter VJ, Strissel KJ, Xie C, Chang E, Bennett G, Defuria J, et al. Adipose tissue inflammation and reduced insulin sensitivity in ovariectomized mice occurs in the absence of increased adiposity. Endocrinology. 2012 Sep;153(9):4266-77. PubMed PMID: 22778213. Epub 2012/07/11. eng. 149 70. D'Eon TM, Souza SC, Aronovitz M, Obin MS, Fried SK, Greenberg AS. Estrogen regulation of adiposity and fuel partitioning. Evidence of genomic and non- genomic regulation of lipogenic and oxidative pathways. J Biol Chem. 2005 Oct 28;280(43):35983-91. PubMed PMID: 16109719. Epub 2005/08/20. eng.

71. Ludgero-Correia A, Jr., Aguila MB, Mandarim-de-Lacerda CA, Faria TS. Effects of high-fat diet on plasma lipids, adiposity, and inflammatory markers in ovariectomized C57BL/6 mice. Nutrition. 2012 Mar;28(3):316-23. PubMed PMID: 22079392. Epub 2011/11/15. eng.

72. Pighon A, Gutkowska J, Jankowski M, Rabasa-Lhoret R, Lavoie JM. Exercise training in ovariectomized rats stimulates estrogenic-like effects on expression of genes involved in lipid accumulation and subclinical inflammation in liver. Metabolism. 2011 May;60(5):629-39. PubMed PMID: 20674948. Epub 2010/08/03. eng.

73. Jackson KC, Wohlers LM, Valencia AP, Cilenti M, Borengasser SJ, Thyfault JP, et al. Wheel running prevents the accumulation of monounsaturated fatty acids in the liver of ovariectomized mice by attenuating changes in SCD-1 content. Appl Physiol Nutr Metab. 2011 Dec;36(6):798-810. PubMed PMID: 22026420. Epub 2011/10/27. eng.

74. Leite RD, Prestes J, Bernardes CF, Shiguemoto GE, Pereira GB, Duarte JO, et al. Effects of ovariectomy and resistance training on lipid content in skeletal muscle, liver, and heart; fat depots; and lipid profile. Appl Physiol Nutr Metab. 2009 Dec;34(6):1079- 86. PubMed PMID: 20029517. Epub 2009/12/24. eng.

75. Maffei M, Halaas J, Ravussin E, Pratley RE, Lee GH, Zhang Y, et al. Leptin levels in human and rodent: measurement of plasma leptin and ob RNA in obese and weight-reduced subjects. Nat Med. 1995 Nov;1(11):1155-61. PubMed PMID: 7584987. Epub 1995/11/01. eng.

76. Nunez NP, Perkins SN, Smith NC, Berrigan D, Berendes DM, Varticovski L, et al. Obesity accelerates mouse mammary tumor growth in the absence of ovarian hormones. Nutr Cancer. 2008;60(4):534-41. PubMed PMID: 18584488. Epub 2008/06/28. eng.

77. Yakar S, Nunez NP, Pennisi P, Brodt P, Sun H, Fallavollita L, et al. Increased tumor growth in mice with diet-induced obesity: impact of ovarian hormones. Endocrinology. 2006 Dec;147(12):5826-34. PubMed PMID: 16959846. Epub 2006/09/09. eng.

78. Barsalani R, Chapados NA, Lavoie JM. Hepatic VLDL-TG production and MTP gene expression are decreased in ovariectomized rats: effects of exercise training. Horm Metab Res. 2010 Nov;42(12):860-7. PubMed PMID: 20938890. Epub 2010/10/13. eng.

150 79. Bruun JM, Nielsen CB, Pedersen SB, Flyvbjerg A, Richelsen B. Estrogen reduces pro-inflammatory cytokines in rodent adipose tissue: studies in vivo and in vitro. Horm Metab Res. 2003 Mar;35(3):142-6. PubMed PMID: 12734773. Epub 2003/05/08. eng.

80. Wohlers LM, Spangenburg EE. 17beta-estradiol supplementation attenuates ovariectomy-induced increases in ATGL signaling and reduced perilipin expression in visceral adipose tissue. J Cell Biochem. 2010 May 15;110(2):420-7. PubMed PMID: 20336671. Epub 2010/03/26. eng.

81. Paquette A, Wang D, Jankowski M, Gutkowska J, Lavoie JM. Effects of ovariectomy on PPAR alpha, SREBP-1c, and SCD-1 gene expression in the rat liver. Menopause. 2008 Nov-Dec;15(6):1169-75. PubMed PMID: 18665039. Epub 2008/07/31. eng.

82. Wohlers LM, Jackson KC, Spangenburg EE. Lipolytic signaling in response to acute exercise is altered in female mice following ovariectomy. J Cell Biochem. 2011 Dec;112(12):3675-84. PubMed PMID: 21815195. Epub 2011/08/05. eng.

83. Cavalcanti-de-Albuquerque JP, Salvador IC, Martins EL, Jardim-Messeder D, Werneck-de-Castro JP, Galina A, et al. Role of estrogen on skeletal muscle mitochondrial function in ovariectomized rats: a time course study in different fiber types. J Appl Physiol (1985). 2014 Apr 1;116(7):779-89. PubMed PMID: 24458744. Epub 2014/01/25. eng.

84. Hong J, Stubbins RE, Smith RR, Harvey AE, Nunez NP. Differential susceptibility to obesity between male, female and ovariectomized female mice. Nutr J. 2009;8:11. PubMed PMID: 19220919. Epub 2009/02/18. eng.

85. American Cancer Society Inc. Breast Cancer 2015 [cited 2015 May 26]. http://www.cancer.org/cancer/breastcancer/detailedguide/breast-cancer-what-is-breastcancer%5D.

86. American Cancer Society. Breast Cancer Facts & Figures 2013-2014. Atlanta: American Cancer Society, Inc. 2013.

87. Allred DC. Ductal carcinoma in situ: terminology, classification, and natural history. J Natl Cancer Inst Monogr. 2010;2010(41):134-8. PubMed PMID: 20956817. Epub 2010/10/20. eng.

88. National Cancer Institute at the National Institutes of Health. Anatomy of the female breast [cited 2015 May 26]. http://www.cancer.gov/PublishedContent/Images/images/cancer- types/cthp/breast_female_anatomy_enlarge.jpg%5D. Available from: http://www.cancer.gov/PublishedContent/Images/images/cancer- types/cthp/breast_female_anatomy_enlarge.jpg.

151 89. Perou CM, Sorlie T, Eisen MB, van de Rijn M, Jeffrey SS, Rees CA, et al. Molecular portraits of human breast tumours. Nature. 2000 Aug 17;406(6797):747-52. PubMed PMID: 10963602. Epub 2000/08/30. eng.

90. Perou CM, Borresen-Dale AL. Systems biology and genomics of breast cancer. Cold Spring Harb Perspect Biol. 2011 Feb;3(2). PubMed PMID: 21047916. Epub 2010/11/05. eng.

91. Sorlie T, Perou CM, Tibshirani R, Aas T, Geisler S, Johnsen H, et al. Gene expression patterns of breast carcinomas distinguish tumor subclasses with clinical implications. Proc Natl Acad Sci U S A. 2001 Sep 11;98(19):10869-74. PubMed PMID: 11553815. Epub 2001/09/13. eng.

92. Staaf J, Ringner M. Making breast cancer molecular subtypes robust? J Natl Cancer Inst. 2015 Jan;107(1):386. PubMed PMID: 25479803. Epub 2014/12/07. eng.

93. Prat A, Parker JS, Karginova O, Fan C, Livasy C, Herschkowitz JI, et al. Phenotypic and molecular characterization of the claudin-low intrinsic subtype of breast cancer. Breast Cancer Res. 2010;12(5):R68. PubMed PMID: 20813035. Epub 2010/09/04. eng.

94. Kohler BA, Sherman RL, Howlader N, Jemal A, Ryerson AB, Henry KA, et al. Annual report to the nation on the status of cancer, 1975-2011, featuring incidence of breast cancer subtypes by race/ethnicity, poverty, and state. J Natl Cancer Inst. 2015 Jun;107(6). PubMed PMID: 25825511. Epub 2015/04/01. eng.

95. Blows FM, Driver KE, Schmidt MK, Broeks A, van Leeuwen FE, Wesseling J, et al. Subtyping of breast cancer by immunohistochemistry to investigate a relationship between subtype and short and long term survival: a collaborative analysis of data for 10,159 cases from 12 studies. PLoS Med. 2010 May;7(5):e1000279. PubMed PMID: 20520800. Epub 2010/06/04. eng.

96. Romond EH, Perez EA, Bryant J, Suman VJ, Geyer CE, Jr., Davidson NE, et al. Trastuzumab plus adjuvant chemotherapy for operable HER2-positive breast cancer. N Engl J Med. 2005 Oct 20;353(16):1673-84. PubMed PMID: 16236738. Epub 2005/10/21. eng.

97. Bertucci F, Finetti P, Cervera N, Esterni B, Hermitte F, Viens P, et al. How basal are triple-negative breast cancers? Int J Cancer. 2008 Jul 1;123(1):236-40. PubMed PMID: 18398844. Epub 2008/04/10. eng.

98. Badve S, Dabbs DJ, Schnitt SJ, Baehner FL, Decker T, Eusebi V, et al. Basal-like and triple-negative breast cancers: a critical review with an emphasis on the implications for pathologists and oncologists. Mod Pathol. 2011 Feb;24(2):157-67. PubMed PMID: 21076464. Epub 2010/11/16. eng.

152 99. Dent R, Trudeau M, Pritchard KI, Hanna WM, Kahn HK, Sawka CA, et al. Triple-negative breast cancer: clinical features and patterns of recurrence. Clin Cancer Res. 2007 Aug 1;13(15 Pt 1):4429-34. PubMed PMID: 17671126. Epub 2007/08/03. eng.

100. Wiseman M. The second World Cancer Research Fund/American Institute for Cancer Research expert report. Food, nutrition, physical activity, and the prevention of cancer: a global perspective. Proc Nutr Soc. 2008 Aug;67(3):253-6. PubMed PMID: 18452640. Epub 2008/05/03. eng.

101. Morimoto LM, White E, Chen Z, Chlebowski RT, Hays J, Kuller L, et al. Obesity, body size, and risk of postmenopausal breast cancer: the Women's Health Initiative (United States). Cancer Causes Control. 2002 Oct;13(8):741-51. PubMed PMID: 12420953. Epub 2002/11/08. eng.

102. Keum N, Greenwood DC, Lee DH, Kim R, Aune D, Ju W, et al. Adult weight gain and adiposity-related cancers: a dose-response meta-analysis of prospective observational studies. J Natl Cancer Inst. 2015 Feb;107(2). PubMed PMID: 25757865. Epub 2015/03/12. eng.

103. Eliassen AH, Colditz GA, Rosner B, Willett WC, Hankinson SE. Adult weight change and risk of postmenopausal breast cancer. JAMA. 2006 Jul 12;296(2):193-201. PubMed PMID: 16835425. Epub 2006/07/13. eng.

104. Huang Z, Willett WC, Colditz GA, Hunter DJ, Manson JE, Rosner B, et al. Waist circumference, waist:hip ratio, and risk of breast cancer in the Nurses' Health Study. Am J Epidemiol. 1999 Dec 15;150(12):1316-24. PubMed PMID: 10604774. Epub 1999/12/22. eng.

105. Calle EE, Rodriguez C, Walker-Thurmond K, Thun MJ. Overweight, obesity, and mortality from cancer in a prospectively studied cohort of U.S. adults. N Engl J Med. 2003 Apr 24;348(17):1625-38. PubMed PMID: 12711737. Epub 2003/04/25. eng.

106. Dal Maso L, Zucchetto A, Talamini R, Serraino D, Stocco CF, Vercelli M, et al. Effect of obesity and other lifestyle factors on mortality in women with breast cancer. Int J Cancer. 2008 Nov 1;123(9):2188-94. PubMed PMID: 18711698. Epub 2008/08/20. eng.

107. Franceschi S, Favero A, La Vecchia C, Baron AE, Negri E, Dal Maso L, et al. Body size indices and breast cancer risk before and after menopause. Int J Cancer. 1996 Jul 17;67(2):181-6. PubMed PMID: 8760584. Epub 1996/07/17. eng.

108. Wildman RP, Sowers MR. Adiposity and the menopausal transition. Obstet Gynecol Clin North Am. 2011 Sep;38(3):441-54. PubMed PMID: 21961712. Epub 2011/10/04. eng.

153 109. Santen RJ, Brodie H, Simpson ER, Siiteri PK, Brodie A. History of aromatase: saga of an important biological mediator and therapeutic target. Endocr Rev. 2009 Jun;30(4):343-75. PubMed PMID: 19389994. Epub 2009/04/25. eng.

110. Bulun SE, Chen D, Moy I, Brooks DC, Zhao H. Aromatase, breast cancer and obesity: a complex interaction. Trends Endocrinol Metab. 2012 Feb;23(2):83-9. PubMed PMID: 22169755. Epub 2011/12/16. eng.

111. McTiernan A, Wu L, Chen C, Chlebowski R, Mossavar-Rahmani Y, Modugno F, et al. Relation of BMI and physical activity to sex hormones in postmenopausal women. Obesity (Silver Spring). 2006 Sep;14(9):1662-77. PubMed PMID: 17030978. Epub 2006/10/13. eng.

112. Key T, Appleby P, Barnes I, Reeves G, Endogenous H, Breast Cancer Collaborative G. Endogenous sex hormones and breast cancer in postmenopausal women: reanalysis of nine prospective studies. J Natl Cancer Inst. 2002 Apr 17;94(8):606-16. PubMed PMID: 11959894. Epub 2002/04/18. eng.

113. Li CI, Malone KE, Daling JR. Interactions between body mass index and hormone therapy and postmenopausal breast cancer risk (United States). Cancer Causes Control. 2006 Jun;17(5):695-703. PubMed PMID: 16633917. Epub 2006/04/25. eng.

114. Suzuki R, Iwasaki M, Inoue M, Sasazuki S, Sawada N, Yamaji T, et al. Body weight at age 20 years, subsequent weight change and breast cancer risk defined by estrogen and progesterone receptor status--the Japan public health center-based prospective study. Int J Cancer. 2011 Sep 1;129(5):1214-24. PubMed PMID: 21064092. Epub 2010/11/11. eng.

115. Surmacz E. Leptin and adiponectin: emerging therapeutic targets in breast cancer. J Mammary Gland Biol Neoplasia. 2013 Dec;18(3-4):321-32. PubMed PMID: 24136336. Epub 2013/10/19. eng.

116. Vona-Davis L, Rose DP. Adipokines as endocrine, paracrine, and autocrine factors in breast cancer risk and progression. Endocr Relat Cancer. 2007 Jun;14(2):189- 206. PubMed PMID: 17639037. Epub 2007/07/20. eng.

117. Shah NR, Braverman ER. Measuring adiposity in patients: the utility of body mass index (BMI), percent body fat, and leptin. PLoS One. 2012;7(4):e33308. PubMed PMID: 22485140. Epub 2012/04/10. eng.

118. Wu MH, Chou YC, Chou WY, Hsu GC, Chu CH, Yu CP, et al. Circulating levels of leptin, adiposity and breast cancer risk. Br J Cancer. 2009 Feb 24;100(4):578-82. PubMed PMID: 19223908. Epub 2009/02/19. eng.

119. Geisler J, Haynes B, Ekse D, Dowsett M, Lonning PE. Total body aromatization in postmenopausal breast cancer patients is strongly correlated to plasma leptin levels. J

154 Steroid Biochem Mol Biol. 2007 Apr;104(1-2):27-34. PubMed PMID: 17350249. Epub 2007/03/14. eng.

120. Maccio A, Madeddu C, Gramignano G, Mulas C, Floris C, Massa D, et al. Correlation of body mass index and leptin with tumor size and stage of disease in hormone-dependent postmenopausal breast cancer: preliminary results and therapeutic implications. J Mol Med (Berl). 2010 Jul;88(7):677-86. PubMed PMID: 20339829. Epub 2010/03/27. eng.

121. Saxena NK, Sharma D. Multifaceted leptin network: the molecular connection between obesity and breast cancer. J Mammary Gland Biol Neoplasia. 2013 Dec;18(3- 4):309-20. PubMed PMID: 24214584. Epub 2013/11/12. eng.

122. Cleary MP, Phillips FC, Getzin SC, Jacobson TL, Jacobson MK, Christensen TA, et al. Genetically obese MMTV-TGF-alpha/Lep(ob)Lep(ob) female mice do not develop mammary tumors. Breast Cancer Res Treat. 2003 Feb;77(3):205-15. PubMed PMID: 12602920. Epub 2003/02/27. eng.

123. Cleary MP, Juneja SC, Phillips FC, Hu X, Grande JP, Maihle NJ. Leptin receptor- deficient MMTV-TGF-alpha/Lepr(db)Lepr(db) female mice do not develop oncogene- induced mammary tumors. Exp Biol Med (Maywood). 2004 Feb;229(2):182-93. PubMed PMID: 14734797. Epub 2004/01/22. eng.

124. Zheng Q, Dunlap SM, Zhu J, Downs-Kelly E, Rich J, Hursting SD, et al. Leptin deficiency suppresses MMTV-Wnt-1 mammary tumor growth in obese mice and abrogates tumor initiating cell survival. Endocr Relat Cancer. 2011 Aug;18(4):491-503. PubMed PMID: 21636700. Epub 2011/06/04. eng.

125. Tworoger SS, Eliassen AH, Kelesidis T, Colditz GA, Willett WC, Mantzoros CS, et al. Plasma adiponectin concentrations and risk of incident breast cancer. J Clin Endocrinol Metab. 2007 Apr;92(4):1510-6. PubMed PMID: 17213279. Epub 2007/01/11. eng.

126. Miyoshi Y, Funahashi T, Kihara S, Taguchi T, Tamaki Y, Matsuzawa Y, et al. Association of serum adiponectin levels with breast cancer risk. Clin Cancer Res. 2003 Nov 15;9(15):5699-704. PubMed PMID: 14654554. Epub 2003/12/05. eng.

127. Visser M, Bouter LM, McQuillan GM, Wener MH, Harris TB. Elevated C- reactive protein levels in overweight and obese adults. JAMA. 1999 Dec 8;282(22):2131- 5. PubMed PMID: 10591334. Epub 1999/12/11. eng.

128. Cancello R, Clement K. Is obesity an inflammatory illness? Role of low-grade inflammation and macrophage infiltration in human white adipose tissue. BJOG. 2006 Oct;113(10):1141-7. PubMed PMID: 16903845. Epub 2006/08/15. eng.

155 129. Iyengar NM, Hudis CA, Dannenberg AJ. Obesity and inflammation: new insights into breast cancer development and progression. Am Soc Clin Oncol Educ Book. 2013:46-51. PubMed PMID: 23714453. Epub 2013/05/30. eng.

130. Pierce BL, Ballard-Barbash R, Bernstein L, Baumgartner RN, Neuhouser ML, Wener MH, et al. Elevated biomarkers of inflammation are associated with reduced survival among breast cancer patients. J Clin Oncol. 2009 Jul 20;27(21):3437-44. PubMed PMID: 19470939. Epub 2009/05/28. eng.

131. Bachelot T, Ray-Coquard I, Menetrier-Caux C, Rastkha M, Duc A, Blay JY. Prognostic value of serum levels of interleukin 6 and of serum and plasma levels of vascular endothelial growth factor in hormone-refractory metastatic breast cancer patients. Br J Cancer. 2003 Jun 2;88(11):1721-6. PubMed PMID: 12771987. Epub 2003/05/29. eng.

132. Simpson ER, Brown KA. Minireview: Obesity and breast cancer: a tale of inflammation and dysregulated metabolism. Mol Endocrinol. 2013 May;27(5):715-25. PubMed PMID: 23518927. Epub 2013/03/23. eng.

133. Balkwill F. TNF-alpha in promotion and progression of cancer. Cancer Metastasis Rev. 2006 Sep;25(3):409-16. PubMed PMID: 16951987. Epub 2006/09/05. eng.

134. Lin WW, Karin M. A cytokine-mediated link between innate immunity, inflammation, and cancer. J Clin Invest. 2007 May;117(5):1175-83. PubMed PMID: 17476347. Epub 2007/05/04. eng.

135. Iyengar NM, Morris PG, Zhou XK, Gucalp A, Giri D, Harbus MD, et al. Menopause is a determinant of breast adipose inflammation. Cancer Prev Res (Phila). 2015 May;8(5):349-58. PubMed PMID: 25720743. Epub 2015/02/28. eng.

136. Morris PG, Hudis CA, Giri D, Morrow M, Falcone DJ, Zhou XK, et al. Inflammation and increased aromatase expression occur in the breast tissue of obese women with breast cancer. Cancer Prev Res (Phila). 2011 Jul;4(7):1021-9. PubMed PMID: 21622727. Epub 2011/05/31. eng.

137. Tan J, Buache E, Chenard MP, Dali-Youcef N, Rio MC. Adipocyte is a non- trivial, dynamic partner of breast cancer cells. Int J Dev Biol. 2011;55(7-9):851-9. PubMed PMID: 21948738. Epub 2011/09/29. eng.

138. Chappell J, Leitner JW, Solomon S, Golovchenko I, Goalstone ML, Draznin B. Effect of insulin on cell cycle progression in MCF-7 breast cancer cells. Direct and potentiating influence. J Biol Chem. 2001 Oct 12;276(41):38023-8. PubMed PMID: 11500498. Epub 2001/08/14. eng.

139. Ish-Shalom D, Christoffersen CT, Vorwerk P, Sacerdoti-Sierra N, Shymko RM, Naor D, et al. Mitogenic properties of insulin and insulin analogues mediated by the

156 insulin receptor. Diabetologia. 1997 Jul;40 Suppl 2:S25-31. PubMed PMID: 9248698. Epub 1997/07/01. eng.

140. Shafie SM, Grantham FH. Role of hormones in the growth and regression of human breast cancer cells (MCF-7) transplanted into athymic nude mice. J Natl Cancer Inst. 1981 Jul;67(1):51-6. PubMed PMID: 6265682. Epub 1981/07/01. eng.

141. Shafie SM, Hilf R. Insulin receptor levels and magnitude of insulin-induced responses in 7,12-dimethylbenz(a)anthracene-induced mammary tumors in rats. Cancer Res. 1981 Mar;41(3):826-9. PubMed PMID: 6780189. Epub 1981/03/01. eng.

142. Maccio A, Madeddu C. Obesity, inflammation, and postmenopausal breast cancer: therapeutic implications. ScientificWorldJournal. 2011;11:2020-36. PubMed PMID: 22125453. Epub 2011/11/30. eng.

143. Gunter MJ, Hoover DR, Yu H, Wassertheil-Smoller S, Rohan TE, Manson JE, et al. Insulin, insulin-like growth factor-I, and risk of breast cancer in postmenopausal women. J Natl Cancer Inst. 2009 Jan 7;101(1):48-60. PubMed PMID: 19116382. Epub 2009/01/01. eng.

144. Papa V, Pezzino V, Costantino A, Belfiore A, Giuffrida D, Frittitta L, et al. Elevated insulin receptor content in human breast cancer. J Clin Invest. 1990 Nov;86(5):1503-10. PubMed PMID: 2243127. Epub 1990/11/01. eng.

145. Mountjoy KG, Finlay GJ, Holdaway IM. Abnormal insulin-receptor down regulation and dissociation of down regulation from insulin biological action in cultured human tumor cells. Cancer Res. 1987 Dec 15;47(24 Pt 1):6500-4. PubMed PMID: 3315189. Epub 1987/12/15. eng.

146. Bodmer M, Meier C, Krahenbuhl S, Jick SS, Meier CR. Long-term metformin use is associated with decreased risk of breast cancer. Diabetes Care. 2010 Jun;33(6):1304-8. PubMed PMID: 20299480. Epub 2010/03/20. eng.

147. Jalving M, Gietema JA, Lefrandt JD, de Jong S, Reyners AK, Gans RO, et al. Metformin: taking away the candy for cancer? Eur J Cancer. 2010 Sep;46(13):2369-80. PubMed PMID: 20656475. Epub 2010/07/27. eng.

148. Foretz M, Guigas B, Bertrand L, Pollak M, Viollet B. Metformin: from mechanisms of action to therapies. Cell Metab. 2014 Dec 2;20(6):953-66. PubMed PMID: 25456737. Epub 2014/12/03. eng.

149. Towler MC, Hardie DG. AMP-activated protein kinase in metabolic control and insulin signaling. Circ Res. 2007 Feb 16;100(3):328-41. PubMed PMID: 17307971. Epub 2007/02/20. eng.

157 150. Dowling RJ, Zakikhani M, Fantus IG, Pollak M, Sonenberg N. Metformin inhibits mammalian target of rapamycin-dependent translation initiation in breast cancer cells. Cancer Res. 2007 Nov 15;67(22):10804-12. PubMed PMID: 18006825. Epub 2007/11/17. eng.

151. Zhuang Y, Miskimins WK. Cell cycle arrest in Metformin treated breast cancer cells involves activation of AMPK, downregulation of cyclin D1, and requires p27Kip1 or p21Cip1. J Mol Signal. 2008;3:18. PubMed PMID: 19046439. Pubmed Central PMCID: 2613390. Epub 2008/12/03. eng.

152. Chun OK, Chung SJ, Song WO. Estimated dietary flavonoid intake and major food sources of U.S. adults. J Nutr. 2007 May;137(5):1244-52. PubMed PMID: 17449588. Epub 2007/04/24. eng.

153. Hertog MG, Feskens EJ, Hollman PC, Katan MB, Kromhout D. Dietary antioxidant flavonoids and risk of coronary heart disease: the Zutphen Elderly Study. Lancet. 1993 Oct 23;342(8878):1007-11. PubMed PMID: 8105262. Epub 1993/10/23. eng.

154. Knekt P, Jarvinen R, Reunanen A, Maatela J. Flavonoid intake and coronary mortality in Finland: a cohort study. BMJ. 1996 Feb 24;312(7029):478-81. PubMed PMID: 8597679. Epub 1996/02/24. eng.

155. Hui C, Qi X, Qianyong Z, Xiaoli P, Jundong Z, Mantian M. Flavonoids, flavonoid subclasses and breast cancer risk: a meta-analysis of epidemiologic studies. PLoS One. 2013;8(1):e54318. PubMed PMID: 23349849. Epub 2013/01/26. eng.

156. Davies JN, Hobson GE. The constituents of tomato fruit--the influence of environment, nutrition, and genotype. Crit Rev Food Sci Nutr. 1981;15(3):205-80. PubMed PMID: 7030623. Epub 1981/01/01. eng.

157. Kawaii S, Tomono Y, Katase E, Ogawa K, Yano M. Quantitation of flavonoid constituents in citrus fruits. J Agric Food Chem. 1999 Sep;47(9):3565-71. PubMed PMID: 10552686. Epub 1999/12/20. eng.

158. Tripoli E, Guardia ML, Giammanco S, Majo DD, Giammanco M. Citrus flavonoids: Molecular structure, biological activity and nutritional properties: A review. Food Chemistry. 2007;104(2):466-79.

159. Horowitz RM. Taste effects of flavonoids. Prog Clin Biol Res. 1986;213:163-75. PubMed PMID: 3520586. Epub 1986/01/01. eng.

160. Peterson JJ, Beecher GR, Bhagwat SA, Dwyer JT, Gebhardt SE, Haytowitz DB, et al. Flavanones in grapefruit, lemons, and limes: A compilation and review of the data from the analytical literature. Journal of Food Composition and Analysis. 2006;19, Supplement(0):S74-S80.

158 161. Gattuso G, Barreca D, Gargiulli C, Leuzzi U, Caristi C. Flavonoid composition of Citrus juices. Molecules. 2007;12(8):1641-73. PubMed PMID: 17960080. Epub 2007/10/26. eng.

162. Paganga G, Miller N, Rice-Evans CA. The polyphenolic content of fruit and vegetables and their antioxidant activities. What does a serving constitute? Free Radic Res. 1999 Feb;30(2):153-62. PubMed PMID: 10193583. Epub 1999/04/08. eng.

163. Mir IA, Tiku AB. Chemopreventive and therapeutic potential of "naringenin," a flavanone present in citrus fruits. Nutr Cancer. 2015;67(1):27-42. PubMed PMID: 25514618. Epub 2014/12/17. eng.

164. Alam MA, Subhan N, Rahman MM, Uddin SJ, Reza HM, Sarker SD. Effect of citrus flavonoids, naringin and naringenin, on metabolic syndrome and their mechanisms of action. Adv Nutr. 2014 Jul;5(4):404-17. PubMed PMID: 25022990. Epub 2014/07/16. eng.

165. Griffiths LA, Barrow A. Metabolism of flavonoid compounds in germ-free rats. Biochem J. 1972 Dec;130(4):1161-2. PubMed PMID: 4656801. Epub 1972/12/01. eng.

166. Abe K-I, Katayama H, Suzuki A, Yumioka E. Biological Fate of Orally Administered Naringin and Naringeninin Rats. Japanese Journal of Pharmacognosy. 1993;47(4):402-7.

167. Choudhury R, Chowrimootoo G, Srai K, Debnam E, Rice-Evans CA. Interactions of the flavonoid naringenin in the gastrointestinal tract and the influence of glycosylation. Biochem Biophys Res Commun. 1999 Nov 19;265(2):410-5. PubMed PMID: 10558881. Epub 1999/11/24. eng.

168. Felgines C, Texier O, Morand C, Manach C, Scalbert A, Regerat F, et al. Bioavailability of the flavanone naringenin and its glycosides in rats. Am J Physiol Gastrointest Liver Physiol. 2000 Dec;279(6):G1148-54. PubMed PMID: 11093936. Epub 2000/11/30. eng.

169. Day AJ, Canada FJ, Diaz JC, Kroon PA, McLauchlan R, Faulds CB, et al. Dietary flavonoid and isoflavone glycosides are hydrolysed by the lactase site of lactase phlorizin hydrolase. FEBS Lett. 2000 Feb 25;468(2-3):166-70. PubMed PMID: 10692580. Epub 2000/02/29. eng.

170. Sesink AL, Arts IC, Faassen-Peters M, Hollman PC. Intestinal uptake of quercetin-3-glucoside in rats involves hydrolysis by lactase phlorizin hydrolase. J Nutr. 2003 Mar;133(3):773-6. PubMed PMID: 12612151. Epub 2003/03/04. eng.

171. Day AJ, DuPont MS, Ridley S, Rhodes M, Rhodes MJ, Morgan MR, et al. Deglycosylation of flavonoid and isoflavonoid glycosides by human small intestine and

159 liver beta-glucosidase activity. FEBS Lett. 1998 Sep 25;436(1):71-5. PubMed PMID: 9771896. Epub 1998/10/15. eng.

172. Nemeth K, Plumb GW, Berrin JG, Juge N, Jacob R, Naim HY, et al. Deglycosylation by small intestinal epithelial cell beta-glucosidases is a critical step in the absorption and metabolism of dietary flavonoid glycosides in humans. Eur J Nutr. 2003 Jan;42(1):29-42. PubMed PMID: 12594539. Epub 2003/02/21. eng.

173. Day AJ, Gee JM, DuPont MS, Johnson IT, Williamson G. Absorption of quercetin-3-glucoside and quercetin-4'-glucoside in the rat small intestine: the role of lactase phlorizin hydrolase and the sodium-dependent glucose transporter. Biochem Pharmacol. 2003 Apr 1;65(7):1199-206. PubMed PMID: 12663055. Epub 2003/03/29. eng.

174. Brand W, van der Wel PA, Rein MJ, Barron D, Williamson G, van Bladeren PJ, et al. Metabolism and transport of the citrus flavonoid hesperetin in Caco-2 cell monolayers. Drug Metab Dispos. 2008 Sep;36(9):1794-802. PubMed PMID: 18515333. Epub 2008/06/03. eng.

175. Silberberg M, Gil-Izquierdo A, Combaret L, Remesy C, Scalbert A, Morand C. Flavanone metabolism in healthy and tumor-bearing rats. Biomed Pharmacother. 2006 Nov;60(9):529-35. PubMed PMID: 16952436. Epub 2006/09/06. eng.

176. Wang M-J, Chao P-DL, Hou Y-C, Hsiu S-L, Wen K-C, Tsai S-Y. Pharmacokinetics and conjugation metabolism of naringin and naringenin in rats after single dose and multiple dose administrations Journal of Food and Drug Analysis. 2006;14(3):247-53.

177. Lin SP, Hou YC, Tsai SY, Wang MJ, Chao PD. Tissue distribution of naringenin conjugated metabolites following repeated dosing of naringin to rats. Biomedicine (Taipei). 2014;4:16. PubMed PMID: 25520929. Epub 2014/12/19. Eng.

178. Manach C, Morand C, Gil-Izquierdo A, Bouteloup-Demange C, Remesy C. Bioavailability in humans of the flavanones hesperidin and narirutin after the ingestion of two doses of orange juice. Eur J Clin Nutr. 2003 Feb;57(2):235-42. PubMed PMID: 12571654. Epub 2003/02/07. eng.

179. Jackson MJ. The assessment of bioavailability of micronutrients: introduction. Eur J Clin Nutr. 1997 Jan;51 Suppl 1:S1-2. PubMed PMID: 9023470. Epub 1997/01/01. eng.

180. Krogholm KS, Bredsdorff L, Knuthsen P, Haraldsdottir J, Rasmussen SE. Relative bioavailability of the flavonoids quercetin, hesperetin and naringenin given simultaneously through diet. Eur J Clin Nutr. 2010 Apr;64(4):432-5. PubMed PMID: 20125185. Epub 2010/02/04. eng.

160 181. Breinholt VM, Svendsen GW, Dragsted LO, Hossaini A. The citrus-derived flavonoid naringenin exerts uterotrophic effects in female mice at human relevant doses. Basic Clin Pharmacol Toxicol. 2004 Jan;94(1):30-6. PubMed PMID: 14725612. Epub 2004/01/17. eng.

182. Assini JM, Mulvihill EE, Sutherland BG, Telford DE, Sawyez CG, Felder SL, et al. Naringenin prevents cholesterol-induced systemic inflammation, metabolic dysregulation, and atherosclerosis in Ldlr(-)/(-) mice. J Lipid Res. 2013 Mar;54(3):711- 24. PubMed PMID: 23269394. Epub 2012/12/28. eng.

183. Mulvihill EE, Assini JM, Sutherland BG, DiMattia AS, Khami M, Koppes JB, et al. Naringenin decreases progression of atherosclerosis by improving dyslipidemia in high-fat-fed low-density lipoprotein receptor-null mice. Arterioscler Thromb Vasc Biol. 2010 Apr;30(4):742-8. PubMed PMID: 20110573. Epub 2010/01/30. eng.

184. Pu P, Gao DM, Mohamed S, Chen J, Zhang J, Zhou XY, et al. Naringin ameliorates metabolic syndrome by activating AMP-activated protein kinase in mice fed a high-fat diet. Arch Biochem Biophys. 2012 Feb 1;518(1):61-70. PubMed PMID: 22198281. Epub 2011/12/27. eng.

185. Yoshida H, Watanabe W, Oomagari H, Tsuruta E, Shida M, Kurokawa M. Citrus flavonoid naringenin inhibits TLR2 expression in adipocytes. J Nutr Biochem. 2013 Jul;24(7):1276-84. PubMed PMID: 23333096. Epub 2013/01/22. eng.

186. Alam MA, Kauter K, Brown L. Naringin improves diet-induced cardiovascular dysfunction and obesity in high carbohydrate, high fat diet-fed rats. Nutrients. 2013 Mar;5(3):637-50. PubMed PMID: 23446977. Epub 2013/03/01. eng.

187. Cho KW, Kim YO, Andrade JE, Burgess JR, Kim YC. Dietary naringenin increases hepatic peroxisome proliferators-activated receptor alpha protein expression and decreases plasma triglyceride and adiposity in rats. Eur J Nutr. 2011 Mar;50(2):81-8. PubMed PMID: 20567977. Epub 2010/06/23. eng.

188. Fresno M, Alvarez R, Cuesta N. Toll-like receptors, inflammation, metabolism and obesity. Arch Physiol Biochem. 2011 Jul;117(3):151-64. PubMed PMID: 21599616. Epub 2011/05/24. eng.

189. Jiao P, Chen Q, Shah S, Du J, Tao B, Tzameli I, et al. Obesity-related upregulation of monocyte chemotactic factors in adipocytes: involvement of nuclear factor-kappaB and c-Jun NH2-terminal kinase pathways. Diabetes. 2009 Jan;58(1):104- 15. PubMed PMID: 18835938. Epub 2008/10/07. eng.

190. Yoshida H, Watanabe H, Ishida A, Watanabe W, Narumi K, Atsumi T, et al. Naringenin suppresses macrophage infiltration into adipose tissue in an early phase of high-fat diet-induced obesity. Biochem Biophys Res Commun. 2014 Nov 7;454(1):95- 101. PubMed PMID: 25450363. Epub 2014/12/03. eng. 161 191. Bok S-H, Shin YW, Bae K-H, Jeong T-S, Kwon Y-K, Park YB, et al. Effects of naringin and lovastatin on plasma and hepatic lipids in high-fat and high-cholesterol fed rats. Nutrition Research. 2000;20(7):1007-15.

192. Huong DT, Takahashi Y, Ide T. Activity and mRNA levels of enzymes involved in hepatic fatty acid oxidation in mice fed citrus flavonoids. Nutrition. 2006 May;22(5):546-52. PubMed PMID: 16483743. Epub 2006/02/18. eng.

193. Kannappan S, Anuradha CV. Naringenin enhances insulin-stimulated tyrosine phosphorylation and improves the cellular actions of insulin in a dietary model of metabolic syndrome. Eur J Nutr. 2010 Mar;49(2):101-9. PubMed PMID: 19727895. Epub 2009/09/04. eng.

194. Kim SY, Kim HJ, Lee MK, Jeon SM, Do GM, Kwon EY, et al. Naringin time- dependently lowers hepatic cholesterol biosynthesis and plasma cholesterol in rats fed high-fat and high-cholesterol diet. J Med Food. 2006 Winter;9(4):582-6. PubMed PMID: 17201649. Epub 2007/01/05. eng.

195. Lee SH, Park YB, Bae KH, Bok SH, Kwon YK, Lee ES, et al. Cholesterol- lowering activity of naringenin via inhibition of 3-hydroxy-3-methylglutaryl coenzyme A reductase and acyl coenzyme A:cholesterol acyltransferase in rats. Ann Nutr Metab. 1999;43(3):173-80. PubMed PMID: 10545673. Epub 1999/11/05. eng.

196. Bulzomi P, Bolli A, Galluzzo P, Leone S, Acconcia F, Marino M. Naringenin and 17beta-estradiol coadministration prevents hormone-induced human cancer cell growth. IUBMB Life. 2010 Jan;62(1):51-60. PubMed PMID: 19960539. Epub 2009/12/05. eng.

197. El Mjiyad N, Caro-Maldonado A, Ramirez-Peinado S, Munoz-Pinedo C. Sugar- free approaches to cancer cell killing. Oncogene. 2011 Jan 20;30(3):253-64. PubMed PMID: 20972457. Epub 2010/10/26. eng.

198. Qin L, Jin L, Lu L, Lu X, Zhang C, Zhang F, et al. Naringenin reduces lung metastasis in a breast cancer resection model. Protein Cell. 2011 Jun;2(6):507-16. PubMed PMID: 21748601. Epub 2011/07/13. eng.

199. Terao J. Dietary flavonoids as antioxidants. Forum Nutr. 2009;61:87-94. PubMed PMID: 19367113. Epub 2009/04/16. eng.

200. Meydani M, Hasan ST. Dietary polyphenols and obesity. Nutrients. 2010 Jul;2(7):737-51. PubMed PMID: 22254051. Epub 2010/07/01. eng.

201. Goldwasser J, Cohen PY, Yang E, Balaguer P, Yarmush ML, Nahmias Y. Transcriptional regulation of human and rat hepatic lipid metabolism by the grapefruit flavonoid naringenin: role of PPARalpha, PPARgamma and LXRalpha. PLoS One. 2010;5(8):e12399. PubMed PMID: 20811644. Epub 2010/09/03. eng.

162 202. Peters J. Animal and Human Calorimetry - Mclean,Ja, Tobin,C. Am J Phys Anthropol. 1989 Jan;78(1):127-. PubMed PMID: ISI:A1989R891100014. English.

203. Folch J, Lees M, Sloane Stanley GH. A simple method for the isolation and purification of total lipides from animal tissues. J Biol Chem. 1957 May;226(1):497-509. PubMed PMID: 13428781. Epub 1957/05/01. eng.

204. Pacheco YM, Perez-Camino MC, Cert A, Montero E, Ruiz-Gutierrez V. Determination of the molecular species composition of diacylglycerols in human adipose tissue by solid-phase extraction and gas chromatography on a polar phase. J Chromatogr B Biomed Sci Appl. 1998 Sep 4;714(2):127-32. PubMed PMID: 9766852. Epub 1998/10/10. eng.

205. Danno H, Jincho Y, Budiyanto S, Furukawa Y, Kimura S. A simple enzymatic quantitative analysis of triglycerides in tissues. J Nutr Sci Vitaminol (Tokyo). 1992 Oct;38(5):517-21. PubMed PMID: 1294711. Epub 1992/10/01. eng.

206. Wendel AA, Purushotham A, Liu LF, Belury MA. Conjugated linoleic acid fails to worsen insulin resistance but induces hepatic steatosis in the presence of leptin in ob/ob mice. J Lipid Res. 2008 Jan;49(1):98-106. PubMed PMID: 17906221. Epub 2007/10/02. eng.

207. Livak KJ, Schmittgen TD. Analysis of relative gene expression data using real- time quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods. 2001 Dec;25(4):402-8. PubMed PMID: 11846609. Epub 2002/02/16. eng.

208. Tschop MH, Speakman JR, Arch JR, Auwerx J, Bruning JC, Chan L, et al. A guide to analysis of mouse energy metabolism. Nat Methods. 2012 Jan;9(1):57-63. PubMed PMID: 22205519. Epub 2011/12/30. eng.

209. Ryan AS, Elahi D. The effects of acute hyperglycemia and hyperinsulinemia on plasma leptin levels: its relationships with body fat, visceral adiposity, and age in women. J Clin Endocrinol Metab. 1996 Dec;81(12):4433-8. PubMed PMID: 8954055. Epub 1996/12/01. eng.

210. Lavoie JM, Pighon A. NAFLD, Estrogens, and Physical Exercise: The Animal Model. J Nutr Metab. 2012;2012:914938. PubMed PMID: 21845221. Epub 2011/08/17. eng.

211. Jou J, Choi SS, Diehl AM. Mechanisms of disease progression in nonalcoholic fatty liver disease. Semin Liver Dis. 2008 Nov;28(4):370-9. PubMed PMID: 18956293. Epub 2008/10/29. eng.

212. Kallwitz ER, McLachlan A, Cotler SJ. Role of peroxisome proliferators-activated receptors in the pathogenesis and treatment of nonalcoholic fatty liver disease. World J Gastroenterol. 2008 Jan 7;14(1):22-8. PubMed PMID: 18176957. Epub 2008/01/08. eng.

163 213. Xue F, Michels KB. Diabetes, metabolic syndrome, and breast cancer: a review of the current evidence. Am J Clin Nutr. 2007 Sep;86(3):s823-35. PubMed PMID: 18265476. Epub 2008/02/13. eng.

214. Brown LM, Clegg DJ. Central effects of estradiol in the regulation of food intake, body weight, and adiposity. J Steroid Biochem Mol Biol. 2010 Oct;122(1-3):65-73. PubMed PMID: 20035866. Epub 2009/12/29. eng.

215. Pfeilschifter J, Koditz R, Pfohl M, Schatz H. Changes in proinflammatory cytokine activity after menopause. Endocr Rev. 2002 Feb;23(1):90-119. PubMed PMID: 11844745. Epub 2002/02/15. eng.

216. Kim WK, Choi EK, Sul OJ, Park YK, Kim ES, Yu R, et al. Monocyte chemoattractant protein-1 deficiency attenuates oxidative stress and protects against ovariectomy-induced chronic inflammation in mice. PLoS One. 2013;8(8):e72108. PubMed PMID: 23977220. Epub 2013/08/27. eng.

217. Borradaile NM, de Dreu LE, Huff MW. Inhibition of net HepG2 cell apolipoprotein B secretion by the citrus flavonoid naringenin involves activation of phosphatidylinositol 3-kinase, independent of insulin receptor substrate-1 phosphorylation. Diabetes. 2003 Oct;52(10):2554-61. PubMed PMID: 14514640. Epub 2003/09/30. eng.

218. Janssen I, Heymsfield SB, Ross R. Low relative skeletal muscle mass (sarcopenia) in older persons is associated with functional impairment and physical disability. J Am Geriatr Soc. 2002 May;50(5):889-96. PubMed PMID: 12028177. Epub 2002/05/25. eng.

219. Ke J-Y, Kliewer KL, Hamad EM, Cole RM, Powell KA, Andridge RR, et al. The flavonoid, naringenin, decreases adipose tissue mass and attenuates ovariectomy- associated metabolic disturbances in mice. Nutrition & Metabolism. 2015;12(1):1.

220. Mason C, Xiao L, Imayama I, Duggan CR, Foster-Schubert KE, Kong A, et al. Influence of diet, exercise, and serum vitamin d on sarcopenia in postmenopausal women. Med Sci Sports Exerc. 2013 Apr;45(4):607-14. PubMed PMID: 23190588. Epub 2012/11/30. eng.

221. Joseph AM, Pilegaard H, Litvintsev A, Leick L, Hood DA. Control of gene expression and mitochondrial biogenesis in the muscular adaptation to endurance exercise. Essays Biochem. 2006;42:13-29. PubMed PMID: 17144877. Epub 2006/12/06. eng.

222. Hood DA. Mechanisms of exercise-induced mitochondrial biogenesis in skeletal muscle. Appl Physiol Nutr Metab. 2009 Jun;34(3):465-72. PubMed PMID: 19448716. Epub 2009/05/19. eng.

164 223. Cartoni R, Leger B, Hock MB, Praz M, Crettenand A, Pich S, et al. Mitofusins 1/2 and ERRalpha expression are increased in human skeletal muscle after physical exercise. J Physiol. 2005 Aug 15;567(Pt 1):349-58. PubMed PMID: 15961417. Epub 2005/06/18. eng.

224. Zygmunt K, Faubert B, MacNeil J, Tsiani E. Naringenin, a citrus flavonoid, increases muscle cell glucose uptake via AMPK. Biochem Biophys Res Commun. 2010 Jul 23;398(2):178-83. PubMed PMID: 20558145. Epub 2010/06/19. eng.

225. Turner N, Cooney GJ, Kraegen EW, Bruce CR. Fatty acid metabolism, energy expenditure and insulin resistance in muscle. J Endocrinol. 2014 Feb;220(2):T61-79. PubMed PMID: 24323910. Epub 2013/12/11. eng.

226. Reagan-Shaw S, Nihal M, Ahmad N. Dose translation from animal to human studies revisited. Faseb J. 2008 Mar;22(3):659-61. PubMed PMID: 17942826. Epub 2007/10/19. eng.

227. Porrini M, Riso P. Factors influencing the bioavailability of antioxidants in foods: a critical appraisal. Nutr Metab Cardiovasc Dis. 2008 Dec;18(10):647-50. PubMed PMID: 18996686. Epub 2008/11/11. eng.

228. Hardwick RN, Ferreira DW, More VR, Lake AD, Lu Z, Manautou JE, et al. Altered UDP-glucuronosyltransferase and sulfotransferase expression and function during progressive stages of human nonalcoholic fatty liver disease. Drug Metab Dispos. 2013 Mar;41(3):554-61. PubMed PMID: 23223517. Epub 2012/12/12. eng.

229. Shimoi K, Saka N, Nozawa R, Sato M, Amano I, Nakayama T, et al. Deglucuronidation of a flavonoid, luteolin monoglucuronide, during inflammation. Drug Metab Dispos. 2001 Dec;29(12):1521-4. PubMed PMID: 11717168. Epub 2001/11/22. eng.

230. Bjursell M, Gerdin AK, Lelliott CJ, Egecioglu E, Elmgren A, Tornell J, et al. Acutely reduced locomotor activity is a major contributor to Western diet-induced obesity in mice. Am J Physiol Endocrinol Metab. 2008 Feb;294(2):E251-60. PubMed PMID: 18029443. Epub 2007/11/22. eng.

231. Baumgartner RN. Body composition in healthy aging. Ann N Y Acad Sci. 2000 May;904:437-48. PubMed PMID: 10865787. Epub 2000/06/24. eng.

232. Funai K, Semenkovich CF. Skeletal muscle lipid flux: running water carries no poison. Am J Physiol Endocrinol Metab. 2011 Aug;301(2):E245-51. PubMed PMID: 21558546. Epub 2011/05/12. eng.

233. Funai K, Song H, Yin L, Lodhi IJ, Wei X, Yoshino J, et al. Muscle lipogenesis balances insulin sensitivity and strength through calcium signaling. J Clin Invest. 2013 Mar 1;123(3):1229-40. PubMed PMID: 23376793. Epub 2013/02/05. eng.

165 234. Badin PM, Vila IK, Louche K, Mairal A, Marques MA, Bourlier V, et al. High-fat diet-mediated lipotoxicity and insulin resistance is related to impaired lipase expression in mouse skeletal muscle. Endocrinology. 2013 Apr;154(4):1444-53. PubMed PMID: 23471217. Epub 2013/03/09. eng.

235. Levin MC, Monetti M, Watt MJ, Sajan MP, Stevens RD, Bain JR, et al. Increased lipid accumulation and insulin resistance in transgenic mice expressing DGAT2 in glycolytic (type II) muscle. Am J Physiol Endocrinol Metab. 2007 Dec;293(6):E1772-81. PubMed PMID: 17940217. Epub 2007/10/18. eng.

236. Erion DM, Shulman GI. Diacylglycerol-mediated insulin resistance. Nat Med. 2010 Apr;16(4):400-2. PubMed PMID: 20376053. Epub 2010/04/09. eng.

237. La Vecchia C, Giordano SH, Hortobagyi GN, Chabner B. Overweight, obesity, diabetes, and risk of breast cancer: interlocking pieces of the puzzle. Oncologist. 2011;16(6):726-9. PubMed PMID: 21632448. Epub 2011/06/03. eng.

238. Decensi A, Puntoni M, Goodwin P, Cazzaniga M, Gennari A, Bonanni B, et al. Metformin and cancer risk in diabetic patients: a systematic review and meta-analysis. Cancer Prev Res (Phila). 2010 Nov;3(11):1451-61. PubMed PMID: 20947488. Epub 2010/10/16. eng.

239. Hatoum D, McGowan EM. Recent advances in the use of metformin: can treating diabetes prevent breast cancer? Biomed Res Int. 2015;2015:548436. PubMed PMID: 25866793. Epub 2015/04/14. eng.

240. de Jager J, Kooy A, Lehert P, Wulffele MG, van der Kolk J, Bets D, et al. Long term treatment with metformin in patients with type 2 diabetes and risk of vitamin B-12 deficiency: randomised placebo controlled trial. BMJ. 2010;340:c2181. PubMed PMID: 20488910. Epub 2010/05/22. eng.

241. Ke J-Y, Cole RM, Hamad EM, Hsiao Y-H, Cotten BM, Powell KA, et al. Citrus flavonoid, naringenin, increases locomotor activity and reduces diacylglycerol accumulation in skeletal muscle of obese ovariectomized mice. Molecular Nutrition & Food Research (under review) 2015.

242. So FV, Guthrie N, Chambers AF, Moussa M, Carroll KK. Inhibition of human breast cancer cell proliferation and delay of mammary tumorigenesis by flavonoids and citrus juices. Nutr Cancer. 1996;26(2):167-81. PubMed PMID: 8875554. Epub 1996/01/01. eng.

243. Meric-Bernstam F, Gonzalez-Angulo AM. Targeting the mTOR signaling network for cancer therapy. J Clin Oncol. 2009 May 1;27(13):2278-87. PubMed PMID: 19332717. Epub 2009/04/01. eng.

166 244. Pullen N, Thomas G. The modular phosphorylation and activation of p70s6k. FEBS Lett. 1997 Jun 23;410(1):78-82. PubMed PMID: 9247127. Epub 1997/06/23. eng.

245. Chan DS, Vieira AR, Aune D, Bandera EV, Greenwood DC, McTiernan A, et al. Body mass index and survival in women with breast cancer-systematic literature review and meta-analysis of 82 follow-up studies. Ann Oncol. 2014 Oct;25(10):1901-14. PubMed PMID: 24769692. Epub 2014/04/29. eng.

246. Rohan TE, Heo M, Choi L, Datta M, Freudenheim JL, Kamensky V, et al. Body fat and breast cancer risk in postmenopausal women: a longitudinal study. J Cancer Epidemiol. 2013;2013:754815. PubMed PMID: 23690776. Epub 2013/05/22. eng.

247. Evans JM, Donnelly LA, Emslie-Smith AM, Alessi DR, Morris AD. Metformin and reduced risk of cancer in diabetic patients. BMJ. 2005 Jun 4;330(7503):1304-5. PubMed PMID: 15849206. Epub 2005/04/26. eng.

248. Chlebowski RT, McTiernan A, Wactawski-Wende J, Manson JE, Aragaki AK, Rohan T, et al. Diabetes, metformin, and breast cancer in postmenopausal women. J Clin Oncol. 2012 Aug 10;30(23):2844-52. PubMed PMID: 22689798. Epub 2012/06/13. eng.

249. Gandini S, Puntoni M, Heckman-Stoddard BM, Dunn BK, Ford L, DeCensi A, et al. Metformin and cancer risk and mortality: a systematic review and meta-analysis taking into account biases and confounders. Cancer Prev Res (Phila). 2014 Sep;7(9):867- 85. PubMed PMID: 24985407. Epub 2014/07/06. eng.

250. Algire C, Zakikhani M, Blouin MJ, Shuai JH, Pollak M. Metformin attenuates the stimulatory effect of a high-energy diet on in vivo LLC1 carcinoma growth. Endocr Relat Cancer. 2008 Sep;15(3):833-9. PubMed PMID: 18469156. Epub 2008/05/13. eng.

251. Algire C, Amrein L, Zakikhani M, Panasci L, Pollak M. Metformin blocks the stimulative effect of a high-energy diet on colon carcinoma growth in vivo and is associated with reduced expression of fatty acid synthase. Endocr Relat Cancer. 2010 Jun;17(2):351-60. PubMed PMID: 20228137. Epub 2010/03/17. eng.

252. Phoenix KN, Vumbaca F, Fox MM, Evans R, Claffey KP. Dietary energy availability affects primary and metastatic breast cancer and metformin efficacy. Breast Cancer Res Treat. 2010 Sep;123(2):333-44. PubMed PMID: 20204498. Epub 2010/03/06. eng.

253. Algire C, Amrein L, Bazile M, David S, Zakikhani M, Pollak M. Diet and tumor LKB1 expression interact to determine sensitivity to anti-neoplastic effects of metformin in vivo. Oncogene. 2011 Mar 10;30(10):1174-82. PubMed PMID: 21102522. Epub 2010/11/26. eng.

254. Mashhedi H, Blouin MJ, Zakikhani M, David S, Zhao Y, Bazile M, et al. Metformin abolishes increased tumor (18)F-2-fluoro-2-deoxy-D-glucose uptake

167 associated with a high energy diet. Cell Cycle. 2011 Aug 15;10(16):2770-8. PubMed PMID: 21811094. Epub 2011/08/04. eng.

255. Thompson MD, Grubbs CJ, Bode AM, Reid JM, McGovern R, Bernard PS, et al. Lack of effect of metformin on mammary carcinogenesis in nondiabetic rat and mouse models. Cancer Prev Res (Phila). 2015 Mar;8(3):231-9. PubMed PMID: 25681088. Epub 2015/02/15. eng.

256. Chen CT, Du Y, Yamaguchi H, Hsu JM, Kuo HP, Hortobagyi GN, et al. Targeting the IKKbeta/mTOR/VEGF signaling pathway as a potential therapeutic strategy for obesity-related breast cancer. Mol Cancer Ther. 2012 Oct;11(10):2212-21. PubMed PMID: 22826466. Epub 2012/07/25. eng.

257. Wahdan-Alaswad R, Fan Z, Edgerton SM, Liu B, Deng XS, Arnadottir SS, et al. Glucose promotes breast cancer aggression and reduces metformin efficacy. Cell Cycle. 2013 Dec 15;12(24):3759-69. PubMed PMID: 24107633. Epub 2013/10/11. eng.

258. Elsheikh S, Green AR, Aleskandarany MA, Grainge M, Paish CE, Lambros MB, et al. CCND1 amplification and cyclin D1 expression in breast cancer and their relation with proteomic subgroups and patient outcome. Breast Cancer Res Treat. 2008 May;109(2):325-35. PubMed PMID: 17653856. Epub 2007/07/27. eng.

259. Stendahl M, Kronblad A, Ryden L, Emdin S, Bengtsson NO, Landberg G. Cyclin D1 overexpression is a negative predictive factor for tamoxifen response in postmenopausal breast cancer patients. Br J Cancer. 2004 May 17;90(10):1942-8. PubMed PMID: 15138475. Epub 2004/05/13. eng.

260. Osborne CK, Schiff R. Mechanisms of endocrine resistance in breast cancer. Annu Rev Med. 2011;62:233-47. PubMed PMID: 20887199. Epub 2010/10/05. eng.

261. Paplomata E, O'Regan R. The PI3K/AKT/mTOR pathway in breast cancer: targets, trials and biomarkers. Ther Adv Med Oncol. 2014 Jul;6(4):154-66. PubMed PMID: 25057302. Epub 2014/07/25. eng.

262. Vazquez-Martin A, Oliveras-Ferraros C, Menendez JA. The antidiabetic drug metformin suppresses HER2 (erbB-2) oncoprotein overexpression via inhibition of the mTOR effector p70S6K1 in human breast carcinoma cells. Cell Cycle. 2009 Jan 1;8(1):88-96. PubMed PMID: 19106626. Epub 2008/12/25. eng.

263. Anisimov VN, Berstein LM, Egormin PA, Piskunova TS, Popovich IG, Zabezhinski MA, et al. Effect of metformin on life span and on the development of spontaneous mammary tumors in HER-2/neu transgenic mice. Exp Gerontol. 2005 Aug- Sep;40(8-9):685-93. PubMed PMID: 16125352. Epub 2005/08/30. eng.

264. Suzuki K, Jayasena CN, Bloom SR. Obesity and appetite control. Exp Diabetes Res. 2012;2012:824305. PubMed PMID: 22899902. Epub 2012/08/18. eng.

168 265. Randolph JF, Jr., Sowers M, Bondarenko IV, Harlow SD, Luborsky JL, Little RJ. Change in estradiol and follicle-stimulating hormone across the early menopausal transition: effects of ethnicity and age. J Clin Endocrinol Metab. 2004 Apr;89(4):1555- 61. PubMed PMID: 15070912. Epub 2004/04/09. eng.

266. Terao J, Mukai R. Prenylation modulates the bioavailability and bioaccumulation of dietary flavonoids. Arch Biochem Biophys. 2014 Oct 1;559:12-6. PubMed PMID: 24736381. Epub 2014/04/17. eng.

267. Mukai R, Horikawa H, Fujikura Y, Kawamura T, Nemoto H, Nikawa T, et al. Prevention of disuse muscle atrophy by dietary ingestion of 8-prenylnaringenin in denervated mice. PLoS One. 2012;7(9):e45048. PubMed PMID: 23028754. Epub 2012/10/03. eng.

268. Nakashima T, Arakawa M, Kojiro M, Sakamoto K, Ikari T, Otsu N, et al. Pathological and morphological studies of hepatocellular carcinoma, particularly on metastases and intracanal tumor growth. Kurume Med J. 1979;26(1):21-33. PubMed PMID: 228113. Epub 1979/01/01. eng.

169

Appendix A

Illustration of different fat depots analyzed by magnetic resonance imaging (MRI)

Figure 23. Illustration of different fat depots analyzed by MRI. Total, intra-abdominal, and subcutaneous adiposity were analyzed by MRI using a Bruker Biospin 94/30 magnet (Billerica, MA, USA) and a 70 mm diameter linear volume coil. T1-weighted coronal images of the whole mouse torso were collected using a respiratory-gated RARE sequence (TR/TE=1570/7.5ms, RARE factor=4, FOV=70x45 mm2, matrix size=256x192, slice thickness=1 mm, navg=2). Mice were anesthetized with

2-2.5% isoflurane mixed with 1 liter per minute carbogen (95%O2+5%CO2) and maintained with 1-1.5% isoflurane. Physiologic parameters such as the electrocardiography, respiration and the temperature of the animals were monitored using a small animal monitoring system (Model 1025, Small Animals Instruments, Inc. Stony Brook, NY, USA). Otsu segmentation (267) was used to segment the mouse body from Continued 170

Figure 23. Continued background. A connected components algorithm (268) was used to label the background objects in the image and ‘fill’ any holes in the segmented body image. The abdominal cavity was manually outlined in the images and a global threshold of 120 grey level intensity was chosen to segment fat from surrounding tissue. The whole body and abdominal masks were then used to calculate the percentage of segmented voxels in the whole body and abdomen, respectively. Subcutaneous fat was calculated by subtracting intra-abdominal fat from total fat.

171