THE EFFECT OF ESTROGEN STATUS ON SELENIUM METABOLISM IN FEMALE RATS DISSERTATION. the Degree Doctor of Philosophy in the Graduate

Transcription

1 THE EFFECT OF ESTROGEN STATUS ON SELENIUM METABOLISM IN FEMALE RATS DISSERTATION Presented in Partial Fulfillment of the Requirements for the Degree Doctor of Philosophy in the Graduate School of The Ohio State University By Xiaodong Zhou, M.S. ***** The Ohio State University 2007 Dissertation Committee: Approved by Dr. Anne M. Smith, Advisor Dr. Mark L. Failla Dr. Steven K. Clinton Dr. Charles L. Brooks Advisor The Ohio State University Nutrition Graduate Program

2 ABSTRACT An association between male and female sex hormones and selenium (Se) status has been reported in animals and humans. These relationships may be important in the regulation of selenium metabolism and relative to the possible use of selenium as an adjunct for treatment of hormone-related diseases such as breast cancer. Insights about impact of estrogen on distribution and metabolism of selenium in multiple tissues are limited. The purpose of the first part of this study was to examine the effect of estrogen status on the absorption, tissue distribution and metabolism of orally administered 75 Se-selenite. Female Sprague Dawley (SD) rats were bilaterally ovariectomized and implanted with either a placebo pellet (OVX, n=16) or pellet with estradiol (OVX+E2, n=16) at 7 weeks of age. At 12 weeks of age, 60 µci (43 ng total) of 75 Se as selenite was orally administered to each rat. Blood and organs were collected 1, 3, 6, and 24h after dosing (4 rats/group at each time). Although apparent absorption of 75 Se was independent of estrogen status, hormone associated differences of 75 Se levels (P<0.05) were noted in plasma, RBC, liver, heart, kidney, spleen, brain, and thymus at certain times. For ii

3 example, total 75 Se in liver was greater in OVX than in OVX+E2 rats after 1 hour (13.1% vs.3.9% of total dose). However, OVX+E2 group had greater 75 Se in liver than OVX group (18.0% vs. 10.9%) after 6h. The relative distribution of 75 Se between cytosol and membrane fractions in organs was independent of estrogen status. Also the relative distribution of 75 Se among the selenoproteins in cytosol of the organs was not influenced by estrogen status. However, plasma selenoprotein P (SelP) in OVX+E2 group contained a greater percentage of administered 75 Se at 3, 6 and 24h after gavage compared to OVX group (P<0.05). 75 Se in plasma glutathione peroxidase (GPx) also was greater in OVX+E2 compared to OVX group at 24 h (P<0.05). The second aim was to investigate the effect of estrogen status on selenium status in blood and tissues and explore whether hepatic levels of SelP mrna and GPx1 mrna were affected by estrogen status. SD female rats (7 weeks of age) were bilaterally ovariectomized and implanted with either a placebo pellet (OVX, n=6) or pellet with estradiol (OVX+E2, n=6). A second set of SD female rats also were sham-operated and implanted with a placebo pellet (Sham, n=24; 6 rats in each 4-day estrous cycle). Blood and tissues were collected at 12 weeks of age. Estrogen significantly increased selenium status as measured by selenium concentration and GPx activity in plasma, liver, and brain. Selenium concentration in RBC was also increased by estrogen treatment. Selenium status in kidney and heart was independent of estrogen treatment. Real-time RT-PCR iii

4 analysis demonstrated that both hepatic SelP and GPx1 mrna were significantly increased by estrogen treatment (P<0.05). In conclusion, these results suggest that estrogen status affects distribution of ingested selenium in tissue- and time-dependent manners. Expression of hepatic SelP and GPx was regulated by estrogen at both mrna and protein levels. As SelP has been shown to function as a selenium transporter, estrogen regulation of SelP may play an important role in whole body metabolism of selenium. iv

5 Dedicated to my wife, Wenyi v

6 ACKNOWLEDGMENTS During my Ph.D. journey at The Ohio State University, there has been so much support and encouragement without which this dissertation could not have been possible. I would like to express my sincere gratitude to my advisor, Dr. Anne Smith. Her intellectual ideas, guidance, and research expertise was a tremendous help throughout my dissertation research. Especially, I am grateful for her patience and encouragement that helped me through each difficulty during my study. I wish to thank Dr. Mark Failla. His insightful advice and continuous intellectual challenge have been key in facilitating my growth as a researcher. I sincerely appreciate my committee members, Dr. Steven Clinton and Dr. Charles Brooks for sharing their time and research expertise with me over the years, and for their valuable advice on my dissertation research. I wish to thank Dr. Zhontang Yu and Dr. Mark Morrison for the incredible opportunity to work in the molecular microbiology lab to conduct the real-time RT-PCR analysis. Thanks to Dr. Jing Chen for training me on the molecular biology techniques, to Dr. Valerie Bergdall for the training on estrus cycle determination in rats and valuable advice on animal handling, to Dr. Michael Darby for helping me set up the gamma vi

7 counter, to Dr. John Bruno for his kindness to let me use his microscope and dissection equipment at Townshend Hall, to Jodi Griffith for her help on animal tissue collection, to Dr. Maureen Geraghty for her encouragement and friendship, and to Jeanne McGuire for the help on radioactive materials handling. My special gratitude goes to Dr. Kristina Hill, and Dr. Raymond Burk at the Vanderbilt University, for their assistance on plasma selenoprotein P measurement, and insights on selenium research over the years. I am grateful to the financial support from OARDC for my dissertation research. I wish to express my appreciation to the Department of Chemistry, who granted me the highly competitive teaching assistantship during my final stage of the Ph.D. study. Lastly, I would like to especially thank my wife, Wenyi and my parents, for their love, faith, and expectation. I would not have reached this point without their consistent support. vii

8 VITA January 21, 1971 Born Mianyang, Sichuan Province, China 1995 Bachelor of Medicine, West China University of Medical Sciences, China 1998 M.S., Human Nutrition, West China University of Medical Sciences, China Clinical Dietitian and lecturer, First University Hospital, West China University of Medical Sciences, China Graduate Research and Teaching Associate, Department of Human Nutrition, The Ohio State University 2006-present Graduate Teaching Associate, Department of Chemistry, The Ohio State University PUBLICATIONS 1. Zhou, X., Smith, A.M., and Failla, M.L. Estrogen Status Alters Tissue Distribution of Oral Dose of 75 Se-Selenite and Liver mrna Levels of SelP and GPx. FASEB J 2007; 21 (5): A717 (Abstract). 2. Zhou, X., Huang, C., Hong, J.,Yao, S., and Zhang, J. A Nested case-control Study on Riboflavin Levels in Blood and Urine and the Risk of Lung Cancer. Journal of Hygiene Research 2003; 32 (6): , 601. viii

9 3. Mao, S., Li, J., Zhou, X., Yuan, H., and Yang Z. Enternal Nutrition Support for the Postoperative Patients with Larynx Tumor. Chinese Journal of Clinical Nutrition 2002; 10 (2): Zhang, J., Zhou, X., Huang, C., Yao, S., Chang, S., Qiao, Y., and Taylor, P. Reproducibility and Validity of a Food Frequency Questionnaire among Male Miners. Modern Preventive Medicine 1999; 26 (2): , Zhou, X., Zhang. J., Huang. C., Yao. S., Taylor, P., and Qiao, Y.. A Case-Control Study of Vitamins and Lung Cancer Risk. Acta Nutrimenta Sinica 1999; 21 (4): Zhou, X., Huang. C., Yu. Q., and Lin. Y. The Effects of Malva Crispa Powder on Immune Function in Normal Mice. Modern Preventive Medicine 1998; 25 (3): FIELDS OF STUDY Major Field: The Ohio State University Nutrition Program ix

10 TABLE OF CONTENTS Page Abstract...ii Dedication...v Acknowledgments.....vi Vita...viii Table of contents.x List of tables.xiii List of figures xiv Chapters: 1 Introduction. 1 2 Literature review General information of selenium Selenoproteins Brief introduction of known selenoproteins Glutathione peroxidase Selenoprotein P Selenium metabolism Absorption Transport Tissue distribution Metabolic pathways Selenium incorporation into selenoproteins Excretion Selenium metabolism models Requirements of selenium Selenium deficiency Keshan disease Kaschin-Beck disease Selenium deficiency in New Zealand and Finland Selenium toxicity.66 x

11 2.7 Selenium and hormone-related cancers Selenium and breast cancer Selenium and prostate cancer Estrogen receptors in estrogen action Estrogens Estrogen receptor Mechanism of estrogen actions Selenium and reproductive hormones Animal studies Human studies Tissue specific influence of sex hormones on selenium status Estrogen status alters tissue distribution and metabolism of oral dose of 75 Se-selenite Introduction Materials and methods Animals Diets SeO 3 2- administration Sample collection Plasma 17β-estradiol concentration Se in tissues and cytosol Determination of 75 Se in selenoproteins Statistical analysis Results Estrogen status, food intake, and body weight Se activity in tissues Distribution of 75 Se among subcellular fractions and selenoproteins Discussion The effect of estrogen on selenium status in tissues and hepatic levels of SelP and GPx mrna Introduction Materials and methods Animals Diets Estrous cycle determination in sham-operated rats Sample collection Laboratory analyses Statistical analyses xi

12 4.3 Results Body and organ weights Plasma 17β-estradiol concentration and ceruloplasmin activity Tissue selenium concentrations and GPx activity Plasma selenoprotein P concentration Hepatic levels of SelP and GPx mrna Discussion Epilogue..173 Bibliography 179 Appendices Appendix A, Animal use protocols approved by ILACUC, The Ohio State University 218 Appendix B, Composition of the custom diet..231 Appendix C, 75 Se use protocols approved by Radiation Safety, EHS, The Ohio State University Appendix D, Procedures of rat tissue collection.243 Appendix E, 17β-estradiol double antibody radioimmunoassay (RIA)..245 Appendix F, The relative distribution of 75 Se among the selenoproteins in cytosol..248 Appendix G, Methods for selenium analysis Appendix H, Methods for glutathione peroxidase activity analysis 260 Appendix I, Polymerase chain reaction (PCR) analysis..265 xii

13 LIST OF TABLES Table Page 3.1 Summary of the 75 Se activity in SD female rat tissues 1, 3, 6, and 24h after gastric administration of 75 Se Se activity in cytosol (dpm/μg protein) Se activity in SelP, TrxR1, TrxR2 and GPx Se distribution among plasma SelP, GPx, and low molecular weight selenium (LMW) 1, 3, 6, and 24h after administration of 75 SeO 3 2- by gavage Initial uptake of 75 Se in tissues of female Wistar rats given oral or intravenous dose of 75 Se-selenite and 75 Se-selenomethionine (Thomson study), or of female SD rats given oral dose of 75 Se-selenite (present study) Primer and probe sequences used for amplication of the SelP, GPx1 and GAPDH genes in rat liver Effect of estrogen status on tissue selenium concentration in female SD rats Effect of estrogen status on tissue glutathione peroxidase (GPx) activity in female SD rats xiii

14 LIST OF FIGURES Figure Page 2.1 Schematic representation of mrna structures from selenoproteins and selenoprotein P Selenium metabolic pathways Human selenite/selenomethionine metabolism model Hypothetical model of selenium transport via SelP and its uptake mechanism by the brain and other tissues Chemical structures of three major naturally occurring estrogens Structure and homology between human ERα and ERβ protein Classical model of the ER activation process Representative plot of molecular weight markers separated by SDS-PAGE Schematic of separation of 75 Se-selenoproteins in cytosol by SDS-PAGE Food intake of female SD rats before and after surgery Body weights of female SD rats before and after surgery Se activity in tissues 1, 3, 6, and 24h after gastric administration of 60 μci 75 SeO 3 2- (0.043 μg Se) Distribution of 75 Se activity among selenoproteins in liver cytosol 6h after gastric administration of 75 Se-selenite..118 xiv

15 Se activity in plasma SelP 1, 3, 6, and 24h after administration of 75 SeO 3 2- by gavage RNA isolated from SD female rats using RNAqueous-4PCR Plate setup for the absolute quantification of the real-time PCR analysis Determination of copy number for GPx1 mrna by using a standard curve Plate setup for the relative quantification of real-time PCR analysis Body weights for female SD rats before and after surgery Organ weights (g/kg BW) for female SD rats from three groups Plasma concentration of 17β-estradiol in different groups of female SD rats Plasma ceruloplasmin activity in different groups of female SD rats Estrogen status does not affect plasma selenoprotein P (SelP) concentration Estrogen replacement in OVX rats increases hepatic SelP mrna Estrogen replacement in OVX rats increases hepatic GPx1 mrna Hypothetical model of estrogen regulation whole body selenium Metabolism 172 Appendix F Distribution of 75 Se activity among selenoproteins in cytosol after gastric administration of 75 Se-selenite 249 xv

16 CHAPTER 1 INTRODUCTION Selenium is an essential nutrient that exerts its biological functions through selenoproteins, i.e., proteins that contain selenium in the form of selenocysteine (Burk and Hill, 1993). At least 25 human selenoproteins have been identified (Gromer et al., 2005). These include cellular and extracellular glutathione peroxidase (GPx), phospholipid hydroperoxide GPx, gastrointestinal GPx, thioredoxin reductase (TR), 5 -iodothyronine deiodinase (5 -DI, type 1, type 2, and type 3), selenoprotein P (SelP), selenoprotein W (SelW), and selenophosphate synthetase. Estrogen has been shown to influence the metabolism of cytokines (Wang et al., 2005; Carruba et al., 2003) and several minerals, including copper (Canaraja et al., 2004; Alkjaersig et al., 1988; Bureau et al., 2002, Dorea and Miazaki, 1999), iron (Milman et al., 1992), magnesium, calcium (Muneyyirci-Delale et al., 1998), and chromium (Bureau et al., 2002). Some relationships between gender/sex hormones and selenium status have been observed in animals and humans, suggesting that male and female reproductive 1

17 hormones may influence the distribution of selenium into organs and blood. It has been shown that female rats have greater hepatic GPx activity than male rats (Capel and Smallwood, 1983; Finley and Kincaid, 1991; Prohaska and Sunde, 1993; Yamamoto et al., 2002). After castration, GPx activity in male liver increased to a level similar to that in female liver (Capel and Smallwood, 1983). This change was proposed to be due to accumulation of selenium in liver that would normally go to the testis. Marano et al. (1991) investigated the effects of human sexual maturation on selenium status. Boys showed a significant decrease in serum selenium concentration during sexual maturation, whereas the serum selenium levels of girls remained constant throughout puberty. These findings suggest that the increase in testosterone during male puberty may cause a redistribution of selenium to the testis. Positive associations between estrogen concentrations during the human menstrual cycle and the plasma selenium, plasma GPx activity (Ha and Smith, 2003), and erythrocyte GPx activity (Massafra et al., 1998; Ha and Smith, 2003) have been observed. The positive association between erythrocyte GPx activity and serum estrogen levels was also observed in both premenopausal and postmenopausal women receiving estrogen replacement therapy (Massafra et al., 2002). Estrogen treatment also significantly increased erythrocyte GPx activity in both premenopausal (L'abbe et al., 1992) and postmenopausal women (Massafra et al., 1997; Massafra et al., 2002; Bednarek-Tupikowska et al., 2006). 2

18 Preliminary findings from our laboratory strongly support the likelihood of a relationship between selenium and estrogen availability. Maternal blood selenium concentrations in rats decline during pregnancy (Smith and Picciano, 1986; Smith and Chang, 1994). Blood selenium and estrogen concentrations are positively correlated during the rat estrous cycle (Smith et al., 1995) and the human menstrual cycle (Ha and Smith, 2003). Moreover, blood selenium concentrations have been shown to decline with the decline in estrogen after menopause (Smith et al., 2000). Breast cancer is the most common malignancy in American women. Breast cancer alone is expected to account for 26% (i.e., 176,296) of all new cancer cases among women in 2007 (American Cancer Society, 2007). Cellular, animal, clinical and epidemiological studies suggest that selenium plays a preventive and/or anticarcinogenic role in the development of breast cancer, although the data are not conclusive. Some studies have shown that women with breast cancer had lower blood selenium levels compared to those of healthy controls (Charalabopoulos et al., 2006; Lopez-Saez et al., 2003; Huang et al., 1999; Gupta et al., 1994; Chaitchik et al., 1988; Schrauzer et al., 1985; McConnell et al., 1980), whereas other studies have failed to confirm this observation (Ghadirian et al., 2000; Mannisto et al., 2000; Dorgan et al., 1998). The induction of breast cancer is a long, complex process, but its rarity among males suggests a role of female sex hormones during the process (Forrest, 1997). Approximately two-thirds of human breast cancers depend on estrogen for growth 3

19 (Lipton, 1999). Estrogen is a modulator of cellular growth and differentiation. Its major targets in females are the mammary gland and uterus. Animal studies have shown that estrogen can induce and stimulate breast cancer not only in adult females, but also in their female offspring (Hilakivi-Clarke et al., 1999). Feeding a high fat diet to pregnant rats increased the circulating estradiol (E2) levels during pregnancy and the risk of developing carcinogen-induced mammary tumors in their female offspring (Hilakivi-Clarke et al., 1999). Estrogen also stimulates the in vitro growth of human breast cancer cells (Dickson and Russo, 2000), and increases breast cancer risk in postmenopausal women (Hankinson et al.,1998). Breast cancer is one of only a few malignancies in which estrogen receptors (ER) are expressed on the surface of the tumor cells (Jensen et al., 2001). When ER is present, the likelihood that breast tumors will respond to hormonal blockade with anti-estrogenic agents such as tamoxifen is great. Selenium may activate and/or deactivate steroid receptors and play a potential role as an environmental estrogen (Shah et al., 2005; Stoica et al., 2000). Selenium in breast fat from breast cancer patients has also been positively correlated with estrogen receptor status (Mussalo-Rauhamaa and Pantzar, 1993). The relationship between selenium and estrogen may also stem from the presence of an estrogen response element (ERE) in the promoter region of some selenoprotein genes such as plasma GPx (GPx3) (Borthwick et al., 2003) and phospholipid GPx (GPx4) (Brigelius-Flohe et al., 1994). An additional suggestion that selenium is somehow 4

20 involved with estrogen and breast cancer risk comes from selenium s preventive effect on breast cancer being more pronounced after menopause (Kumar et al., 1991; Hardell et al., 1993; Huang et al., 1999). Use of the anti-estrogen tamoxifen also appears to affect selenium status. Postmenopausal women with breast cancer receiving long-term tamoxifen had greater plasma selenium levels than similar patients who were not receiving tamoxifen (Schrijver et al., 1987). These findings suggest that the interaction between selenium and estrogen may play a role in the etiology of breast cancer. Due to the inconclusive data on selenium and breast cancer risk and the approval of anti-estrogens for use in breast cancer prevention (Fisher et al., 1998), information about the impact of estrogen on the tissue distribution and metabolism of selenium is required before considering the use of selenium in breast cancer prevention and/or treatment. Previous research has focused on the effect of estrogen status or estrogen administration on selenium concentration and GPx activity in plasma and erythrocyte. Little is known about the effect of estrogen status on selenium metabolism in other organs, such as liver, kidney, brain, heart, lung, thymus, and spleen. Recent studies suggest that SelP plays a central role in whole body selenium metabolism by serving as a transport protein to shuttle selenium from liver to other tissues (Hill et al., 2003; Schomburg et al., 2003). Information on how SelP is regulated by estrogen may provide insights about selenium metabolism. 5

21 Experimental animal models, specifically rats and mice, have been used effectively to investigate selenium distribution among tissues. In the present study, estrogen status was manipulated by subjecting female Sprague Dawley (SD) rats to the following treatment: 1) ovariectomy with estrogen replacement (OVX+E2); 2) ovariectomy with a placebo pellet; and 3) sham operation with a placebo pellet (Sham). As the rats have different estrogen status, it is an ideal model to be used for evaluating of the impact of estrogen availability on selenium metabolism and distribution to tissues. Experiments were designed to investigate the following questions: 1) Does estrogen affect tissue distribution and metabolism of oral dose of 75 Se-Selenite? 2) Does estrogen affect selenium status, as assessed through measurements of selenium concentrations and GPx activity, in blood and organs? 3) Does estrogen affect hepatic levels of SelP mrna and GPx mrna? The research hypotheses for the study were: 1) Estrogen status will affect the whole body distribution and metabolism of orally administered 75 Se in a tissue-specific manner. 2) Estrogen status will affect selenium concentration and GPx activity in selective tissues. 3) Estrogen will upregulate hepatic levels of SelP mrna and GPx mrna. The increase in hepatic SelP mrna is expected to result in increased synthesis of SelP in liver and increased transport of selenium from liver to other tissues. The specific aims of my dissertation project were: 1) examination of the effect of estrogen status on absorption of 75 Se; 2) examination of the effect of estrogen status on 6

22 75 Se activity in blood and organs; 3) examination of the effect of estrogen status on the distribution of 75 Se between cytosol and membrane fractions in organs; 4) examination of the effect of estrogen status on the distribution of 75 Se among the selenoproteins in cytosol and plasma; 5) examination of the effect of estrogen status on the selenium concentrations and GPx activity in blood and organs; 6) examination of the effect of estrogen status on hepatic levels of SelP mrna and GPx mrna. The results of this study will provide information about selenium transport and metabolism during altered estrogen status, and suggest a mechanism by which SelP may be the link between estrogen and selenium status. The acquisition of such knowledge is critical to the development of effective strategies to assess selenium status relative to a woman s estrogen status, and for the possible use of selenium as a chemopreventive or chemotherapeutic agent for estrogen-responsive breast and perhaps, ovarian cancer. 7

23 CHAPTER 2 LITERATURE REVIEW 2.1 General information of selenium In 1818, the Swedish chemist Jons Jacob Berzeliu first discovered selenium as an element. He named it Selene, Greek for the moon goddess (Krehl, 1970). Selenium (Se) is element 34 (Group VI A) on the periodic table. It lies between the group VA metal arsenic and the group VII nonmetal bromine. Thus, selenium is considered a metalloid, having both metallic and nonmetallic properties. Selenium has an atomic weight of 78.96, a boiling point of 958 Kelvin, and a melting point of 494 Kevin. It can exist in various oxidation states: -2, 0, +2, +4, and +6. The +2 state, however, is not commonly found in nature (Krehl, 1970; Wilber, 1980). Elemental selenium (0) can be oxidized to selenite (+4) or selenate (+6), or reduced to selenide (-2). There are six stable isotopes of selenium in nature, including 74 Se, 76 Se, 77 Se, 78 Se, 80 Se, and 82 Se. These isotopes have been used to study the biological utilization of selenium in food (Janghorbani et al., 1981). 75 Se is a radioisotope with a half-life 120 8

24 days. Because of its emission of gamma-radiation and relatively long half-life, 75 Se has been widely used as a tracer in selenium metabolic studies. Selenium is one of the rarest elements (a trace element) that naturally occur in the earth s crust. Its concentration in soil varies depending on geochemical factors, especially the nature of the parent rocks. Thus, it is not surprising that selenium concentration in the crops varies greatly by geographic area. In the early 1970s, the biological function of selenium was reported to be as part of glutathione peroxidase, an enzyme that has antioxidant properties (Rotruck et al., 1973; Flohe et al., 1973). Since then, evidence for its importance to human health has been rapidly accumulating. Many new selenoproteins have been identified, and their biological functions have been explored. 2.2 Selenoproteins Brief introduction of known selenoproteins Selenoproteins are specific proteins that contain selenium in the form of genetically encoded selenocysteine, now recognized as the 21 st amino acid (Stadtman, 1996). Selenium exerts its biological functions mainly through selenoproteins. More than thirty selenoproteins have been identified (Surai, 2006; Gromer et al., 2005; Behne and Kyriakopoulos, 2001; Kryukov et al., 2003), including: 9

25 7 forms of glutathione peroxidase (GPx), a group of antioxidant enzymes that have been widely used as selenium markers. An association between tissue GPx activity and estrogen has been documented. The details of GPx will be reviewed in this chapter. Selenoprotein P (SelP), a selenoprotein that can act as a selenium-transporter. SelP plays an important role in whole-body selenium metabolism. The details of SelP will be reviewed in this chapter. 3 forms of iodothyronine deiodinase, a group of enzymes that regulate the metabolism of thyroid hormones, e.g., T3 and T4. Selenoprotein W (SelW), a selenoprotein mainly expressed in skeletal muscle, and has antioxidative property. 3 forms of thioredoxin reductase (TrxR), enzymes that catalyze NADPH-dependent reduction of thioredoxin. TrxRs regulate the synthesis of DNA, and also act as antioxidative enzymes. Selenophosphate synthetase (SPS2), an enzyme catalyzes the formation of selenophosphate from selenide and ATP. Two forms of SPS in humans have been identified with SPS2 being a selenoprotein. 15-kDa selenoprotein (Sep15), a selenoprotein expressed in the prostatic gland, testes, brain, kidney, and liver. Certain prostate cancer cell lines have reduced 10

26 level of Sep15. However, the beneficial effect of Sep15 on cancer prevention is still speculative. Selenoprotein H (SelH), a globular protein expressed in numerous tissues. Its biological function is unknown at present. Selenoprotein I (SelI), a membrane protein expressed in a number of tissues, yet its function is unknown. Selenoprotein K (SelK), a membrane protein with unknown function. Selenoprotein M (SelM), a protein with the highest mrna levels in the brain and the lowest mrna levels in liver and spleen. SelM has been suggested to be a redox active protein. Selenoprotein N (SelN), a glycoprotein found in high levels in several human fetal tissues and at much lower levels in adult tissues. SelN has been suggested to play a role in regulation of early development and in cell proliferation or regeneration. Mutations in the SelN gene are associated with congenial muscular dystrophy with spinal rigidity and restrictive respiratory syndrome. Selenoprotein O (SelO), a large selenoprotein (73.4 kda) with a possible redox-dependent activity. Selenoprotein R (SelR), a cytosolic and nucleic protein with a possible role on protect brain against oxidative stress. 11

27 Selenoprotein S (SelS), a protein whose expression is inversely correlated to the plasma glucose concentration in rats, indicating a role in regulating glucose metabolism. Selenoprotein T (SelT), a selenoprotein that has possible redox properties. Selenoprotein V (SelV), a selenoprotein expressed in the testes, may have a redox-related function Glutathione peroxidase The glutathione peroxidase (GPx) family includes 7 selenium dependent enzymes that differ in their molecular weight, structure and functions, i.e., cytosolic GPx, gastrointestinal GPx, plasma (or extracellular) GPx, phospolipid hydroperoxide GPx, sperm nuclei GPx, GPx6, and GPx7. A general function of GPx is to catalyze the reduction of hydrogen peroxide and organic hydroperoxide and thus protect cells from oxidative damage Cytosolic glutathione peroxidase (cgpx or GPx1) The cytosolic GPx (cgpx) was the first identified selenoprotein (Rotruck et al., 1973). It catalyzes the reduction of hydrogen peroxide and various soluble organic peroxides to water and alcohols, respectively (Burk and Hill, 1993). cgpx is composed of four identical 22-kDa subunits each containing one selenocysteine residue (Burk and 12

28 Hill, 1993). Mullenbach et al. (1987) demonstrated that the human cgpx gene encodes a 201 amino acid polypeptide. This structure is similar to that of the mouse GPx, which is also a 201 amino acid polypeptide (Chambers et al., 1986). cgpx is present in nearly all tissues, but its activity varies greatly among different species and tissues. In rats (Smith and Picciano, 1986; Smith and Picciano, 1987) and mice (Schisler and Singh, 1988), liver and kidney had the highest GPx activity compared to other tissues. Muscle and brain have been shown to have the lowest GPx activity compared to other tissues (Behne and Wolters, 1963). cgpx activity is dependent on selenium status of tissues. In rats (Arthur et al., 1987) and mice (Reiter and Wendel, 1984) fed a selenium-deficient diet, liver cgpx activity dropped to less than 1% of the controls. However, the losses in cgpx activity did not cause any negative effects on the health of these animals (Arthur et al., 1987; Reiter and Wendel, 1984; Sunde et al., 1997b). Also, the pathological changes of cgpx knockout mice were not observed and the mice showed normal development (Ho et al., 1997). These studies suggest that animals could maintain their healthy status with very low cgpx activity under normal conditions. However, when challenged with oxidative stress, pathological changes have been observed in cgpx knockout mice (Fu et al., 1999). cgpx activity in various tissues also decreased at high oxidative stress (Lee et al., 2003; Goering et al., 2002). 13

29 Dietary selenium supplementation has been shown to increase cgpx activity in rats, mice, and some other animals (Brigelius-Flohe, 1999). For selenium-deficient rats, the cgpx activity could be recovered by feeding a diet containing 0.02 to 0.1 microgram selenium/gram of diet (Sunde et al. 1989). The effect of selenium status on mrna levels of cgpx has been investigated. In rats fed selenium-deficient diet, hepatic mrna levels of cgpx decreased to 6-15% of selenium adequate levels (Saedi et al., 1988; Hill et al., 1992). Hepatic cgpx mrna levels elevated with an increase in dietary selenium (Weiss et al., 1997). Thus, dietary selenium regulates cgpx synthesis at both mrna and protein levels Gastrointestinal glutathione peroxidase (GI-GPx, or GPx2) Gastrointestinal glutathione peroxidase was isolated from human hepatoma cells (Chu et al., 1993). This is a 22-kDa tetrametric selenoenzyme that has the same enzymatic properties as those of cytosolic GPx. It catalyzed the reduction of hydrogen peroxide and fatty acid hydroperoxide, but not of phospholipids hydroperoxide or cholesterol hydroperoxide (Esworthy et al., 1993). Unlike cgpx, which is present in almost every tissue, GI-GPx is found only in the GI tract of rats, and the GI tract and liver in humans (Chu et al., 1993). The distribution of GI-GPx varies in the intestine with a higher concentration in the crypts and a lower concentration at the luminal surface (Florian et al., 2001). mrna of GI-GPx was detected throughout the GI tract with the 14

30 greatest level of expression in the ileum and cecum (Chu and Esworthy, 1995). The universe presence of GI-GPx in the GI tract suggests that it may be a major component in the defense system that protects the GI tract from damage caused by ingested lipid hydroperoxides (Chu et al., 1993). Inflammatory bowl disease and bacteria-induced tumors were observed in cgpx and GI-GPx double knockout mice (Chu et al., 2004) Plasma or extracellular glutathione peroxidase (pgpx, or GPx3) Plasma or extracellular glutathione peroxidase was purified and characterized from human plasma with four identical 23-kDa subunits, each containing one selenocysteine residue encoded by a UGA (Takahashi et al, 1987). This enzyme is a glycoprotein and is present in the extracellular fluids. Besides plasma, it is also present in breast milk (Bhattacharya et al., 1988) and intestinal extracellular fluid (Tham et al., 1988). It is mainly synthesized in the kidney, and the kidney has the highest level of pgpx mrna (Chu et al., 1992; Avissar et al., 1994; Yoshimura et al., 1991). In vitro experiments have shown that pgpx could catalyze the reduction of hydrogen peroxide and various organic peroxides using glutathione as a substrate. But its specific enzymatic activity is only 10% that of cgpx (Takahashi et al., 1987). The biological function of pgpx is still not clear. The known function of GPx does not seem to be possible because glutathione concentration in blood plasma is too low to serve as a reducing substrate for pgpx. It has been suggested that pgpx could reduce lipid 15

31 hydroperoxides in LDL, however, it is not active against peroxidized cholesterol esters (Flohe and Brigelius-Flohe, 2001). pgpx activity is affected by dietary selenium intake. The activity decreases in selenium deficiency and increases with enhanced selenium intake (Whanger et al., 1988a). Thus, pgpx activity has been used as a biomarker of selenium nutritional status. Positive associations between estrogen concentrations during the human menstrual cycle and the pgpx activity have been observed (Ha and Smith, 2003) Phospholipid hydroperoxide glutathione peroxidase (PHGPx, or GPx4) Phospholipid hydroperoxide glutathione peroxidase (PHGPx) was identified by Ursini et al. (1985) as a kda monomer. PHGPx differs from cgpx in the amino acid composition. Arginine residues surround the reaction center of cgpx and are involved in the binding of GSH with cgpx, but are not present in PHGPx (Mauri et al., 2003). It uses glutathione as a reducing substrate and exhibits a broad specificity in reducing different phospholipid hydroperoxides to their corresponding alcohols. It can also reduce hydroperoxides still integrated in membranes, thus acting as a universal antioxidant in the protective system against the oxidative destruction of biomembranes (Ursini et al., 1982; Ursini et al., 1985; Schnurr et al. 1996). PHGPx also has important functions in the redox regulation of a variety of processes. PHGPx has been shown to inhibit the synthesis of fatty acid hydroperoxide by inhibiting lipoxygenases 16

32 (Brigeliius-Flohe et al., 1999, Brigeliius-Flohe et al., 1997). PHGPx can also inhibit apoptosis of certain cells (Tang et al., 2005; Brigeliius-Flohe et al., 1999). PHGPx is present in almost every tissue, and its levels vary in different tissues with the highest level in testis. In rat, testis had a 15- fold higher PHGPx activity and a 45-fold higher PHGPx mrna level than those in liver (Lei et al., 1995). The high activity of PHGPx in testis is related to its biological role in spermatogenesis. Earlier studies have shown that selenium is highly enriched in the spermatozoa in the form of a selenoprotein located in the outer mitochondrial membrane (Calvin et al., 1981). Later studies identified this selenoprotein as 19.7kDa PHGPx (Behene et al., 1988; Ursini et al., 1999). The sharp rise of the selenium concentration in testis during puberty is due to the abundant expression of PHGPx in the round spermatids (Maiorino et al., 1998). The role of PHGPx in spermatogenesis is supported by the findings that low selenium levels and possibly PHGPx polymorphisms are associated with male infertility (Vogt, 2004). Testosterone causes a redistribution and accumulation of selenium to the testis during male puberty (Marano et al., 1991). This redistribution provides more selenium for incorporation into PHGPx, which has particular relevance to male fertility (Foresta et al., 2002). PHGPx is more resistant to selenium deficiency compared to other forms of GPx. In weaning male Wistar rats, selenium deficiency caused a much less decrease in PHGPx levels than that of the cgpx (Guan et al., 1995). 17

33 Other glutathione peroxidase Sperm nuclei glutathione peroxidase (SnGPx, GPx5) was identified in rat testis as a 34-kDa selenoprotein with similar properties to PHGPx (Pfeifer et al., 2001). It is present in the sperm nuclei and comprises around 80% of total selenium present in the sperm nuclei (Behne et al., 2000). It differs from PHGPx in its N-terminal sequence. An alternative exon responsible for formation of the SnGPx was identified in the first intron of the PHGPx gene. SnGPx is necessary for male fetility due to its role in stabilizing the condensed chromation by cross-linked protamine disulfides. SnGPx level decreased to one third of normal in selenium-depleted rats, and chromatin condensation was disturbed (Pfeifer et al., 2001). Glutathione peroxidase 6 (GPx6) is a close homolog to plasma GPx3. It was identified by Kryukov et al. (2003) and the gene encoding human and porcine GPx6 was cloned. So far, GPx 6 has only been detected in embryos and olfactory epithelium. It is suggested that GPx6 has a function in olfaction because it expressed in or near the Bowman s glands, where several olfactory-specific biotransformation enzymes are localized. Glutathione peroxidase 7 (GPx7) is a 22-kDa cytoplasmic protein sharing a similar structural domain to GPx4 (Utomo et al., 2004). It is expressed in many tissues, including mammary gland. It has been suggested to be involved in breast cancer cell defense by alleviating oxidative stress generated from consumption of polyunsaturated fatty acids.. 18

34 2.2.3 Selenoprotein P Selenoprotein P (SelP) was first reported in the early 1970s (Burk, 1973; Millar et al., 1972). It was shown that 75 Se rapidly incorporated into a plasma protein in rats. Herrman (1977) identified this plasma protein to be distinct from glutathione peroxidase. In 1982, two research groups further characterized this protein. It was found that in selenium deficient rats, 75 Se was incorporated into this selenoprotein more efficiently than into glutathione peroxidase (Burk and Gregory, 1982). Tappel s group (Mostenbocker and Tappel, 1982) demonstrated that this protein contained selenium in the form of selenocysteine, thus this protein was accepted as the second selenoprotein identified in animals, and named selenoprotein P because of its plasma location Properties of selenoprotein P It was not until 1987 that SelP was purified successfully from rat plasma with immunoaffinity chromatography (Yang et al., 1987). SelP has also been isolated from human plasma (Akesson et al., 1994). Subjection of purified SelP to SDS-PAGE resulted in a 57-kDa protein band. Selenoprotein P is one of the major selenoproteins in plasma (along with plasma glutathione peroxidase). It is present in rat plasma at a concentration of 30 μg peptide/ml (Burk et al., 1991). In rats, it accounts for 60%-65% of the selenium in plasma (Read et al. 1990, Deagen et al. 1993). In humans, 40-60% of plasma selenium has been found to 19

35 be associated with SelP (Deagen et al. 1993, Akesson et al. 1994, Daher and van Lente 1994, Hill et al. 1996a, Saito and Takahashi, 2002). Besides plasma, SelP is widely expressed in other tissues, including liver, heart, brain, kidney, testis and muscle (Burk and Hill, 1994). Selenoprotein P is unique in its high selenium content. It contains 10 (human, murine, and rat) or 12 (bovine) selenocysteines in its polypeptide chain, while most selenoproteins contain only one selenocysteine per polypeptide chain (Gromer et al., 2005). In addition to the presence of selenocysteines, SelP has a high content of cysteine and histidine residues. The rat and human SelP sequences each encode two histidine-rich regions. These regions, in conjunction with the cysteine and selenocysteine content, are likely responsible for the metal binding function of SeP (Tujebajeva et al., 2000) Molecular biology Selenoprotein P contains selenium in the form of selenocysteine. Selenocysteine incorporation into the polypeptide chain is accomplished by a unique co-translational process, which requires the presence of specific secondary structures in selenoprotein mrnas, termed SECIS (selenocysteine insertion sequences) (Berry et al.,1991). The structure of SelP-mRNA is different from other selenoprotein-mrnas. Fig 2.1 shows different mrna structures of prokaryotic and eukaryotic selenoproteins, and of human 20

36 SelP (Mostert, 2000). In bacteria, a stem-loop structure is present immediately downstream from the UGA codon. Thus, the stem-loop structure appears to interact with the ribosome directly or indirectly to mediate selenocysteine incorporation. In eukaryotes, a stable stem-loop structure is present in the 3 -untranslated region (3 UTR) of the respective mrna. It acts at a considerable distance in directing the decoding of an in-frame UGA codon with selenocysteine. The structure of SeP mrna is unique. It contains two functional SECIS in its 3 UTR (Hill et al., 1993) (Fig 2.1 C), rather than a single stem-loop structure as other selenoproteins. Selenocysteine is synthesized from serine on a unique trna that recognizes certain UGA codons in the open reading frame of mrna. The secondary structure of mrna may mediate the recognition of UGA as coding for selenocysteine. Thus, the unique Sep-mRNA structure seems to be related to the regulation of SelP. 21

37 Figure 2.1 Schematic representation of mrna structures from selenoproteins and selenoprotein P. (A) mrna of prokaryotic selenoproteins; (B) mrna of eukaryotic selenoproteins; (C) mrna of human SelP. The black verticals bars indicate the positions of in-frame UGA codons (not to scale). (Volker, 2000) 22

38 Isoforms Chittum et al. (1996) has identified 5 isoforms of SelP in rat plasma by phosphorimaging of separated proteins with SDS-PAGE. Three forms were 57-kDa (designated as Se-P 57A, Se-P 57B, and Se-P 57C ), and the other two forms were 45-kDa (designated as Se-P 45A, Se-P 45B ). A later study using conventional peptide sequencing and mass spectrometry further characterized the isoforms of SelP in rat plasma (Ma et al., 2002). It has been shown that the three shorter isoforms shared the same N terminus but had C termini at the positions of the second, third, and seventh UGAs. They were given the names Se-P 1, Se-P 2, and Se-P 6, the subscript indicates the number of selenium atoms in the isoform. The presence of the isoforms of SelP implies that the second UGA codon in SelP mrna has alternative functions: coding for the incorporation of selenocysteine or coding for termination of translation. Such a regulatory mechanism would allow the cell to produce either the short protein or the long one from a single mrna. For example, in rat, Se-P 57B is a full-length product of the mrna, and Se-P 45B terminates at the second UGA in the open reading frame. The major difference between Se-P 57B and Se-P 45B appears to be the length of their polypeptide chains (Himeno et al., 1996) Regulation of SelP SelP level in plasma was affected by several factors, including age, gender, cigarette smoking, geographic location, and dietary selenium (Moschos, 2000). Many studies have 23

39 focused on the response of SelP to dietary selenium. Janghorbani, et al. (1999) investigated the effect of dietary selenium restriction on selenium status, including plasma SelP concentration, in Chinese adult males with high life-long intake of selenium. Ten adult male volunteers (Enshi cith, Hubei Province) whose habitual daily Se intake is approximately 480 µg/day (several-fold higher than habitual intake of adult US residents, which is approximately 100 µg/day) were recruited in this study. They were moved to Lichuan county, Hubei province, where they stayed for 70 days and consumed locally grown food, which provided Se around 43.3 µg/day. After 62 days of Se restriction, the total plasma Se decreased by 20%, and the plasma SelP decreased by 35%. Supplementation of selenium to selenium-deficient population in China (Xia et al., 2005) and U.S. (Duffield et al., 1999) for 20 weeks increased their plasma SelP concentration. And the increase was dose-dependent. In animal studies, similar results were obtained. Weanling male rats were fed a Torula yeast-based selenium-deficient diet, and the controls were fed the identical diet containing 0.25 mg selenium as sodium selenite/kg of diet for one year (Chittum et al., 1997). The level of plasma SelP of selenium-adequate rats (control) was 29.2 µg/ml, while plasma SelP in selenium-deficient rats decreased to 1.0 µg/ml. Some research has been carried out to investigate the effect of dietary selenium intake on SelP mrna content. Weanling male rates were fed a basal torula yeast diet (0.07 µg Se/g diet) supplemented with graded levels of dietary selenium (0 to 0.2 µg Se/g 24

40 as Na 2 SO 3 ) (Weiss et al., 1997). After 33 days, liver SelP mrna levels in the unsupplemented rats were 69±2% of the levels in rats fed 0.1 µg Se/g diet. In another similar designed study, weaning rats were also fed a selenium-deficient diet supplemented with 0, 0.1 or 2.0 mg Se/kg of diet as sodium selenite (Christensen et al., 1995). After feeding Se deficient diet for 91 days, Northern blot analysis shows that reductions in SelP mrna levels were 50% in kidney, and 14% in liver. To determine the effects of different selenium intakes on the relative rates of selenoproteins transcription, a run-on assay was conducted with liver nuclei from rats fed three levels of dietary selenium (0, 0.1 or 2.0 mg/kg diet of Se supplement) (Christensen et al., 1995). α- 32 P-UTP was used to label nascent transcripts. The results suggest that transcription of the gene for GPx1, 5 -ID, and SelP was unaffected by the levels of dietary selenium intakes. Since selenium deficiency leads to the reduction of SelP mrna level, and it does not alter transcription rate of SelP mrna, the regulation of SelP mrna level is not a transcriptional control, but a posttranscriptional control. But how selenium deficiency increases the degradation of SelP mrna is still not clear. Selenocysteine is synthesized from serine on a unique trna that recognizes certain UGA codons in the open reading frame of mrna. Due to the deficiency of Se, selenocysteine is not available for insertion at UGA codons during translation, protein synthesis cannot proceed and no functional selenoprotein is generated (Mostert, 2000). In adult Chinese males, after 62 days of selenium restriction, plasma SelP decreased by 25

41 35%, while glutathione peroxidase decreased by 44% (Janghorbani et al., 1999). This indicates a certain hierarchy of synthesis among different selenoproteins when selenium intake is not sufficient, with SelP synthesis having priority over glutathione peroxidase. The differences in the translational control between SelP and other selenoproteins may due to the differences in the secondary structure of SelP mrna. As shown in Fig 2.1, SelP mrna contains two functional SECIS elements in its 3 -UTR, while other selenoproteins have a single loop-stem structure. As protein synthesis can be regulated by binding of a protein to secondary structure in their mrna, the secondary structure of SelP mrna may have a regulatory function for controlling translation. In addition to the cis-acting SECIS elements in selenoprotein mrnas, selenocysteine incorporation requires several trans-acting factors. These include the selenophosphate synthetase, an enzyme required for biosynthesis of selenocysteine on the trna, and the selenocysteine-specific trna, trna (Ser)Sec. The mammalian SectRNA (Ser)Sec population consists of two major isoacceptors that differ by a single methyl group attached to the ribosyl moiety at position 34. One trna contains 5 -methylcarboxymethyluridine (mcm 5 U) at position 34 and the other contains 5- methylcarboxymethyluridine-2 -O-methylribose (mcm 5 Um) (Chittum et al., 1997). It has been shown that repletion of selenium-deficient rats with selenium resulted in a gradual, tissue-dependent shift in the distribution of the different selenocysteine trna (Ser)Sec isoacceptors (Chittum et al., 1997). The redistribution of Sec-tRNA was 26

42 found in each of the four tissues examined, muscle, kidney, liver and heart, although the redistribution was tissue-specific. The steady state level of Sec-tRNA (Ser)Sec population was higher in each tissue of the control group (fed selenium-adequate diet) than in the Ose group (fed selenium-deficient diet). The Sec-tRNA (Ser)Sec population in each tissue of control group had a greater abundance of the mcm 5 Um isoacceptor, while the OSe group had a greater abundance of the mcm 5 U isoacceptor. Changes in selenium status affected the distribution of Sec-tRNA (Ser)Sec population in a time dependent manner. The changing of the ratio between these two isoacceptors depends on the time of selenium supplementation (2h-72h). The tissue-dependent shift in the distribution of Sec-tRNAs isoaccepors suggests another mechanism that regulates the tissue levels of SelP and other selenoproteins. This tissue- specific regulation of SelP plays an important role in whole-body selenium metabolism, because SelP has been shown to function as a selenium transport protein (Schweizer et al., 2005). Besides dietary selenium intake, other factors affecting SelP expression have also been investigated. A group in Germany examined how SelP expression is regulated by cytokines. A 1.8-kilobase (kb) genomic clone from a human placenta library was isolated, cloned, and sequenced (Dreher et al., 1997). The major start site of SelP in human Hep G2 hepatocarcinoma cells was identified at position 70. Sequence analysis of the 5 region of the SelP gene found the sequence CTTCCAGGAAG at nucleotides 742 to 732, which is an interferon γ (IFN-γ)-responsive element. The effects of cytokines 27

43 treatment on SelP promoter were investigated using luciferase reporter gene. HepG2 cells were treated with cytokines for 20 h after transfection of the 1.8-kb SelP promoter construct. IL-1β (interleukin 1β), IFN-γ, and TNF-α (tumor necrosis factor α) significantly repressed luciferase activity (46%, 40%, 55% of control respectively, P<0.05). It has also been reported that an anti-inflammatory cytokine TGF-β1 (Transforming growth factor β1) inhibited SelP secretion when HepG2 cells were incubated for 48h with TGF-β1, and the inhibition effect was dose dependent (Mostert et al., 1999). Treatment of TGF-β1 also reduced SelP mrna levels. A maximum reduction of SelP mrna levels to 40% was observed after treatment with 100 pm TGF-β1 for 24h. Reporter gene assays under control of the human SelP promoter were conducted to characterize the regulation mechanism of TGF-β1 regulation on SelP expression. TGF-β1 treatment of HepG2 cells resulted in a marked dose-dependent decrease of luciferase activity (Mostert et al., 1999), indicating a down-regulation by TGF-β1 on human SelP promoter. TGF-β1-responsive element in the SelP promoter has been identified (Mostert et al., 2001). These data suggest that SelP expression in the human HepG2 cells is inhibited on a transcriptional level. It seems that TGF-β1 induces the transcription and synthesis of one or several factors which inhibit SelP transcription. However, to date Smad proteins are the only identified transcription partners for transduction of TGF-β1 signals (Mostert et al., 1999; Mosterv et al., 2001). 28

44 In summary, IL-1β, IFN-γ, TNF-α, and TGF-β1 inhibits SelP expression in the human HepG2 cells by a down-regulation on SelP promoter, which is a transcriptional control. Feeding selenium deficient diet to weaning rats caused the reduction of SelP mrna levels in liver and kidney (Weiss et al., 1997; Christensen et al., 1995) with no effect on the transcription rate of SelP mrna (Christensen et al., 1995), which indicates a posttranscriptional control. Selenium deficiency may increase the degradation of SelP mrna. Selenium intake may also affect selenocysteine incorporation. When selenium is insufficient, selenocysteine is not available for insertion at UGA codons during translation, and protein synthesis cannot proceed. The secondary structure of SelP mrna may have the regulatory function in this translational level control. SelP is unique in that its mrnas in different mammalian species encode selenocysteine residues and contain two SECIS elements in the 3 UTR region. While it is clear that the 3 UTR of SelP mrna is important for the incorporation of selenocysteine into SelP, the elements of the mrna involved have not been thoroughly characterized. It is still not clear how the two SECIS elements in the 3 UTR mediate selenocysteine insertion at UGA. What is the relationship between the two SECIS elements and UGA codons? Whether the two SECIS elements serve different roles, or function for specific UGA codons, are to be investigated. The answers of these questions will surely shed light on the mechanism by which the 3 UTR mediates selenocysteine incorporation by UGA codons. 29

45 Besides dietary selenium intake and cytokines, estrogen may also affect SelP synthesis. SelP promoter contains 2 ERE half-sites, 2 AP1 binding sites, and 1 Sp1 binding site (Al-Taie et al., 2002; Dreher et al., 1997), indicating a possibility that SelP gene transcription is regulated by estrogen Function of SelP Although the physiological function of selenoprotein P has been unclear since its identification in early 1970s, two possible functions have been suggested. The first function is to act as a selenium transport protein. Because of its high selenium content and extracellular location, SelP has been suggested to play a role in selenium transport. SelP has been identified as a bioavailability source of selenium for cultured cells (Yan and Barrett, 1998; Saito and Takahashi, 2002). Adding SelP to SelP-depleted human serum or a serum-free medium was the most effective way to recover the cellular glutathione peroxidase (cgpx) activity, compared with other selenium-containing proteins (Saito and Takahashi, 2002). A follow-up study conducted by this group demonstrated that the C-terminal fragment, but not the N-terminal fragment, of SelP supplied selenium to cultured cells (Saito et al., 2004).. In selenium-deficient rats, selenium was taken from the liver to peripheral tissues, such as testis, kidney, and spleen (Motsenbocker and Tappel, 1982), supposedly in the form of selenoprotein P (Motsenbocker and Tappel, 1984). Ducros et al. (1994) 30

46 demonstrated that albumin (or similar protein) was the main plasma acceptor of selenium during the first 4 h post ingestion, but by 8 h, selenium was primarily incorporated into SelP after processing by the liver. Using a SelP-knockout mouse model, researchers have found that selenium concentration in tissues, such as testis, brain, and kidney was depressed in SelP deletion male mice, and glutathione peroxidase activity also dropped markedly in these tissues (Hill et al., 2003; Schomburg et al., 2003). These findings strongly suggest that SelP has the function to transport selenium from liver to other tissues, especially testis and brain. Selenoproteins play an important role in brain function (Nakayama et al., 2007; Schweizer et al., 2004). It has been shown that SelP gene knockout mice displayed a neurological phenotype with a movement disorder and epileptic seizure associated with markedly decreased selenium content in brain (Schomburg et al. 2003; Schweizer et al., 2004; Hill et al., 2003; Hill et al., 2004). A conditional knockout mice model for hepatic SelP gene only has been developed, in which mice are unable to synthesize and secrete hepatic SelP into circulation due to lack of the trna [Ser]Sec gene in the liver (Schweizer et al., 2005). Researchers found that brain function was not impaired in this model, and brain selenium levels were unaffected, though there was a sharp decrease of SelP in the plasma. These results suggest that brain can use local SelP expression for appropriate functioning in the absence of hepatic SelP. The brain selenium level appears to be independent of plasma selenium level as long as local SelP expression in brain is 31

47 preserved, indicating a novel role for SelP as an extracellular selenium storage and recycling protein. Despite the above evidence, however, the hypothesis that SelP functions as a selenium transport protein has been challenged by the fact that selenium is covalently bound in SelP. Thus, the transport would require breakdown of the protein for release of its selenium, which is not an efficient way for selenium transport. Such challenge has partially been resolved by identification of selenocysteine β-lyase, an enzyme that specifically liberates selenium from selenocysteine (Mihara et al., 2000). The second proposed biological function of SelP is to act as an antioxidant. This function is supported by the observation that after administration of selenium to selenium-deficient rats, there was an increase in SelP level, but not GPx level, and the rats showed no signs of lipid peroxidation after diquat-induced liver lesions (Burk et al., 1995). These data suggests that SelP may protect the cells against lipid peroxidation and thereby prevent liver necrosis. It has also been shown that in human plasma, SelP contributed to the destruction of peroxynitrite, an important factor in inflammatory toxicity (Arteel et al., 1998). 2.3 Selenium metabolism Selenium metabolism in the body is complex, and is largely dependent on the chemical forms of selenium. Selenium is present in foods in both inorganic and organic 32

48 forms. Selenocysteine and selenomethionine are the most typical organic forms, while selenite (SeO 3 2- ) and selenate (SeO 4 2- ) are common inorganic selenium compounds Absorption Selenium is freely absorbed in the gastrointestinal tract. Using ligated intestinal segments in rats, Whanger et al. (1976) found that selenite and selenomethionine were absorbed from all segments of the small intestine, with slightly higher absorption from the duodenum than from the jejunum or ileum. Virtually no absorption of selenite or selenomethionine occurred in the stomach (Whanger et al., 1976). Generally, selenium is well absorbed and, under normal feeding conditions, absorption is not a limiting factor to bioavailability (Mutanen, 1986). In the rat, selenium deficiency had no effect on absorption of any selenocompound in any intestinal segment (Vendeland et al., 1992). Actually, the absorption rate does not vary even with a ten fold difference in selenium intake, suggesting no gastrointestinal homoeostatic control of selenium absorption (Diplock, 1987). Thomson and Stewart (1973) found that rats absorbed 95-98% of an oral dose of selenomethionine and 91-93% of an oral dose of selenite. Another study done by the same research group (Thomson and Stewart, 1974) indicated that selenite appears to be less well absorbed by human than by rats. The absorption was 70%, 64%, and 44% in three young women given oral doses of 75 Se-selenite. More recently balance studies (Holben et al., 2002) demonstrated that the 33

49 apparent selenium absorption ranged from 71% to 76% in adolescent girls consumed a diet with 100 µg selenium/day. The manner in which selenium is absorbed by the intestine is dependent on its chemical form. The organic form of selenium, selenomethionine is better absorbed than the inorganic form, selenite (Swanson et al., 1991). When different forms of selenium were added to infant formulas, Lonnerdal et al. (1992) found that the uptake and retention of selenium by suckling rat pups was most rapid for selenomethionine (70%), followed by selenate (51%) and selenite (29%). The low uptake of selenite by pups given the selenite supplement was found to be due to a higher retention of selenite in the stomach and small intestinal wall. It is postulated that low diffusibility of selenite during digestion (Shen et al., 1997) and greater binding to brush border membrane vesicles (Vendeland et al., 1994) cause lower absorption of selenite. A human study (Van Dael et al., 2001) confirmed that selenate is better absorbed than selenite. In this study, ten healthy men were fed a milk-based formula labeled with 40 µg selenium as SeO 3 (selenite) or SeO 4 (selenate) on two consecutive days. Selenium absorption for selenate was significant greater than that of selenite (91.3% vs. 50.2%, P<0.05). Using an in vitro model composed of Caco-2 monolayers, the researchers further demonstrated that selenate and selenomethionine were the most efficiently transported. The apparent permeability coefficients (Papp) were in the following rank order: selenite selenocystine<selenomethionine<selenate (Leblondel et al., 2001). 34

50 In gut sac experiments (McConnell and Cho, 1965), selenomethionine was transported against a concentration gradient by an energy-dependent process and the process was inhibited by methionine. These results indicated that the seleno- amino acid uses the same pathway as methionine. Similarly, selenate is actively absorbed and may share the same absorption pathway as sulfur. The active transport of selenate is dependent on the sodium gradient maintained by Na + K + -ATPase (Arduser et al., 1985). However, an in vitro study did not support the active transport of selenate (Leblondel et al., 2001). The authors observed unspecific action of thiosulfate in monolayers of Caco-2 cells, which inhibited both selenate and selenite. Selenite is not actively transported and does not compete with the corresponding sulfur analog for transportation (Sunde and Hoekstra, 1980). Selenite is rapidly absorbed by simple diffusion, with labeled selenite appearing in the plasma within 30 minutes (Patterson et al., 1989). But results of the selenocysteine transport are inconclusive: a passive absorption in everted intestinal sacs (McConnel and Chao, 1965) or a competitive transport with cysteine in brush-border membranes (Wollffram S et al., 1989) have been both reported. Selenium status of an animal has little effect on intestinal absorption of selenite. The status of other nutrients and the nutrient composition of the diet, however, may affect selenium absorption. Dietary magnesium deficiency has been shown to decrease selenium absorption after 7 weeks feeding in Wistar rats (Jimenez et al., 1997), while a high protein diet enhances selenium absorption (Daniels, 1996). Yu and Beynen (2001) 35

51 demonstrated that the effects of dietary copper concentration on selenium absorption depends on the amount of selenium in the diet. Although the effects of copper on selenium absorption were not seen in rats fed high or low selenium diet, an increase in dietary copper concentrations elevated selenium concentrations in the liver and kidney of rats fed diets with the normal level of selenium. Vitamin A, C, E, and reduced glutathione (GSH) in the diet appeared to promote absorption of selenium in the intestinal lumen (Vendeland et al., 1992; Robinson et al., 1985a; Anundi et al., 1984). Vitamin C, however, could reduce selenite to elemental Se, which is not absorbed. In young women, when selenite and 1 gram ascorbic acid were taken together, the absorption of Se was reduced almost to zero. (Robinson et al., 1985a). The presence of heavy metals and phytates inhibits selenium absorption by chelation and precipitation (Burk and Hill, 1993; Forbes and Erdman, 1983). Selenium absorption varies among different food sources. By labeling food with stable isotope 77 Se or 82 Se and measuring the plasma appearance of the isotopes, researchers demonstrated that selenium absorption was significantly higher from wheat (81%) and garlic (78%) than from fish (56%) (Fox et al., 2005). Feeding chickens 1000 ppm Pb for 3 weeks prior to measurement of absorption of 75 Se-selenite caused a reduction in transfer of selenite into the body by increasing retention in intestinal tissue (Mykkanen and Humaloja., 1984). Increasing the dietary selenium alleviated the inhibition. It is proposed that long-term exposure to Pb may 36

52 cause synthesis of proteins that can bind selenium, and when selenium intake is increased, these binding sites may become saturated Transport Once absorbed, selenium is transported in plasma bound to proteins. The proteins that bind selenium vary with the species. Albumin has been suggested to be associated with plasma selenium transport in mice (Sandholm, 1974), rats (Suzuki and Itoh, 1997; Shiobara and Suzuki, 1998a), and humans (Ducros et al., 1994). Selenoprotein P (SelP) also has been suggested to play a role in selenium transport among the tissues (Burk and Hill, 2005). Studies indicate that it is responsible for the transport of selenium from liver to other tissues, such as kidney, brain, testis, and heart (Hill et al., 2003; Schomburg et al., 2003). Transport role of SelP has been shown to be for only the hepatic SelP (Schweizer et al., 2005). The details of SelP function in circulating transport are included in section 2.2.2, i.e., Selenoprotein P. Erythrocytes appear to be essential for initial uptake and reduction of selenite, a process which is dependent upon the reduced glutathione concentrations in the erythrocytes. Suzuki et al. (1998) demonstrated that selenite injected intravenously into rats was taken up selectively by red blood cells (RBCs) through the anion-exchange carrier, the major integral membrane protein of RBCs. GSH deprivation inhibited the uptake of selenite by RBCs. In RBCs, selenite is reduced to selenide by thiol groups such 37

53 as GSH, and then released back into plasma and bound selectively to albumin (Shiobara and Suzuki, 1998a). Selenide bound to albumin is transported to the liver and taken up rapidly by the liver (Suzuki et al., 1999). In contrast, selenate is taken up directly by the liver without being reduced in the RBCs (Suzuki and Ogra, 2002) Tissue distribution Selenium is transported into organs and tissues in a protein-bound form. The main factors that affect the concentration of selenium in organs and tissues include the particular tissue considered, the amount and chemical form of selenium in the diet, and the hormone status. The tissue distribution of selenium has been studied by supplementation of rats with radioactive isotope ( 75 Se)- or humans with stable isotopes ( 74 Se or 82 Se) -labeled organic or inorganic selenium. After supplementation, liver and kidney of rats were consistently higher in 75 Se concentrations than any other tissue (Beilstein and Whanger, 1985; Janghorbani et al., 1990a). In humans, the highest concentrations of selenium also occur in kidney and liver (Ingro et al., 1990). Because these organs contain a relatively small proportion of total body selenium (8% for liver and 4% for kidney) (Oster et al., 1988), they are thought to be involved in metabolism and excretion of selenium rather than only as storage organs. The high concentration of selenium in kidney may be related to the kidney being the primary organ of excretion. Plasma and erythrocytes contain an 38

54 estimated 3.4% and 4.3% of total body selenium, respectively. The lowest selenium content occurs in the brain, lung, and muscles. Muscle tissue, however, accounts for the highest proportion of total body selenium (nearly 40%) because of its relatively large mass (Oster et al., 1988). In a study carried out in Poland (Zachara et al., 2001), selenium concentrations in tissue samples taken at autopsy from 46 healthy individuals killed in accidents and from 75 corpses of victims of various diseases were in the following decreasing order: Kidney (469 ng Se/mg wet tissue) >liver>spleen>pancreas>heart>brain>lung>bone>skeletal muscle (51 ng Se/mg wet tissue). Another study determined the selenium concentration in the renal cortex, liver, and hair among people who died suddenly (Hac et al., 2003). No correlation of selenium concentration among the renal cortex, liver, and hair was observed, and selenium concentration in the investigated tissues was not age-dependent. In humans and animals, blood selenium levels are influenced by dietary intake. The lowest blood selenium values reported in humans are 0.01 to 0.03 ppm for people suffering from Keshan disease in People s Republic of China (Keshan Disease Research Group, 1979a). Selenium distribution into different organs and tissues is also dependent on chemical forms of selenium. Selenium is deposited in tissues at higher concentrations when present in diet as organic rather than as inorganic selenium. Whanger and Butler (1988b) found that selenium accumulated in all tissues in rats at higher levels when 39

55 selenomethionine was fed than when selenite was given, especially in muscle and brain. Incorporation of selenium into the proteins of liver and muscle was investigated after oral administration of normal and large amounts of 75 Se-labled selenite and selenomethionine in selenium depleted male rats (Behne et al., 1991). Greater tissue selenium contents after administration of selenomethionine were mainly due to nonspecific incorporation into a large number of proteins. In monkeys, fed selenomethionine for 11 months, there was significantly more selenium in liver, muscle and hair than in monkeys fed selenite (Butler et al., 1990). Thomson et al. (1993) compared the effects of long-term supplementation of selenate and selenomethionine on blood selenium levels and glutathione peroxidase activities in New Zealand women aged years. They found that selenomethionine raised blood selenium levels and platelet glutathione peroxidase to a much greater extent than selenate. Similarly, a study on selenium-deficient men in China (Xia et al., 1992) suggested that taking selenomethionine increased selenium levels in both plasma and erythrocytes at a significantly faster rate than taking selenate. Besides affecting the total selenium concentration, chemical form of selenium in the diet also affects the binding form of selenium in the tissues. Selenocysteine was the predominant form of the element in rat tissues after administration of selenite, whereas 70% of the selenium was present in the form of selenomethionine in muscle after selenomethionine was given (Beilstein and Whanger, 1988b). In women taking selenomethionine, the majority of the selenium was with hemoglobin (Hb) but was about 40

56 equally distributed between Hb and glutathione peroxidase after taking selenate (Butler et al., 1991). There is evidence of gender related differences in the tissue distribution of selenium. Results of several studies suggest that the gender differences of selenium are related to the reproductive hormones. The relationships between selenium and reproductive hormones will be discussed in section 2.8, i.e., Selenium and reproductive hormones Metabolic pathways Selenium metabolism in the body is influenced by chemical forms of selenium. Selenide is the key intermediate for both organic and inorganic selenium metabolism (Figure 2.2). The two inorganic types of selenium are reduced to selenide by GSH, while the two organic types of selenium are cleaved at carbon-se bonds by β-lyase. Selenide is either used for the synthesis of selenoproteins and Se-binding proteins or excreted as methylated metabolites (Suzuki and Ogra, 2002; Sunde, 1997a). The general metabolic pathways of selenium are shown in Figure 2.2. The selenate to selenite conversion (path 1) is believed to involve adenosine phosphoselenate (APSe) or phosphoadenosine phosphoselenate (PAPSe) activated intermediates (Axley and Stadtman, 1989). Selenite is reduced nonenzymically (path 2) by glutathione (GSH) to zero oxidation state Se 0 in the form of seleno-diglutathione (GS-Se-SG), which is further reduced to selenopersulfide (GSSeH) by glutathione reductase in the presence of 41

57 General body protein Selenoproteins Selenite (1) Selenate (11) (19) (2) Selenomethionine Sec-t RNA SER UCA GS-Se-SG (12) (18) Ser-tRNA SER UCA (17) Ser (3) Selenocysteine Selenophosphate GS-SeH (13) β-lyase (16) (4) Se 0 (14) (10) (5) CH 3 SeH (15) (6) (CH 3 ) 2 Se (8) (7) H 2 Se breath (CH 3 ) 3 Se + (9) urine Se-binding Proteins Figure 2.2 Selenium metabolic pathways. (Adapted from Sunde 1997a, Medina et al. 2001, and Suzuki and Ogra 2002) 42

58 nicotinamide adenine dinucleotide phosphate (reduced form, NADPH) (path 3). Under anaerobic conditions, GSSeH can be converted to an acid-volatile selenide (path 4), either by glutathione reductase in the presence of NADPH or by nonenzymic reduction by excess GSH (Hsieh and Ganther, 1975). Selenide can be methylated using S-adenosylmethionine by either microsomal or cytosolic methyltransferases in liver to form methaneselenol (path 5) and dimethylselenide (path 6) or trimethylselenonium ion (path 7) (Heieh and Ganther, 1977), and then be eliminated via breath (path 8) or urine (path 9). In addition, selenide can bind nonenzymatically to selenium-binding proteins (path 10), which is proposed to account for acute toxicity of selenium. Selenomethionine is incorporated into proteins (path 11) or converted to selenocysteine (path 12), or is metabolized to methaneseleno (path 15) through a transamination pathway (Steele and Benevenga, 1979). Vitamin B 6 status affects selenomethionine metabolism, because B6 dependent enzymes are involved in the metabolic activation of selenomethionine (Soda et al., 1999). Any selenomethionine that is not immediately metabolized is incorporated into organs with a high rate of protein synthesis (path 11), such as the skeletal muscles, erythrocytes, pancreas, liver, kidney, stomach and the gastrointestinal mucosa. The conversion of selenomethionine to selenocysteine (path 12) involves two steps. Selenomethionine is readily metabolized to [Se]-adenosyl methionine (SeAM) (Markham et al., 1980), an excellent methyl donor in mammalian systems. SeAH is a ready substrate for cystathionine β-synthase and 43

59 cystathionine γ-lyase and is thus converted to selenocysteine in mammalian tissues. Selenocysteine is degraded to elemental selenium by selenocysteine lyase (path 13). The element selenium is reduced nonenzymatically to selenide (path 14) by glutathione or other thiols (Esaki et al., 1982) Selenium incorporation into selenoproteins Selenoproteins are proteins that contain selenium in the form of selenocysteine (Burk and Hill, 1993). In prokaryotes and eukaryotes, it has been established that selenocysteine is the 21st amino acid in ribosome-mediated protein synthesis (Bock et al., 1991). Selenocysteine incorporation into protein is a cotranslational event directed by the UGA codon. Four unique gene products are required for selenocysteine incorporation: sela, selb and seld are proteins, and selc is a unique trna (Stadtman, 1996). The initial reactants of this process are selenide, serine, and ATP. Selenophosphate synthetase (seld gene product) catalyzes the formation of selenophosphate from selenide and ATP (Figure 2.2, path 16). Selenophosphate is a key selenium donor compound for prokaryotes as well as eukaryotes, and is the substrate for synthesis of selenocysteinyl-trna (Figure 2.2, path 18). Serine, the source of the carbon skeleton, is esterified to the 3 terminal adenosine of the trna SER UCA (selc gene product) by the usual cellular seryl-trna synthases to form the corresponding seryl-trna (Figure 44

60 2.2, path 17). Seryl-tRNA reacts with selenocysteine synthase (sela gene product), a pyridoxal phosphate-dependent enzyme to form aminoacrylyl-trna Sec, followed by a C2-C3 addition of selenium as selenophosphate, to form the selenocysteine moiety still esterified to the trna Sec (path 18) (Forchhammer et al 1991). Thus, sela, selc, and seld gene products are involved in the synthesis of Sec-t RNA SER UCA, the selenocysteine donor. The cotranslational selenocysteine incorporation (Figure 2.2, path 19) requires the fourth unique protein, a 68-kDa selb gene product, selenocysteine-specific elongation factor (SEL-B). SEL-B protein is very specific for Sec-t RNA SER UCA and for the stem-loop structure (SECIS element) on the selenoprotein mrnas. It transports the Sec-t RNA SER UCA to the ribosome, thus increasing the Sec-t RNA SER UCA concentration on the mrnas. Recognition of UGA as a selenocysteine codon rather than a stop codon requires the fifth element. In bacteria, it is a psecis (prokaryotic selenocysteine insertion sequence), a stem-loop structure in the RNA immediately downstream from the UGA (Heider et al., 1992). In mammalian, however, it is a esecis element, a stemloop structure in the 3 UTR (untranslated region) of the mrna (Berry et al., 1991). It is proposed that the selenoprotein mrna, SEL-B, Sel-C, and guanosine triphosphate (GTP) assemble in a quaternary complex on the ribsome for selenocysteine cotranslational insertion (Ringquist et al., 1994). Catalyzed by peptidyltransferase, a peptide bond between selenocysteine and the nascent polypeptide is then formed, thus finishing the insertion of selenocysteine into selenoproteins. 45

61 The number of newly identified selenoproteins has a sharp increase during the last decade. To date, at least 35 human selenoproteins have been identified (Surai et al., 2006; Gromer et al., 2005) Excretion Selenium may be eliminated from the body by all three major elimination pathways: the urine, the feces, and the expired air. The absolute level of intake and the chemical form in which the selenium is absorbed are factors that affect the excretion of selenium. Urinary excretion is the primary route of excretion. Generally, 50-75% of total ingested selenium is excreted in the urine (Robinson and Thomson, 1983; Alaejos and Romero, 1993; Holben et al., 2002). Fecal losses are usually not large and are independent of the dose (Thomson and Robinson, 1986). The elimination of volatile selenium (H 2 Se) by expired air becomes significant only with toxic intakes (Robinson and Thomson,1983). Methylation has been considered to be the reaction leading to excretion of selenium (Figure 2.2, path 5-9, 15). Excessive selenium is excreted into the urine and breath in the form of mono-, di- and trimethylated Se. Both the doses and the selenium status of the animals influence the form and amount of urinary selenium excretion. Monomethylselenol (MMSe) is the only metabolite detected in urine from rats fed deficient to low-toxic doses of selenium diet (Suzuki et al., 1995; Suzuki, 1996; Itoh and Suzuki, 1997). After MMSe reaches a plateau in the urine, the trimethylselenonium ion 46

62 (TMSe, (CH3) 3 Se+) increases in the urine. When rats were injected with selenite (0, 0.1, 0.3, 0.5, and 1.0 mg Se/kg body weight) into the tail vein, three types of Se-metabolites-MMSe, TMSe, and inorganic Se, were detected in urine (Itoh and Suzuki, 1997). Urinary MMSe increased rapidly at first and was slowly followed by linear dose-dependent excretion of TMSe. When selenite was administered at a dose of 250 µg/kg body weight, MMSe was the major urinary Se-metabolite (Janghorbani et al., 1990a; Suzuki et al., 1995; Suzuki, 1996). Administration of additional selenium in the water (4 µg Se/mL) increased TMSe from 2% of urinary selenium to 35-45% within 3 days of selenium supplementation (Janghorbani et al., 1990a). In humans, trimethylselenonium ion is also a minor component of urine. Janghorbani and Young (1987) reported that 7% to 17% of urinary selenium is preset as trimethylselenonium ion. But in New Zealand women supplemented with 200 µg/day selenium as selenomethionine or selenate for 32 weeks, TMSe only accounted for 1% of the total selenium in urine (Robinson et al., 1997). Based on the above findings, it is proposed that methylated forms of Se are excreted from urine in the following order: excretion of MMSe precedes at a low dose and upon reaching the limit, TMSe and inorganic Se begin to be excreted more than MMSe (Itoh and Suzuki,1997). Although the monomethylated selenium excreted into the urine has been believed to be simply monomethylselenol (MMSe, CH 3 SeH), more recent studies by Suzuki s group from Japan using mass spectrometry identified the monomethylated selenium as a new 47

63 selenosugar, 1β-methylseleno-N-acetyl-D-galactosamine (Kobayashi et al., 2002). A metabolic pathway from the glutathione-s-conjugated selenosugar to the methylated one has been proposed. The authors remarked that urinary monomethylated selenium (the selenosugar) is a biomarker indicating that selenium intake has increased within the required to low-toxic range, while the appearance of trimethylated selenium suggests a toxic level of selenium intake. Urinary excretion is the principal way by which selenium homeostasis is achieved at physiological selenium levels. When selenium intake increases, urinary excretion tends to increase in order to maintain selenium homeostasis. Daily urinary selenium excretion is positively related to dietary selenium in populations with different dietary selenium consumption all over the world (Alaejos and Romero, 1993). In China, urine selenium concentrations of residents in areas of chronic selenosis were 400 times higher than those in low-se areas where Keshan disease occurred (Yang et al., 1983; Yang et al., 1989). The 24-h urinary selenium concentrations in US populations are µg/l, while in New Zealand, an area low in selenium, the urinary selenium concentrations are less than 20 µg/l (Alaejos and Romero, 1993). Several weeks of supplementation with high-se bread, however, increased New Zealanders renal plasma clearances of selenium (CSe). When supplementation ceased, CSe returned to the basal range within a few days (Robinson et al., 1985b). The kidney of the NZ residents thus exhibited a more 48

64 conservative model than others to adapt to chronic low selenium intakes. Renal regulation is important since there is no gastrointestinal regulation of selenium. Chemical forms of selenium appear to affect urinary excretion. Urinary selenium is lower when organic selenium is supplemented rather than inorganic selenium (Thomson and Stewart,1973; Robinson et al., 1985b; Robinson et al., 1997). After supplementing rats with 75 Se-labeled selenite and selenomethione for 7 days, 12.7% of the selenite was excreted in urine, while only 4.2% of the selenomethionine was found in the urine (Thomson and Stewart, 1973). In New Zealand women supplemented with 200 µg Se/day as selenomethionine or selenate, the selenate group excreted more selenium than the selenomethionine group, with 123 vs. 66 µg/d excreted respectively at week 2 (Robinson et al., 1997). Robinson et al. (1985b) suggested that the different urinary excretion between inorganic and organic selenium is due to the different reabsorption rate from the glomerular filtrate, i.e. inorganic selenium is less likely to be reabsorbed from glomerular filtrate than selenomethionine. For inorganic forms, selenium from selenate is more readily eliminated in the urine. Hirooka and Galambos (1966) compared the disposition of selenite and selenate in rats (1.4 mg Se/Kg, ip) and found that 51% of the 75 Se from selenate was excreted in urine while only 37% of the 75 Se from selenite was eliminated during the 24-h period. Though most studies found that chemical forms of selenium affect the urinary excretion of selenium, Suzuki s research group (Shiobara et al., 1998b) reported that in 49

65 rats fed selenomethionine or selenite, the concentration of urinary selenium was not dependent on the chemical species but on the doses. These authors remarked that the urinary amount of selenium appears to reflect the total amount of selenium absorbed from the diet, thus independent of selenium forms. The seeming conflicting results from the studies of urinary selenium excretion may be due to differences in the protein status of the rats used in the experiments. In Suzuki s studies, the lack of a difference in urinary selenium excretion after selenomethionine and selenite administration may have been related to the methionine status of the rats studied. Rats may have adequate methionine status, which led to less selenomethionine incorporated into general proteins, thus more selenomethionine would be available to be excreted from urine. Besides doses and chemical forms of selenium, some other factors may affect urinary selenium status as well. A Spanish research group (Rodriguez et al., 1995) found that in healthy Canadian people, females excreted higher amount of selenium per kg of body weight in urine than males. They also found that children (<10 years old) had greater daily urinary selenium excretion per kg of body weight than older persons (>10 years old). In pregnant females, urinary excretion of Se was lower than in non-pregnant controls (Swanson et al., 1982). Cancer also plays a role in urinary selenium status. Navarrete et al. (2001) observed a significant increase of urinary selenium excretion in cervical uterine cancer patients, and the excretion was highest for patients in the intermediated stages of the disease. 50

66 Fecal excretion of selenium represents the unabsorbed selenium in the gastrointestinal tract and bile sections (Levander and Baumann, 1966; Bopp et al., 1982). In rats, fecal excretion accounted for 10%-15% of the intraperitoneally or intravenous administrated selenium (Burk et al., 1972; Burk et al., 1973; Thomson and Stewart, 1973). Balance studies using oral administered selenium have shown that the fecal excretion ranged from 13%-23% (Cary et al., 1973). In humans, fecal excretion after intravenous administration of selenite was minimal (<1%) (Kuikka and Nordman, 1978). Balance studies in adolescent girls consuming a diet with 100 µg selenium/day demonstrated that the fecal urinary selenium excretion ranged from 22-29% (Holben et al., 2002). However, when selenium is not well absorbed (e.g. after high doses of selenium) from the gastrointestinal tract, the fecal excretion is higher, ranging from 33% to 58% of the dose (Thomson and Stewart, 1974). Selenium excretion through expired air is a minor pathway when selenium intake does not reach the toxic level. Thomson and Stewart (1974) found that less than 0.02% of an orally administered 75 Se-Selenite was excreted via expired air in young women. However, respiratory excretion may reach 50% when persons are exposed accidentally to toxic levels of selenium (Bopp et al., 1982). The respiratory excretion of selenium results in a garlic-like odor of breath. 51

67 2.3.7 Selenium metabolism models Janghorbani s model Using tracer 75 Se or stable 74 Se isotopes, researchers have been able to monitor selenium flux in metabolism, and thus have proposed some kinetic models of selenium metabolism. In order to accurately determine selenium status in humans, Janghorbani s group (Janghorbani et al., 1990b; Janghorbani et al., 1990c; Janghorbani et al., 1991) developed a metabolic model using an in vivo isotope-dilution approach. In this model, they proposed that whole-body selenium can be divided into two distinct metabolic pools (Janghorbani et al., 1990b): 1) Pool 1, the selenite-exchangeable metabolic pool (Se-EMP), which incorporates all forms of selenium derived from inorganic selenite/selenide, including intermediate forms of selenium such as GS-Se-SG, and end products like glutathione peroxidase, selenoprotein P, DMSe and/or TMSe. 2) Pool 2, comprises all selenomethionine-containing proteins. Se-EMP provides all known functionally important selenocompounds, whereas pool 2 has no known function other than it may act as a storage compartment for selenium. Although pool 2 may contribute selenium to pool 1 by degradation of selenomethionine-containing proteins, pool 1 does not contribute selenium to pool 2. Human studies have shown that the pool size (W Se-EMP ) positively correlates with daily selenium intake, and thus may provide a sensitive means for assessment of selenium status (Janghorbani et al., 1990b). 52

68 The quantitative relationship between W Se-EMP and whole body selenium has been investigated in adult rats by administering labeled selenite, 74 SeO 2-3 (Janghorbani et al., 1990c). Plasma or urine specific activity (µg tracer isotope/µg endogenous Se) was initially high but continued to decrease as the labeled selenite was incorporated into the Se-EMP in a time-dependent manner. Based on this observation, a formula using plasma or urine specific activity values to calculate the average size of Se-EMP at the time of interest was then proposed. Both short-term (7 d) and long-term (60 d) experiments have been carried out. The results indicated that the time course of organ specific activities varied among organs. It increased rapidly in plasma, liver and kidney, reaching its peak in 1 week. But in skeletal muscle, the increase of specific activity was slower, reaching its peak at week 5 (Janghorbani et al., 1990c). Further studies demonstrated that muscle selenium was far less exchangeable than selenium in other tissues (Janghorbani et al., 1991). Skeletal muscle contributed 41.5% of the endogenous total selenium, but only 16.7% of the selenite exchangeable pool. In contrast, liver contributed 7.4% and 12.6% of the endogenous total selenium and selenite exchangeable pool, respectively. In rats, W Se-EMP has been established to accurately reflect total body selenium content and the selenium content in different organs (Janghorbani et al., 1991). However, the quantitative relationship between the measured pool size (W Se-EMP ) and total body or liver selenium is to be determined in humans. 53

69 Zech s model An additional model has been developed to describe the pharmacokinetics of both inorganic (selenite) and organic selenium (selenomethionine) metabolism in humans (Patterson et al., 1989; Swanson et al., 1991; Patterson and Zech, 1992). The model was based on the appearance and disappearance of the stable isotope tracer, 74 Se, in plasma, urine, and feces in adults who received a single oral 200 µg dose of 74 Se as selenite or selenomethionine. A major assumption of the model is that subjects are in steady state. The model includes the following compartments in human body: gastrointestinal tract, enterocyte, four kinetically distinct plasma components, liver-pancreas, and two tissue components (Figure 2.3). According to this model (Patterson et al., 1989; Swanson et al., 1991; Patterson and Zech, 1992), labeled seleniumis absorbed along a chain of three small intestine compartments, G1-G3. Unabsorbed selenium passes through large intestine (LI) and appears in the feces. Material is absorbed into enterocytes (ENT), a pool that is turned over rapidly. The absorbed material leaves the ENT by two pathways. The first pathway is from ENT to the first plasma component (P1) and to liver and pancreas (L/P). This pathway represents the earliest appearance of material in the peripheral circulation. The second pathway is from ENT, passing through HPL (hepatopancretic subsystem or lymphatic system), to the second plasma component (P2). The tracer in P1 and P2 can be excreted in the urine or it can move into L/P compartments. Material leaves L/P by two pathways. The first 54

70 Figure 2.3 Human selenite/selenomethionine metabolism model. G1, G2, and G3, three gut compartments, probably small intestine; ENT, enterocytes (intestine cells); HPL, compartment in hepatopancreatic subsystem or lymphatic system; L/P, liver and pancreas; LI, large intestine; T1, T2, T3, T4, peripheral tissues, eg, skeletal muscle, bone, kidney. The heavy solid lines indicate the major modifications of selenomethionine model to the selenite model (Swanson et al. 1991). 55

71 pathway is enterohepatic recirculation. Material passes through the liver bile and pancreatic secretions compartment (BILE) and returns back to the gut. The second pathway is to the third plasma component (P3). P3 may represent proteins or selenoenzymes secreted or excreted by the liver or pancreas. From P3, the material can be excreted into the urine or can move into TISSUE, a large pool that is slowly turned over. The TISSUE pool includes peripheral tissues such as kidney, muscle, and bone and has two compartments that exchange with each other. Material in P3 passes through TISSUE pool and reaches P4, the fourth component of the plasma pool. Thus, P4 represents material emerging from the peripheral tissues and it probably contains the final metabolic products moving through the plasma before clearance into the urine. Differences within the metabolic model for selenomethionine (Swanson et al., 1991) and selenite (Patterson et al., 1989; Patterson and Zech, 1992) are listed as following (indicated by heavy solid lines in Figure 2.3): (1) Selenomethionine is absorbed more completely than that of selenite, (2) In the selenomethionine model, there is a path to L/P from ENT, (3) There is a second distinct peripheral tissue pool for the selenomethionine model, and (4) In selenomethionine mode, selenium in the P4 compartment is recycled back to the liver, whereas all of the selenite-derived selenium is excreted into the urine. 56

72 Schweizer s model The selenium-transport role of selenoprotein P (SelP) has been confirmed by the SelP gene knockout mouse model (Hill et al., 2003; Schomburg et al., 2003). SelP could also serve as a local selenium storage and recycling protein to maintain brain selenium levels (Schweizer et al., 2005). A model of selenium transport via SelP and its uptake mechanism by the brain and other tissues has been proposed (Figure 2.4, Schweizer et al., 2005; Richardson, 2005). This model states that dietary selenium is absorbed from the gastrointestinal tract (GI) tract, and is taken up by the liver to be used for the synthesis of SelP. SelP is secreted into plasma and transports selenium to other tissues. SelP transfers selenium to extrahepatic tissues by binding to specific membrane receptors (Wilson and Tappel, 1993). SelP may release selenium to endothelial cells of the blood-brain barrier via a receptor-mediated mechanism, but this needs to be validated experimentally. Selenocysteine β-lyase (SCL) (Miahara et al., 2000) catalyzes the release of selenium from SelP. It is not known whether the SelP and its receptors can be recycled to the cell surface for re-utilization after selenium is released from SelP. After it passes the blood brain barrier, SelP is internalized in brain cells and releases selenium via the selenocysteine β-lyase (SCL). De novo synthesis of SelP helps the brain maintain Se levels during Se-deficient status (Schweizer et al., 2005). 57

73 Figure 2.4 Hypothetical model of selenium transport via SelP and its uptake mechanism by the brain and other tissues (Richardson, 2005; Schweizer et al., 2005). Dietary selenium is taken up by the liver for the synthesis of SelP. SelP transfers selenium to extrahepatic tissues by binding to specific membrane receptors. SelP may release selenium to endothelial cells of the blood brain barrier via a receptor-mediated mechanism. After it passes the blood brain barrier, SelP is internalized in brain cells and releases selenium via selenocysteine lyase (SCL). 58

74 In summary, these whole-body metabolic models of selenium provide very useful information on selenium flux among various compartments. It can be used to estimate parameters, to check assumptions, and to make predictions. However, the effects of other metabolic components (e.g. hormones, immune factors) on the selenium flux among various compartments are to be determined. More work needs to be done to better characterize the nature of different components and make these metabolic models more appropriate to represent the whole-body selenium metabolism situation. 2.4 Requirements of selenium Since the relationship between selenium and Keshan disease, an endemic fatal cardiomyopathy, was reported in 1970s (Keshan Disease Research Group, 1974, 1979a), evidence for its importance to human health has been rapidly accumulating. It has been demonstrated that selenium is associated with a number of health outcomes, including cardiovascular disease, cancer, immune function, viral infection, reproduction, and inflammatory conditions (Surai, 2006). Thus, there has been much interest to determine the requirements of selenium for the prevention of chronic diseases. In 1980, the 9 th edition of the Recommended Dietary Allowances (RDA) established an estimated safe and adequate daily selenium intake for adults of μg/day (National Research Council, 1980). This recommendation was based on data extrapolation from animal experiments because of limited data available from human 59

75 studies at that time. These ranges of intake were believed to be adequate to prevent overt signs of nutritional selenium deficiency. After more human studies were conducted, the National Research Council published the first recommended dietary allowance (RDA) for selenium in 1989 (National Research Council, 1989). The recommendations for adult men and women were 70μg/day and 55 μg/day, respectively. This level of dietary intake was obtained by using the dietary selenium intake needed to maximize the activity of GPx activity in Chinese people living in Keshan disease area, and by taking into account body weight and safety factors. The Institute of Medicine (USA) analyzed data from two intervention studies to determine the Estimated Average Requirement (EAR) of Americans. EAR is defined as a nutrient intake value that is estimated to meet the requirement of half the healthy individuals in a life stage and gender group (Institute of Medicine, 2000). The Chinese intervention study (Yang et al., 1987) suggests that a selenium intake of 41 μg/day was sufficient to reach a plateau of plasma GPx activity. It would be 52 μg/day for American males after adjusting for body weight. A New Zealand study suggested that an EAR for Americans was 38 μg/day (Duffield et al., 1999). The average of the two studies, 45 μg/day, has been set as the EAR. The RDA is defined as equal to the EAR plus twice the CV, with CV being 10% of EAR. Thus, the RDA for selenium is 120% of EAR. Rounding the calculated RDA to the nearest 5 μg, the RDA of 55 μg/day was obtained for adults (19-50 years) including men and women (Institute of Medicine, 2000). During 60

76 pregnancy (14-50 years), the EAR is increased by 4 μg/day to meet the fetus needs, thus the EAR for pregnancy is 49 μg/day, and the RDA is 60 μg/day. RDA for lactation (14-50 years) has been set as 70 μg/day to account for the selenium content of human milk. The recommendations for selenium intake at other life stages are as follows (Institute of Medicine, 2000): Infants (0-6 months) 15 μg/day Infants (7-12 months) 20 μg/day Children (1-3 years) 20 μg/day Children (4-8 years) 30 μg/day Adolescents (9-13 years) Adolescents (14-18 years) 40 μg/day 55 μg/day Adults (51-70 years) 55 μg/day Adults (>70 years) 55 μg/day It is difficult to accurately estimate the dietary selenium intake because there are enormous geographic variations in the selenium content of food. The main dietary sources of selenium are meat, poultry, fish and cereals. The wide ranges of selenium content of the principal food groups are listed as follows (Reilly, 1996): Meat, meat product, and eggs ng/g Fish and marine ng/g 61

77 Cereals and cereal products ng/g Dairy products ng/g Fruit and vegetables 1-20 ng/g Because of the wide variations, it is essential to use local food data to assess the dietary selenium intake. Also, the interpretion of selenium intake should be based on the local RDA, as RDA may be different among the countries. For example, the Australian RDA is 85 μg/day for adult males, and 70 μg/day for adult females (Truswell et al., 1990), while the United Kingdom s RDA is 75 μg/day for adult males, and 60μg/day for adult females (Department of Health, 1991). 2.5 Selenium deficiency Selenium deficiency has been public concern among populations in parts of the world notable for their low soil selenium, including China, New Zealand, and Finland Keshan disease Keshan disease (KD) is an endemic cardiomyopathy in China mainly occurs in children aged 2-10 years and women of childbearing age, living in hilly and mountainous areas with heavily eroded soils where soil is low in selenium. It affects a narrow zone forming a long belt running from northeast to southwest China (Tan, 1982). Typical manifestations of KD are fatigue after even mild exercise, cardiac arrhythmia and 62

78 palpitations, loss of appetite, shortness of breath, cardiac enlargement, and congestive heart failure. Pathologic changes include a multifocal myocardial necrosis and fibrosis (Chen et al., 1980). Selenium deficiency was identified as a major risk factor of KD. Selenium deficiency was widespread in habitants of the KD area (Keshan Disease Research Group, 1974). Hair and blood selenium concentration, blood GPx activity, and urinary selenium excretion were significantly lower in people from the endemic area than those from the nonendemic area (Ge and Yang, 1993). Supplementation of selenium (as sodium selenite, mg/week) to children of susceptible age (1-9 years) for 2 years significantly decreased the risk (13.5% vs. 2.2%) (Keshan Disease Research Group, 1979b). Although selenium is effective in the prevention of KD, it may not be the only factor responsible for the development of KD. Because some features of KD could not be explained solely by selenium deficiency. For example, people in New Zealand and Finland also had very low selenium status, but there was no KD cases reported in these areas. Second, the incidence of KD fluctuated seasonally and annually, peaked in winter in the northeast and in summer in southern part of China (Ge and Yang, 1993). These findings suggested that an infectious cofactor was required along with a deficiency in selenium for the development of KD. Scientists in China isolated coxsackie virus from the blood and tissues of KD patients. When inoculated with coxsackie virus, mice fed Se-deficient diet developed 63

79 myocarditis, while mice fed Se-sufficient diet did not (Ge, 1987; Beck et al., 1994). Beck s group (Beck et al., 1994; Beck et al., 1995) further demonstrated that this virus can become virulent in a Se-deficient host. When Se-deficient mice were inoculated with a benign strain of the coxsackie virus (CVB3/0), mutations occurred in the genome to give a cardiovirulent form of the virus that caused myocarditis with similarities to that seen in human beings. When the virus from these mice was inoculated into mice that had adequate selenium, it still induced heart damage, showing the irreversibility of the mutation. Six point mutations of coxsackie virus were identified with the development of virulence, causing myocarditis in the host. Coxsackie virus also caused immune dysfunction in Se-deficient mice (Beck 1999). Using the GPx1-knockout mouse model, it has been demonstrated that GPx is essential for the avoidance of oxidative damage to the genome of coxsackie virus (Beck et al., 1998). As for Se-deficient mice, CVB3/0-infected GPx-1 KO mice developed myocarditis, whereas the wild-type mice did not. Virus isolated from the hearts of GPx-1 KO mice that developed myocarditis also had six mutations. The mechanisms for the increased viral mutations are not currently understood. Beck (2007) proposed that decreased intake of antioxidant nutrients such as selenium causes increased oxidative stress, leading to decreased immune response of the host. Interplay between host immune dysfunction and oxidative stress results in increased viral mutations. 64

80 2.5.2 Kaschin-Beck disease Kaschin-Beck disease (KBD) is another endemic disease mainly detected in children aged 5-13 years. This disease is endemic in areas which overlap KD belt, Siberia, and North Korea (Sokoloff, 1989). KBD is an osteoarthropathy characterized by a chronic disabling degeneration and necrosis of the joints. Clinically, weakness is followed by joint stiffness and pain and often the enlargement of joints and deformity of limbs occur in advanced cases (Ge and Yang 1993). An association of KBD and selenium deficiency has been suggested. KBD occurs in areas where the availability of soil selenium concentration is low. Hair and blood selenium concentrations and blood GPx activity were lower in KBD patients compared to healthy people (Li, 1989; Wang et al., 1987). Oral supplementation of sodium selenite (1.0 mg/week) for 3 years significantly decreased the KBD incidence (39.6% vs. 10.7%), while the control group did not show any change of the incidence rates (42.1% vs. 38.6%) (He et al., 1988). A number of other ecological factors apart from selenium deficiency have been suggested to be associated with KBD, including low iodine status (Moreno-Reyes et al., 1998), calcium deficiency, sulfur deficiency, and the presence of toxic compounds in food (Yang, 1982) Selenium deficiency in New Zealand and Finland Most New Zealand soils contain relatively low concentration of selenium. In the 1950s white muscle disease in cattle and sheep in certain regions was found to be due to 65

81 selenium deficiency. In humans, low selenium status has been shown to be related to the risk of cardiovascular disease (CVD) and respiratory complications of premature newborn (Thompson, 2004). Because of the increase in the importation of Australian wheat and other cereal products, selenium intakes of New Zealanders have increased during the past 20 years (Thomson and Robinson, 1996). However, their selenium status remains low compared with populations of many other counties and may still be considered marginal (de Jong et al., 2001). Finland is a low-se area due to geochemical reasons. Dietary intake of selenium was exceptionally low (25 micrograms/day) before wheat was fortified with selenium (Mutanen, 1985). Epidemiology studies have shown a link between low selenium status and the risk of CVD, rheumatoid arthritis, and cancer (Rayman, 2002). Starting in 1984, sodium selenate has been added to the main fertilizers to increase the selenium content of domestic grain. This leads to the increase of the average selenium intake to 56 micrograms/d (Koivistoinen and Varo, 1987). 2.6 Selenium toxicity Selenium toxicity, or selenosis, has been reported in animals and humans. Animals experience hair loss of the mane and tail/or tail, distemper, retarded growth, swelling and tenderness of the feet, and blind staggers (Spallholz, 1994; Wilber 1980). In humans, chronic selenosis has been reported to cause hair and nail brittleness and loss, 66

82 gastrointestinal disturbances, skin rash, garlic breath odor, fatigue, irritability, and nervous system abnormalities (Yang et al., 1983; CDC, 1984; Yang et al., 1989). To prevent the symptoms of selenosis, the Tolerable Upper Intake Level (UL) has been set for selenium intake. UL is defined as the highest level of daily nutrient intake that is likely to pose no risk of adverse health effects in almost all individuals (Institute of Medicine, 2000). UL for the Americans has been set as follows (Institute of Medicine, 2000): Infants (0-6 months) 45 μg/day Infants (7-12 months) 60 μg/day Children (1-3 years) 90 μg/day Children (4-8 years) 150 μg/day Adolescents (9-13 years) Adolescents (14-18 years) 280 μg/day 400 μg/day Adults (>19 years) 400 μg/day Pregnancy (>14 years) 400 μg/day Lactation (>14 years) 400 μg/day As the benefits and an appropriate dose for selenium supplementation is still controversial, supplementation should be recommended with caution to avoid selenosis. 67

83 2.7 Selenium and hormone-related cancer Cancer is second to cardiovascular disease as the leading cause of death in the United States, accounting for 23.1% of deaths in 2004 (American Cancer Society, 2007). It has been estimated that 35% of cancer deaths in United States are attributed to diet (Doll and Peto, 1981; Doll, 1992). The first suggestion that selenium was associated with cancer risk appeared in geographic studies, in which an inverse relationship between selenium content in the local soil and cancer mortality rate was observed (Shamberger and Frost, 1969; Shamberger and Willis, 1971). Since then, a substantial body of research has established that selenium may have anticarcinogenic activities. These studies include in vitro studies, animal experiments, epidemiological studies, and clinical intervention trials. Several mechanisms have been postulated for selenium s anticancer effects which include: protection against oxidative stress by functioning within GPx and other antioxidative selenoprotein enzymes (e.g., thioredoxin reductase), inhibition of the initiation stage of carcinogenesis by the induction of tumor suppressor gene p53, effects on immune system, production of antitumorigenic selenium metabolites, stimulation of apoptosis, inhibition of angiogenesis, alteration in DNA methylation, and inactivation of protein kinase C (Gerald and Willam, 1998; Rayman, 2005). Many studies have found an association between selenium and the risk of the hormone-related cancers of the breast and prostate. 68

84 2.7.1 Selenium and breast cancer Breast cancer is regarded as the most common malignancy in American women. Breast cancer alone is expected to account for 26% (176,296) of all new cancer cases among women in 2007 (American Cancer Society, 2007). Some possible risk factors of developing breast cancer include a) Inheritance of BRCA1 and BRCA2 genes, b) ionizing radiation exposure, c) total cumulative estrogen exposure, and d) alcohol intake (Grover and Martin, 2002). Evidence from epidemiological, cell, animal, and clinical studies suggests that selenium may play a preventive and/or anticarcinogenic role in the development of breast cancer Epidemiology studies Epidemiology studies have shown that women with breast cancer had lower blood selenium levels compared to those of healthy controls. Lopez-Saez et al. (2003) conducted a case-control study to compare serum concentrations of selenium in women with breast cancer, healthy women, and women with chronic diseases. The mean serum selenium concentration of the breast cancer patients (81 μg/l) was significantly lower than that in the healthy women (96 μg/l) and women with chronic disease (100 μg/l). A number of studies have confirmed that women with breast cancer have lower blood selenium concentrations than those of healthy controls (Charalabopoulos et al., 2006; Huang et al., 1999; Gupta et al., 1994; Chaitchik et al., 1988; Schrauzer et al., 1985; 69

85 McConnell et al., 1980). However, the inverse association between high blood selenium and low breast cancer risk was not observed in some observation studies (Ujiie and Kikuchi, 2002; Ghadirian et al., 2000; Mannisto et al., 2000; Dorgan et al., 1998). In addition, it is difficult to determine whether the lower selenium concentration is the cause or result of the breast cancer in case-control studies. Prospective studies could provide more direct evidence regarding whether low selenium status is the cause of the breast cancer. In a prospective study by Dorgan et al. (1998), no association between serum selenium concentrations and breast cancer risk was observed during a 9.5-year follow-up study. Several other prospective studies also failed to find any association between blood or toenail selenium concentration and breast cancer development (Hunter et al., 1990; van den Brandt et al., 1994; van Noord et al., 1993). In addition, data from the Nutritional Prevention of Cancer Trial (NPCT) suggest that selenium supplementation (200 μg/d) may increase the risk of breast cancer (RR 1.82; 95% CI ) during an average of 7.4-year follow up (Duffield-Lillico et al., 2002). Although the result was not statistically significant (P=0.24), it is worth future attention since only a small number of the subjects have developed breast cancer (11 in selenium group and 6 in placebo group). 70

86 Animal studies Laboratory animal models have been used to examine the inhibitory effect of selenium on many types of cancer, including liver, mammary gland, colon, pancreas, and skin (Kise et al., 1990). 7,12-Dimethylbenz[a]anthracene (DMBA), a polycyclic aromatic hydrocarbon (PAH) has been widely used to induce mammary tumors in rodents (Thompson and Singh, 2000). The inhibitory effect of selenium on DMBA-induced breast tumors in rodents has been reported by a number of studies (Kocdor et al., 2005; El-Bayoumy et al., 2003a; El-Bayoumy et al., 2003b; El-Bayoumy et al., 1995; Ip et al., 1992; Birt et al., 1986, Medina and Shepard, 1981). In a recent rat study (Kocdor et al., 2005), rats were given 20 μg selenomethionine daily via gavage, starting 2 wk before the single oral dose of DMBA administration (12mg) and continued for 1 wk. Rats were sacrificed 120 days after DMBA administration. Malignant tumor frequency was 33% and death occurred in 30% of the DMBA group, while no malignant tumors and deaths occurred in the DMBA Se-treated group. Though data from animal studies cannot be applied directly to humans due to the differences of species, doses, and forms of selenium administered, these data do help to explore the possible protective role of selenium on breast cancer and its underlying mechanisms. 71

87 In vitro studies In vitro studies have been carried out to determine whether selenium can inhibit the growth of breast cancer cells, and explore possible mechanisms of the inhibitory effect. Estrogen can promote the growth of breast cancer by binding to the estrogen receptors (ER). Upon binding the activated ER to the estrogen response elements (ERE) in the promoter region of the target gene, the gene transcription is affected (Hall and McDonnell, 2005). Prolonged estrogen exposure is associated with increased risk of breast cancer (Key et al., 2002). Methylseleninic acid (MSA) is a monomethylated selenium metabolite and has potent inhibitory effect on the growth of a number of cancer cells (Shah et al., 2005; Ip et al., 2000). It is an ideal compound for investigating the anticancer effect of selenium in vitro. Shah et al. (2005) demonstrated that the selenium compound methylseleninic acid (MSA) decreased ER-positive MCF-7 breast cancer cell growth and gene expression via inhibiting estrogen receptor α (ERα) signaling. MSA downregulated ERα expression at both the mrna and protein levels. This study suggests that selenium may have an inhibitory effect on ER-positive breast cancer. MSA has also been reported to inhibit TM6 mouse mammary epithelial tumor cell growth (Sinha et al., 2001). Cells treated with 5 μm MSA were arrested in G1 phase of the cell cycle. In addition, MSA and other forms of selenium have been shown to inhibit mammary cancer cell line growth by inhibiting cell proliferation (El-Bayoumy et al., 2003b), inducing 72

88 apoptosis (Sinha et al., 2001; Ronai et al., 1995; Thompson et al., 1994), or inhibiting angiogenesis (Jiang et al., 1999). In summary, selenium can inhibit the mammary tumor cell growth as well as mammary carcinogenesis in DMBA-induced rodent. However, data from the epidemiology studies on selenium and breast cancer do not provide strong evidence to support the protective role of selenium on breast cancer Selenium and prostate cancers Prostate cancer is the most common non-skin malignancy in U.S. men, accounting for 9% of cancer deaths (the second leading cause of cancer death); 29% of all cancer cases are estimated to be prostate cancer for 2007 (American Cancer Society, 2007). Family history, age, and race are well-defined risk factors of prostate cancer. American blacks have a higher risk than other races (Fair et al., 1997). However, the evidence that risk can be increased through migration from areas of low risk (e.g. Japan) to areas of high risk (e.g. USA) suggests that environmental elements, such as diet and life style are also important in its etiology (Wolk, 2005). There is increasing epidemiological, experimental, and metabolic evidence, suggesting that oxidative stress may play an important role in the development and progression of prostate cancer. These observations are derived from: a) the inverse association of prostate cancer risk and dietary fat consumption, a major substrate for oxidative stress (Fair et al., 1997), b) 73

89 oxidative biomarker data, suggesting increased oxidative stress among patients with prostate cancer (Rao et al., 1999), c) ubiquitous defects of prostate cancer in the glutathione-s-transferase pi pathway, a major endogenous antioxidant mechanism (Brooks et al., 1998), and d) evidence that androgens (important promoters of prostate cancer growth) work in part via generation of reactive oxygen species (ROS) (Ripple et al., 1997). Because oxidative damage appears to be important in prostatic carcinogenesis, it is hypothesized that selenium, in its antioxidant role can be a chemoprevention agent to prevent prostate cancer. Evidence of a relationship between selenium and prostate cancer has accumulated markedly during last two decades Epidemiology studies The National Cancer Institute (NCI) conducted a case-control study to examine the relationship between blood selenium concentrations and prostate cancer risk in U.S. black and white men between the ages of 40 and 79 years (Vogt et al., 2003). Serum selenium concentrations of 212 cases and 233 controls were measured. An inverse association between serum selenium and risk of prostate cancer was observed (comparing highest to lowest quartiles, OR=0.71, 95% CI ), with similar patterns seen in both blacks and whites. Data from another case-control study using subjects from the Baltimore Longitudinal Study (52 prostate cancer patients and 96 age matched controls) found that compared to the lowest quartile of plasma selenium, the 74

90 odds ratios of the second, third and fourth quartiles were 0.15, 0.21 and 0.24, respectively, suggesting that low plasma selenium was associated with a 4 to 5-fold increased risk of prostate cancer (Brooks et al., 2001). An earlier study also reported the reverse association between plasma selenium and prostate cancer risk (Hardell et al., 1995). In contrast, the association between selenium and prostate cancer risk was not observed in some other case-control studies (Allen et al., 2004; Ghadirian et al., 2000; Jian et al., 1999). A number of prospective observation studies have examined the association between prostate cancer incidence and pre-diagnostic selenium concentrations in biological samples. In these studies, the association between the selenium content of biological samples and the development of prostate cancer is more likely to indicate risk because the specimens were obtained years before the development of the prostate cancer. A nested case-control study consisting of 724 prostate cancer patients and 879 control subjects failed to find an association between serum selenium and prostate cancer risk (P=0.07) during 8 years of follow-up (Peters, et al., 2007). Subset analysis, however, demonstrated that greater serum selenium concentrations were associated with reduced prostate cancer risks in men who reported a higher intake of vitamin E, in multivitamin users, and in smokers. An inverse association between prostate cancer risk and serum selenium levels was found in a cohort of 249 Hawaiian Japanese men who were diagnosed with prostate cancer during more than 20 years of follow-up (4 th quartile vs. 1 st 75

91 quartile OR=0.2, 95% CI= ) (Nomura et al., 2000). The inverse association between plasma or toenail selenium levels and prostate cancer risk has also been observed in a number of other prospective epidemiologic studies (Li et al., 2004; van den Brandt et al., 2003; Helzlsouer et al., 2000; Yoshizawa et al., 1998). However, no association between selenium and prostate cancer risk was observed in two prospective studies (Hartman et al., 1998; Goodman et al., 2001). These conflicting results may result from the differences of subjects recruitment, years of follow-up, and controls of confounding factors such as supplement intake and dietary intake. The most persuasive evidence for the protective effect of selenium against prostate cancer comes from the Nutritional Prevention of Cancer Trail (NPCT) study at the Arizona Cancer Center. NPCT is a randomized, double-blind, placebo-controlled clinical trial designed to determine whether a nutritional supplement of selenium (200 μg/d in the form of selenized yeast) would decrease the risk of skin cancer. Selenium supplementation had no effect on the primary skin cancer endpoint. However, after a mean of 4.5 years of treatment and 6.4 years of follow-up, a 63% reduction (P=0.02) in prostate cancer incidence was observed in men receiving selenium supplements (Clark et al., 1996). A reanalysis of the data including a further 25 months of blinded treatment and follow-up indicated a 52% reduction (P=0.005) of the prostate cancer incidence in men supplemented with selenium (Duffield-Lillico et al., 2002). The complete NPCT data also demonstrated that for men with plasma PSA concentrations 4 ng/ml, 76

92 selenium treatment was associated with a 65% reduction in prostate cancer risk (P = 0.01). However, there was no significant effect of selenium treatment for men entering the trial with PSA > 4 µg/l (RR = 0.88, P = 0.86) (Duffield-Lillico et al., 2003), suggesting that the protection by selenium for prostate cancer is only in the early stage(s) of carcinogenesis. The NPCT provided compelling rationale for the ongoing Selenium and Vitamin E Cancer Prevention Trial (SELECT), a prospective, 2 2 factorial double-blinded randomized study designed to examine whether selenium and/or vitamin E supplementation can decrease the risk of prostate cancer in healthy men (Lippman et al., 2005). This study is taking place in the United States, Puerto Rico, and Canada. Over 35,000 men have been enrolled as of June 2004, when the recruiting was closed. This large cancer prevention trial will be completed in 2013, and is expected to provide more direct evidence on the association between selenium and prostate cancer risk Animal studies There have been relative fewer animal studies on selenium and prostate cancer. Corcoran et al. (2004) investigated whether supranutritional selenium supplementation could inhibit prostate cancer progression in male nude mice established PC3 tumors. It was found that 6-week treatment with inorganic selenium (sodium selenate, 3 ppm) significantly inhibited the growth of prostatic tumors. To determine whether the levels of 77

93 selenoproteins can affect the development of prostate cancer, a unique mouse model was developed by breeding two transgenic animals: mice with reduced selenoprotein levels because of the expression of an altered selenocysteine-trna (i 6 A - ) and mice that develop prostate cancer because of the targeted expression of the SV40 large T and small t oncogenes to that organ[c3(1)/tag](diwasdlkar-navsariwala et al., 2006). The selenoprotein-deficient mice (i 6 A - /Tag) exhibited accelerated development of lesions associated with prostate cancer progression compared to the controls (WT/Tag). The results indicate that selenium may exert its chemoprevention effect on prostate cancer via its presence in selenoproteins. Selenium has also been shown to inhibit prostate cancer development in a canine model (Waters et al., 2005) In vitro studies A number of in vitro studies have been conducted to explore the possible mechanisms through which selenium inhibits prostate cancer development. The cellular and molecular responses of prostate cancer cells to Methylseleninic acid (MSA) treatment have been reported in a number of studies (Dong et al. 2005; Dong et al., 2004; Dong et al., 2003; Jiang et al., 2002; Zhao et al., 2004). Dong et al. (2003) examined the effects of MSA on androgen receptor (AR) negative PC-3 human prostate cancer cells. Physiological dose of MSA inhibited the growth of PC-3 cells in a dose- and time-dependent manner. MSA retarded cell cycle progression and markedly induced 78

94 apoptosis of cancer cells. A later study by this group (Dong et al., 2004) demonstrated that MSA inhibited AR positive LNCaP human prostate cancer cells growth as well by suppressing prostate specific antigen (PSA) expression. MSA also suppressed the expression of androgen receptor (AR) as well as the binding of AR to the androgen responsive element site. The authors further reported that MSA not only decreased the expression of androgen receptor and PSA in LNCaP cells, but also in other four prostate cancer cell lines (LAPC-4, CWR22Rv1, LNCaP-C81, and LNCaP-LN3) (Dong et al., 2005). A similar study was conducted to investigate the effects of MSA on LNCaP human prostate cancer cells by using high-density cdna microarrays (Zhao et al., 2004). The authors identified 951 genes whose expression shows striking dose- and time-dependent changes in response to MSA treatment over the time course of 48h. MSA caused an accumulation of cells at G0/G1 phase, suppressed androgen receptor (AR) expression at both mrna and protein level, and decreased levels of prostate specific antigen secreted into the medium. Recently, the same research group (Zhao and Brooks, 2007) determined the effects of selenomethionine on AR positive LNCaP human prostate cancer cells. Selenomethionine induced G2/M arrest in LNCaP cells, and also affected the expression of androgen responsive genes. The inhibition effect of LNCaP cell growth by selenomethionine characterized by G(1) phase has also been reported by other studies (Venkateswaran et al., 2002; Bhamre et al., 2003). 79

95 In summary, in vitro studies suggest that selenium may protect against prostate cancer by inhibiting cell growth, by inducing apoptosis, or by modulating the expression of AR and AR-regulated genes. The hypothesis that selenium could prevent the development of prostate cancer is supported by the consistent findings from studies with animal tumor and cell culture models, and by most of the epidemiologic observations. The ongoing large clinical trial, SELECT, is expected to provide more direct evidence to support this hypothesis. 2.8 Estrogen receptors in estrogen action Estrogens Estrogens are a group of steroid compounds functioning as the primary female sex hormone. Naturally estrogens include three major compounds, 17β-estradiol (E2), estrone, and estriol (Figure 2.5). 17β-estradiol (E2) is the principal physiological ligand for eliciting estrogenic biological responses. The primary function of estrogens is to promote the development of secondary sexual characteristics and induce menstruation in women. In addition to their critical role in the reproductive system, estrogens also play important regulatory roles in a wide variety of biological processes in other systems, including the skeletal, cardiovascular and central nervous systems (Couse and Korach, 1999). Estrogens have been implicated in reducing the incidence of coronary heart disease, osteoporosis, colon cancer, prostate cancer, and neurodegenerative diseases 80

96 A. 17β-estradiol (E2) B. estrone C. estriol Figure 2.5 Chemical structures of three major naturally occurring estrogens. Panel A, 17β-estradiol; Panel B, estrone; Panel C, estriol. 81

97 including stroke, Parkinson disease (PD), and Alzheimer disease (AD) (Pearce and Jordan, 2004; Deroo and Korach, 2006). Estrogens act as important free-radical scavengers due to their ability to donate hydrogen from their phenol-hydroxyl ring. Estrogens have been shown to have antioxidative effects by inhibiting membrane phospholipids peroxidation and low density lipoprotein (LDL) oxidation (Sugioka et al., 1987;.Maziere et al., 1991; Sack et al., 1994). Despite these health benefits, some cancers are stimulated by estrogen and estrogen use is associated with the risk of breast and ovarian cancer (Pearce and Jordan, 2004; Deroo and Korach, 2006). Antiestrogens, such as tamoxifen have been used for the treatment of breast cancer and ovarian cancer (Treek et al., 2006; Hasan et al., 2005) Estrogen receptor Biological effects of estrogens are mediated by two estrogen receptor (ER) proteins, ERα and ERβ. The ERα and ERβ are members of the nuclear hormone receptor superfamily that can initiate or enhance the transcription of genes containing specific hormone response elements, including receptors for progestins, glucocorticoids, androgens, thyroid hormones, retinoids and bile acids (Evans, 1988). ERα was first identified in the early 1960s in the uterus of rats on the basis of its ability to bind radiolabeled estrogen (Jensen and Jacobson, 1962). ERα protein has a molecular weight of approximately 66 kda with 595 amino acids (Green et al., 1986). During the following 82

98 decades, this receptor has been recognized as the single mediator of the physiological effects of estrogens until a novel ER was discovered in the rat prostate in 1996 (Kuiper et al., 1996). The newly identified receptor was named ERβ to distinguish it from the earlier receptor, ERα. ERβ protein has a molecular weight of 54.2 kda with 485 amino acids. ERα and ERβ are coexpressed in the breast, brain, thyroid, cardiovascular system, urogenital tract, and bone (Pearce and Jordan, 2004; Mattews and Gustafsson, 2003). ERα is the main subtype in the liver, kidney, uterus, pituitary, and testis, whereas ERβ is the main ER in the lung, prostate and gastrointestinal tract (Pearce and Jordan, 2004; Kuiper et al., 1997). The difference of tissue distribution of ERα and ERβ suggests that the two receptors may have distinct functions. Human ERα and ERβ share a conserved functional structure consisting of six domains (Figure 2.6) defined based on the putative functions that are contained in each area. The A/B domain located in the N-terminal region contains activation function 1 (AF1), a constitutive activation function that contribute to the transcriptional activity of the ER. This domain is one of the least conserved domains between ERα and ERβ, with only 30% identity between the two receptors. The C domain, or DNA binding domain (DBD), is the most conserved region between ERα and ERβ, with 96% identity. Two zinc fingers in the DBD are responsible for the binding of ER to the estrogen response elements (ERE) (Schwabe et al., 1993). When the ER lacks the DBD, it cannot bind DNA either in vitro or in vivo (Kumar and Chambon, 1988; Kumar et al., 1987). The D 83

99 AF-1 AF-2 Figure 2.6 Structure and homology between human ERα and ERβ protein. The functional A to E/F domains are schematically represented, with the numbers of amino acid residues indicated. Percentage of amino acid identity between the individual domains at the amino acid level is depicted (adapted from Ogwa et al., 1998). 84

100 domain, or hinge region, is involved in ER dimerization and in binding to heat shock protein hsp90. This region contains a signal area that is important for the movement of the receptor to the nucleus following synthesis in the cytoplasm. The D domain is not well conserved between ERα and ERβ, with only 30% identity. The E/F domains include the ligand binding domain (LBD), the dimerization domain, a second nuclear localization signal, and activation function 2 (AF2). The LBD is located in the C-terminus and is responsible for specific ligand recognition. AF2 requires binding of ligand to activate transcription, whereas AF1 can activate transcription in a ligand-independent manner (Berry et al., 1990; Webster et al., 1988). ERα and ERβ share a sequence identify of 53% in E/F domain Mechanism of estrogen action Estrogens affect the transcription of target genes by two pathways, i.e., ligand-dependent and ligand-independent mechanisms of ER activation Ligand-dependent activation of transcription by ER The classical model of ligand-dependent activation of ER is illustrated in Figure 2.7 (Lonard and Smith, 2002). Prior to binding, the ER exists as an inactive complex that includes a variety of proteins, such as hsp90, hsp70, and hsp56. hsp90 is an important protein to maintain the inactive state of ER. Binding of estrogen (ligand) to ER leads to 85

101 conformation changes of ER, which causes dissociation of chaperones such as hsp90 and hsp70, and phosphorylation of ER (Liberman, 1997). The ligand-bound, phosphorylated ER undergoes dimerization and is associated with co-regulatory proteins that modulate ER activity either by stimulating (so-called co-activators) or by repressing it (so-called co-repressors) (Hall and McDonnell, 2005). The ER : ligand : co-factor complex binds to the estrogen response element (ERE), a 13 bp DNA sequence in the promoter region of the target gene, thus stimulating or inhibiting the gene transcription. It has been shown that some genes with ERE half-sites (5'-AGGTCA-3') are responsive to estrogen (Zhu et al., 2006; Bahadur et al., 2005; Das et al., 2004; Klinge et al., 1997). However, whether ER binding to an ERE half-site as a monomer in certain genes remains controversial. For genes that lack ERE in their promoter regions, estrogen can modulate their transcription via a non-classical model of protein-protein interactions. According to this model, E2 binding to ER leads to the interaction of ER with transcription factors such as AP-1, Sp1, or NF-kB (Cascio et al., 2007; Lu et al., 2006; Maor et al., 2006). This complex binds to DNA that has AP-1, Sp1, or NF-kB sites, thus influencing transcription of the target genes. 86

102 Figure 2.7 Classical model of the ER activation process. Estradiol (E2) action is exerted through ER located in the nucleus. Estrogen binding induces a conformation change of ER, which causes ER homodimerization. The E2-bound ER interacts with estrogen response elements (ERE) in the promoter region of target genes and affects the gene expression (Lonard and Smith, 2002). 87

103 Ligand-independent activation of transcription by ER In addition to the ligand activation of ER, a number of ligand-independent pathways for activation of ER have been reported. Serine and tyrosine residues in the AF-1 and AF-2 domains are phosphorylated by signaling pathways downstream of growth factor receptors, such as EGF-, IGF-, and insulin- receptors (Mendez et al., 2006; Nicholson et al., 2005). This phosphorylation causes activation of ER and ER dimerization without estrogen binding. Other molecules such as camp (Schreihofer et al., 2001; Aronica and Katzenellenbogen, 1993), dopamine (Gangolli et al., 1997), c-myc, cyclin D1 and cyclin E (Butt et al., 2005) have also been reported to activate ER in the absence of ligand. 2.9 Selenium and reproductive hormones Estrogen affects the metabolism of several minerals, including copper (Canaraja et al., 2004; Alkjaersig et al., 1988; Bureau et al., 2002, Dorea and Miazaki, 1999), iron (Milman et al, 1992), magnesium, calcium (Muneyyirci-Delale et al., 1998), and chromium (Bureau et al., 2002). An effect of estrogen on selenium metabolism is also possible based on known associations between gender and/or sex hormones and selenium status observed in animals and humans. 88

104 2.9.1 Animal studies Studies in rats have shown that liver GPx activity of females were greater than that of males (Pinto and Bartley, 1969; Capel and Smallwood, 1983; Igarashi et al., 1984; Finley and Kincaid, 1991; Prohaska and Sunde, 1993; Yamamoto et al., 2002). An increase in liver GPx activity after the onset of sexual maturation in female rats (Pinto and Bartley, 1967) suggests that estrogen is responsible for the increase. A decline in blood parameters of selenium status during pregnancy in rats also suggests a hormone influence on selenium metabolism (Behne et al., 1978; Smith et al.,1986). Blood selenium and estrogen concentrations have also been positively correlated during the rat estrous cycle (Smith et al., 1995). Estrogen administration to chickens for 5 weeks significantly increased serum selenium concentrations (Halifeoglu et al., 2003). Hepatic GPx activity in male rats increased after castration to a level that was similar to that in liver of females, suggesting that testosterone plays a role in selenium distribution (Capel and Smallwood, 1983; Igarashi et al., 1984). More recently, testosterone in male rats has been shown to mediate expression of phospholipid hydroperoxide glutathione peroxidase (PHGPx) in the testis, a selenoprotein that is essential for male fertility (Maiorino et al., 1998). 89

105 2.9.2 Human studies A significant decrease in serum selenium concentration has been shown in boys during sexual maturation, whereas the serum selenium levels of girls remained constant throughout puberty (Marano et al., 1991). These findings support the role of testosterone in causing a redistribution and accumulation of selenium to the testis during male puberty. This redistribution provides more selenium for incorporation into PHGPx, which has particular relevance to male fertility (Foresta et al., 2002). The PHGPx gene is expressed in three different ways in testicular tissue. By alternate use of initiation codons, either a cytosolic protein or a mitochondria enzyme is generated (Arai et al., 1996). Just as testosterone has been shown to cause the redistribution of selenium in males, estrogen may play a similar role in females. As in animals, the decrease in parameters of selenium status in pregnant women may be caused by changes in reproductive hormone levels (Lopes et. al., 2004; Navarro et al., 1996; Golubkina et al., 2002; Zachara et al., 1993). The positive relationship between blood selenium parameters and estrogen during the menstrual cycle and after menopause is also a strong indication that estrogen plays a role in selenium metabolism. Earlier studies demonstrated differences in plasma selenium concentration (Das and Chowdhury. 1997; McAdam et al., 1994), plasma GPx activity (McAdam et al., 1994), and erythrocyte GPx activity (Larsen et al., 1996) throughout the menstrual cycle. More recent studies also have shown a more specific positive association between estrogen concentrations and plasma selenium, plasma GPx 90

106 activity (Ha and Smith, 2003), and erythrocyte GPx activity (Massafra et al., 1998; Ha and Smith, 2003) during the menstrual cycle. However, erythrocyte selenium content does not appear to be affected by hormonal fluctuations during the menstrual cycle (Ha and Smith, 2003; Peiker et al., 1991). The decline in estrogen during menopause also appears to affect selenium metabolism. Erythrocyte GPx activity was lower among postmenopausal women compared to that of the premenopausal women (Smith et al., 2000). Massafra et al. (2002) found that erythrocyte GPx activity was positively correlated with serum estrogen levels in pre- and post-menopausal women receiving estrogen replacement therapy. Several studies show that hormonal replacement therapy significantly increased erythrocyte GPx activity in postmenopausal women (Massafra et al., 1997; Massafra et al., 2002; Bednarek-Tupikowska et al., 2006), whereas hormonal replacement therapy failed to alter plasma and erythrocyte GPx activity in postmenopausal women in another study (Bureau et al., 2002). The inconsistent findings may be due to the different doses and duration of estrogen use. The use of estrogen-containing oral contraceptives also has been shown to increase erythrocyte GPx activity in premenopausal women (Capel et al., 1981; L'abbe et al., 1992; Massafra et al., 1993). 91

107 2.9.3 Tissue specific influence of sex hormones on selenium status Erythrocyte GPx activity (McMaster et al., 1990; Guemouri et al., 1991) in humans and liver selenium concentration and GPx activities in rats (Finley and Kincaid 1991) are greater in females than in males. However, plasma selenium content and GPx activity, as well as renal GPx activity, are greater in male than in female rats (Finley and Kincaid 1991). These findings suggest a tissue- specific effect of gender and/or sex hormones on selenium status. 92

108 CHAPTER 3 ESTROGEN STATUS ALTERS TISSUE DISTRIBUTION AND METABOLISM OF ORAL DOSE OF 75 Se-SELENITE 3. 1 INTRODUCTION Selenium is an essential nutrient of fundamental importance to human health. It is an integral component of several selenoproteins including glutathione peroxidase (GPx), an antioxidant enzyme protecting tissues against oxidative damage from hydrogen peroxide. Selenium has been associated with the risk of hormone- related diseases such as breast cancer (Huang et al., 1999; Lopez-Saez et al., 2003) and prostate cancer (Clark et al., 1996; Lee et al., 2006; Duffield et al., 2002). An understanding of the interaction between selenium and hormones is needed before considering the use of selenium in the prevention or treatment of these hormone-related diseases. 93

109 An association between gender and/or sex hormones and selenium status has been observed in animals and humans. Liver GPx activity of female rats has been shown to be greater than that of male rats (Pinto and Bartley, 1969; Capel and Smallwood, 1983; Igarashi et al., 1984; Finley and Kincaid, 1991; Prohaska and Sunde, 1993; Yamamoto et al., 2002). An increase in liver GPx activity after the onset of sexual maturation in female rats (Pinto and Bartley, 1967) suggests that estrogen is responsible for the increase. A decline in blood parameters of selenium status during pregnancy in rats also suggests a hormone influence on selenium metabolism (Behne et al., 1978; Smith et al.,1986). Blood selenium and estrogen concentrations have also been positively correlated during the rat estrous cycle (Smith et al., 1995). Estrogen administration to chickens for 5 weeks significantly increased serum selenium concentrations (Halifeoglu et al., 2003). Hepatic GPx activity in male rats increased after castration to a level that was similar to that in liver of females, suggesting that testosterone plays a role in selenium distribution (Capel and Smallwood, 1983; Igarashi et al., 1984). More recently, testosterone in male rats has been shown to mediate expression of phospholipid hydroperoxide glutathione peroxidase (PHGPx), a selenoprotein that is essential for male fertility (Maiorino et al., 1998). The effect of gender and sex hormones on selenium status is tissue specific. Erythrocyte GPx activity (McMaster et al., 1990; Guemouri et al., 1991) in humans and liver selenium concentration and GPx activities in rats (Finley and Kincaid 1991) are 94

110 greater in females than in males. However, plasma selenium content and GPx activity, as well as renal GPx activity, are greater in male than in female rats (Finley and Kincaid 1991). A significant decrease in serum selenium concentration has been shown in boys during sexual maturation, whereas the serum selenium levels of girls remained constant throughout puberty (Marano et al., 1991). These findings support the role of testosterone in causing a redistribution and accumulation of selenium to the testis during male puberty. This redistribution provides more selenium for incorporation into PHGPx, which has particular relevance to male fertility (Foresta et al., 2002). Just as testosterone has been shown to cause the redistribution of selenium in males, estrogen may play a similar role in females. As in animals, the decrease in parameters of selenium status in pregnant women may be caused by changes in reproductive hormone levels (Lopes et. al., 2004; Navarro et al., 1996; Golubkina et al., 2002; Zachara et al., 1993). The positive relationship between blood selenium parameters and estrogen during the menstrual cycle and after menopause is also a strong indication that estrogen plays a role in selenium metabolism. Earlier studies demonstrated differences in plasma selenium concentration (Das and Chowdhury. 1997; McAdam et al., 1994), plasma GPx activity (McAdam et al., 1994), and erythrocyte GPx activity (Larsen et al., 1996) throughout the menstrual cycle. More recent studies also have shown a more specific positive association between estrogen concentrations and plasma selenium, plasma GPx 95

111 activity (Ha and Smith, 2003), and erythrocyte GPx activity (Massafra et al., 1998; Ha and Smith, 2003) during the menstrual cycle. However, erythrocyte selenium content does not appear to be affected by hormonal fluctuations during the menstrual cycle (Ha and Smith, 2003; Peiker et al., 1991). The decline in estrogen during menopause also appears to affect selenium metabolism. Erythrocyte GPx activity was lower among postmenopausal women compared to that of the premenopausal women (Smith et al., 2000). Massafra et al. (2002) found that erythrocyte GPx activity was positively correlated with serum estrogen levels in pre- and post-menopausal women receiving estrogen replacement therapy. Several studies show that hormonal replacement therapy significantly increased erythrocyte GPx activity in postmenopausal women (Massafra et al., 1997; Massafra et al., 2002; Bednarek-Tupikowska et al., 2006), whereas hormonal replacement therapy failed to alter plasma and erythrocyte GPx activity in postmenopausal women in another study (Bureau et al., 2002). The inconsistent findings may be due to the different doses and duration of estrogen use. The use of estrogen-containing oral contraceptives also has been shown to increase erythrocyte GPx activity in premenopausal women (Capel et al., 1981; L'abbe et al., 1992; Massafra et al., 1993). The apparent relationship between estrogen status and selenium metabolism may have important implications for the prevention and treatment of certain estrogen responsive diseases, e.g., breast cancer (Huang et al., 1999; Lopez-Saez et al., 2003). 96

112 Previous research has focused on the effect of estrogen status or estrogen administration on selenium concentration and GPx activity in plasma and erythrocyte. Little is known about the effect of estrogen status on selenium metabolism in other organs. Systematic studies are needed to investigate the effect of estrogen on selenium metabolism in different tissues. 75 Se has been widely used as a tracer to investigate selenium metabolism in animal tissues. The purpose of this study was to investigate the effect of estrogen on the tissue distribution and metabolism of an oral dose of 75 Se-selenite in an ovariectomized rat model. Ovariectomized rats were subcutaneously implanted with estradiol pellets or placebo pellets to generate two groups of animals with significantly different estrogen status. Blood and organs were collected 1, 3, 6, and 24 hours after 75 Se administration. The hypothesis for this study was that estrogen status would affect the tissue distribution and metabolism of selenium. The effect of estrogen status on 75 Se absorption, distribution in blood, organs, and between cytosol and membrane fractions, as well as incorporation of 75 Se into abundant selenoproteins in cytosol and plasma were investigated. 97

113 3.2 MATERIALS AND METHODS Animals Female Sprague Dawley rats (Taconic, Germantown, NY) weighing 45-55g were purchased at weaning (3 weeks). At 7 weeks of age, animals were bilaterally ovariectomized and randomly assigned to one of the following two groups: Group 1, ovariectomized receiving estrogen replacement (OVX+E2, n=16); Group 2, Ovariectomized receiving a placebo pellet (OVX, n=16). Rats in OVX+E2 group were implanted subcutaneously with estradiol pellets containing 1.5 mg 17β-estradiol (60-day release, Innovative Research of America, Sarasota, FL). Rats in the OVX group were subcutaneously implanted with a placebo pellet that contained the same components (cholesterol, cellulose, lactose, phosphates, and stearates) but lacked 17β-estradiol. Pellets were implanted using a stainless steel precision 10 gauge trochar. Animals were housed in individual plastic cages at a controlled temperature (20-22 ) and exposed to a 12-h light-dark cycle. (lights off from 1800 to 0600 h). All procedures involving animals were approved by the Institutional Animal Care and Use Committee (ILACUC) of The Ohio State University (Appendix A). 98

114 3.2.2 Diets Rats were fed casein-based AIN-93G diet containing the NRC (National Research Council) selenium requirement of 150 μg/kg diet as sodium selenate (Research Diets, New Brunswick, NJ). The complete composition of the custom diets is provided in Appendix B. Animals were fed this semi-purified diet from weaning and throughout the study. Immediately after surgery and implantation with estrogen or placebo pellet at 7 weeks of age, dietary intake by rats in OVX group was significant higher than those of the OVX+E2 group (16.4 g/day vs g/day, respectively). In order to maintain similar calorie and selenium intake between the groups, rats in OVX+E2 group were given free access to the diets, while rats in OVX group were fed the average intake of OVX+E2 group during the previous day for the remainder of the study. Dietary intakes were recorded daily and body weights were recorded weekly SeO 3 2- administration 75 Se as sodium selenite ( 75 SeO 3 2- ) with specific activity of 1,400 Ci/g was purchased from the University of Missouri Research Reactor Facility (Columbia, MO). Five weeks after implantation of either estrogen or placebo pellets, each rat was administered µci of 75 Se as sodium selenite in 1.0 ml 150mM NaCl by gavage. The exact amount of 75 Se gavaged was determined by weighing the syringe containing the radioactive solution before and after injection. Eight rats (4 from OVX+E2 group and 4 from OVX group) 99

115 were killed 1, 3, 6, and 24 hours after 75 Se administration. All procedures involving 75 SeO 3 2- use were approved by Radiation Safety, the Office of Environmental Health and Safety at The Ohio State University (Appendix C) Sample collection Rats were anesthetized by brief exposure to carbon dioxide prior to terminal cardiac puncture for collection of blood samples into syringes containing 10mg Na 2 EDTA in 100 µl saline. Liver, kidney, heart, brain, lung, thymus, spleen, and the intact gastrointestinal tract (GI tract) and its contents were removed from all animals, rinsed in saline, weighed, and frozen in liquid nitrogen. Blood was centrifuged at 3,000g for 30 min at 4 to separate plasma and cells. All blood and tissue samples were stored at 20 until analysis. See Appendix D for a flow chart summary of the sample collection procedure Plasma 17β-estradiol concentration Concentration of 17β-estradiol in plasma was determined in samples from OVX and OVX+E2 rats without 75 Se-selenite administration (n=6 for each group) from another part of the study (chapter 4), using a double-antibody estradiol 125 I radioimmunoassay kit (Diagnostic Products Corp., Los Angeles, CA). Plasma 17β-estradiol was measured by the ability of the sample to compete with 125 I-labeled 17β-estradiol for antibody sites. 125 I activity was determined with a Cobra II auto-gamma counter (Packard Instrument 100

116 Company, Meriden, CT). The detailed procedure for the measurement of plasma 17β-estradiol concentration is summarized in Appendix E Se in tissues and cytosol 75 Se was determined in aliquots of plasma, erythrocytes, and organs with a Cobra II auto-gamma counter (Packard Instrument Company, Meriden, CT). All counts were corrected for radioactive decay. The instrument was calibrated daily with 137 Cs, and 75 Se standards also were counted with samples to determine counting efficiency (66% - 68%). All counts were corrected for radioactive decay using the known half-life of 75 Se ( days). Slight fluctuations in daily counting efficiency were corrected by reference to 75 Se standards. When calculating the total activity of 75 Se in blood, the volume of plasma was estimated as 3.82% body weight, and the volume of RBC was estimated as 2.37% body weight (Wang, 1959). The GI tract and its contents were homogenized for 75 Se measurement in aliquots. The apparent absorption of the 75 Se was estimated as (Total 75 Se dose 75 Se of GI)/Total 75 Se dose 100% Because organs of rats had not been perfused after excision, correction for residual blood containing 75 Se activity was done by measuring hemoglobin (Hb) concentrations in organs using Drabkin reagent (Drabkin, 1949). To determine the actual 75 Se activity, 75 Se 101

117 activity contamination of erythrocytes and plasma in organs was calculated and subtracted from the total 75 Se activity in organs. For organs in which 75 Se activity differed for OVX group and OVX+E2 group, the distribution of 75 Se in total membrane (organelle) and cyoplasmic compartments was determined. Organs were mechanically homogenized using Brinkman Polytron homogenizer (Brinkman Instruments Co., Westbury, NY) in ph 6.3, 0.05 M phosphate buffer (25% homogenate) containing 10% sucrose and centrifuged at 4 for 90 min at 110,000 g (Ti 50 rotor, Beckman Model L7-65, Palo Alto, CA) (Chambers and Rickwood, 1978). The volumes of supernatants (cytosol) were recorded and an aliquot was removed to measure 75 Se activity. 75 Se activity in membrane fraction was calculated by subtracting activity in cytosol from the total activity in organs. In order to normalize the 75 Se activity, the protein concentrations in cytosol was determined by bicinchoninic acid method (BCA protein assay kit, Pierce, Rockford, IL). Proteins in cytosol were collected by precipitation with trichloroacetic acid (TCA; 20% w/v final concentration). After incubation for 10 min at 4 C, tubes were centrifuged at 16,100 g for 5 min (Eppendorf Model 5415D centrifuge, Hamburg, Germany). Supernatant was removed and the pellet was washed with 200 μl cold acetone, before repeating centrifugation. Supernatant was removed and 75 Se activity in washed pellet was measured. 102

118 3.2.7 Determination of 75 Se in selenoproteins The incorporation of 75 Se into selenoproteins was determined using sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE). Proteins in plasma and organ cytosol were separated with 5% staking gel and 12% resolving gel. 20 μl cytosol containing μg protein was loaded to each well. Electrophoresis was performed at constant voltage (120V) for 1.5h. Gels were stained with Coomassie Blue R-250 for 4h and destained with 40% methanol in 10% acetic acid for 4h. The relative molecular weights of the bands were calibrated by comparison of their relative mobility, R f, with the following marker proteins: myosin (MW=193.5 KDa); β-galactosidase (MW=112.4 KDa); bovine serum albumin (MW=64.2 KDa); carbonic anhydrase (MW=30.4 KDa); soybean trypsin inhibitor (MW=26.1 KDa); lysozyme (MW=12.9 KDa); and aprotinin (MW=6.5 KDa) (Bio-Rad, Hercules, CA). The R f was calculated as the ratio of the distance migrated by the molecule to that migrated by a marker dye-front. A standard curve of R f vs. log 10 (MW) for the marker proteins was plotted for each gel (Figure 3.1 is representative of a standard curve). Gels were sliced according to known molecular weight of well characterized selenoproteins (Figure 3.2) and the 75 Se activity of the gel slices was measured. Total 75 Se activity in gel slices were about 90-95% 75 Se activity of cytosol proteins. This procedure provided a convenient method to examine the 75 Se distribution among selenoproteins. 103

119 log 10 MW y = x R 2 = R f Figure 3.1 Representative plot of molecular weight markers separated by SDS-PAGE. The R f was calculated as the ratio of the distance migrated by the molecule to that migrated by a marker dye-front. The marker proteins in the plot were: myosin (MW=193.5 KDa); β-galactosidase (MW=112.4 KDa); bovine serum albumin (MW=64.2 KDa); carbonic anhydrase (MW=30.4 KDa); soybean trypsin inhibitor (MW=26.1 KDa); lysozyme (MW=12.9 KDa); and aprotinin (MW=6.5 KDa). 104

120 MW (KD) 193 Marker Estimated location of selenoproteins Figure 3.2 Schematic of separation of 75 Se-selenoproteins in cytosol by SDS-PAGE. Female Sprague-Dawley rats were administered 60 μci 75 SeO 3 2- by gavage. Cytosol was prepared and 20 μl was loaded in each well. Gels were sliced according to known molecular weight of selenoproteins. Well characterized selenoproteins and their MW in each gel slice are as follows: 1) 73,000 (SelO). 70,000 (SelN); 2) 64,900 (TrxR3, thioredoxin reductase 3); 3) 57,000 (TrxR1, SelP), 56,000 (TrxR2), 45,000 (SelP); 4) 34,000 (SnGPx, sperm nuclei GPx); 5) 32,000 (ID3, type 3 deiodinase), 30,000 (ID2), 27,800 (ID1); 6) 23,000 (egpx, extracellular GPx), 22,000 (cgpx, cytosolic GPx and GI-GPx, gastrointestinal GPx), 19,700 (PHGPx); 7) 15,600 (Sep15); 8) 10,000 (SelW). 105

121 3.2.8 Statistical analysis Data are reported as means±sem (standard error of the mean). The Student t-test was used to detect the differences of the indexes between OVX+E2 and OVX groups. Analysis of Variance (ANOVA) was utilized to determine the differences of the indexes at different times after administering 75 Se (1, 3, 6, and 24 hours). Differences were considered significant at P<0.05. When significant differences among the means was indicated, Tukey s post hoc probability test was used to determine which means were significantly different. The statistical analyses were performed using SPSS V15.0 (SPSS Inc., Chicago, IL). 3.3 RESULTS Estrogen status, food intake, and body weight Plasma 17β-estradiol concentration in OVX+E2 group was significantly (P<0.05) greater than that in OVX group (64.5 pg/ml vs pg/ml). Food intake (g/d) increased from 6.2 g/d at 3 weeks of age to 13.4 g/d at 6 weeks of age (Figure 3.3). There were no significant differences in mean body weight between rats assigned to OVX+E2 group and OVX group prior to surgery (P > 0.05, Figure 3.4). After surgery (7 weeks), estrogen treated rats (OVX+E2) had decreased food intake (12.5 g/day), whereas food intake increased in OVX group (16.4 g/day). Rats in OVX group were then pair-fed the 106

122 20 Food intake (g/d) Surgery Week Figure 3.3 Food intake of female SD rats before and after surgery. Ovariectomy was performed at 7 weeks of age and rats were implanted with either a placebo (OVX) or estrogen (OVX+E2) pellet. After surgery, rats in OVX group were fed the average intake of OVX+E2 group during the previous day. Data are means ± SEM for 79 rats before surgery and for 22 rats in OVX+E2 group after surgery. The additional rats in OVX group were used to address other specific aims in this project. 107

123 Body Weight (g) OVX+E2 OVX Surgery Week Figure 3.4 Body weights of female SD rats before and after surgery. Ovariectomy was performed at 7 weeks of age. Rats were randomized into two treatment groups, i.e., OVX with estrogen replacement (OVX+E2) and OVX with a placebo pellet (OVX). Data are means ± SEM for 16 rats in each group. From weeks 8 to 12, body weight in OVX group was significantly greater than that of the OVX+E2 group (P < 0.05). 108

124 average amount of diet consumed by the OVX+E2 group beginning at week 7 and throughout the remainder of study. Body weight of OVX group was significantly (P < 0.05) greater than that of the OVX+E2 group. Thus, chronic estrogen administration to OVX rats decreased both food intake and body weight gain Se activity in tissues The effects of estrogen administration to OVX rats on 75 Se activity in the GI tract, blood and organs are presented in Figure 3.5. Data were determined as percent of 75 Se dose in order to normalize the slight differences in the amount of 75 Se administered by gavage. There were no significant differences in 75 Se activity in the GI tract (Figure 3.5A) between the OVX group and OVX+E2 group at any test time. 75 Se activity in the GI tract decreased from 1h to 24h (Figure 3.5A). By 24h, less than 5% of administered dose remained in the GI tract. Because the amount of selenium excreted from feces during the first 24h is known to be relatively small (around 4.5% dose) (Thomson and Stewart, 1973), our results suggest efficient absorption of 75 Se. There was a significant increase (P<0.05) of 75 Se activity from 1h to 3h (Figure 3.5) in blood and most organs in both treatment groups. 75 Se activity differed (P<0.05) between the OVX+E2 and OVX groups in plasma (Figure 3.5B), RBC (Figure 3.5C), liver (Figure 3.5D), heart (Figure 3.5E), kidney (Figure 3.5F), spleen (Figure 3.5G), thymus (Figure 3.5H), and brain (Figure 3.5I) at one or more times. The differences of 109

125 A. GI tract A % of dose a B b B b C OVX OVX+E2 0 c Hours B. Plasma 15 b % of dose 10 5 a b *B *B c B OVX OVX+E2 0 *A Hours Continued Figure Se activity in tissues 1, 3, 6, and 24h after gastric administration of 60 μci 75 SeO 2-3 (0.043 μg Se). Ovariectomized rats were implanted with either placebo (OVX) or estradiol (OVX+E2). Data are means ± SEM (n=4). Statistically significant difference (P<0.05) between OVX group and OVX+E2 group are denoted by presence of an asterisk (*). Within the same treatment (OVX/OVX+E2), means that do not share the same letter are significantly different at P<0.05. Panel A, GI tract; panel B, plasma; panel C, RBC; panel D, liver; panel E, heart; panel F, kidney; panel G, spleen; panel H, thymus; panel I, brain; panel J, lung. 110

126 Figure 3.5 continued C. RBC 2.5 % of dose b b *C BC a *AB ab *AB Hours OVX OVX+E2 D. Liver % of dose a a B *C a B a OVX OVX+E2 5 0 *A Hours Continued 111

127 Figure 3.5 continued E. Heart 0.25 % of dose a b b C *B C b OVX OVX+E *A Hours F. Kidney % of dose a b *B b C D b OVX OVX+E2 0 A Hours Continued 112

128 Figure 3.5 continued G. Spleen % of dose a b AB b BC C b OVX OVX+E *A Hours H. Thymus 0.20 % of dose b b b *B a *AB *AB *A OVX OVX+E2 Hours Continued 113

129 Figure 3.5 continued I. Brain 0.08 D % of dose a b B *C b c OVX OVX+E A Hours J. Lung b % of dose a ab B B ab B OVX OVX+E A Hours 114

130 75 Se activity between groups at each time are summarized in Table 3.1. Plasma and thymus in OVX group had greater 75 Se activity at 6h than OVX+E2 group. Liver and brain in OVX+E2 group had greater 75 Se activity than in OVX group at 6h. Thymus had greater and RBC had less 75 Se activity in OVX group than OVX+E2 group at 24h. In contrast, there was no difference in 75 Se activity between OVX and OVX+E2 groups in lung at any time (Figure 3.5J). OVX>OVX+E2 OVX<OVX+E2 No difference 1h plasma, RBC, liver, none kidneys, lung, brain heart, spleen, thymus 3h plasma, RBC, kidney, none liver, lung, brain heart, thymus spleen 6h plasma, thymus liver, brain RBC, heart, kidney, lung, spleen 24h thymus RBC plasma, liver, kidney, heart, spleen,brain,lung Table 3.1 Summary of the 75 Se activity in SD female rat tissues 1, 3, 6, and 24h after gastric administration of 75 Se. Ovariectomized rats were implanted with either placebo pellet (OVX) or pellet with estradiol (OVX+E2). 75 Se activity was expressed as percent of 75 Se administered. There were 4 rats per group at each time. The differences between OVX+E2 and OVX were determined by student t test. P<0.05 was considered as statistically significant. 115

131 3.3.3 Distribution of 75 Se among subcellular fractions and selenoproteins For those organs that differed in 75 Se activity between OVX and OVX+E2 groups at indicated times, 75 Se activity in cytosolic and membrane fraction (includes all other fractions except cytosol) was determined (Table 3.2). Estrogen status similarly affected 75 Se activity in both cytosolic and membrane fraction. For example, estrogen treatment proportionally increased 75 Se activity in both cytosolic and membrane fractions of liver at 6h. Thus, the percentage of the 75 Se activity in each fraction was not significantly affected by estrogen treatment. The percentages of 75 Se activity for cytosolic fraction were 45.4% and 48.9% in OVX rats and OVX+E2 rats, respectively (P>0.05 between the two groups). Cytosol contained 41-49% of the total 75 Se activity in liver, heart, and thymus, 54% in spleen, and only 30% in brain at 6h (data not shown), demonstrating the tissue-specific nature of intracellular distribution. Cytosol protein was precipitated in tissues to examine if estrogen status affected molecular distribution of recently administered selenium. The majority (80-90%) of 75 Se in cytosol was associated with proteins and independent of estrogen status (data not shown). Estrogen status also did not appear to markedly affect the relative distribution of 75 Se among the selenoproteins in cytosol prepared from organs collected from OVX and OVX+E2 rats. The one exception was a significantly greater percentage of 75 Se in liver GPx (Figure 3.6) from OVX group than OVX+E2 group at 6h (34.7% vs. 31.2% of total dpm in gel slices, respectively). The relative distribution of 75 Se among the 116

132 Organs Time after dose OVX OVX+E2 Liver 1h 22.4 ± ± 1.7 * Liver 6h 29.9 ± ± 3.6 * Heart 1h 0.3 ± ± 0.1 * Heart 3h 3.7 ± ± 0.2 * Kidney 3h 37.0 ± ± 2.4 * Spleen 1h 1.8 ± ± 0.1 * Thymus 1h 0.6 ± ± 0.1 * Thymus 3h 3.5 ± ± 0.5 * Brain 6h 0.3 ± ± 0.1 * Table Se activity in cytosol (dpm/μg protein). Samples were prepared for organs with significantly different 75 Se activity between OVX group and OVX+E2 group at indicated times. 75 Se activity in the cytosol was normalized to protein content. Data are means ± SEM (n=4). Statistical significance (P<0.05) between OVX group and OVX+E2 group is denoted by presence of an asterisk (*). 117

133 % of 75 Se in gel slices OVX OVX+E2 GPx * Gel slices Figure 3.6 Distribution of 75 Se activity among selenoproteins in liver cytosol 6h after gastric administration of 75 Se-selenite. Ovariectomized rats were implanted with either placebo pellet (OVX) or pellet with estradiol (OVX+E2) and gavaged after 5 weeks. Cytosolic proteins in liver were separated by SDS-PAGE. The gels were sliced according to known molecular weight of selenoproteins and the 75 Se activity in slices was measured by gamma ray spectrometry. 75 Se activity in gel slices is expressed as percentage of total 75 Se activity in proteins. Data are means ± SEM (n=4). Statistical significance (P<0.05) between OVX group and OVX+E2 group is denoted by presence of an asterisk (*). Molecular weights of selenoproteins in each gel slice are: 1) 73,000 (SelO), 70,000 (SelN); 2) 64,900 (TrxR3, thioredoxin reductase 3); 3) 57,000 (TrxR1, SelP), 56,000 (TrxR2), 45,000 (SelP); 4) 34,000 (SnGPx, sperm nuclei GPx); 5) 32,000 (ID3, type 3 deiodinase), 30,000 (ID2), 27,800 (ID1); 6) 23,000 (egpx, extracellular GPx), 22,000 (cgpx, cytosolic GPx and GI-GPx, gastrointestinal GPx), 19,700 (PHGPx); 7) 15,600 (Sep15); 8) 10,000 (SelW). 118

134 selenoproteins in cytosol of other organs was independent of estrogen status. Data are provided in Appendix F. For all selenoproteins, there was no difference of the percentage 75 Se activity between OVX and OVX+E2 groups. These results indicate that estrogen treatment similarly affected the 75 Se activity in selenoproteins, i.e., it proportionally increased (or decreased) 75 Se incorporation into all selenoproteins in general rather than affecting incorporation into specific selenoproteins. GPx contained the highest percentage of 75 Se activity among the selenoproteins in all tissues, accounting for 31-38% of the total protein 75 Se activity in liver, 28%-31% in heart, 24%-26% in kidney, spleen, thymus, and brain (Table 3.3). SelP, TrxR1 and TrxR2 together accounted for 6%-15% of the total protein 75 Se activity in cytosol of all organs examined. In plasma, a greater percent of 75 Se was present in SelP in OVX+E2 group than in OVX group at 3, 6 and 24h (Figure 3.7). 75 Se was not detected in plasma GPx at 1, 3, and 6h. However, GPx in OVX+E2 group accounted for 8.6% of total protein 75 Se activity at 24h, which was significantly higher than that in OVX group (6.5%). Previous studies have shown that plasma 75 Se was distributed between SelP, GPx, and low molecular weight selenium (LMW) (Burk et al., 2003). The percentage of 75 Se in LMW was calculated by subtracting the 75 Se percentage of SelP and GPx from the total (100%). Table 3.4 summarizes the distribution of 75 Se among SelP, GPx, and LMW in plasma. At 1h, approximately 70% 75 Se was present as low molecular weight selenium, whereas only 21%-32% of 75 Se was associated with LMW at 3, 6, and 24h. 119

135 Organs Time after SelP+TrxR1+TrxR2 (%) GPx (%) dose OVX OVX+E2 OVX OVX+E2 Liver 1h 6.8 ± ± ± ± 2.1 Liver 6h 10.5 ± ± ± ± 0.9 * Heart 1h 10.7 ± ± ± ± 0.9 Heart 3h 10.5 ± ± ± ± 0.6 Kidney 3h 6.3 ± ± ± ± 1.6 Spleen 1h 14.9 ± ± ± ± 1.0 Thymus 1h 11.1 ± ± ± ± 0.9 Thymus 3h 10.0 ± ± ± ± 0.8 Brain 6h 14.5 ± ± ± ± 1.3 Table Se activity in SelP, TrxR1, TrxR2 and GPx. Data are expressed as % of the total 75 Se activity in protein. Cytosol were analyzed in organs that differed in 75 Se activity between OVX group and OVX+E2 group at indicated times. Data are expressed as percent of the total protein 75 Se activity in cytosol. Data are means ± SEM (n=4). Statistical significance (P<0.05) between OVX group and OVX+E2 group is denoted by presence of an asterisk (*) as superscript. 120

136 Hours OVX (%) OVX+E2 (%) SelP GPx LMW SelP GPx LMW 1h h h h Table Se distribution among plasma SelP, GPx, and low molecular weight selenium (LMW) 1, 3, 6, and 24h after administration of SeO 3 by gavage. Ovariectomized rats were implanted with either placebo pellet (OVX) or pellet with estradiol (OVX+E2). Plasma proteins were separated by SDS-PAGE and 75 Se activity in gel slices containing proteins with MW in the range of SelP and GPx determined. LMW 75 Se (%) was calculated by subtracting the percentage of SelP and GPx from the total (100%). Data are means of 4 animals in each group, and are expressed as percentage of 75 Se activity in plasma. 121

137 % 75 Se in plasma proteins * B *B OVX OVX+E2 *C b b c A a Hours after 75 Se gavaged Figure Se activity in plasma SelP 1, 3, 6, and 24h after administration of SeO 3 by gavage. Animals are ovariectomized rats receiving either a placebo pellet (OVX) or pellet with estradiol (OVX+E2) for 5 weeks. Plasma proteins were acid precipitated and 75 Se activity measured as described in methods. Data are means ± SEM (n=4). Statistically significant (P<0.05) between OVX group and OVX+E2 group are denoted by presence of an asterisk (*). Within the same treatment (OVX/OVX+E2), means do not share the same letter differ significantly (P<0.05). 122

138 3.4 DISCUSSION An ovariectomized rat model has been used in present study. By implanting ovariectomized female SD rats with either estrogen or placebo pellets, we generated two groups of rats with significantly different levels of 17-β estradiol concentrations in plasma (64.5 pg/ml in OVX+E2 rats vs pg/ml in OVX rats). Although a number of studies have suggested a relationship between estrogen and selenium status, the present study is the first in which radioactive tracer 75 Se was used to compare the impact of estrogen status on distribution of absorbed selenium in organs, subcellular fractions, and selenoproteins in an ovariectomized rat model. In order to measure the apparent absorption of 75 Se from GI tract, the whole GI tract and contents was removed, and the total 75 Se activity of the whole GI tract and its contents was counted. 75 Se activity in the gut continued to decrease from 1h (25%-29% of dose) to 24h (3.4%-4.4% of dose) (Figure 3.5A), indicating efficient absorption of administered 75 Se. It has been reported that in rats, less than 50% of Zn, Fe, Cu, and Mn was absorbed (Chen et al., 2006). The present study confirms that selenium is well absorbed compared to other minerals. 75 Se activity in feces was not measured in the present study due to the relatively small amount of selenium typically excreted in the feces. Previously in rats, only 4.5% of 75 Se given orally as 75 Se-selenite was excreted through feces during the first 24 hours, 123

139 although 21% of administered 75 Se was excreted in feces after 7 days (Thomson and Stewart, 1973). In the present study, 3.4%-4.4% of administered 75 Se was in the GI at 24h. Thus, around 90% of administered 75 Se was absorbed. These values were in general agreement with results from Thomson and Stewart (1973) in which 91-93% oral dose of 75 Se-selenite was absorbed by rats during 24 hour period after administration. Some selenium is secreted in bile, but this represents a relatively minor pathway as urine is the primary route for excretion. Previously, approximately 1-2% of the dose was eliminated in the bile 1h after the subcutaneous injection of 75 Se-selenite to male rats. By 3h after injection of 75 Se-selenite, the bile contained 4% of the administered dose (Leavander and Baumann, 1966). Absorption of selenium is influenced by a number of factors, such as chemical speciation, and co-consumption with other nutrients (e.g. Vitamin A, C, E, and copper) (Venderland et al., 1992; Yu and Beynen, 2001). To our knowledge, the present study is the first to determine the effect of estrogen status on the absorption of oral dose of selenium. No differences in apparent 75 Se absorption were observed between OVX rats and OVX+E2 rats at any of the test times, suggesting the absorption of selenium is independent of estrogen status. Thomson and Stewart (1973) observed the distribution of 75 Se in female rats given oral or intravenous doses of 75 Se-selenite and 75 Se-selenomethionine for 16 weeks. In their study, the initial uptake was defined as the maximum amount of tracer observed in 124

140 the tissue during the first week. In rats, 75 Se activity from both an oral dose (the present study) and an intraperitoneal dose (Brown and Burk, 1973) was maximum within 24 hours in most tissues. Table 3.5 compares the initial 75 Se uptake data from the earlier (Thomson and Stewart, 1973) and the present study. The greatest amount of radioactivity was detected in liver in both studies. Among other organs, the kidney contained the second greatest amount of 75 Se. Other studies also confirm liver and kidney are primary sites in rats that accumulate 75 Se-labled organic or inorganic selenium (Beilstein and Whanger,1985; Hopkins et al., 1966; Hansen and Kristensen, 1979). Considerably lower activity has been found in the heart, spleen, thymus and lung, while the brain generally has the least accumulation. Previous studies have demonstrated that, depending on the dose, 5-50% of administered 75 Se could be excreted through urine in methylated form (Hopkins et al., 1966; Burk et al., 1972; Hirooka and Galambos, 1966; Thomson and Stewart, 1973). Although the selenium concentration in skeletal muscle (per gram) is very low, 20-50% of whole body selenium could be deposited in this tissue because of its large mass (Oster et al., 1988; Beilstein and Whanger, 1988; Thomson and Stewart, 1973). The 75 Se activity in muscle was not determined in present study because of the difficulty in estimating total muscle mass. 125

141 Tissues Thomson s study* Present study % dose %dose/g % dose %dose/g Plasma RBC Liver Heart Kidneys Spleen Thymus Brain Lungs Table 3.5 Initial uptake of 75 Se in tissues of female Wistar rats given oral or intravenous dose of 75 Se-selenite and 75 Se-selenomethionine (Thomson study), or of female SD rats given oral dose of 75 Se-selenite (present study). The initial uptake was defined as the maximum amount and concentration of tracer observed in the tissue during the first week (Thomson and Stewart study) or during the first 24h (present study). * See reference Thomson and Stewart,

142 The present study was designed to determine the effect of estrogen on distribution of selenium among tissues. Previously, Brown and Burk (1973) investigated the effect of gender on tissue distribution of 75 Se in rats given an intraperitoneal tracer dose of 75 Se-selenite. The 75 Se content was expressed as a percentage of dose per gram tissue. During the 10-week period, female rats consistently retained more 75 Se in all tissues except the brain and reproductive organs than males. It was expected that females would have higher 75 Se retention in tissues outside the reproductive tract because ovary and uterus had much lower 75 Se retention than those of the testis and epididymis. Differences in estrogen status resulted in a similar trend in the present study between the OVX+E2 and OVX group at 24 hours. OVX+E2 rats had significantly higher (P<0.05) 75 Se in blood (0.53% dose/g vs. 0.38% dose/g), liver (1.7% dose/g vs. 1.2% dose/g), and spleen (0.4% dose/g vs. 0.3% dose/g) compared to OVX rats. In contrast, 75 Se activity in heart (0.19% dose/g vs. 0.16% dose/g), kidney (2.3% dose/g vs. 2.2% dose/g), lung (0.42% dose/g vs. 0.37% dose/g), thymus (0.27% dose/g vs. 0.20% dose/g), and brain (0.36% dose/g vs. 0.30% dose/g) from OVX+E2 and OVX rats were not significantly different. These results suggest estrogen-dependent 75 Se distribution is tissue-specific in rats. The results of this study suggest that estrogen facilitated post-absorptive 75 Se transport into liver during the first 6h and 75 Se efflux from liver to other tissues after 6h. The liver has an important role in selenium metabolism. It takes up selenite, selenate, selenomethionine, and selenocysteine from portal vein and also is the principal organ 127

143 responsible for the removal of excess selenium as it synthesizes metabolites of selenium for excretion. In the present study, total 75 Se activity in liver was greater in OVX than in OVX+E2 rats at 1h (13.1% vs.3.9% of total dose, Figure 3.5D). However 75 Se activity in OVX+E2 group increased markedly from 1 to 6h when it was greater than OVX group (18.0% dose vs. 10.9% dose). By 24h, the two groups had similar 75 Se activity in liver. The effect of estrogen on the distribution of 75 Se within tissues was also investigated. In the present study, 45% to 49% 75 Se was deposited in liver cytosolic fraction at 6h. This result agrees with those of Brown and Burk (1973) who found that 47% to 56% 75 Se was deposited in liver cytosol fraction of male rats intraperitoneally administered 75 Se-selenite. Compared to cytosol from other organs with 41%-54% of administered 75 Se, brain had a lower percentage of 75 Se activity in cytosol (29%-32%). This indicates that more 75 Se was deposited in membrane-bound organelles in brain (around 70%). Brain is a tissue that retains selenium well when the trace element is limited in the diet (Behne et al., 1988). It is not clear whether low 75 Se activity in brain cytosol is related to such retention. To determine possible effects of estrogen availability on 75 Se distribution among selenoproteins, SDS-PAGE was used to separate tissue proteins. Gel sections were cut and 75 Se activities were measured. In plasma, 61% to 79% 75 Se was associated with SelP from 3h to 24h. These results agree well with previous studies using the similar method in which 68% 75 Se was deposited in SelP when female rats were injected 128

144 intraperitoneally with 75 Se-selenomethione (Deagen et al., 1993). Estrogen status did not affect the relative distribution of 75 Se among the selenoproteins in organ cytosol. However, estrogen status affected the relative distribution of 75 Se in SelP and GPx in plasma. A greater percentage of 75 Se was incorporated into plasma SelP at 3, 6, and 24h in OVX+E2 group than in OVX group (Figure 3.7). OVX+E2 group also had a greater percentage of 75 Se incorporated into plasma GPx at 24 hour than that OVX group (8.6% vs. 6.5%). It has been reported that plasma 75 Se was distributed between SelP, GPx, and low molecular weight selenium (Burk et al., 2003). In the present study, approximately 30% plasma 75 Se was incorporated into SelP, with 70% 75 Se present in low molecular weight fraction at 1h. There was a rapid increase of 75 Se that was incorporated into SelP after 1h with 69% to 79% of plasma 75 Se present in SelP by 3h. From 6h to 24h, there was a decrease of 75 Se in SelP. SelP has been shown to function as a selenium transport protein (Hill et al., 2003; Schomburg et al., 2003). It transfers selenium to extrahepatic tissues by binding to specific membrane receptors, and release the selenium via a receptor-mediated mechanism (Richardson, 2005; Schweizer et al., 2005). The decrease of 75 Se in SelP suggests that SelP delivered selenium to other organs, including kidney, the site of plasma GPx synthesis. 75 Se was detected in plasma GPx at 24h, and the activity was greater in OVX+E2 group than that of OVX group. This suggests increased newly synthesized plasma SelP was transported to the kidney, leading to synthesis of 75 Se-GPx. 129

145 To our knowledge this is the first study in which gavaged 75 Se was used in an ovariectomized rat model to investigate estrogen-dependent selenium metabolism. It is a comprehensive study that not only investigated 75 Se absorption, but also determined the 75 Se distribution at tissue level, subcellular fraction level, and selenoprotein level. Another strength of this study is that the data were collected at four times, so that the influence of estrogen status on transport and metabolism of oral selenium could be explored chronologically. In summary, although absorption of 75 Se was independent of estrogen status, 75 Se activity differed (P<0.05) in a number of tissues during the 24h period following absorption. Estrogen did not affect the relative distribution of 75 Se between cytosol and membrane fractions in organs, or the relative distribution of 75 Se among the selenoproteins in organ cytosol. However, plasma SelP in OVX+E2 group was enriched in 75 Se at 3, 6 and 24h compared to OVX group (P<0.05). 75 Se in plasma GPx also was greater in OVX+E2 compared to OVX group at 24h (P<0.05). These results suggest that estrogen status affects distribution of ingested 75 Se-selenite in tissues. The effect of estrogen status on other selenium biomarkers, such as selenium concentration and GPx activity in tissues remains unclear. This issue is the subject of the next chapter. 130

146 CHAPTER 4 THE EFFECT OF ESTROGEN ON SELENIUM STATUS IN TISSUES AND HEPATIC LEVELS OF SelP AND GPx mrna 4. 1 INTRODUCTION A relationship between estrogen status and selenium metabolism has been supported by the following evidence obtained from both animal and human studies. First, males and females differed in their selenium status. Liver GPx activity of females was greater than that of the males in rat (Pinto and Bartley, 1969; Capel and Smallwood, 1983; Igarashi et al., 1984; Finley and Kincaid, 1991; Prohaska and Sunde, 1993; Yamamoto et al., 2002) and mice (Prohaska and Sunde, 1993). Female mice had 15 25% higher GPx activity in brain compared to age-matched males throughout the life cycle (Sobocanec et al., 2003). However, GPx activity in brain is similar in male and female rats (Capel and Smallwood, 1983). In humans, boys significantly decreased in serum selenium concentration during 131

147 sexual maturation, whereas the serum selenium levels of girls remained constant throughout puberty (Marano et al.,1991). These findings support the hypothesis that estrogen increases liver GPx activity and plasma selenium concentration. Second, there was a decline in blood parameters of selenium status during pregnancy in rats (Behne et al., 1978; Smith et al.,1986), and humans (Lopes et. al., 2004; Navarro et al., 1996; Golubkina et al., 2002; Zachara et al., 1993). Since estrogen levels were low during pregnancy, these findings suggest that estrogen upregulated selenium status in blood. Third, blood selenium and GPx activity were positively correlated with estrogen concentrations during the rat estrous cycle (Smith et al., 1995) and the human menstrual cycle (Ha and Smith, 2003). Fourth, estrogen administration elevated selenium status. Halifeoglu et al. (2003) reported that administration of estrogen to chickens for 5 weeks significantly increased serum selenium concentrations. Estrogen administration also significantly increased erythrocyte GPx activity in both premenopausal (L'abbe et al., 1992) and postmenopausal women (Massafra et al., 1997; Massafra et al., 2002; Bednarek-Tupikowska et al., 2006). Selenium exerts its biological effects largely through its presence in selenoproteins, i.e., proteins that contain selenium in the form of selenocysteine (Burk and Hill, 1993). At least 25 human selenoproteins have been identified (Gromer et al., 2005), including glutathione peroxidase (GPx) and selenoprotein P (SelP). GPx is a selenium dependent enzyme with antioxidant activity. GPx activity has been widely used as a biomarker to 132

148 assess selenium status because it is correlated to selenium concentrations in tissues. As discussed previously, GPx activity is affected by gender or estrogen status in some tissues, including plasma, RBC, liver, and brain. A recent study demonstrated that GPx4 mrna levels in the bovine oviduct were upregulated by 17β-estradiol administration. However, mrna levels of GPx1 and GPx3 were not altered in response to estrogen supplementation (Lapointe et al., 2005). The basis for the selective up-regulation of GPx4 mrna in the oviduct merits further investigation. Estrogen also had no effect on GPx mrna levels in cultured human endothelial cells (Strehlow et al., 2003). However, an increase of 40% in GPx mrna expression has been observed in hepatoma H4IIE cells exposed to the dietary phytoestrogen daidzein (Rohrdanz.et al., 2002). Using a SelP knockout mouse model, researchers have found decreased selenium in tissues such as testis, brain, and kidney in male SelP-/- mice, while selenium accumulated in liver (Hill et al., 2003; Schomburg et al., 2003). These results suggest that SelP serve as a transport protein to shuttle selenium from liver to other tissues. Thus, SelP may play an important role in selenium metabolism because of its function in selenium transport. SelP synthesis is regulated by dietary selenium intake. Inadequate dietary selenium intake decreased plasma SelP levels in rats (Chittum et al., 1997) and humans (Janghorbani et al., 1999). There also was a 60-90% reduction in SelP levels during selenium depletion in HepG2 and H4IIE liver cell lines (Hill et al., 1996b). The effect of dietary selenium intake on SelP mrna content has also been investigated. 133

149 Compared with rats fed selenium supplemented diet, rats fed selenium deficient diet had significant lower SelP mrna levels in liver (Weiss et al., 1997; Christensen et al., 1995) and kidney (Christensen et al., 1995). There is little known about regulation of SelP by sex hormones. Rats are an ideal model for investigating changes in selenium status occurring during the reproductive cycle because of the short length of the estrous cycle. The estrous cycle in rats lasts four days and is characterized as proestrus, estrus, metestrus and diestrus stages. The estrous cycle may be determined according to the cell types observed in the vaginal smear. We have demonstrated that estrogen affected the metabolism of oral dose of 75 Se-selenite in a tissue-specific manner (Chapter 3). 75 Se activity in tissues, cytosol fraction, and selenoproteins was measured to trace the distribution and utilization of newly administered selenium. However, total selenium concentrations and GPx activity in tissues were not measured. Also, the effect of estrogen status in a 4-day estrous cycle on the selenium distribution among tissues in female rats was not investigated. The purpose of this study was to investigate 1) effects of estrogen on selenium concentrations and GPx activity in different tissues, and 2) effects of estrogen on the hepatic levels of SelP mrna and GPx mrna. The working hypothesis was that estrogen status will affect selenium status as measured by selenium concentration and GPx activity in selected tissues and that SelP and GPx moleculars in liver will also be affected by estrogen. Estrogen availability was manipulated using 134

150 ovariectomized Sprague Dawley rats with (OVX+E2) or without estrogen replacement (OVX). The effect of the OVX procedure was controlled by including a sham-operated group of rats. The effect of the pellet implantation procedure was controlled by including OVX rats with implanted placebo pellet. The effect of the estrogen pellet on circulating estrogen concentrations was verified by measurement of plasma 17-β estradiol concentration. Plasma ceruloplasmin activity was analyzed as positive control since estrogen is known to increase the amount of the cuproprotein (Canaraja et al., 2004; Alkjaersig et al., 1988). The results of this study demonstrated that estrogen modulates selenium status in different tissues, and also provided useful insights about the possible mechanisms by which estrogen regulates selenium metabolism. 4.2 MATERIALS AND METHODS Animals Female Sprague Dawley weaning rats (3 weeks of age) weighing 45-55g (Taconic, Germantown, NY) were fed a standard semi-purified diet containing 150 µg Se/kg throughout the study. At 7 weeks of age, rats were randomly divided into three groups. The first two groups underwent OVX with or without estrogen replacement (n=6 rats/group). The third group was sham operated (n=24, with 6 rats to be harvested for each day of the estrous cycle). One group of OVX rats with estrogen replacement 135

151 (OVX+E2) were subcutaneously implanted with a pellet containing 1.5mg 17β-estradiol for release with 60 days (Innovative Research of America, Sarasota, FL). Rats in another two groups (OVX and Sham) were subcutaneously implanted with a placebo pellet that released its contents in 60 days. All pellets were positioned subcutaneously prior to skin closure for the ovariectomy (or sham) procedure. Animals were housed in individual plastic cages at a controlled temperature (20-22 ) and exposed to a 12h light-dark cycle. (lights off from 1800 to 0600 h). All procedures involving animals were approved by the Institutional Animal Care and Use Committee of the Ohio State University (Appendix A) Diets Rats were fed the casein-based AIN-93G diet containing NRC (National Research Council) selenium requirement of 150 μg/kg diet as sodium selenate (Research Diets, New Brunswick, NJ). The complete composition of the custom diets is provided in Appnedix B. Rats in OVX+E2 group were given free access to the diets. Rats in OVX group and sham-operated group were pair-fed, e.g. fed the average food intake of OVX+E2 group during the previous day. Food intakes were recorded daily and body weights were recorded weekly. 136

152 4.2.3 Estrous cycle determination in sham-operated rats Estrous cycles of the animals in sham-operated group were determined by examining the vaginal smears taken between 8am to 9am each day. Cells types in the smear were subsequently examined microscopically to determine estrous stage. The consecutive stages of the estrous cycle were characterized as following: proestrus, scattered distribution of cornified cells and the presence of nucleated cells; estrus, presence of only cornified cells; metestrus, presence of leukocytes and fewer cornified cells; and, diestrus, mostly polymorponuclear leukocytes (Turner and Bagnara, 1976). In order to precisely monitor the patterns of the estrous cycles, the estrous cycles of the animals were examined for at least two consecutive 4-day cycles before killing. There were 6 rats killed each of the four days of the estrous cycle Sample collection At 12 weeks of age, rats were anesthetized by brief exposure to carbon dioxide prior to terminal cardiac puncture for collection of blood samples into syringes containing 10mg Na 2 EDTA in 100 μl saline. Liver, kidney, heart, and brain were removed from all animals, rinsed in saline, weighed, and frozen in liquid nitrogen. Blood samples were centrifuged at 3,000g for 30 min at 4 to separate plasma and blood cells. All samples were stored at 80 before analysis. 137

153 4.2.5 Laboratory analyses Selenium: Selenium concentrations in diet, plasma, red blood cells, liver, kidney, heart, and brain were measured by a gas chromatography technique described by McCarthy et al. (1981) using an Agilent 6890 Series gas chromatograph with an electron capture detector and a 225 Durabond Megabor column. Instrument operating temperatures were 190 for oven (column), 220 for front inlet (injector), and 300 for front detector. Nitrogen (~60 psi) and helium (~80 psi) were used as carrier gases. Data were integrated using an Agilent 6890 Series Integrator. Bovine liver (Standard Reference Material 1577b, National Institute of Standard & Technology, Gaithersburg, Maryland) was used as the reference standard and sodium selenite (Na 2 SeO 3 ) was used as the working standard. The detailed procedure for selenium assay is summarized in Appendix G. Glutathione peroxidase activity: Glutathione peroxidase (GPx) activity of plasma, red blood cells, liver, kidney, brain, and heart was determined by the coupled assay of Paglia and Valenine (1967) using a Shimadzu UV visible recording spectrophotometer (model UV160U). Liver, kidney, brain, and heart were mechanically homogenized (25% homogenate) using Brinkman Polytron homogenizer (Brinkman Instruments Co., Westbury, NY) in 50 mm phosphate buffer containing 10% sucrose, ph 6.3. The homogenates were centrifuged at 4 for 90 min at 110,000 g (Ti 50 rotor, Beckman 138

154 Model L7-65, Palo Alto, CA). The supernatants (cytosol) were removed and used to measure GPx activity. GPx activity was calculated based on the decreasing rate of NADPH absorbance at 340 nanometers. The activity was expressed as units per gram of protein. 1 unit of activity is equivalent to 1 µmol NADPH oxidized per minute. The detailed procedure for GPx activity assay is summarized in Appendix H. Protein and hemoglobin concentration: The protein concentration in plasma and cytosol fraction of tissues was measured spectrometrically at 562 nanometers by bicinchoninic acid method (Pierce, Rockford, IL) using bovine serum albumin as a standard. The total hemoglobin concentration of RBC was determined spectrometrically at 540 nanometers using Drabkin s reagent (Drabkin, 1949). Plasma selenoprotein P concentration: Plasma SelP concentration was determined by a competitive radioimmunoassay in the laboratory of Dr. Kristina E. Hill, Vanderbilt University, Nashville, Tennessee, using a modified method of Hill et al. (1996a). 75 Se-SelP was obtained from a selenium-deficient rat injected with 75 Se-selenite 18 h previously. Polyclonal antiserum 143 was raised in New Zealand White female rabbits immunized with rat selenoprotein P. Rabbit antiserum 143 (100 µl) was mixed with 75 Se-labeled rat SelP (100 µl) and the unknown rat plasma samples (100 µl) in phosphate-buffered saline (PBS), ph 7.4. After overnight incubation at 4, 500 µl goat 139

155 anti-rabbit IgG precipitating antiserum was added. The mixture was centrifuged at 3,500 g for 1h after incubation at room temperature for 1h. The supernatant was removed, and 75 Se activity in the pellet was determined. Plasma SelP concentration was determined using a standard curve constructed daily using standard rat plasma (Harlan, Indianapolis, Indiana). Plasma 17β-estradiol concentration: Concentration of 17β-estradiol in plasma was determined using a double-antibody estradiol 125 I radioimmunoassay kit (Diagnostic Products Corp., Los Angeles, CA). Plasma 17β-estradiol was measured by the ability of the sample to compete with 125 I-labeled 17β-estradiol for antibody sites. 125 I activity was determined with a Cobra II auto-gamma counter (Packard Instrument Company, Meriden, CT). The detailed procedure for the measurement of plasma 17β-estradiol concentration is summarized in Appendix E. Plasma ceruloplasmin activity: Ceruloplasmin activity in plasma was measured by the method developed by Schosinsky et al. (1974) with o-dianisidine dihydrochloride as substrate. The substrate is converted to a yellowish-brown product in the presence of ceruloplasmin and oxygen at ph 5.5. Enzyme activity was terminated by addition of 9M sulfuric acid. Absorbance of the purple-red solution was determined spectrometrically at 540 nanometers. 140

156 RNA extraction: Total RNA was extracted from liver using RNAqueous-4PCR kit as described in the manufacture s manual (Ambion, Austin, Texas). Briefly, mg tissue was disrupted in a 1.5 ml microfuge tube using a small plastic pestle. Ten to 12 volumes of lysis buffer were added to homogenate. After mixing with an equal volume of 64% ethanol, lysates were transferred to an Ambion column and centrifuged at 10,000 g for 1 minute. The column was washed once with 700 μl wash buffer #1 and twice with 500 μl wash buffer #2/3. Elution buffer (50 μl) was added to column lysate with collection of eluate by centrifugation for 30 seconds at 10,000 g. The procedure was repeated and elutes pooled. RNA was then treated with DNase I for 25 minutes in a 37 C water bath to remove genomic DNA contamination. DNase I was removed by adding 0.1 volume DNase inactivation reagent. RNA was collected by centrifugation at 10,000 g for 1 minute. Purity and concentration of the RNA was estimated by measuring absorbance at 260 nm and 280 nm. A range of the ratio (A 260 : A 280 ) between 1.8 and 2.1 suggests high purity. The RNA integrity was confirmed by electrophoresis on 1% agarose gel, then stained with ethidium bromide (0.5 µg/ml TAE) solution for 20 minutes before visualizing by ultraviolet fluorescence. The presence of clear 28S and 18S rrna bands indicates intact RNA (Figure 4.1). A polymerase chain reaction (PCR) was performed to check if there was DNA contamination in extracted RNA samples. The detailed procedure of the PCR is described in Appendix I. 141

157 1kb ladder 28S-rRNA 18S-rRNA Figure 4.1 RNA isolated from SD female rats using RNAqueous-4PCR. After electrophoresis on 1% agarose gel, gel was stained at ethidium bromide and was visualized by ultraviolet fluorescence. The presence of clear 28S and 18S rrna bands indicates that RNA was intact. 142

158 Reverse transcription-polymerase chain reaction (RT-PCR): First stand cdna was synthesized from 1 µg RNA sample using the SuperScript First-Strand Synthesis System for RT-PCR (Invitrogen, Carlsbad, CA). In order to check the quality of cdna and whether the designed primers work in the PCR assay, cdna product was amplified by regular PCR using the procedure described in Appendix I. The published primers for SelP (Tabuchi et al., 2005), GPx1(Knoll et al., 2005), and GAPDH (Mikula et al., 2003) (Table 4.1) were used in the present study. The primers were synthesized by Alpha DNA (Montreal, Quebec, Canada). 1 µl of the first strand cdna was used as template in 50 µl of reaction buffer. The PCR products were checked by electrophoresis on 1.5% agarose gel at 100 V for 50 minutes. The gel was stained with ethidium bromide (0.5 µg/ml TAE) solution for 20 minutes and was visualized by ultraviolet fluorescence to verify single product formation at the expected MW. PCR products were purified by using QIAquick PCR purification kit (Qiagene; Santa Clarita, CA) according to the manufacturer s instructions (eluted volume 30 µl). Purified PCR products were quantified using the Quant-iT TM DNA assay kit (Molecular Probes, Eugene, OR) by reading fluorescence of the standards and the PCR products using an Mx3000P QPCR system (Stratagene, La Jolla, CA). 143

159 Gene GenBank Primer/probe Sequence, 5' 3 Amplicon Accession No. (bp) SelP a X99807 F: AGCACAGGCAGGGTCACTTAGA 95 R: TGCACCCCCTTCGTCAGA P: FAM-CTTGCACCTTTCACTTGCCCAGAGGA-BHQ-1 GPx1 b X07365 F: TGAGAAGTGCGAGGTGAATGG 70 R: GTGCTGGCAAGGCATTCC P: FAM-AAGGCTCACCCGCTCTTTACCTTCCTG-BHQ GAPDH c M32559 F: CCAGAACATCATCCCTGCAT 66 R: GTTCAGCTCTGGGATGACCTT P: HEX-CACTGGTGCTGCCAAGGCTGTG-BHQ-2 Table 4.1 Primer and probe sequences used for amplication of the SelP, GPx1, and GAPDH genes in rat liver. F, forward primer; R, reverse primer; P, probe. a : sequences published by Tabuchi et al. (2005). b : sequences published by Knoll et al. (2005). c : sequences published by Mikula et al. (2003).

160 Real-time PCR assay: Quantitative real-time PCR analysis was performed by using the specific primers and probes listed in Table 4.1. The Master Mix contained 10 PCR buffer, 50 mm MgCl 2, 3.36% bovine serum albumin, 100 mm of each deoxynucleoside triphosphate (dntp mixture), 5 U/µl platinum Taq DNA polymerase (Invitrogen, Carlsbad, California), 100 µm forward primer, 100 µm reverse primer, and 100 µm probe (Alpha DNA, Montreal, Quebec, Canada), in a volume of 24 µl. 1 µl of sample was added into the Master Mix. The reaction was performed in triplicate in clear 96-well plates using an Mx3000P QPCR system (Stratagene, La Jolla, CA). Non-template controls (NTC) were included to determine baseline noise. Temperature cycling conditions were 95 C for 2 min, 45 cycles of 95 C for 30 s, 60 C for 1 min. Housekeeping gene GAPDH was used as the internal standard. Two approaches, absolute quantification and relative quantification, have been used to determine the SelP/GAPDH, and GPx1/GAPDH. In the absolute quantification analysis, a standard curve was made from a dilution series of purified PCR products. Gene copies for each standard are plotted in log scale on the X-axis. Threshold cycles (Ct, i.e., the amplification cycle at which fluorescence is significant above the background signal) are plotted on the Y-axis, The copies of the SelP, GPx-1, and GAPDH were obtained from the standard curve, and the ratios of SelP/GAPDH, and GPx-1/ GAPDH were calculated. Figure 4.2 shows an example of the plate setup for the absolute quantification of genes (GAPDH). Figure 4.3 shows an example of using the standard curve to quantitate gene copies in real-time PCR. 145

161 In the relative quantification analysis, no standard curve was made. The ratios of SelP/GAPDH, and GPx-1/GAPDH were calculated by the differences in Ct value between the SelP(or GPx) and GAPDH. Figure 4.4 shows an example of the plate setup for the relative quantification of genes (SelP/GAPDH) Statistical analysis All data were reported as means±sem. Differences in selenium concentration, GPx activity, plasma estrogen and ceruloplasmin activity among the 6 groups (OVX, OVX+E2, proestrus, estrus, metestrus, diestrus) was assessed by one-way ANOVA. Differences were considered significant at P<0.05. When analysis of variance indicated significant differences among the means, Tukey s post hoc probability test was used to determine which means were significantly different. The differences of liver mrna levels of SelP and GPx between OVX group and OVX+E2 group were determined by Student s t-test. P< 0.05 was considered significant. The statistical analyses were performed using SPSS V15.0 (SPSS Inc., Chicago, IL). 146

162 Figure 4.2 Plate setup for the absolute quantification of the real-time PCR analysis. A standard curve was made to determine the copies of gene in the cdna samples. The reaction was performed in triplicate in clear 96-well plates using an Mx3000P QPCR system. NTC, no template control. 147

163 Figure 4.3 Determination of copy number for GPx1 mrna by using a standard curve. The log of the initial template quantity was plotted against the Ct values. The copies of the gene in the samples were determined from the standard curve by its Ct values. The reaction was performed in triplicate in clear 96-well plates using an Mx3000P QPCR system. 148

164 Figure 4.4 Plate setup for the relative quantification of real-time PCR analysis. Each well contained two sets of primers and probes. The reaction was performed in triplicate in clear 96-well plates using an Mx3000P QPCR system. NTC, no template control. 149

165 4.3 RESULTS Body and organ weights There were no significant differences in body weight among the three experimental groups prior to surgery at 7 weeks of age (P > 0.05, Figure 4.5). After surgery, rats in the OVX group had the most rapid increase in body weight, followed by Sham group, and then the OVX+E2 group despite all animals ingesting identical quantities of food. At week 8, 9, and 10, there was no difference in body weight between OVX group and Sham group (P > 0.05), but body weight in both groups was significantly greater than for OVX+E2 group (P < 0.05). At week 11, and 12, body weight was significantly different for the three groups (P < 0.05) with OVX > Sham > OVX+E2. Weights of liver, kidney, heart, and brain normalized to total body weight (g/kg BW) are shown in Figure 4.6. Relative weights of kidney and brain in OVX+E2 group were significantly greater than in OVX and Sham groups. Relative weight of liver in OVX+E2 group also was greater than in Sham group. Relative weight of heart was not statistically different among the three groups. 150

166 Body Weight (g) OVX+E2 OVX Sham Surgery Week Figure 4.5 Body weights for female SD rats before and after surgery. The surgery was performed at 7 weeks of age. Rats were randomized into three treatment groups: OVX (n=6), OVX+E2 (n=6), and Sham (n=24). Data are means ± SEM. At week 8, 9, and 10, body weight in OVX+E2 group was significantly smaller than that of the other two groups (P < 0.05). At week 11 and 12, body weight of OVX>Sham>OVX+E2 (P < 0.05). 151

167 a ab b OVX+E2 OVX Sham g/kg BW a b b a b b a a a liver kidney heart brain Figure 4.6 Organ weights (g/kg BW) for female SD rats from three groups. The experimental groups are: OVX (n=6), OVX+E2 (n=6), and Sham (n=24). Data are means ± SEM and different letters above each bar indicate significant differences between experimental groups (P < 0.05). 152

168 4.3.2 Plasma 17β-estradiol concentrations and ceruloplasmin activity Ovariectomy significantly reduced 17β-estradiol concentration in plasma (Figure 4.7). Plasma 17β-estradiol concentrations in the OVX group were 46%, 58%, 62%, and 64% that of the Sham group during proestrus, metestrus, estrus, and diestrus stages, respectively. Estrogen replacement with the subcutaneous pellets significantly increased 17β-estradiol in plasma to levels greater than in sham operated rats. Plasma 17β-estradiol levels were significantly greater during proestrus than those in other stages of the estrus cycle (P<0.05) in sham-operated rats. There were no significant differences in plasma 17β-estradiol levels among sham operated rats during metestrus, diestrus, and estrus (P>0.05). Estrogen replacement significantly increased plasma ceruloplasmin activity (Figure 4.8) and the activity was significantly greater than that of the sham operated rats and OVX rats (P<0.05). There were no differences in the plasma ceruloplasmin activity between OVX rats and those sham-operated rats in all estrus stages except proestrus. There was a positive association (r=0.78, P<0.01) between 17β-estradiol concentration and ceruloplasmin activity in plasma. 153

169 80 d 60 pg/ml 40 b c b b 20 a 0 OVX+E2 OVX D P E M Figure 4.7 Plasma concentration of 17β-estradiol in different groups of female SD rats. The groups are: OVX with estrogen replacement (OVX+E2); OVX with placebo replacement (OVX); sham-operated with placebo replacement in diestrus (D); proestrus (P); estrus (E); and metestrus (M) stages. Data are means ± SEM for 6 animals in each group. Presence of different letters above each bar indicates significant differences (P<0.05). 154

170 150 a bc c bc bc U/L 60 b 30 0 OVX+E2 OVX D P E M Figure 4.8 Plasma ceruloplasmin activity in different groups of female SD rats. The groups are: OVX with estrogen replacement (OVX+E2); OVX with placebo replacement (OVX); sham-operated with placebo replacement in diestrus (D); proestrus (P); estrus (E); and metestrus (M) stages. Data are means ± SEM for 6 animals in each group. Presence of different letters above each bar indicates significant differences (P<0.05). 155

171 4.3.3 Tissue selenium concentrations and GPx activity Estrogen status affected selenium concentrations in plasma, RBC, liver, and brain (Table 4.2). Plasma selenium concentration was significantly greater in OVX+E2 than sham-operated rats; OVX rats had significantly lower plasma selenium than sham-operated rats at all stages of estrus cycle. Among sham-operated rats, plasma concentrations were greatest during proestrus. Plasma selenium levels were not significantly different during metestrus, diestrus, and estrus stages. Selenium concentrations in RBC, liver, and brain were significantly higher in OVX+E2 rats than sham-operated rats at all stages of the estrus cycles, except proestrus. OVX rats had significantly lower selenium concentrations in RBC, liver, and brain compared to OVX+E2 and sham-operated rats in the proestrus stage. Selenium concentrations in kidney and heart were not affected by plasma estrogen. However, there was a trend that selenium concentration was greater in kidney of OVX rats than in Sham rats, and lowest in OVX+E2 rats (P=0.061 for ANOVA test). 156

172 OVX+E2 OVX Diestrus Proestrus Estrous Metestrus Plasma, ng/ml 443 ± 16 a 256 ± 6 d 325 ± 18 c 382 ± 10 b 321 ± 12 c 321 ± 14 c RBC, ng/ml 642 ± 46 a 362 ± 26 c 411 ± 45 bc 555 ± 19 ab 384 ± 31 c 402 ± 36 c Liver, ng/g 1,760 ± 37 a 1,295 ± 44 c 1,505 ± 67 bc 1,692 ± 27 ab 1,482 ± 68 bc 1,370 ± 67 c Kidney, ng/g 1,357 ± 27 a 1,624 ± 47 a 1,563 ± 97 a 1,461 ± 15 a 1,587 ± 87 a 1,494 ± 62 a Heart, ng/g 423 ± 25 a 411 ± 19 a 405 ± 22 a 450 ± 14 a 434 ± 21 a 432 ± 13 a Brain, ng/g 441 ± 9 a 344 ± 9 c 363 ± 21 bc 421 ± 16 ab 363 ± 21 bc 366 ± 21 bc 157 Table 4.2 Effect of estrogen status on tissue selenium concentration in female SD rats. Test groups include OVX with estrogen replacement (OVX+E2), OVX with placebo replacement (OVX), sham-operated with placebo replacement in diestrus (D), proestrus (P), estrus (E), and metestrus (M) stages. Data are means ± SEM for 6 animals in each group and those not sharing a common superscript among estrogen treatment are significantly different (P<0.05).

173 Tissue GPx activity is shown in Table 4.3. Plasma estrogen affected GPx activity in plasma, liver, and brain, but not in RBC, kidney, and heart. Plasma and liver GPx activity in OVX+E2 rats were significantly greater than those in OVX rats and in sham-operated rats in all estrous stages other than proestrus. There were no significant differences of plasma and liver GPx activity between OVX rats and sham-operated rats in all estrous stages other than proestrus. OVX was associated with decreased GPx activity in brain, with significant lower GPx activity than that of sham-operated rats in proestrus stage and OVX+E2 rats (P<0.05). OVX+E2 rats had greater GPx activity in brain than that of sham-operated rats in all estrous stages except for proestrus (P<0.05) Plasma selenoprotein P concentration Plasma selenoprotein P (SelP) concentrations are shown in Figure 4.9. No differences were observed among groups for plasma SelP concentration, suggesting that estrogen availability did not affect the protein levels of SelP in plasma. However, there was a positive association (r=0.62, P=0.01) between plasma 17β estradiol concentration and SelP concentration in sham-operated rats. 158

174 OVX+E2 OVX Diestrus Proestrus Estrous Metestrus Plasma 146 ± 10 a 70 ± 6 b 71 ± 12 b 119 ± 10 a 75 ± 11 b 75 ± 6 b RBC 470 ± 12 a 410 ± 39 a 427 ± 16 a 460 ± 42 a 419 ± 18 a 417 ± 43 a Liver 1,702 ± 60 a 1,202 ± 90 b 1,230 ± 66 b 1,658 ± 82 a 1,265 ± 96 b 1,290 ± 90 b Kidney 272 ± 22 a 364 ± 48 a 364 ± 43 a 291 ± 18 a 316 ± 26 a 321 ± 41 a Heart 289 ± 31 a 242 ± 27 a 250 ± 21 a 328 ± 26 a 296 ± 24 a 250 ± 34 a Brain 199 ± 9 a 102 ± 7 c 133 ± 13 bc 169 ± 5 ab 135 ± 14 bc 137 ± 17 bc 159 Table 4.3 Effect of estrogen status on tissue glutathione peroxidase (GPx) activity in female SD rats. Test groups include OVX with estrogen replacement (OVX+E2), OVX with placebo replacement (OVX), sham-operated with placebo replacement in diestrus (D), proestrus (P), estrus (E), and metestrus (M) stages. GPX activity is expressed as U/g protein. 1 unit of activity is equivalent to 1 µmol NADPH oxidized per minute. Data are means ± SEM for 6 animals in each group and those not sharing a common superscript among estrogen treatment are significantly different (P<0.05).

175 30 20 µg/ml 10 0 OVX+E2 OVX D P E M Figure 4.9 Estrogen status does not affect plasma selenoprotein P (SelP) concentration. Groups of female Sprague-Dawley rats include OVX with estrogen replacement (OVX+E2), OVX with placebo replacement (OVX), sham-operated with placebo replacement in diestrus (D), proestrus (P), estrus (E), and metestrus (M) stages. Data are means ± SEM for 6 animals in each group. Plasma SelP concentration was not significantly different among groups. 160

176 4.3.5 Hepatic levels of SelP and GPx mrna Two approaches, absolute quantification and relative quantification, have been used to determine the levels of SelP mrna and GPx1 mrna in liver. The results from both approaches were consistent. Using relative quantification, the level of SelP mrna was normalized to GAPDH mrna and was 88% higher in OVX+E2 rats compared to OVX rats (0.94 vs. 0.50, P<0.05) (Figure 4.10). Similarly, using absolute quantification, the level of SelP mrna was 74% higher in OVX+E2 rats than that of the OVX rats (0.75 vs. 0.43, P<0.05). Levels of GPX1 mrna in liver were also affected by estrogen availability (Figure 4.11). The level of GPX1 mrna normalized to GAPDH mrna was significantly (P<0.05) greater in OVX+E2 rats compared to OVX rats, using either relative quantification (2.8 vs. 2.2), or absolute quantification (2.9 vs. 2.2). 161

177 SelP/GAPDH * * OVX OVX+E Relative Q Absolute Q Figure 4.10 Estrogen replacement in OVX rats increases hepatic SelP mrna. RT reaction was carried out with extracted RNA isolated from liver of OVX rats receiving estrogen replacement (OVX+E2) and OVX rats with placebo pellet (OVX). Real-time qpcr was performed using an Mx3000P QPCR system. mrna levels were normalized by GAPDH. Data are means ± SEM for 6 animals in each group. Relative Q, data obtained by relative quantification. Absolute Q, data obtained by absolute quantification. * P<0.05 between OVX and OVX+E2 by Student s t test. 162

178 GPx1/GAPDH * * OVX OVX+E2 1 0 Relative Q Absolute Q Figure 4.11 Estrogen replacement in OVX rats increases hepatic GPx1 mrna. RT reaction was carried out with extracted RNA isolated from liver of OVX rats receiving estrogen replacement (OVX+E2) and OVX rats with placebo pellet (OVX). Real-time qpcr was performed using an Mx3000P QPCR system. mrna levels were normalized by GAPDH. Data are means ± SEM for 6 animals in each group. Relative Q, data obtained by relative quantification. Absolute Q, data obtained by absolute quantification. * P<0.05 between OVX and OVX+E2 by Student s t test 163

179 4.4 DISCUSSION An ovariectomized rat model was used in this study to investigate the effect of estrogen status on selenium metabolism. Estrogen status was manipulated by subjecting female Sprague Dawley (SD) rats to the following treatments: 1) ovariectomy with estrogen replacement (OVX+E2); 2) ovariectomy with a placebo pellet (OVX); and 3) sham operation with a placebo pellet (Sham). Estrogen levels were verified by measuring 17β-estradiol concentrations in plasma. While ovariectomy resulted in a significant reduction of 17β-estradiol levels, implantation with estrogen pellets elevated 17β-estradiol to a concentration greater than that present throughout the estrus cycle in sham-operated rats (Figure 4.7). Estrogen treatment also suppressed body weight gain compared to that of the Sham and OVX rats (Figure 4.5). This result agrees well with observations from other studies (Clinton et al., 1995; Roesch, 2006; Ke et al., 1995; McCormick et al., 2004, Latour et al., 2001). Previous studies have demonstrated that the weight gain following ovariectomy and its suppression in response to estradiol treatment are due mainly to altered food intake. In our study, the food intake was 12.5 g/day and 16.4 g/day in OVX+E2 rats and OVX rats, respectively, at 7 weeks of age immediately after the surgery and implantation of pellets with and without 17β-estradiol. Cholecystokinin (CCK) released from the small intestine during meals is an endocrine negative-feedback signal controlling meal size (i.e. satiation) in rats (Gibbs et al., 1973). 164

180 Estradiol increases the activity of CCK satiation-signaling pathway, resulting in decreased food intake (Geary, 2001). As the rats in OVX group consumed the same amount of diet as that of the OVX+E2 group (pair-fed) in the present study, food intake is not the factor that caused the differences in body weight. The observation that rats in OVX+E2 group were more active than those in OVX and Sham group suggests that decreased weight gain in OVX+E2 group may result from increased physical activity. Others have reported increased and decreased activity levels in female rats during the proestrus (elevated blood estrogen levels) and diestrus stages (blood estrogen levels reduced), respectively (Brobeck et al., 1947; Colvin and Sawyer, 1969). Plasma ceruloplasmin, a copper containing protein, has been shown to be upregulated by estrogen (Canaraja et al., 2004; Alkjaersig et al., 1988). Thus, plasma ceruloplasmin activity was measured in the present study as a positive control for estrogen-mediated alteration of trace element metabolism. As expected, estrogen treatment significantly increased the plasma ceruloplasmin activity in rats (Figure 4.8). Data for body weight, dietary intake, plasma estradiol concentration, and plasma ceruloplasmin activity confirm that the rat model was appropriate to examine impact of estrogen status on selenium content and utilization in tissues. The results of the present study demonstrate that measures of selenium status are affected by estrogen in blood. Estrogen treatment significantly increased plasma GPx activity and RBC selenium compared to these parameters in sham-operated and OVX 165

181 rats (P<0.05). These results agree with data from the 75 Se tracer study (Chapter 3), in which OVX+E2 rats had greater 75 Se activity in plasma GPx and RBC 24 hours after administered 75 Se-selenite by gavage. Plasma selenium, plasma GPx activity and RBC selenium fluctuated throughout the estrus cycle in sham-operated rats. RBC GPx activity was not affected by either ovariectomy or estrogen treatment, and the levels were relatively constant during the estrus cycle in sham-operated rats (Table 4.3). Previous studies support the observations that plasma selenium and GPx activity fluctuated during rat estrous cycle (Smith et al., 1995), and human menstrual cycle (Ha and Smith, 2003). However, data on selenium status in RBC is inconclusive. An earlier study in chronically-catheterized SD rats demonstrated that RBC concentration, but not RBC GPx activity, fluctuated throughout the estrus cycles (Smith et al., 1995), which agrees our observation. In humans, cyclic variations in RBC GPx activity has been observed during the human menstrual cycle (Ha and Smith, 2003, Massafra et al., 1998). However, Larsen et al. (1996) failed to observe any phase-related fluctuation in RBC GPx activity during the menstrual cycle. The inconclusive data on RBC GPx activity may be due to the followings. First, there are no uniform units in the RBC GPx assay (Diplock, 1993), and no standardized assay method, causing wide variability between laboratories (Litov and Comb, 1991). Second, the distribution of selenium between GPx and hemoglobin is affected by the form of dietary selenium in RBC (Bulter et al., 1991). It has been shown that more selenium was associated with GPx in women ingesting selenate than in those 166

182 ingesting selenomethionine. Thus, the inconsistent data may due to different dietary form of selenium used. Third, RBC has a life span of 120 days in human and 61 days in SD rats (Derelanko, 1987), and the human menstrual cycle is days, which is much longer that 4-day estrous cycle in rat. These differences may contribute to different results in human and rat studies. Previous studies with rats have shown that liver GPx activity in females is greater than males (Pinto and Bartley, 1969; Capel and Smallwood, 1983; Igarashi et al., 1984; Finley and Kincaid, 1991; Prohaska and Sunde, 1993; Yamamoto et al., 2002). These observations suggested that hormonal differences may be responsible for the differences. The present study directly examined the effect of estrogen availability on liver selenium status. Both selenium concentrations and GPx activity in the liver were greater when estrogen was available. As liver is the central organ for selenium metabolism, the effect of estrogen on hepatic selenium status has important implications for selenium metabolism. Interestingly, we observed a trend that selenium concentration in kidney was inversely related to estrogen status, with OVX rats having the highest selenium concentration followed by Sham rats, and OVX+E2 having the lowest selenium concentration (P=0.061 for ANOVA test). Kidney is the primary organ for synthesis of extracellular plasma GPx (egpx) (Chu et al., 1992). Estrogen treatment may enhance renal synthesis of egpx, markedly reducing selenium content in kidney. 167

183 For the first time, we demonstrated that estrogen upregulated liver mrna levels of both SelP and GPx1. There was an increase of SelP mrna (88%, Figure 4.10) and GPx1 mrna (32%, Figure 4.11) in OVX+E2 rats compared to OVX rats. The upregulation of hepatic GPx1 mrna by estrogen may lead to more GPx1 synthesis, and thus higher hepatic GPx activity in OVX+E2 rats. Although hepatic SelP mrna was upregulated by estrogen, there was no difference of plasma SelP concentration between OVX+E2 and OVX groups. This may be due to more SelP was in other tissues, and thus less SelP remained in the plasma. It is known that estrogen can modulate the expression of estrogen-dependent genes that have estrogen response elements (ERE) in their promoter region. Estrogen binding to the estrogen receptor (ER) activates ER, which allows dimerization of the ER, dissociation of heat shock protein (HSP90), and association of co-regulatory proteins. Upon binding to estrogen response elements (ERE) in the promoter of the estrogen dependent gene, the estrogen-estrogen receptor-cofactor complex stimulates or inhibits gene transcription (Katzenellenbogen et al., 1996; McKenna et al., 1999). Of the four types of GPx, GPx3 (Borthwick et al., 2003) and GPx4 (Brigelius-Flohe et al., 1994) have EREs in the promoter region. By analyzing the gene sequence obtained from the GeneBank database using BioEdit Sequence Alignment Editor software (V , Hall, 1999), we found that GPx1, GPx2, and SelP do not contain EREs, but do contain ERE half-sites (5'-AGGTCA-3') in the promoter regions. Although a number of genes that have regulatory ERE half-sites are responsive to 168

184 estrogen, ER binding to an ERE half-site as a monomer in some genes remains controversial (Das et al., 2004). Thus, whether estrogen regulates GPx1 and SelP by binding of ER to ERE-half site remains to be investigated. For genes having no ERE, estrogen may regulate their expression by a non-classical pathway. Estrogen binding to ER activates transcription factors such as AP1, Sp1 and NF-kB. These proteins bind to cognate recognition sequences in DNA directly and influence the transcription of genes (Cerillo et al., 1998; Porter et al., 1997; McKay and Cidlowski, 1998). SelP promoter contains 2 binding sites for AP1, and 1 binding site for Sp1 (Al-Taie et al., 2002; Dreher et al., 1997), but no binding site for NF-kB. In the present study, estrogen may also have regulated transcription of GPx1 and SelP through binding to AP1 and Sp1. Recent evidence indicates that SelP functions as a selenium transport protein. Deletion of SelP resulted in sharp decrease of selenium concentration in murine testis and brain, suggesting that testis and brain mainly rely on SelP for the supply of selenium (Hill et al., 2003; Schomburg et al., 2003). A modest decrease in selenium concentration in kidney and other tissues suggests that these tissues also depend on both SelP and other transport forms of selenium (e.g. low molecular weight species of selenium) to supply selenium. The present study demonstrated that levels of SelP mrna in liver were upregulated by estrogen treatment (Figure 4.10) and plasma SelP concentration was positively associated with plasma 17β estradiol concentration in sham-operated rats (Figure 4.9). It would be interesting to determine SelP concentrations in tissues, because 169

185 this would provide more direct evidence of the relationship between estrogen and SelP. Unfortunately, efforts to measure SelP in tissues have had limited success in immunoprecipitation experiments, either because antibody does not react with the tissue form of the protein, or SelP is processed through the cell very quickly and the concentration is too low to be detected (Personal communication, Dr. Kristina E. Hill and Dr. Raymond Burk at Vanderbilt University). Estrogen upregulated the transcriprion of SelP gene in liver, and incorporation of 75 Se into SelP and GPx in plasma increased by estrogen treatment (Chapter 3). Plasma SelP concentration was not affected by estrogen despite greater 75 Se was incorporated into SelP, this may be due to estrogen facilitated the delivery of selenium by SelP to other tissues, such as brain. These data suggest that estrogen may affect selenium metabolism in selected tissues by regulating synthesis and secretion of the selenium transport protein, SelP. Based on these findings and the fact the there are ERE half sites, AP1 site and Sp1 site in the promoter region of SelP gene, we thus proposed a model on how estrogen regulates whole body selenium metabolism by regulating the synthesis of SelP (Figure 4.12). We hypothesized that estrogen may activate SelP transcription in liver by ER binding to half-site estrogen response elements (ERE) in the promoter region of SelP gene, or by activation of transcription factors such as AP1 and Sp1. And the synthesis and secretion of SelP in liver is upregulated. After SelP is synthesized in liver, SelP may deliver selenium to extrahepatic tissues via a receptor-mediated mechanism, 170

186 i.e., by binding to SelP receptor on the membrane of the tissues, and releasing selenium via receptor-mediated endocytosis and lysosomal degeneration (Richardson, 2005). In summary, the present study demonstrated that estrogen significantly increased selenium status, as measured by selenium concentration and GPx activity, in plasma, liver, and brain. Selenium concentration in RBC was also increased by estrogen treatment. Estrogen upregulated the hepatic levels of GPx1 mrna and SelP mrna. Mechanisms responsible for this upregulation are not clear. Gel mobility shift assay studies and reporter gene assays are needed to determine whether this upregulation is mediated by the interaction of ER with ERE half-sites in the promoter region of the genes or by ER induced transcription factors such as AP1 and Sp1. 171

187 Liver cell E 2 Nucleus ER Transcription AP1, Sp1 Coactivator Half-site ERE? AP1, Sp1 site? SelP gene SelP mrna Translation 75 Se increased, Se and GPx increased SelP Protein Plasma: Se and GPx increased RBC: 75 Se increased; Se increased. Blood SelP-R Extrahepatic Tissues, e.g. brain Figure 4.12 Hypothetical model of estrogen regulation whole body selenium metabolism. Estrogen binding to estrogen receptor (ER) activates ER. ER may activate SelP gene transcription in liver by binding to half-site estrogen response elements (ERE) in the promoter region of SelP gene, or by activation of transcription factors such as AP1 and Sp1. After synthesis in liver, SelP may deliver selenium to extrahepatic tissues via a receptor-mediated mechanism. 172

188 CHAPTER 5 EPILOGUE The association between estrogen and selenium status has been documented in both animal and human studies. The present study is the first in which the effect of estrogen on selenium metabolism has been investigated in multiple tissues. Estrogen status was manipulated by implanting pellets either with estrogen (OVX+E2), or without estrogen (OVX) in ovariectomized Sprague Dawley (SD) rats, or implanting placebo pellets in sham-operated rats. The purpose of the first study was to investigate the effect of estrogen on the absorption, tissue distribution and metabolism of orally administered 75 Se-selenite. The effect of estrogen on selenium metabolism was investigated at the tissue, subcellular, and protein levels. Although the apparent absorption of 75 Se was independent of estrogen status, 75 Se activity differed in plasma, RBC, liver, heart, kidney, spleen, brain, and thymus at various times after gavage. The incorporation of 75 Se into plasma SelP and GPx was increased in animals with elevated plasma levels of estrogen at selected times after dosing. Plasma selenoprotein P (SelP) in OVX+E2 group contained a greater 173

189 percentage of administered 75 Se at 3, 6 and 24h after gavage compared to OVX group. 75 Se in plasma glutathione peroxidase (GPx) also was greater in OVX+E2 compared to OVX group at 24h. The second study was performed to determine the effect of estrogen on selenium status in tissues and hepatic levels of SelP mrna and GPx1 mrna. Estrogen significantly increased selenium concentration and GPx activity in plasma, liver, and brain. Selenium concentration in RBC was also increased by estrogen treatment. Real-time RT-PCR analysis demonstrated that both hepatic SelP and GPx1 mrna were significantly upregulated by estrogen treatment. The results of the present study strongly support a role of estrogen on the regulation of selenium metabolism, especially, its effect on the selenium distribution among tissues. Animals may initiate the physiological response under certain healthy status via the action of hormones, including the reproductive hormones. Selenium status in blood and some tissues decreased in male animals (Capel and Smallwood, 1983; Finley and Kincaid, 1991; Prohaska and Sunde, 1993; Yamamoto et al., 2002) and humans (Marano et al., 1991) during sexual maturation, while increased in the testis. Testosterone has been shown to mediate the sharp rise of selenium concentration in rat testis during puberty (Maiorino et al., 1998). This redistribution of selenium provides more selenium for incorporation into PHGPx, a selenoprotein carrying an important role in spermatogenesis (Vogt, 2004). Similar to testosterone, estrogen likely has a central role in regulating 174

190 selenium status in females at certain stages of the lifecycle, such as pregnancy, menstrual cycle, and menopause. Indeed, estrogen has been shown to influence the metabolism of several minerals, including copper (Canaraja et al., 2004; Alkjaersig et al., 1988; Bureau et al., 2002, Dorea and Miazaki, 1999), iron (Milman et al., 1992), magnesium, calcium (Muneyyirci-Delale et al., 1998), and chromium (Bureau et al., 2002). Thus, it is not surprising that estrogen is also a regulator of selenium metabolism. Sex-based differences in the anticarcinogenic effects of selenium have been observed in some prospective epidemiology studies. Water et al. (2004) hypothesized that cancer risk in men is more profoundly influenced by selenium status than cancer risk in women. They further pointed out that sex-based differences in metabolism or tissue distribution of selenium could be one of the factors that cause the differences. The present study has demonstrated a role of estrogen in selenium metabolism. Thus, estrogen status should be considered when developing effective strategies to assess selenium status, and for the possible use of selenium as a chemopreventive or chemotherapeutic agent for cancers, especially estrogen-responsive cancers of the breast and ovaries. The mechanisms on how estrogen affects selenium metabolism are not clear. Recent studies have demonstrated that SelP acts as a selenium transport protein to mobilize selenium from liver to other tissues (Hill et al., 2003; Schomburg et al., 2003). It has been shown in this study that estrogen upregulated hepatic SelP gene expression and 175

191 enhanced the incorporation of 75 Se into plasma SelP, suggesting that estrogen may affect selenium metabolism in selected tissues by regulating synthesis and secretion of the selenium transport protein, SelP. A hypothetical model on how estrogen regulates whole body selenium metabolism by regulating the synthesis of hepatic SelP has been proposed (Figure 4.12). Estrogen is known to modulate the expression of estrogen-dependent genes that have one or more estrogen response elements (ERE) at the promoter region. It is likely that GPx3 and GPx4 are regulated by estrogen via this pathway since both genes contain ERE in their promoter regions. In the present study, liver mrna levels of GPx1 and SelP were upregulated by estrogen treatment. However, GPx1 and SelP only contain ERE half-sites in the promoter region. To date, there is still controversy regarding whether ER binds to ERE half-sites as a monomer to modulate transcriptional activity. Estrogen could also regulate gene transcription by ER-mediated modulation of the amounts and/or activities of transcription factors such as AP1, Sp1 and NF-kB. SelP promoter itself contains 2 binding sites for AP1, and 1 binding site for Sp1 (Al-Taie et al., 2002; Dreher et al., 1997), but no binding site for NF-kB. In the present study, estrogen may also have regulated transcription of GPx1 and SelP indirectly through binding of the E2: ER complex to AP1 and Sp1. Further studies are needed to determine whether estrogen regulates GPx1 and SelP expression by binding of ER to an ERE-half site, or through other pathways, e.g. modulation of transcription factors such as AP-1 and Sp1, which in 176

192 turn may affect expression of selective selenoproteins. The following studies may be conducted to explore the mechanism on estrogen regulation of SelP gene. First, ERE-half, AP1, or Sp1 binding sites which are involved in estrogen mediated activation of SelP promoter will be determined by luciferase reporter assay. Plasmid with the full length of SelP promoter and the SelP promoter-deletion constructs (deletion of ERE-half, AP1, or Sp1 sites singularly and in combinations) will be prepared. Luciferase reporter vector with the full length SelP promoter (positive control) or promoter-deletion constructs will be transfected into cells, e.g., HepG2 cells. Cells will be treated with 10 nm E2. The cells will be lysed and the substrate of luciferase, luciferin, is introduced into the cellular extract along with Mg and excess ATP. The luciferase activities and SelP mrna levels of the transfected cells will be determined. It is expected that cells transfected with plasmid containing the full length SelP promoter will have greater luciferase activity and SelP mrna levels compared to the promoter-deletion constructs. A significant decrease of the luciferase activity and SelP mrna of the promoter-deletion constructs indicates that the deleted site is involved in the estrogen mediated activation of SelP promoter. Second, the estrogen effect on selenium status in tissues will be determined using SelP knockout mice model. The SelP knockout mice will be generated by deletion of those binding sites that have been shown to be involved in the estrogen mediated activation of SelP promoter in the above luciferase reporter assay. Selenium 177

193 concentration and GPx activity of blood, liver, kidney, heart, and brain will be measured in both knockout and wild-type mice (SelP+/+ mice, Hill et al., 2003). The SelP+/+ mice will serve as the positive control to assess influence of truncated promoter on selenium metabolism. It is expected that the knockout mice will have lower selenium concentration and GPx activity in blood, liver, kidney, heart, and brain compared to the wild-type mice. These in vitro and in vivo studies together will help to determine whether estrogen regulates SelP gene expression by binding of ER to ERE-half site, or by activation of transcription factors, such as AP1 and Sp1, or by both pathways. 178

194 BIBLIOGRAPHY Akesson B, Bellew T, Burk RF. Purification of selenoprotein P from human plasma. Biochim Biophys Acta 1994; 1204: Alaejow MS, Romero CD. Urinary selenium concentrations. Clin Chem 1993; 39: Alkjaersig N, Fletcher AP, de Ziegler D, Steingold KA, Meldrum DR, Judd HL. Blood coagulation in postmenopausal women given estrogen treatment: comparison of transdermal and oral administration. J Lab Clin Med 1988; 111: Allen NE, Morris JS, Ngwenyama RA, Key TJ. A case-control study of selenium in nails and prostate cancer risk in British men. Br J Cancer 2004; 90: Al-Taie OH, Seufert J, Mörk H, Treis H, Mentrup B, Thalheimer A, Starostik P, Abel J, Scheurlen M, Köhrle J, Jakob F. A complex DNA-repeat structure within the Selenoprotein P promoter contains a functionally relevant polymorphism and is genetically unstable under conditions of mismatch repair deficiency. Eur J Hum Genet 2002; 10: American Cancer Society, Cancer Statistics Atlanta. Anundi I, Hoberg J, Stahl A. Absorption of selenite in the rat small intestine: interactions with glutathione, Acta Pharmacol. Toxicol 1984; 54:273. Arai M, Imai H, Sumi D, Sumi D, Imanaka T, Takano T, Chiba N, Nakagawa Y. Import into mitochondria of phospholipid hydroperoxide glutathione peroxidase requires a leader sequence. Biochem Biophys Res Commun 1996; 227: Arduser F, Wolffram S, Scharrer E. Active absorption of selenate by ileum. J Nutr 1985; 53:

195 Aronica SM, Katzenellenbogen BS. Stimulation of estrogen receptor-mediated transcription and alteration in the phosphorylation state of the rat uterine estrogen receptor by estrogen, cyclic adenosine monophosphate, and insulin-like factor-i. Mol Endocrinol 1993; 7: Arteel GE, mostert V, Oubrahim H, Briviba K, Abel J, Sies H. Protection by selenoprotein P in human plasma against peroxynitrite-mediated oxidation and nitration. Biol Chem 1998; 379: Arthur JR, Morrice PC, Nicol F, Beddows SE, Boyd R, Hayes JD, Beckett GJ. The effects of selenium and copper deficiencies on glutathione S-transferase and glutathione peroxidase in rat liver. Biochem J 1987; 248: Aumais JP, Lee HS, DeGannes C, Horsford J, White JH. Function of directly repeated half-sites as response elements for steroid hormone receptors. J Biol Chem 1996; 271: Avissar N, Ornt DB, Yagil Y, Horowitz S, Watkins RH, Kerl EA, Takahashi K, Palmer IS, Cohen HJ. Human kidney proximal tubules are the main source of plasma glutathione peroxidase. Am J Physiol 1994; 266: C Bahadur U, Ganjam GK, basudevan N, Kondaiah P. Estrogen regulation of chicken riboflavin carrier protein gene is mediated by ERE half sites without direct binding of estrogen receptor. Mol Cell Endocrinol 2005; 231:1-11. Beck MA, Kolbeck PC, Rohr LH, Shi Q, Morris VC, Levander OA. Benign human enterovirus becomes virulent in selenium-deficient mice. J Med Virol 1994; 43: Beck MA, Shi Q, Morris VC, Levander OA. Rapid genomic evolution of a non-virulent coxackievirus B3 in selenium-deficient mice results in selection of identical virulent isolates. Nat Med 1995; 1: Beck MA, Esworthy RS, Ho Y-S, Chu F-F. Glutathione peroxidase protects mice from viral induced myocarditis. FASEB J 1998; 12: Beck MA. Selenium and host defense towards viruses. Proc Nutr Soc 1999; 58: Beck MA. Selenium and vitamin E status: Impact on viral pathogenicity. J Nutr 2007; 137:

196 Behne D, von Berswordt-Wallrabe R, Elger W, Hube G, Wolters W: Effects of pregnancy and lactation on the serum selenium content of rats. Experimentia 1978; 34: Behne D, Wolters W. Distribution of selenium and glutathione peroxidase in the rat. J Nutr 1983; 113: Behne D, Hilmert H, Scheid S, Gessner H, Elger W. Evidence for specific selenium target tissues and new biologically important selenoproteins. Biochim Biophys Acta 1988; 966: Behne D, Kyriakopoulos A, Scheid S, Gessner H. Effects of chemical form and dosage on the incorporation of selenium into tissue proteins in rats. J Nutr 1991; 121: Behne D, Rothlein D, Pfeifer H, Kyriakopoulos A. Identification and characterization of new mammalian selenoproteins. J Trace Elem Med Biol 2000; 14: 117. Behne D, Kyriakopoulos. Mammalian selenium-containing proteins. Annu Rev Nutr 2001; 21: Bednarek-Tupikowska G, Tworowska U, Jedrychowska I, Radomska B, Tupikowski K, Bidzinska-Speichert B, Milewicz A. Effects of oestradiol and oestroprogestin on erythrocyte antioxidative enzyme system activity in postmenopausal women. Clin Endocrinol 2006; 64: Beilstein MA, Whanger PD. Selenium accumulation in tissues, tissue fractions and cytosolic proteins in rats. Biochemical Archives 1985; 1: Beilstein MA, Whanger PD. Glutathione peroxidase activity and chemical forms of selenium in tissues of rats given selenite or selenomethionine. J Inorg Biochem 1988b; 33: Berry M, Metzger D, Chambon P. Role of the two activating domains of the estrogen receptor in the cell-type and promoter-context dependent agonist activity of the antiestrogen 4-hydoxytamoxifen. EMBO J 1990; 9: Berry MJ, Banu L, Chen Y. Recognition of a UGA as a selenocysteine codon in Type I deiodinase requires sequences in the 3 untranslated region. Nature 1991; 353: Bhamre S, Whitin JC, Cohen HJ. Selenomethionine does not affect PSA secretion independent of its effect on LNCaP cell growth. Prostate 2003; 54:

197 Bhattacharya ID, Picciano MR, Milner JA. Characteristics of human milk glutathione peroxidase. Biol Trace Elem Res 1988; 18: Birt DF, Julius AD, Runice CE, Salmasi S. Effects of dietary selenium on bis(2-oxopropyl)nitrosamine-induced carcinogenesis in Syrian golden hamsters. J Natl Cancer Inst 1986; 77: Bock A, Forchhammer K, Heider J, Leinfelder W, Sawers G, Veprek B, Zinoni F. Selenocysteine: the 21st amino acid. Mol Microbiol 1991; 5: Bopp BA, Sonders RC, Kesterson JW. Metabolic fate of selected selenium compounds in laboratory animals and man. Drug Met. Rev 1982; 13(2): Borthwick JM, Charnock-Jones DS, Tom BD, Hull ML, Teirney Raewyn, Phillips SC, Smith S. Determination of the transcript profile of human endometrium. Mol Hum Reprod 2003; 9: Brigelius-Flohe R, Aumann KD, Blocker H, Gross G, Kiess M, Kloppel KD, Maiorino M, Roveri A, Schuckelt R, Usani F. Phospholipid-hydroperoxide glutathione peroxidase. Genomic DNA, cdna, and deduced amino acid sequence. J Biol Chem 1994; 269: Brigelius-Flohe R, Friedrichs B, Maurer S, Schultz M, Streicher R. Interleukin-1-induced nuclear factor kappa B activation is inhibited by overexpression of phospholipid hydroperoxide glutathione peroxidase in a human endothelial cell line. Biochem J 1997; 328: Brigelius-Flohe R. Tissue-specific functions of individual glutathione peroxidases. Free Radic Biol Med 1999; 27: Brobeck J, Wheatland M, Stronminger J. Variations in regulation of energy exchange associated with estrus, diestrus and pseudopregnancy in rats. Endocrinolgo 1947; 40: Brooks JD, Weinstein M, Lin X, Sun Y, Pin SS, Bova GS, Epstein JI, Isaacs WB, Nelson WG. CG island methylation changes near the GSTP1 gene in prostatic intraepithelial neoplasia. Cancer Epidemiol Biomarkers Prev 1998; l 7:

198 Brooks JD, Metter EJ, Chan DW, Sokoll LJ, Landis P, Nelson WG, Muller D, Andres R, Carter HB. Plasma selenium level before diagnosis and the risk of prostate cancer development. J Urol 2001;166: Brown DG, Burk RF. Selenium retention in tissues and sperm of rats fed a Torula yeast diet. J Nutr 1973; 103: Bureau I, Anderson R, Arnaud J, Raysiguier Y, Favier AE, Roussel AM. Treace mineral status in post menopausal women: impact of hormonal replacement therapy. J Trace Elem Med Biol 2002; 16: Burk RF, Brown DG, Seely RJ, Scaief CC. Influence of dietary and injected selenium on whole-body retention, route of excretion, and tissue retention of 75 SeO3 2- in the rat. J Nutr 1972; 102: Burk RF, Seely RJ, Kiker KW. Selenium: dietary threshold for urinary excretion in the rat. Proc. Soc. Exp. Bio. Med. 1973; 142(1): Burk RF, Gregory PE. Some characteristics of 75 Se-P, a selenoprotein found in rat liver and plasma, and comparison of it with seleno-glutathione peroxidase. Arch Biochem Biophys 1982; 213: Burk RF, Hill KE, Read R, Bellew T. Response of rat selenoprotein P to selenium administration and fate of its selenium. Am Physiol Soc 1991;E Burk RF, Hill KE. Regulation of selenoproteins. Annu Rev Nutr 1993; 13: Burk RF, Hill KE. Selenoprotein P. A selenium-rich extracellular glycoprotein. J Nutr 1994; 124: Burk RF, Hill KE, Awad JA, Morrow JD, Kato T, Cockell KA, Lyons PR. Pathologenesis of diquat-induced liver necrosis in selenium-deficient rats: assessment of the roles of lipid peroxidation and selenoprotein P. Hepatology 1995; 21: Burk RF, Hill KE, Motley Ak. Selenoprotein metabolism and function: evidence for more than one function for selenoprotein P. J Nutr 2003; 133: 1517S-20S. Burk RF and Hill KE. Selenoprotein P: an extracellular protein with unique physical characteristics and a role in selenium homeostasis. Annu Rev Nutr 2005; 25:

199 Bureau I, Anderson R, Arnaud J, Raysiguier Y, Favier A, Roussel AM. Trace mineral status in post menopausal women: impact of hormonal replacement therapy. J Trace Elem Med Biol 2002; 16: Butler JA, Whanger PD, Kaneps AJ and Patton NM. Metabolism of selenite and selenomethionine in the rhesus monkey. J Nutr 1990; 120: Butler JA, Thomson CD, Whanger PD, and Robinson M. Selenium distribution in blood fractions of New Zealand women taking organic or inorganic selenium. Am J Clin Nutr 1991; 53: Butt AJ, McNeil CM, Musgrove EA, Sutherland RL. Downstream targets of growth factor and oestrogen signaling and endocrine resistance: the potential roles of c-myc, cyclin D1 and cyclin E. Endocr Relat Cancer 2005;12: S47-59 Calvin HI, Cooper GW, Wallace E. Evidence that selenium in rat sperm is associated with a cysteine-rich structural protein of the mitochondrial capsules. Gamete Res 1981; 4: Capel ID, Jenner M, Williams DC, Donaldson D, Nath A. The effect of prolonged oral contraceptive steroid use on erythrocyte glutathione peroxidase activity. J Steroid Biochem 1981; 14: Capel ID, Smallwood AE. Sex differences in the glutathione peroxidase activity of various tissues of the rat. Res Comm Chem Path Pharm 1983; 40: Carruba G, D Agostino P, Miele M, Calabro M, Barbera C, Bella GD, Milano S, Ferlazzo V, Caruso R, Rosa ML, Cocciadiferro L, Campisi I, Castagnetta L, Cillari E. Estrogen regulates cytokine production and apoptosis in PMA-differentiated, macrophage-like U937 cells. J Cell Biochem 2003; 90: Cary EE, Allaway WH, Miller M. Utilization of different forms of dietary selenium. J. Anim. Sci. 1973; 36(2): Cascio S, Bartella V, Garofalo C, Russo A, Giordano A, Surmacz E. Insulin-like growth factor 1 differentially regulates estrogen receptor-dependent transcription at estrogen response element and AP-1 sites in breast cancer cells. J Biol Chem 2007; 282: CDC (Centers for Disese Control and Prevention). Selenium intoxication- New York. Morbid Mortal Wkly Rep 1984; 33:

200 Cerillo G, Rees A, Manchanda N, Reilly C, Brogan I, White A, Needham M. The oestrogen receptor regulates NFkappaB and AP-1 activity in a cell-specific manner. J Steroid Biochem Mol Biol 1998; 67: Chaitchik S, Shenberg F, Nir-El Y, Mantel M. The distribution of selenium in human blood samples of Israeli population-comparison between normal and breast cancer cases. Biol Trace Elem Res 1988; 15: Chambers, JAA, Rickwood D. in Centrifugation: A Practical Approach (Rickwood, D et al.) pp , Information Retrieval, London. Chambers I, Frampon J, Goldfarb P, Affara N, McBain W, Harrison PR. The structure of the mouse glutathione peroxidase gene: the selenocysteine in the active site is encoded by the termination codon, TGA. EMBO J 1986; 5: Charalabopoulos K, Kotsalos A, Batistatou A, Charalabopoulos A, Vezyraki P, Peschos D, Kalfakakou V, Evangelou A. Selenium in serum and neoplastic tissue in breast cancer: correlation with CEA. Br J Cancer 2006; 95: Chen X, Yang GQ, Chen J, Chen X, Wen Z, Ge K. Studies on the relations of selenium and Keshan disease. Biol Trace Elem Res 1980; 2; Chen G, Zhao L, Bao S, Cong T. Effects of different proteins on the metabolism of Zn, Cu, Fe, and Mn in rats. Biol Trace Elem Res 2006; 113: Chittum HS, Himeno S, Hill KE, Burk RF. Multiple forms of selenoprotein P in rat plasma. Arch Biochem & Biophys 1996; 325: Chittum HS, Hill KE, Carlson BA, Lee BJ, Burk RF. Replenishment of selenium deficient rats with selenium results in redistribution of the selenocysteine trna population in a tissue specific manner. Biochim Biophys Acta 1997; 1359: Christensen MJ, Cammack PM, Wray CD. Tissue specificity of selenoprotein gene expression in rats. J Nutr Biochem 1995; 6: Chu F, Esworthy RS, Doroshow JH, Doran D, Liu X. Expression of plasma glutathione peroxidase in human liver in addition to kidney, heart, lung, and breast in humans and rodents. Blood 1992; 79:

201 Chu FF, Doroshow JH, Esworthy RS. Expression, characterization, and tissue distribution of a new cellular selenium-dependent glutathione peroxidase, GSHPx-GI. J Biol Chem 1993; 268: Chu FF, Esworthy RS. The expression of an intestinal form of glutathione peroxidase (GSH-GI) in rat intestinal epithelium. Arch Biochem Biophysi 1995; 323: Chu FF, Esworthy RS, Chu PG, Longmate JA, Huycke MM, Wilczynski S. et al. Bacteria-induced intestinal cancer in mice with disrupted GPx and GPx2 genes. Cancer Res 2004; 64: Clark LC, Combs GF, Turnbull BW, Slate EH, Chalker DK, Chow J, Davis LS, Glover RA, Graham GF, Gross EG, Krongrad A, Lesher JL, Park HK, Sanders BB, Smith CL, Taylor JR. Effects of selenium supplementation for cancer prevention in patients with carcinoma of the skin. J.A.M.A. 1996; 276: Clinton SK, Li PS, Mulloy AL, Imrey PB, Nandkumar S, Visek WJ. The combined effects of dietary fat and estrogen on survival, 7,12-dimethylbenz(a)anthracene-induced breast cancer and prolactin metabolism in rats. J Nutr 1995; 125: Colvin G, Sawyer C. Induction of running activity by intracerebral implants of estrogen in ovariectomized rats. Neuroendocrinology 1969; 4: Corcoran NM, Najdovska M, Costello AJ. Inorganic selenium retards progression of experimental hormone refractory prostate cancer. J Urol 2004; 171: Couse JF, Korach KS. Estrogen receptor null mice: what have we learned and where will they lead us? Endocrinol Rev 1999; 20: Daher R, van Lente F. Concanavalin A-bound selenoprotein in human serum analyzed by graphite furnace atomic absorption spectrometry. Clin Chem1994; 40: Daniels LA. Selenium metabolism and bioavailability. Biol Trace Elem Res 1996; 54: Das D, Peterson RC, Scovell WM. High mobility group B proteins facilitate strong estrogen receptor binding to classical and half-site estrogen response elements and relax binding selectivity. Mol Endocrinol 2004; 18:

202 Das K, Chowdhury AR. Metallic ion concentration during menstrual cycle in normally menstruating women. Indian J Med Sci 1997; 51 : de Jong N, Gibson RS, Thomson CD, Ferguson EF, McKenzie JE, Green TJ, Horwath CC. Selenium and zinc status are suboptimal in a sample of older New Zealand women in a community-based study. J Nutr 2001; 131: Deagen JT, Bulter JA, Zachara BA, Whanger PD. Determination of the distribution of selenium between glutathione peroxidase, selenoprotein P, and albumin in plasma. Anal Biochem 1993; 208: Department of Health. Dietary reference values for food energy and nutrients for the United Kingdom. Report on Health and Social Subjects London: Her Majesty s Stationery Office. Derelanko MJ. Determination of erythrocyte life span in F-344, Wistar, and Sprague-Dawley rats using a modification of the [ 3 H]Diisopropylfluorophosphate ([ 3 H]DFP) Method. Toxicol Sci 1987; 9: Deroo BJ,. Korach KS. Estrogen receptors and human disease. J Clin Invest 2006; 116: Dickson RB, Russo J. Biochemical control of breast development. Harris J. R. eds. Diseases of the Breast 2000:15-31 Lippincott Williams & Wilkins Philadelphia, PA. Diplock AT. Trace elements in human health with special reference to selenium. Am J Clin Nutr 1987; 45: Diplock AT. Indexes of selenium status in human populations. Am J Clin Nutr 1993; 57: 256S-8S. Diwasdlkar-Navsariwala V, Prins GS, Swanson SM, Birch LA, Ray VH, Hedayat S, Lantvit DL, Diamond AM. Selenoprotein deficiency accelerates prostate carcinogenesis in a transgenic model. PNAS 2006; 103: Doll R, Peto R. The cause of cancer: quantitative estimates of avoidable risks of cancer in the United States today. J Natl Cancer Inst 1981; 66: Doll R. The lessons of life: keynote address to the nutrition and cancer conference. Cancer Res 1992; 52: 2024S-2029S. 187

203 Donaldson GM, Braune BM. Sex-related levels of selenium, heavy metals, and organochlorine compounds in American white pelicans (Pelecanus erythrorhyncos). Arch Environ Contam Toxicol 1999; 37: Dong Y, Zhang H, Hawthorn L, Ganther HE, Ip C. Delineation of the molecular basis for selenium-induced growth arrest in human prostate cancer cells by oligonucleotide array. Cancer Res 2003; 63: Dong Y, Lee SO, Zhang H, Marshall J, Gao AC, Ip C. Prostate specific antigen expression is down-regulated by selenium through disruption of androgen receptor signaling. Cancer Res 2004; 64: Dong Y, Zhang H, Gao AC, Marshall JR, Ip C. Androgen receptor signaling intensity is a key factor in determining the sensitivity of prostate cancer cells to selenium inhibition of growth and cancer-specific biomarkers. Mol Cancer Ther 2005; 4: Dorea JG, Miazaki ES. The effects of oral contraceptive use on iron and copper concentrations in breast milk. Fertil Steril 1999; 72: Dorgan JF, Sowell A, Swanson CA, Potischman N, Miller R, Schussler N, Stephenson HE Jr. Relationships of serum carotenoids, retinol, alpha-tocopherol, and selenium with breast cancer risk: results from a prospective study in Columbia, Missouri (United States). Cancer Causes Control 1998; 9: Drabkin DL. The standardization of hemoglobin measurement. Am J Med Sci 1949; 217: Dreher I, Jakobs TC, Kohrle J. Cloning and characterization of the human selenoprotein P promoter. J Biol Chem 1997; 272: Ducros V, Richard MJ, Favier A. The distribution of selenium in human plasma proteins for 24 hours after ingestion of 74Se (in sodium selenite form). J Inorg Biochem 1994; 55: Duffield-Lillico AJ, Thomson CD, Hill KE, Williams S. An estimation of selenium requirements for New Zealanders. Am J Clin Nutr 1999; 70:

204 Duffield-Lillico AJ, Reid ME, Turnbull BW, Combs GF, Slate EH, Fischbach LA, Marshall JR, Clark LC. Baseline characteristics and the effect of selenium supplementation on cancer incidence in a randomized clinical trial: a summary report of the Nutritional Prevention of Cancer Trial. Cancer Epidemio Biomarkers Prev 2002; 11: Duffield-Lillico AJ, Dalkin BL, Reid ME, Turnbull BW, Combs GF, Jr Slate EH., Marshall JR, Clark LC. Selenium supplementation, baseline plasma selenium status, and influence on prostate cancer: an analysis of the complete treatment period of the Nutritional Prevention of Cancer Trial. Br. J. Urol 2003; 91: El-Bayoumy K, Upadhyaya P, Chae YH, Sohn OS, Rao CV, Fiala ES, Reddy BS. Chemoprevention of cancer by organoselenium compounds. J Cell Biochem 1995; 22: El-Bayoumy K, Narayanan BA, Desai DH, Narayanan NK, Pittman B, Amin SG, Schwartz J, Nixon DW. Elucidation of molecular targets of mammary cancer chemoprevention in the rat by organoselenium compounds using cdna microarray. Carcinogenesis 2003a; 24: El-Bayoumy K, Das A, Boyiri T, Desai D, Sinha R, Pittman B, Amin S. Comparative action of 1,4-phenylenebis(methylene)selenocyanate and its metabolites against 7,12-dimethylbenz[a]anthracene-DNA adduct formation in the rat and cell proliferation in rat mammary tumor cells. Chem Biol Interact 2003b; 146: Esaki N, Nakamura T, Tanaka H, Soda K. Selenocysteine lyase, a novel enzyme that specifically acts on selenocysteine. J Biol Chem 1982; 257: Esworthy RS, Chu FF, Geiger P, Girotti AW, Doroshow JH. Reactivity of plasma glutathione peroxidase with hydroperoxide substrates and glutathione. Arch Biochem Biophs 1993; 307: Evans RM. The steroid and thyroid hormone receptor family. Science 1988; 240: Fair W, Fleshner N, Heston W. Carcinoma of the prostate: a nutritional disease? Urology 1997; 50: Finley JW, Kincaid RL. Effect of sex and time of sampling on selenium and glutathione peroxidase activity in tissues of mature rats. Bio Trace Elem Res 1991;3:

205 Fisher B, Costantino JP, Wickerham DL, Redmond CK, Kavanah M, Cronin WM, Vogel V, Robidoux A, Dimitrov N, Atkins J, Daly M, Wieand S, Tan-Chiu E, Ford L, Wolmark N. Tamoxifen for prevention of breast cancer: Report of the National Surgical Adjuvant Breast and Bowel Project P-1 Study. J Natl Cancer Inst 1998;90: Flohe L, Gunzler WA, Shock HH. Glutathione peroxidase: A selenoenzyme. FASEB lett 1973; 32: Flohe L, Brigelius-Flohe R. Selenoproteins of glutathione system. In: Selenium. Its molecular Biology and Role in Human Health (Hatfield DL, Ed.) Kluwer Academic Publishers, Boston-Dordrecht-London, PP Florian S, Wingler K, Schmehl K, Jacobasch G, Kreuzer OJ, Meyerhof W, Brigeliu-Flohe R. Cellular and subcellular localization of gastrointestinal glutathione peroxidase in normal and malignant human intestinal tissue. Free Radic Rec 2001; 35: Forbes RM, Erdman Jr. JW. Bioavailability of trace mineral elements. Ann. Rev. Nutr. 1983; 3: Forchhammer K, Leinfelder W, Boesmiller K, Veprek B, Bock A. Selenocysteine synthase from Escherichia coli: nucleotide sequence of the gene (sela) and purification of the protein. J Biol Chem 1991; 266: Foresta C, Flohe L, Garolla A, Roveri A, Ursini F, Maiorino M. Male fertility is linked to the selenoprotein phospholipids hydroperoxide glutathione peroxidase. Biol Reprod 2002; 67: Forrest P. Introduction to breast cancer. In: S.P. Langdon, W. R. Miller, and A. Berchunck (eds.). Biology of Female Cancers. Boca Raton. FL: CRC Press LLC pp Fox TE, Atherton C, Dainty JR, Lewis DJ, Langford N,J Baxter MJ, Crews HM, Fairweather-Tait SJ. Absorption of selenium from wheat, garlic, and cod intrinsically labeled with Se-77 and Se-82 stable isotopes. Int J Vitam Nutr Res 2005; 75: Fu Y, Cheng WH, Porres JM, Ross DA, Lei XG. Knockout of cellular glutathione peroxidase gene renders mice susceptible to diquat-induced oxidative stress. Free Radi Biol Med 1999; 27:

206 Ganaraja B, Pavithran P, Ghosh S. Effect of estrogen on plasma ceruloplasmin level in rats exposed to acute stress. Indian J Med Sci 2004; 58: Gangolli EA, Conneely OM, O Malley BW. Neurotransmitters activate the human estrogen receptor in a neuroblastoma cell line. J Steroid Biochem Mol Biol 1997; 61: 1-9. Ge K, Bai J, Deng XJ et al. The protective effects of selenium against viral myocarditis in mice. In: Combs G, Spallholz J, Levander O, Oldfield J, editors, Proceedings of the Third International Symposium on Selenium Biology Medicine (Part B), pp Van Hodtrand Reinhold Company, New York, NY. Ge K, Yang G. The epidemiology of selenium deficiency in the etiological study of endemic diseases in China. Am J Clin Nutr 1993; 57: 259s-263s. Geary N. Estradiol, CCK and satiation. Peptides 2001; 22: Gerald FC, Jr, Willam. Chemopreventive Agent: Selenium. Pharmacol. Ther. 1998; 79: Ghadirian P, Maisonneuve P, Perret C, Kennedy G, Boyle P,Krewski D, Lacroix A. A case-control study of toenail selenium and cancer of the breast, colon, and prostate. Cancer Detect Prev. 2000; 24: Gibbs J, Young RC, Smith GP. Cholecystokinin decreases food intake in rats. J Comp Physiol Psychol 1973; 84: Goering PL, Morgan DL, Ali SF. Effects of mercury vapor inhalation on reactive oxygen species and antioxidant enzymes in rat brain and kidney are minimal. J App Toxi 2002; 22: Golubkina NA, Alfthan G.. Selenium status of pregnant women and newborns in the former Soviet Union. Biol. Trace Element Res. 2002; 89: Goodman GE, Schaffer S, Bankson DD, Hughes MP, Omenn GS; Carotene and Retinol Efficacy Trial Co-Investigators. Predictors of serum selenium in cigarette smokers and the lack of association with lung and prostate cancer risk. Cancer Epidemiol Biomarkers Prev 2001;10:

207 Green S, Walter P, Kumar V, Krust A, Borner JM, Argos P, Chambon P. Human oestrogen receptor cdna: sequence, expression and homology with v-erb. Nature 1986; 320: Gromer S, Eubel JK, Lee BL, Jacob J. Human selenoproteins at a glance. Cell Mol Life Sci. 2005; 62: Grover PL, Martin FL. The initiation of breast and prostate cancer. Carcinogenesis 2002; 23: Guan JY, Komura S, Ohishi N, Yagi K. Difference in effects of classic and phopholipid hydroperoxide glutathione peroxidase on liver lipid peroxide levels in selenium-deficient rats. Biochem Mole Biol Interntl 1995; 37: Guemouri L, Artur Y, Herbeth B, Jeandel C, Cuny G, Siest G. Biological variability of superoxide dismutase, glutathione peroxidase, and catalase in blood. Clin Chem 1991; 37: Gupta S, Narang R, Krishnaswami K, Yadav S. Plasma selenium level in cancer patients. Indian J Cancer 1994; 31: Ha EJ, Smith AM. Plasma selenium and glutathione peroxidase activity increase during the pre-ovulatory phase of the menstrual cycle. J Am Coll Nutr 2003; 22: Hac E, Krechniak J, Szyszko M, Krzyzanowski M. Selenium concentrations in human renal cortex, liver, and hair in northern Poland. Biol Trace Elem Res 2003; 92: Hankinson SE, Willett WC, Manson JE, Colditz G.A, Hunter DJ, Spiegelman D, Barbieri RL, Speizer FE. Plasma sex steroid hormone levels and risk of breast cancer in postmenopausal women. J Natl Cancer Inst 1998; 90: Halifeoglu I, Karatas F, Canatan H, Colak R, Karadas E. Investigation of antioxidant vitamins (A, E and C) and selenium levels in chickens receiving estrogen or testosterone. Cell Biochem Funct 2003; 21: Hall TA. BioEdit: a user-friendly biological sequence alignment editor and analysis program for Windows 95/98/NT. Nucleic Acids Symposium Series 1999; 41: Hall JM, McDonnell, DP. Coregulators in nuclear estrogen receptor action: from concept to therapeutic targeting. Mol Interv 2005; 5:

208 Hansen JC, Kristensen P. The kinetics of 75Se-selenium in relation to dose and mode of administration to mice. J Nutr 1979; 109: Hardell L, Danell M, Angqvist C-A, Marklund SL, Fredriksson M, Zakari A-L, Kjellgren A. Levels of selenium in plasma and glutathione peroxidase in erythrocytes and the risk of breast cancer. Biol Tr Ele Res 1993;36: Hardell L, Degerman A, Tomic R, Marklund SL, Bergfors M. Levels of selenium in plasma and glutathione peroxidase in erythrocytes in patients with prostate cancer or benign hyperplasia. Eur J Cancer Prev 1995; 4: Hartman TJ, Albanes D, Pietinen P, Hartman AM, Rautalahi M, Tangrea JA, Taylor PR. The association between baseline vitamin E, selenium, and prostate cancer in the alpha-tocopherol, beta-carotene cancer prevention study,. Cancer Epidemiol Biomarkers Prev 1998; 7: Hasan J, Ton N, Mullamitha S, Clamp A, McNeilly A, Marshall E, Jayson GC. Phase II trial of tamoxifen and goserelin in recurrent epithelial ovarian cancer. Br J Cancer 2005; 19: He CK, Liu SG, Wu ZS. The observation on the effect of oral supplementation of sodium selenite on Kaschin-Beck disease. Chin J Endemiol 1988; 7: Helzlsouer KJ, Huang H, Alberg AJ, Hoffman S, Burke A, Norkus EP, Morris JS, Comstocket GW. Association between alpha-tocopherol, gamma-tocopherol, selenium, and subsequent prostate cancer. J Natl Cancer Inst 2000; 92: Heider J, Baron C, Bock A. Coding from a distance: dissection of the mrna determinants required for the incorporation of selenocysteine into protein. EMBO J 1992; 11: Herrman JL. The properties of a rat serum protein labeled by the injection of sodium selenite. Biochim Biophys Acta 1977; 500: Hilakivi-Clarke L, Clarke R, Lippman M. The influence of maternal diet on breast cancer risk among female offspring. Nutrition 1999, 15, Hill KE, Lyons R, Burk RF. Differential regulation of rat liver selenoprotein mrnas in selenium deficiency. Biochm Biophs Res Commun 1992; 185:

209 Hill KE, Lloyd RS, Burk RF. Conserved nucleotide sequences in the open reading frame and 3 untranslated region of selenoprotein P mrna. Proc Natl Acad Sci USA 1993; 90: Hill KE, Xia Y, Akesson B, Boeglin ME, Burk RF. Selenoprotein P concentration in plasma is an index of selenium status in selenium-deficient and selenium-supplemented Chinese subjects. J Nutr 1996a; 126: Hill KE, Chittum HS, Lyons PR, Boeglin ME, Burk RF. Effect of selenium on selenoprotein P expression in cultured liver cells. Biochim Biophys Acta 1996b; 1313: Hill KE, Zhou J, McMahan WJ, Motley AK, Atkins JF, Gesteland RF, Burk RF. Deletion of selenoprotein P alters distribution of selenium in the mouse. J Biol Chem 2003; 278: Hill KE, zhou J, McMahan WJ, Motley AK, Burk RF. Neurological dysfunction occurs in mice with targeted deletion of the selenoprotein P gene. J Nutr 2004; 134: Himeno S, Chittum HS, Burk RF. Isoforms of selenoprotein P in rat plasma. J Biol Chem 1996; 271: Hirooka T, Galambos. Selenium metabolism. II. Effect of injected selenium compounds and of liver injury. Proc Soc Exp Biol Med 1966; 121: Ho YS, Magnenat JL, Bronson RT, Cao J, Gargano M, Sugawara M, Funk CD. Mice deficient in cellular glutathione peroxidase develop normally and show no increase sensitivity to hyperoxia. J Biol Chem 1997; 272: Holben DH, Smith AM, Ilich JZ, Landoll JD, Holcomb JP, Matkovic V. Selenium intakes, absorption, retention, and status in adolescent girls. J Am Diet Assoc 2002; 102: Hopkins LL, Pope AL, Baumann CA. Distribution of microgram quantities of selenium in the tissue of the rats, and effects of pervious selenium intake. J Nutr 1966; 88: Hsieh HS, Ganther HE. Acid-volatile selenium formation catalyzed by glutathione reductase. Biochemistry 1975; 14: Hsieh HS, Ganther HE. Biosynthesis of dimethyl selenide from sodium selenite in rat liver and kidney cell-free systems. Biochim Biophys Acta 1977; 497:

210 Huang Y-L, Sheu J-Y, Lin T-H. Association between oxidative stress and changes of trace elements in patients with breast cancer. Clin Biochem 1999;32: Hunter DJ, Morris JS, Stampfer MJ, Colditz GA, Speizer FE, Willet WC. A prospective study of selenium status and breast cancer risk. JAMA 1990; 264: Igarashi T, Satoh T, Ono S, Iwashita K, Hosokawa M, Ueno K, Kitagawa H. Effect of steroid sex hormones on the sex-related differences in the hepatic activities of gamma-glutamyl-transpeptidase, glutathione S-transferase and glutathione peroxidase in rats. Res Commun Chem Pathol Pharmacol 1984; 45: Institute of Medicine (U.S.). Dietary Reference Intakes for Vitamin C, Vitamin E, Selenium, and Carotenoids. Washington, DC. National Academy Press C2000. Ingrao G, Belloni P, Di Piero S, Santaroni GP. Levels of some trace elements in selected autopsy organs, and in hair and blood samples from adult subjects of the Italian population. Biol Trace Elem Res 1990; 7: Ip C, Lisk DJ, Stoewsand GS. Mammary cancer prevention by regular garlic and selenium-enriched garlic. Nutr Cancer 1992; 17: Ip C, Thompson HJ, Zhu Z, Ganther HE. In vitro and in vivo studies of methylseleninic acid: evidence that a monomethylated selenium metabolite is critical for cancer chemoprevention. Cancer Res 2000; 60: Itoh M, Suzuki KT. Effects of dose on the methylation of selenium to monomethylselenol and trimethylselenonium ion in rats. Arch Toxicol 1997; 71: Janghorbani M, Ting BTG, Young V. Use of stale isotopes of selenium in human metabolic studies: development of analytical methodology. Am J Clin Nutr 1981; 34: Janghorbani M, Young VR. Selenium metabolism in North Americans: studies based on stable isotope tracers. In: Combs GFJ, Spallholz JE, Levander OA, Oldfield JE, eds. Selenium in Biology and Medicine. New York: AVI, 1987: Janghorbani M, Rockway S, Mooers CS, Roberts EM, Ting BT, Sitrin MD. Effect of chronic selenite supplementation on selenium excretion and organ accumulation in rats. J Nutr 1990a; 120:

211 Janghorbani M, Martin RF, Kasper LJ, Sun XF, Young VR. The selenite-exchangeable metabolic pool in humans: a new concept for the assessment of selenium status. Am J Clin Ntr 1990b; 51: Janghorbani M, Lynch NE, Mooers CS, Ting BTG. Comparison of the magnitude of the selenite-exchangeable metabolic pool and whole body endogenous selenium in adult rats. J Nutr 1990c; 120: Janghorbani M, Mooers CS, Smith MA, Hazell T, Blanock K, Ting BTG. Correlation between the size of the selenite-exchangeable metabolic pool and total body of liver selenium in rats. J Nutr 1991; 121: Janghorbani M, Xia Y, Ha P, Whanger PD, Butler JA., Olesik JW, Grunwald E. Effect of dietary selenium restriction on selected parameters of selenium status in men with high life-long intake. J Nutr Biochem 1999; 10: Jensen EV, Jacobson HI. Basic guides to the mechanism of estrogen action. Recent Prog Horm Res 1962; 18: Jensen EV, Cheng G, Palmieri C, Saji S, Makela S, Van Noorden S, Wahlstrom T, Warner M, Coombes RC, Gustafsson JA. Estrogen receptors and proliferation markers in primary and recurrent breast cancer. PNAS 2001; 98, Jian MG, Hislop GT, Howe GR, Ghadirian P. Plant foods, antioxidants, and prostate cancer risk: findings from case-control studies in Canada. Nutr Caner 1999; 34; Jiang C, Jiang W, Ip C, Ganther H, Lu J. Selenium-induced inhibition of angiogenesis in mammary cancer at chemopreventive levels of intake. Mol Carcinog 1999; 26: Jiang C, Wang Z, Ganther H, Lu J. Distinct effects of methylseleninic acid versus selenite on apoptosis, cell cycle, and protein kinase pathways in DU145 human prostate cancer cells. Mol Cancer Ther 2002; 1: Jimenez A, Planells E, Aranda P, Sanchez-Vinas M, Llopis J. Changes in bioavailability and tissue distribution of selenium caused by magnesium deficiency in rats. J Am Coll Nutr. 1997; 16 (2): Katzenellenbogen BS. Estrogen receptors: bioactivities and interactions with cell signaling pathways. Biol Reprod 1996; 54:

212 Ke HZ, Simmons HA, Pirie GM, Crawford DT, Thompson DD. Droloxifene, a new estrogen antagonist/agonist, prevents bone loss in ovariectomized rats. Endocrinology 1995; 136: Keshan Disease Research Group. The relationship between selenium and the prevalence of Keshan disease. In: Proceedings of the First National Symposium on the Etiology of Keshan Disease in : Keshan Disease Research Group. Epidemiologic studies on the etiologic relationship of selenium and Keshan disease. Chin Med J 1979a; 92: Keshan Disease Research Group. Observation on effect of sodium selenite in prevention of Keshan disease. Chinese Med J 1979b; 91: Key T, Appleby P, Barnes I, Reeves G, Endogenous Hormones and Breast Cancer Collaborative Group. Endogenous sex hormones and breast cancer in postmenopausal women: reanalysis of nine prospective studies. J Natl Cancer Inst 2002; 94: : Kise Y, Yamamura M, Kogata M, Uetsuji S, Takada H, Hioki K, Yamamoto M. Inhibitory effect of selenium on hamster pancreatic cancer induction by N -nitrosobis (2-oxopropyl) amine. Int J Cancer 1990; 46: Klinge CM, Bodenner DL, Desai D, Niles RM, Traish AM. Binding of type II nuclear receptors and estrogen receptor to full and half-site estrogen response elements in vitro. Nucl Acids Res 1997; 25: Knoll KE, Pietrusz JL, Liang M. Tissue specific transcriptome responses in rats with early streptozotocin-induced diabetes. Physiol Genomics 2005; 21: Kobayashi Y, Ogra Y, Ishiwata K, Takayama H, Aimi N, Suzuki KT. Selenosugars are key and urinary metabolites for selenium excretion within the required to low-toxic range. PNAS 2002; 29: Kocdor H, Cehreli R, Kocdor MA, Sis B, Yilmaz O, Canda T, Demirkan B, Resmi H, Alakavuklar M, Harmancioglu O. Toxicity induced by the chemical carcinogen 7,12-Dimethylbenz [a] anthracene and the protective effects of selenium in Wistar rats. J Toxicol Environ Health. Part A 2005; 68:

213 Koivistoinen P, Varo P. Selenium in Finnish food. In: Selenium in Biology and Medicine. Cambs JF, Spallholz JE, Lavander OA, p New York, Oldfield. Van Nostrand. Krehl WA. Selenium-The maddening mineral. Nutr Today Win 1970; Kryukov GV, Castellano S, Novoselov SV, Lobanov AV, Zehtab O, Guigo R, Gladyshev VN. Characterization of mammalian selenoproteomes. Science 2003; 300: Kuikka J, Nordman E. Measurement of 75Se-sodium selenite in the human body. Int J Nucl Med Biol 1978; 5(1): Kuiper GGJM, Enmark E, Pelto-Huikko M, Nilsson S, Gustafsson J-A. Cloning of a novel estrogen receptor expressed in rat prostate and ovary. Pro Natl Acad Sci USA 1996; 93: Kuiper GG, Carlsson B, Grandien K, Enmark E, Haggblad J, Nilsson S, Gustafsson JA. Comparison of the ligand binding specificity and transcript tissue distribution of estrogen receptors α and β. Endocrinology 1997; 138: Kumar V, Green S, Stack G, Berry M, Jin J-R, Chambon P. Functional domains of the estrogen receptor. Cell 1987; 51: Kumar V, Chambon P. The estrogen receptor binds tightly to its responsive element as a ligand-induced homodimer. Cell 1988; 55: Kumar K, Thangaraju M, Sachdanandam P. Changes observed in antioxidant system in the blood of post-menopausal women with breast cancer. Bioch Intl 1991; 25: L'abbe MR, Collins MW, Trick KD, Laffey PJ. Glutathione peroxidase activity in a healthy Canadian population. Effects of age, smoking and drinking habits, exercise and oral contraceptive use. Trace Elem Med 1992; 9: Lapointe J, Kimmins S, MacLaren LA, Bilodeau JF. Estrogen selectively up-regulates the phospholipids hydroperoxide glutathione peroxidase in the oviducts. Endocrinology 2005; 146: Larsen LF, Andersen HR, Hansen AB, Andersen O. Variation in risk indicators of cardiovascular disease during the menstrual cycle: an investigation of within-subject variations in glutathione peroxidase, haemostatic variables, lipids and lipoproteins in healthy young women. Scand J Clin Lab Invest 1996; 56 :

214 Latour MG, Shinoda M, Lavoie JM. Metabolic effects of physical training in ovariectomized and hyperestrogenic rats. J Appl Physiol 2001; 90: Leblondel G, Mauras Y, Cailleux A, Allain P. Transport measurements across Caco-2 monolayers of different organic and inorganic selenium. Biological Trace Element Research 2001; 83: Lee SO, Yeon CJ, Nadiminty N, Trump D, Ip C, Dong Y, Gao AC. Monomethylated selenium inhibits growth of LNCaP human prostate cancer xenograft accompanied by a decrease in the expression of androgen receptor and prostate-specific antigen (PSA). Prostate 2006; 66: Lei XG, Evenson JK, Thompson KM, Sunde RA. Glutathione peroxidase and phospholipid hydroperoxide glutathione peroxidase are differentially regulated in rats by dietary selenium. J Nutr 1995; 125: Lee YK, Mok Kim S, Han S. Ozone-induced inactivation of antioxidant enzymes. Biochimie 2003; 85: Levander OA, Baumann CA. Selenium metabolism. VI. Effect of arsenic on the excretion of selenium in the bile. Toxicol Appl Pharmacol 1966; 9: Li H, Stampfer MJ, Giovannucci EL, Morris JS, Willett WC, Gaziano JM, Ma J. A prospective study of plasma selenium levels and prostate cancer risk. J Natl Cancer Inst 2004; 96: Li JR. Investigation on the relationship between environmental selenium level and the incidence rate of KBD. Chin J Endemiol 1989; 8: Liberman BA. The estrogen receptor activity cycle: dependence on multiple protein-protein interactions. Crit Rev Eukaryotic Gene Expression 1997; 7: Lipton A. Hormonal influences on oncogenisis and growth. In Breast Cancer. Roses DF, ed., New York: Churchill Livingsone, Litov RE, Combs GF. Selenium in pediatric nutrition. Pediatr 1991; 87: Lonard DM, Smith CL. Molecular perspectives on selective estrogen receptor modulators (SERMs): progress in understanding their tissue-specific agonist and antagonist actions. Steroids 2002; 67:

215 Lonnerdal B, Glazier CE, Lien EL. Selenium supplementation of infant formula: uptake and retention of various forms of selenium in suckling rats. J Nutr biochem 1992; 3: Lopes PA, Santos MC, Vicente L, Rodrigues MO, Pavão ML, Nève J, Viegas-Crespo AM. Trace Element Status (Se, Cu, Zn) in Healthy Portuguese Subjects of Lisbon Population: A Reference Study. Biol Trace Elem Res 2004; 101: Lopez-Saez JB, Senra-Varela A, Pousa-Estevez L. Selenium in breast cancer. Oncology 2003; 64: Lu T, Achari Y, Sciore P, Hart DA. Estrogen receptor alpha regulates matrix metalloproteinase-13 promoter activity primarily through the AP-1 transcriptional regulatory site. Biochim Biophys Acta 2006; 1762: Ma SG, Hill KE, Caprioli RM, Burk RF. Mass spectrometric characterization of full-length rat selenoprotein P and three isoforms shortened at the C terminus. Evidence that three UGA codons in the mrna open reading frame have alternative functions of specifying selenocysteine insertion or translation termination. J Biol Chem 2002; 277: Maiorino M, Wissing JB, Brigelius-Flohe R, Calabrese F, Roveri A, Steinert P, Ursini F, Flohe L. Testosterone mediates expression of the selenoprotein PHGPx by induction of spermatogenesis and not by direct transcriptional gene activation. FASEB J 1998; 12: Mannisto S, Alfthan G, Virtanen M, Kataja V, Uusitupa M, Pietinen P. Toenail selenium and breast cancer-a case-control study in Finland. Eur J Clin Nutr 2000; 54: Maor S, Mayer D, Yarden RI, Lee AV, Sarfstein R, Werner H, Papa MZ. Estrogen receptor regulates insulin-like growth factor-i receptor gene expression in breast tumor cells: involvement of transcription factor Sp1. J Endocrinol 2006;191: Marano G, Spagnolo A, Morisi G, Menotti A. Changes of serum selenium and serum cholesterol in children during sexual maturation. J Trace Ele Elec Heal. Dis 1991; 5: Markham GD, Hafner EW, Tabor CW, Tabor H. Adenosylmethionine synthetase from Escherichia coli. J Bio Chem 1980; 255:

216 Massafra C, Buonocore G, Berni S, Gioia D, Giuliani A, Vezzosi P. Antioxidant erythrocyte enzyme activities during oral contraception. Contraception 1993; 47: Massafra C, Buonocore G, Gioia D, Sargentini I, Farina G. Effects of estradiol and medroxyprogesterone-acetate treatment on erythrocyte antioxidant enzyme activities and malondialdehyde plasma levels in amenorrhoic women. J Clin Endocrinol Metab 1997; 82: Massafra C, Felice C, Gioia D, Buonocore G. Variations in erythrocyte antioxidant glutathione peroxidase activity during the menstrual cycle. Clin Endocrinol 1998; 49: Massafra C, Gioia D, Fellice C, Musceola M, Longini M, Buonocore G. Gender-related differences in erythrocyte glutathione peroxidase activity in healthy subjects. Clin Endocrinol 2002; 57: Massafra C, Felice C, Gioia D, Buonocore G. Variations in erythrocyte antioxidant glutathione peroxidase activity during the menstrual cycle. Clin Endocrinol 1998; 49: Matthews J, Gustafsson J-A. Estrogen signaling: a subtle balance between ERα and ERβ. Mol Interventions 2003; 3: Mauri P, Benazzi L, Flohe L, Maiorino M, Pietta PG, Pilawa S, Roveri A, Ursini F. Versatility of selenium catalysis in PHGPx unraveled by LC/ESI-MS/MS. Biol Chem 2003; 384: Maziere C, Auclair M, Ronveaux MF, Salmon S, Santus R, Maziere JC. Estrogen inhibits copper and cell-mediated modification of low density lipoprotein. Atherosclerosis 1991; 89: McCarthy TP, Brodie B, Milner JA, Bevill RF. Improved method for selenium determination in biological sample by gas chromatography. J Chromat 1981; 225: McConnell KP, Chao GJ. Transmucosal movement of selenium. Am J Pysiol 1965; 208, McAdam PA, Morris VC, Forman MR, Lanza E, Campbell W, Beecher GR, Levander OA. Differences in plasma selenium and glutathione peroxidase activity during three phases of the menstrual cycle in normal healthy women. FASEB 1994; J8 :A

217 McConnell KP, Jager RM, Bland KI, Blotcky AJ. The relationship of dietary selenium and breast cancer. J Surg Oncol 1980; 15: McCormick KM, Burns KL, Piccone CM, Gosselin LE, Brazeau GA. Effects of ovariectomy and estrogen on skeletal muscle function in growing rats. J Muscle Res Cell Motil 2004; 21: Mckay L, Cidlowski J. Cross-talk between nuclear factor-kb and the steroid hormone receptors: mechanisms of mutual antagonism. Mol Endocrinol 1998; 12: McKenna N, Lanz R, O Malley B. Nuclear receptor coregulators: cellular and molecular biology. Endocr Rev 1999; 20: McMaster D, Bell N, Anderson P, Love AHG. Automated measurement of two indicators of human selenium status, and applicability to population studies. Clin Chem1990; 35: Medina D, Shepard R. Selenium-mediated inhibition of 1,12-dimethylbenz(a)anthracene induced mouse mammary tumorigenesis. Carcinogenesis 1981; 2: Medina D, Thompson H, Ganther H, Ip C. Se-Methylselenocysteine: A new compound for chemoprevention of breast cancer. Nutr Cancer 2001; 40(1): Mendez P, Wandosell F, Garcia-Segura LM. Cross-talk between estrogen receptors and insulin-like growth factor-i receptor in the brain: cellular and molecular mechanisms. Front Neuroendocrinol. 2006; 27: Mihara H, Kurihara T, Watanabe T, Yoshimura T, Esaki N. cdna cloning, purification, and characterization of mouse liver selenocysteine lyase. Candidate for selenium delivery protein in selenoprotein synthesis. J Biol Chem 2000; 75: Mikula M, Dzwonek A, Jagusztyn-Krynicka K, Ostrowski J. Quantitative detection for low levels of Helicobacter pylori infection in experimentally infected mice by real-time PCR. J Microbiol Methods 2003; 55: Millar KR, Sheppard AD, Gardiner MA. A comparison of the distribution of 75 Se in proteins of blood, liver, and kidney from rats differing in selenium status. N.Z.J. Agric Res 1972; 15:

218 Milman N, Kirchhoff M, Jorgensen T. Iron status markers, serum ferritin and hemoglobin in 1359 Danish women in relation to menstruation, hormonal contraception, parity, and postmenopausal hormone treatment. Ann Hematol 1992; 65: Moreno-Reyes R, Suetens C, Mathieu F, Begaux F, Zhu D, River MT, Boelaert M, Neve J, Perlmutter N, Vanderpas J. Kaschin-Beck osteoarthropathy in rural Tibet in relation to selenium and iodine status. N Engl J Med 1998; 339: Moschos MP. Selenoprotein P. Cell Mol Life Sci 2000; 57: Mostert V, Dreher I, Kohrle J, Abel J. Transforming growth factor-β1 inhibits expression of selenoprotein P in cultured human liver cells. FEBS Lett 1999; 460: Mostert V. Selenoprotein P: properties, functions, and regulation. Arch Biochem Biophys 2000; 376: Mostert V, Wolff S, Dreher I, Kohrle J, Abel J. Identification of an element within the promoter of human selenoprotein P responsive to transforming growth factor-beta. Eur J Biochem. 2001; 268: Motsenbocker MA, Tappel AL. A selenocysteine-containing selenium-transport protein in rat plasma. Biochem. Biophys Acta 1982; 719: Motsenbocker MA, Tappel AL. Effect of dietary selenium on plasma selenoprotein P, selenoprotein P1, and glutathione peroxidase in the rat. J Nutr 1984; 114: Mullenbach GT, Tabrizi A, Irvine BD, Bell GI, Hallewell RA. Sequence of a cdna coding for human glutathione peroxidase confirms TGA encodes active site selenocysteine. Nucleic Acids Res 1987; 15: Muneyyirci-Delale O, Nacharaju VL, Altura BM, Altura B. Sex steroid hormones modulate serum ionized magnesium and calcium levels throughout the menstrual cycle in women. Fertil Seril 1998; 69: Mussalo-Rauhamaa H, Pantzar P. Selenium and DDE in breast fat of breast cancer patients: Their relationship to hormone receptors in breast tissue. J Natl Cancer Inst 1993; 85: Mutanen M. Comparison of chemical analysis and calculation method in estimating selenium content of Finnish diets. Nutr Res 1985; 5:

219 Mutanen M. Bioavailability of selenium. Ann Clin Res 1986; 18: Mykkanen H, Humaloja T. Effect of lead on the intestinal absorption of sodium selenite and selenomethionine (75Se) in chicks. Biol TraceElemtnt Res 1984; 6: 11. Nakayama A, Kill KE, Austin LM, Motley AK, Burk RF. All regions of mouse brain are dependent on selenoprotein P for maintenance of selenium. J Nutr 2007; 17: National Research Council. Recommended Dietary Allowance, Ninth Edition National Academy Press, Wathington, D.C. National Research Council. Selenium. In: Recommended Dietary Allowances, Tenth edition pp National Academy Press, Washington, D.C. Navarrete M, Gaudry A, Revel G, Martinez T, Cabrera L. Urinary selenium excretion in patients with cervical uterine cancer. Biol Trace Elem Res 2001; 79(2): Navarro M, Lopez H, Perez V, Lopez MC. Serum selenium levels during normal pregnancy in healthy Spanish women. Sci Total Environ 1996; 186: Nicholson RI, Hutcheson IR, Britton D, Knowlden JM, Jones HE, Harper ME, Hiscox SE, Barrow D, Gee JM. Growth factor signaling networks in breast cancer and resistance to endocrine agents: new therapeutic strategies. J Steroid Biochem Mol Biol 2005; 93: Nomura AMY, Lee J, Stemmermann GN, CombsGF Jr. Serum selenium and subsequent risk of prostate cancer. Cancer Epidemiol Biomarkers Prev 2000; 9: Ogawa S, Inoue S, Watanabe T, Hiroi H, Orimo A, Hosoi T, Ouchi Y, Muramatsu M. The complete primary structure of human estrogen receptor β (herβ) and its heterodimerization with ER α in vivo and in vitro. Biochem Biophys Res Commun 1998; 243: Oster O, Schmiedel G, Prellwitz W. The organ distribution of selenium in German adults. Biol Trace Elem Res 1988; 15: Paglia DE, Valentine WN. Studies on the quantitative and qualitative characterization of erythrocyte glutathione peroxidase. J Lab Clin Med 1967;70:

220 Patterson BH, Levander OA, Helzlsouer K, McAdam PA, Lewis SA, Taylor PR, Veillon C, Zech LA. Human selenite metabolism: A kinetic model. Am J Physiol 1989; 257: R Patterson BH, Zech LA. Development of a model for selenite metabolism in humans. J Nutr 1992; 122: Pearce ST, Jordan VC. The biological role of estrogen receptors α and β in cancer. Crit Rev Oncol Hematol 2004; 50: Pfeifer H, Conrad M, Roethlein D, Kyriakopoulos A, Brielmeier M, Bornkamm GW, Behne D. Identification of a specific sperm nuclei selenoenzyme necessary for protamine thiol cross-linking during sperm maturation. FASEB J 2001; 15: Peiker G, Muller B, Dawczynski H, Winnefeld K. Behavior of selenium in serum and erythrocytes in females with normal and risk pregnancy (pregnancy-induced hypertension, fetal retardation and hepatosis).zentralbl Gynakol 1991; 113: Peters U; Foster CB, Chatterjee N, Schatzkin A, Reding D, Andriole GL, Crawford ED, Sturup S, Chanock SJ, Hayes RB. Serum selenium and risk of prostate cancer-a nested case-control study. Am J Clin Nutr 2007; 85: Pinto RE, Bartley W. Effect of age and sex on glutathione reductase and glutathione peroxidase activities and on aerobic glutathione oxidation in rat liver homogenates. Biochem J 1967; 115: Pinto RE, Bartley W. Effect of age and sex on glutathione reductase and glutathione peroxidase activities and on aerobic glutathione oxidation in rat liver homogenates. Biochem J 1969; 112; Porter W, Saville B, Hoivik D, Safe S. Functional synergy between the transcription factor Sp1 and the estrogen receptor. Mol Endocrinol 1997; 11; Prohaska J, Sunde R. Comparison of liver glutathione peroxidase activity and mrna in female and male mice and rats. Comp Biochem Physical 1993;105B: Rao AV, Fleshner N, Agarwal S. Serum and tissue lycopene and biomarkers of oxidation in prostate cancer patients: a case-control study. Nutr Cancer 1999; 33:

221 Rayman MP. Selenium in cancer prevention: a review of the evidence and mechanism of action. Proc Nutr Soc 2005; 64: Read R, Bellew T, Yang JG, Hill KE, Palmer IS, Burk RF. Selenium and amino acid composition of selenoprotein P, the major selenoprotein in rat serum. J Bio Chem 1990; 265: Reilly C. Selenium in food and health. London, Blackie Academic and Professional Reiter R, Wendel A. Selenium and drug metabolism-ii. Independence of glutathione peroxidase and reversibility of hepatic enzyme modulations in deficient mice. Biochem Pharm 1984; 33: Richardson DR. More roles for selenoprotein P: local selenium storage and recycling protein in the brain. Biochem J 2005; 386: e5-e7. Rigquist S. Schneider D, Gibson T, Baron C, Bock A, Gold L. Recognition of the mrna selenocysteine insertion sequence by the specialized translational elongation factor SELB. Genes Dev 1994; 8: Ripple MO, Henry WF, Rago RP, Wilding G. Prooxidant-antioxidant shift induced by androgen treatment of human prostate carcinoma lines. J Natl Cancer Inst 1997; 89: Robinson MF, Thomson CD. The role of selenium in the diet. Nutr Abst Rev Clin Nutr 1983; 53:3-26. Robinson MF, Thomson CD, Huemmer PK. Effect of a megadose of ascorbic acid, a meal and orange juice on the absorption of selenium as sodium selenite. New Zeal Med J 1985a; 98: Robinson JR, Robinson MF, Levander OA, Thomson CD. Urinary excretion of selenium by New Zealand and North American human subjects on differing intakes. Am J Clin Nutr 1985b; 41(5): Robinson MF, Thomason CD, Jenkinson CP, Luzhen G, Whanger PD. Long-term supplementation with selenate and selenomethionine: urinary excretion by New Zealand women. Br J Nutr 1997; 77(4):

222 Rohrdanz E, Ohler S, Tran-Thi QH, Kahl R. The phytoestrogen daidzein affects the antioxidant enzyme system of rat hepatoma H4IIE cells. J Nutr 2002; 132: Rodriguez ER, Alaejos MS, Romero CD. Urinary selenium status of healthy people. Eur J Clin Chem Clin Biochem 1995; 33(3): Roesch DM. Effects of selective estrogen receptor agonists on food intake and body weight gain in rats. Physiol Behav 2006; 87: Ronai Z, Traganos F, Darzynkiewicz Z, Tillotson JK, Conaway CC, Upadhyaya P, El-Bayoumy K. Effects of organic and inorganic selenium compounds on rat mammary tumor cells. Int J Cancer 1995; 63: Rotruck JT, Pope AL, Ganther HE, Swanson AB, Hafeman DG, Hoekstra WG. Selenium: biochemical role as a component of glutathione peroxidase. Science 1973; 179: Roy D, Liehr JG. Changes in activities of free radical detoxifying enzymes in kidneys of male Syrian hamsters treated with estradiol. Cancer Res 1989; 49: Sack MN, Rader DJ. Cannon RO III. Oestrogen and inhibition of oxidation of low-density lipoproteins in postmenopausal women. Lancet 1994; 343: Saedi MA, Smith CG, Frampton J, Harrison PR, Sunde RA. Effect of selenium status on mrna levels for glutathione peroxidase in rat liver. Biochem Biophys Res Commun 1988; 153: Saito Y, Takahashi K. Characterization of selenoprotein P as a selenium supply protein. Eur J Biochem 2002; 269: Saito Y, Sato N, Hirashima M, Takebe G, Nagasawa S, Takahashi K. Domain structure of bi-functional selenoprotein P. Biochem J 2004; Sandholm M. Selenium carrier proteins in mouse plasma. Acta Pharmacol Toxicol 1974; 35: Schisler NJ, Singh SM. Modulation of selenium glutathione peroxidase (Se-GSH-Px) activity in mice. Free Radic Biol Med 1988; 4: Schnurr K, Belkner J, Ursini F, Schewe T, Kuhn H. The selenoenzyme phospholipid hydroperoxide glutathione peroxidase controls the activity of the 15-lipoxygenase with 207

223 complex substrates and preserves the specificity of the oxygenation products. J Biol Chem 1996; 271: Schomburg L, Schweizer U, Holtmann B, Flohe L, Sendtner Michael, Kohrle J. Gene disruption discloses role of selenoprotein P in selenium delivery to target tissues. Biochem J 2003; 370: Schosinsky KH, Lehmann HP, Beeler MF. Measurement of ceruloplasmin from its oxidase activity in serum by use of o-dianisidine dihydrochloride. Clin Chem 1974; 20: Schrauzer GN, Molenaar T, Mead S, Kuehn K, Yamamoto H, Araki E. Selenium in the blood of Japanese and American women with and without breast cancer and fibrocystic disease. Jpn J Cancer Res 1985; 76: Schreilhofer DA, Resnick EM, Lin VY, Shupnik MA. Ligand-independent activation of pituitary ER: dependence on PKA-stimulated pathways. Endocrinology 2001; 142: Schrijver J, Alexieva-Figusch J, van Breederode N, van Gilse HA. Investigation on the nutritional status of advanced breast cancer patients. The influence of long-term treatment with megestrol acetate or tamoxifen. Nutr Cancer 1987;10: Schwabe JW, Chapman L, Finch JT, Rhodes D. The crystal structure of the estrogen receptor DNA-binding domain bound to DNA: How receptors discriminate between their response elements. Cell 1993; 75: Schweizer U, Brauer AU, Kohrle J, Nitsch R, Savaskan NE. Selenium and brain function: a poorly recognized liaison. Brain Res Rev 2004; 45: Schweizr U, Streckfuss F, Pelt P, Carlson BA, Hatfield DL, Kohrle J, Schomburg L. Hepatically derived selenoprotein P is a key factor for kidney but not for brain selenium supply. Biochem J 2005; 386: Shah YM, Kaul A, Dong Y, Ip C, Rowan BG. Attenuation of estrogen receptor α (ERα) signaling by selenium in breast cancer cells via downregulation of ERα gene expression. Breast Cancer Res Treat 2005; 92: Shamberger RJ, Frost DV. Possible protective effect of selenium against human cancer. Can Med Assoc J 1969; 100:

224 Shamberger RJ, Willis CE. Selenium distribution and human cancer mortality. CRC Crit Rev Clin Lab Sci 1971; 2: Shen L, van Dyck K, Luten J, Deelstra H. Diffusibility of selenate, selenite, selenomethionine, and seleno-cystine during simulated gastrointestinal digestion. Bio Trace Elem Res 1997; 58: Shiobara Y, Suzuki KT. Binding of selenium (administrated as selenite) to albumin after efflux from red blood cells. J Chromatogr B 1998a; 710: Shiobara Y, Yoshida T, Suzuki KT. Effects of dietary selenium species on Se concentrations in hair, blood and urine. Toxicol Appl Pharmacol 1998b; 152: Sinha R, Unni E, Ganther HE, Medina D. Methylseleninic acid, a potent growth inhibitor of synchronized mouse mammary epithelial tumor cells in vitro. Biochem Pharmacol 2001; 61: Smith AM, Picciano MF, Milner JA. Selenium intakes and status of human milk and formula-fed infants. Am J Clin Nutr 1982; 35: Smith AM, Picciano MF. Evidence for increased selenium requirement for the rat during pregnancy and lactation. J Nutr l986; ll6:l Smith AM, Picciano MF. Relative bioavailability of seleno-compounds in the lactating rat. J Nutr 1987; 117: Smith AM, Cha C, Kimura RE. Plasma selenium and glutathione peroxidase activity fluctuate during the rat estrous cycle. Nutr Res 1995;15: Smith AM, Chang MP-H, Medeiros LC. Generational differences in selenium status of women. Biol Tr Ele Res 2000; 75: Sobocanec S, Balog T, Sverko V, Marotti T. Sex-dependent antioxidant enzyme activities and lipid peroxidation in ageing mouse brain. Free Radic Res 2003; 37: Soda K, Oikawa T, Esaki N. Vitamin B6 enzymes participation in selenium amino acid metabolism. Biofators 1999; 10: Sokoloff L. The history of Kaschin-Beck disease. N Y State J Med 1989; 89:

225 Spallholz JE. On the nature of selenium toxicity and carcinostatic activity. Free Radic Biol Med 1994; 17: Steele RD, Benevenga NJ. The metabolism of 3-methylthiopropionate in rat liver homogenates. J Bio Chem 1979; 254: Stadtman TC. Selenocysteine. Annu Rev Biochem 1996; 65: Stoica A, Pentecost E, Martin MB. Effects of selenite on estrogen receptor- expression and activity in MCF-7 breast cancer cells. J Cell Biochem 2000; 79: Strehlow K, Rotter S, Wassmann S, Adam O, Grohe C, Laufs K, Bohm M, Nickenig G. Modulation of antioxidant enzyme expression and function by estrogen. Circ Res 2003;93: Sugioka K, Shimmosegawa Y, Nakano M. Estrogens as natural antioxidants of membrane phospholipid peroxidation. FEBS Letters 1987; 210, Sunde RA, Hoekstra WG. Structure, synthesis and function of glutathione peroxidase. Nutr Rev 1980; 38: Sunde RA, Saedi MS, Knight SAB, Smith CG, Evenson JK. Regulation of expression of glutathione peroxidase by selenium. In selenium in Biology and Medicine. A. Wendel, ed. 1989; 8-13, Springer-Verlag, Heidlberg, Germany. Sunde RA. Selenium. In: Handbook of Nutritionally Essential Mineral Elements (O Dell, BL and Sunde RA eds). 1997a; PP Marcel Dekker, New York. Sunde RA, Thompson RM, Palm MD, Weiss SL, Thompson KM, Evenson JK. Selenium regulation of selenium-dependent glutathione peroxidases in animals and transfected CHO cells. Biomed Environ Sci 1997b; 10: Surai PF. Selenium in nutrition and health. Nottingham University Press Suzuki KT, Itoh M, Ohmichi M. Detection of selenium-containing biological constituents by high-performance liquid chromatography-plasma source mass spectrometry. J Chromatogr B 1995; 666: Suzuki KT. Simultaneous speciation of endogenous and exogenous elements by HPLC/ICP-MA with enriched stable isotopes. Tohoku J Exp Med 1996; 178:

226 Suzuki KT, Itoh M. Metabolism of selenite labeled with enriched stable isotope in bloodstream. J Chromatogr B 1997; 692: Suzuki KT, Shiobara Y, Itoh M, Ohmichi M. Selective uptake of selenite by red blood cells. Analyst 1998; 123: Suzuki KT, Ogra. Metabolic pathway for selenium in the body: speciation by HPLC-ICP MS with enriched Se. Food Addit Contam 2002; 19(10): Swanson CA, Reamer DC, King JC, Levander OA, Veillon C. Metabolism of selenium by pregnant and non-pregnant woman fed controlled diets. Fed Proc 1982;41: 757. Swanson CA, Patterson BH, Levander OA, Veillon C, Taylor PR, Helzlsouer K, McAdam PA, Zech LA. Human [ 74 Se] selenomethionine metabolism: a kinetic model. Am J Clin Nutr 1991; 54: Tabuchi Y, Kondo T, Suzuki Y, Obinata M. Genes involved in nonpermissive temperature-induced cell differentiation in Sertoli TTE3 cells bearing temperature-sensitive simian virus 40 large T-antigen. Biochem Biophys Res Commun 2005; 329: Takahashi K, Cohen HJ. Selenium-dependent glutathione peroxidase protein and activity: immunological investigations on cellular and plasma enzymes. Blood 1986; 68: Takahashi K, Avissar N, Whitin J, Cohen H. Purification and characterization of human plasma glutathione proxidase: a selenoglycoprotein distinct from the known cellular enzyme. Arch Biochem Biophys 1987; Tan JA. The Keshan disease in China: a study of ecological chemio-geography. Nat Georg J Ind 1982; 28: 15. Tang R, Liu H, Wang T, Huang K. Mechanisms of selenium inhibition of cell apoptosis induced by oxysterols in rat vascular smooth muscle cells. Arch Biochem Biophys 2005; 441: Tham DM, Whitin JC, Kim KK, Zhu SX, Cohen HJ. Expression of extracellular glutathione peroxidase in human and mouse gastrointestinal tract. Am J physiol 1998; 275: G

227 Thomson CD, Stewart RDH. Metabolic studies of [ 75 Se] selenomethionine and selenite in the rat. Br J Nutr 1973; 30: Thomson, CD; Stewart, R D. The metabolism of [75Se] in young women. Br J Nutr 1974; 32: Thomson CD, Robinson MF. Urinary and fecal excretions and absorption of a large supplement of selenium: superiority of selenate over selenite. Am J Clin Nutr 1986; 44: Thomson CD, Robinson MF, Butler JA, Whanger PD. Long-term supplementation with selenate and selenomethionine: selenium and glutathione peroxidase (EC ) in blood components of New Zealand women. Br J Nutr 1993; 69 (2): Thomson CD, Robinson MF. The changing selenium status of New Zealand residents. Eur J Clin Nutr 1996; 50: Thomson CD. Selenium and iodine intakes and status in New Zealand and Australia. Br J Nutr 2004; 91: Thompson HJ, Wilson A, Lu J, Singh M, Jiang C, Upadhyaya P, El-Bayoumy K, Ip C. Comparison of the effects of an organic and an inorganic form of selenium on a mammary carcinoma cell line. Carcinogenesis 1994; 15: Thompson HJ, Singh M. Rat models of premalignant breast diseases. J. Mammary Gland Biol.Neoplasia 2000; 5: Treeck O, Haldar C, Ortmann O. Antiestrogens modulate MT1 melatonin receptor expression in breast and ovarian cancer cell lines. Oncol Rep 2006; 15: Truswell AS, Dreosti IE, English RM, Rutishauser IHE, Palmer N. Recommended Nutrient Intakes. Australian Papers Sydney: Australian Professional Publications. Tujebajeva R M, Harney JW, Berry MJ. Selenoprotein P expression, purification, and immunochemical characterization. J Biol Chem 2000; 275: Turner CD, Bagnara JT. General Endocriology, 6th ed. W. G. Philadelphia: Saunders Co. 1976:

228 Ujiie S, Kikuchi H. The relation between serum selenium value and cancer in Miyagi, Japan: 5-year follow up study. Tohoku J Exp Med 2002; 196: Urisini F, Maiorino M, Valente M, Ferri L, Gregolin C. Purification from pig liver of a protein which protects liposomes and biomembranes from peroxidative degradation and exhibits glutathione peroxidase activity on phosphatidylcholine hydroperoxides. Biochim Biophys Acta 1982; 710: Ursini F, Maiorino M, Gregolin C. The selenoenzyme phopholipid hydroperoxide glutathione peroxidase. Biochim Biophys Acta 1985; 839: Ursini F, Heim S, Kiess M, Maiorino M, Roveri A, Wissing J, Flohe L. Dual function of the selenoprotein PHGPx during sperm maturation. Science 1999; 285: Utomo A, Jiang X, Furuta S, Yuan J, Levin DS, Wang YC, Desai KV, Green JE, Chen PL, Lee WH. Identification of a novel putative non-selenocysteine containing phospholipid hydroperoxide glutathione peroxidase (NPGPx) essential for alleviating oxidative stress generated from polyunsaturated fatty acids in breast cancer cells. J Biol Chem 2004; 279: Van Dael P, Davidsson L, Munoz-box R, Fay LB, Barclay D. Selenium absorption and retention from a selenite- or selenate-fortified milk-based formula in men measured by a stable-isotope technique. Br J Nutr 2001; 85: van den Brandt PA, Goldbohm RA, van't Veer P, Bode P, Dorant E, Hermus RJ, Sturmans F. Toenail selenium levels and the risk of breast cancer. Am J Epidemiol 1994; 140: van den Brandt PA, Zeegers MP, Bode P, Goldbohm RA. Toenail selenium levels and the subsequent risk of prostate cancer: a prospective cohort study. Cancer Epidemiol Biomarkers Prev 2003; 12: van Noord PAH, Maas MJ, van der Tweel I, Collette C. Selenium and the risk of postmenopausal breast cancer in the DOM cohort. Breast Cancer Res Treat 1993; 25: Vendeland SC, Butter JA, Whanger PD. Interstinal absorption of selenite, selenate, and selenomethionine in the rat. J Nutr Biochem 1992; 3:

229 Vendeland SC, Deagen JT, Butler JA, Whanger PD. Uptake of selenite, selenomethionine and selenate by brush border membrane vesicles isolated from rat small intestine. Biometals 1994; 7: Venkateswaran V, Klotz LH, Fleshner NE. Selenium modulation of cell proliferation and cell cycle biomarkers in human prostate carcinoma cell lines. Cancer Res 2002; 62: Vikoren T, Bernhoft A, Waaler T, Handeland K. Liver concentrations of copper, cobalt, and selenium in wild Norwegian red deer (Cervus elaphus). J Wildl Dis 2005; 41: Vogt PH. Molecular genetics of human male infertility: from genes to new therapeutic perspectives. Curr Pharm Des 2004; 10: Vogt TM, Ziegler RG, Graubard BI, Swanson CA, Greenberg RS, Schoenberg JB, Swanson GM, Hayes RB, Mayne ST. Serum selenium and risk of prostate cancer in U.S. blacks and whites. Int J Cancer 2003; 103: Wang DS, Yang WZ, Sun QY, Gao JP. The studies on Se in environment and human body in Kaschin-Beck disease endemic area in Henan province and exploration of etiology. Chin J Endemiol 1987; 6; Wang L. Plasma volume, cell volume, total blood volume and F cells factor in the normal and splenectomized Sherman rat. Am J Physiol 1959; 196: Wang Y, Wang L, Zhao J, Qiao Z. Estrogen, but not testosterone, down-regulates cytokine production in nicotine-induced murine macrophage. Methods Find Exp Clin Pharmacol 2005; 27: Waters DJ, Chiang EC, Cooley DM, Morris JS. Making sense of sex and supplements: differences in the anticarcinogenic effects of selenium in men and women. Mutat Res 2004; 551: Waters DJ, Shen S, Glickman LT, Cooley DM, Bostwick DG, Qian J, Combs GF, Morris JS. Prostate cancer risk and DNA damage: translational significance of selenium supplementation in a canine model. Carcinogenesis 2005; 26: Webster NJG, Green S, Jin JR, Chambon P. the hormone-binding domains of the estrogen and glucocorticoid receptors contain an inducible transcription activation function. Cell 1988; 54:

230 Whanger PD, Pedersen ND, Hatfield J, Wewig PH. Absorption of selenite and selenomethionine from ligated digestive tract segments in rats. Proc Soc Exper Biol Md 1976; 153:295. Whanger PD, Beilstein MA, Thomson CD, Robinson MF, Howe M. Blood selenium and glutathione peroxidase activity of populations in New Zealand, Oregon, and South Dakota. FASEB J 1988a; 2: Whanger PD, Butler JA. Effects of various dietary levels of selenium as selenite or selenomethionine on tissue selenium levels and glutathione peroxidase activity in rats. J Nutr 1988b; 118: Wilber CG.. Toxicology of selenium: a review. Clin Toxicol 1980; 17: Wilson DS, Tappel AL. Binding of plasma selenoprotein P to cell membranes. J Inorg Biochem 1993; 51: Wolffram S, Berger B, Grenacher B, Scharrer E. Transport of selenoamino acids and their sulfur analogues across the intestinal brush border membrane of pigs. J Nutr 1989; 119: Wolk A. Diet, lifestyle and risk of prostate cancer. Acta Oncol 2005; 44: Weiss SL, Evenson JK, Thompson KM, Sunde RA. Dietary selenium regulation of glutathione peroxidase mrna and other selenium-dependent parameters in male rats. J Nutr Biochem 1997; 8: Wilber CG. Toxicology of selenium: a review. Clin Toxicol 1980; 17: Xia Y, Zhao X, Zhu L, Whanger PD. Metabolism of selenate and selenomethionine by a selenium-deficient population of men in China. J Nutr Biochem 1992; 3: Xia Y, Hill KE, Byrne DW, Xu, J, Burk RF. Effectiveness of selenium supplements in a low-selenium area of China. Am J Clin Nutr 2005; 81: Yan J, Barrett JN. Purification from bovine serum of a survival-promoting factor for cultured central neurons and its identification as selenoprotein-p. J Neurosci 1998; 18:

231 Yang JB. Experimental pathological studies of the pathogenic effect of oxysporin on Kascin-Beck disease. Chin Med J 1982; Yang JC, Morrison-Plummer J, Burk RF. Purification and quantitation of a rat plasma selenoprotein distinct from glutathione peroxidase using monoclonal antibodies. J Biol Chem 1987; 262: Yamamoto T, Ohkuwa T, Itoh H, Sato Y, Naoi M. Effect of gender differences and voluntary exercise on antioxidant capacity in rats. Biochem Biophys Res Commun 2002; 132: Yang G, Wang S, Zhou R, Sun S. Endemic selenium intoxication of humans in China. Am J Nutr. 1983: 37: Yang G, Zhu L, Liu S, Gu L, Qian P, Huang J, Lu M. Human selenium requirements in China. In: Combs GF Jr, Levander OA, Spallholz JE, Oldfield JE, eds. Selenium in Biology and Medicine New York: Avi. Pp Yang G, Zhou R, Yin S, Gu L, Yan Y, Liu Y, Liu Y, Li X. Studies of safe maximal daily dietary selenium intake in a seleniferous area in China. I. Selenium intake and tissue selenium levels of the inhabitants. J Trace Electrolytes Health Dis 1989; 3: Yoshimura S, Watanabe K, Suemizu H, Onozawa T, Mizoguchi J, Tsuda K, Hatta H, Moriuchi T. Tissue specific expression of the plasma glutathione peroxidase gene in rat kidney. J Biochem 1991; 109: Yoshizawa K, Willett WC, Morris JS, Stampfer MJ, Spiegelman D, Rimm EB, Giovannucci E. Study of prediagnostic selenium level in toenails and the risk of advanced prostate cancer. J Natl Cancer Inst 1998; 90: Yu S, Beynen AC. The lowering effect of high copper intake on selenium retention in weanling rats depends on the selenium concentration of the diet. J Anim Physiol Anim Nutr 2001; 85(1-2): Zachara BA, Wardak C, Didkowski W, Maciag A, Marchaluk E. Changes in blood selenium and glutathione concentrations and glutathione peroxidase activity in human pregnancy. Gynecol Obstet Invest 1993; 35:

232 Zachara B, Pawluk H, Bloch-Boguslawska E, Sliwka K, Korenkiewicz J, Skok Z, Ryc, K. Tissue level, distribution, and total body selenium content in healthy and diseased humans in Poland. Arch Environ Health 2001; 56(5): Zhao H, Whitfield ML, Xu T, Botstein D, Brooks JD. Diverse effects of methylseleninic acid on the transcriptional program of human prostate cancer cells. Mol Biol Cell 2004; Zhao H, Brooks JD. Selenomethionine induced transcriptional programs in human prostate cancer cells. J Urol 2007;177: Zhu Y, Sulivan LL, Nair SS, Williams CC, Pandey AK, Marrero L, vadlamudi RK, Jones FE. Coregulation of Estrogen Receptor by ERBB4/HER4 Establishes a Growth-Promoting Autocrine Signal in Breast Tumor Cells. Cancer Res 2006; 66:

233 APPENDIX A, ANIMAL USE PROTOCOLS APPROVED BY ILACUC, THE OHIO STATE UNIVERSITY 218

234 219

235 220

236 221

237 222

238 223

239 224

240 225

241 226

242 227

243 228

244 229

245 230

246 APPENDIX B, COMPOSITION OF THE CUSTOM DIET 231

247 Formulated by Research Diets, Inc. 232

248 APPENDIX C, 75 Se USE PROTOCOLS APPROVED BY RADIATION SAFETY, EHS, THE OHIO STATE UNIVERSITY 233

249 234

250 235

251 236

252 Determination of the effects of estrogen availability on selenium distribution into tissues using tracer 75 Se. Methods Animal model. Female Sprague Dawley rats will be ovariectomized at 7 weeks of age, and be randomized into the following two groups: Group 1-bilaterally ovariectomized receiving a placebo pellet (n=16) Group 2-ovariectomized receiving estrogen replacement (n=16) 75 SeO3 2- administration: 75 Se as sodium selenite will be purchased from the University of Missouri Research Reactor Facility (Columbia, MO). After 5 weeks treatment of estrogen/placebo, each rat will be administered 60 µci of 75 Se as sodium selenite (about 200 µg Ci/kg body mass) by gavage at 12 weeks of age. 1 hour, 3 hour, 6 hours, and 24 hours after 75 Se administration, 8 rats (4 from group 1 and 4 from group 2) will be killed at each time point. Blood, liver, kidney, heart, brain, mammary tissue, and the gastrointestinal (GI) tract will be taken for the measurement of 75 Se activity. Measurement of the 75 Se tracer. The 75 Se activity in plasma, erythrocytes, GI tract, and tissues will be measured by Cobra II auto-gamma counter (Packard Instrument Company, Meriden, CT). SDS-PAGE analysis of 75 Se incorporation into selenoproteins: Plasma and tissue samples will be subjected to SDS-PAGE on 12% gels. After electrophoresis, the gels will be stained with Coomassie Blue R-250 (2h) and then destained (3-4h). The gels will then 237

253 be dried and exposed to Kodak XAR film. The 75 Se-labeled proteins will be identified by autoradiography. The autoradiographs will be scanned and quantitated using Scion Image software (Scion, Frederick, MD). In immunoprecipitation assay, plasma will be subjected to immunoprecipitation by polyclonal antibodies against rat selenoprotein P and human GSHPx-3. The immunoprecipitates will be separated by SDS-PAGE. The 75 Se-labeled proteins will be identified by autoradiography. 238

254 239

255 240

256 241

257 242