Effects of selenium in the intracellular peroxideremoval

University of Iowa Iowa Research Online Theses and Dissertations Fall 2011 Effects of selenium in the intracellular peroxideremoval system Weipeng Bian University of Iowa Copyright 2011 Weipeng Bian This thesis is available at Iowa Research Online: http://ir.uiowa.edu/etd/2674 Recommended Citation Bian, Weipeng. "Effects of selenium in the intracellular peroxide-removal system." MS (Master of Science) thesis, University of Iowa, 2011. http://ir.uiowa.edu/etd/2674. Follow this and additional works at: http://ir.uiowa.edu/etd Part of the Other Biochemistry, Biophysics, and Structural Biology Commons

EFFECTS OF SELENIUM IN THE INTRACELLULAR PEROXIDE-REMOVAL SYSTEM by Weipeng Bian A thesis submitted in partial fulfillment of the requirements for the Master of Science degree in Free Radical and Radiation Biology in the Graduate College of The University of Iowa December 2011 Thesis Supervisor: Professor Garry R. Buettner

Graduate College The University of Iowa Iowa City, Iowa CERTIFICATE OF APPROVAL MASTER S THESIS This is to certify that the Master s thesis of Weipeng Bian has been approved by the Examining Committee for the thesis requirement for the Master of Science degree in Free Radical and Radiation Biology at the December 2011 graduation. Thesis Committee: Garry R. Buettner, Thesis Supervisor Prabhat C. Goswami Apollina Goel

ACKNOWLEDGMENTS I would like to express my sincere gratitude to my advisor, Dr. Garry R. Buettner. My academic and professional progress benefited from his profound insight, patience and guidance. His encouragement and kind support made this thesis possible. I would like to thank my committee members, Dr. Prabhat C. Goswami and Apollina Goel. Their contribution, guidance and critical advice made this thesis complete. I would like to express my appreciation to all my lab members, Brett A. Wagner, Juan Du, Jordan R. Witmer, Thomas Joost Van t Erve, Malvika Rawal and Cameron Cushing for their assistance. Without their help, group discussions and support, I couldn t learn so much and finish this thesis. In addition, I would like to thank all members of the Free Radical and Radiation Biology program for their discussions, help and support. I would like to thank Laura Hefley and Jennifer Dewitte for their secretarial support and assistance. Finally, I would like to express my happiness to be in Dr. Buettner s lab in the past three years. He is the one who gave me the chance and I could meet so many nice people, learn so much new knowledge, and make so much progress in my life. This experience is inestimable and will benefit me for all rest of my life. I am glad it happened and I will remember this forever. ii

TABLE OF CONTENTS LIST OF FIGURES...v CHAPTER I INTRODUCTION...1 1.1 Cellular Peroxide-Removal System...1 1.2 Superoxide Dismutase...2 1.3 Glutathione Peroxidase...2 1.4 Thioredoxin Reductase...3 1.5 Peroxiredoxin...3 1.6 Catalase...4 1.7 Selenium...5 1.8 Overview of Thesis...6 CHAPTER II CELLULAR RESPONSES TO SELENIUM EXPOSURE...8 2.1 Introduction...8 2.2 Materials And Methods...9 2.2.1 Clonogenic Survival Assay...9 2.2.2 Glutathione Peroxidase Activity Assay...9 2.2.3 TrxR Activity Assay...9 2.2.4 Western Blotting Assay...10 2.2.5 H 2 O 2 -Removal Assay...10 2.3 Results And Discussion...11 2.3.1 Cytotoxicity...11 2.3.2 Enzyme Expressions...12 2.3.3 GPx and TrxR Activities...12 2.3.4 H 2 O 2 Removal...14 2.4 Conclusions...15 CHAPTER III SELENIUM AND GSH PRODUCE REACTIVE OXYGEN SPECIES...31 3.1 Introduction...31 3.2 Materials And Methods...31 3.2.1 Oxygen Consumption Assay...31 3.2.2 GSSeSG Detection Assay...32 3.3 Results And Discussion...32 3.3.1 Selenite and GSH react to consume oxygen....32 3.3.2 Seleno-L-methionine and GSH reacted to consume oxygen....33 3.3.3 GSSeSG was formed during the reactions of GSH with selenite or SeM....34 3.3.4 Peroxide generation during reaction of GSH and selenite and SeM. Superoxide, O 2 -, was detected by SOD and catalase slowed the OCR by converting H 2 O 2 to H 2 O and O 2....34 3.4 Conclusions...35 iii

CHAPTER IV THEORY OF ENZYMATIC COOPERATION...42 4.1 Introduction...42 4.2 Results And Discussion...43 4.3 Conclusions...46 REFERENCES...49 iv

LIST OF FIGURES Figure 1.1 The hydrogen peroxide-removal system....7 Figure 2.1 The paradox of selenium effects vs. concentration....16 Figure 2.2 Selenite of 30 nm shows neither cytotoxicity nor proliferative effect.....17 Figure 2.3 HL60 cell growth curves with treatment of different doses of selenite or SeM....18 Figure 2.4 Selenium increased protein expressions of GPx-1 and TrxR, but not CAT or Prdx-1 in MB231 cells....19 Figure 2.5 Selenium increased TrxR-1 and GPx-1 in PC3 cells, but had no effects on Prdx-1, Prdx-3 or CAT....20 Figure 2.6 Prdx-1 and CAT were not changed in ASPC, Panc-1, or MCF7 cells....21 Figure 2.7 Prdx-1 and Prdx-3 were not increased by addition of selenium in PC3 cells....22 Figure 2.8 Number of effective active glutathione peroxidase enzymes after treatment of selenite in PC3 cells and HL60 cells....23 Figure 2.9 Similar trends on GPx activities on other cells....25 Figure 2.10 TrxR activities with treatment of different SeM doses in HL60 cells....26 Figure 2.11 Similar trends on TrxR activities on other cells with treatment of selenite....27 Figure 2.12 GPx activities were increased in ASPC, MB231, MCF-7 and Panc-1 cells....28 Figure 2.13 Capacity for removal of extracellular H 2 O 2 by PC3 cells is not affected significantly by increased media Se....29 Figure 2.14 H 2 O 2 -removal in MCF7, MB231, Panc-1 and ASPC cells....30 Figure 3.1 Possible mechanism of the reaction between selenite and GSH.....37 Figure 3.2 Influence of ratios of GSH to selenite on their reaction.....38 Figure 3.3 Influence of ratios of GSH to SeM on their reaction.....39 Figure 3.4 Reactions of GSH and selenite or SeM show different patterns.....40 Figure 3.5 Formation of O 2 - and H 2 O 2 in the reaction of GSH and SeM or selenite.....41 v

Figure 4.1 Simplified mechanism of the balance needed for an appropriate cellular redox environment....47 Figure 4.2 Changes to a new steady-state and optimal bundle...48 vi

1 CHAPTER I INTRODUCTION Cells can produce reactive oxygen species (ROS) such as hydrogen peroxide (H 2 O 2 ) after exposure to diverse intracellular or environmental radical-generating conditions. The role of ROS in human health is being currently recognized. ROS at normal concentrations are essential to cell growth. For instance, H 2 O 2 is important in cell signaling and appropriate concentrations of H 2 O 2 are required for cell proliferation. However, high-level ROS, such as H 2 O 2, generates oxidative stress and causes damage to cells. Therefore, the oxidative stress has to be reduced and levels of cellular peroxides must be lowered to keep cells healthy [1, 2]. Some important enzymes, such as superoxide dismutase (SOD), glutathione peroxidase (GPx), thioredoxin reductase (TrxR), peroxiredoxin (Prdx), catalase (CAT), participate in this process to keep the appropriate balance of ROS. In this introductory chapter, cellular peroxide-removal system is introduced. An overview of the peroxide-removal is presented in Figure 1.1 [3]. Superoxide dismutase (SOD), glutathione peroxidase (GPx), thioredoxin reductase (TrxR), peroxiredoxin (Prdx), catalase (CAT), selenium, and their roles in peroxideremoval system are discussed. 1.1 Cellular Peroxide-Removal System Superoxide dismutase, glutathione peroxidase, catalase, peroxiredoxin and thioredoxin reductase are involved directly with peroxide formation and consumption [2, 3]. As seen in Figure 1.1, there are three nodes for removal of H 2 O 2, CAT, GPx and Prdx, while SOD is acting to produce H 2 O 2. A brief discussion on these enzymes follows.

2 1.2 Superoxide Dismutase Superoxide dismutase (SOD) is a family of enzymes that catalyzes the dismutation of superoxide into oxygen and H 2 O 2. It includes three different types, Mn- SOD (which binds manganese), Cu/Zn-SOD (which binds both copper and zinc) and ECSOD (which is extracellular superoxide dismutase). Because of their function, this family behaves as an important antioxidant defense to oxygen exposure in cells [4, 5, 6, 7]. Reactions 1 and 2 show the reaction mechanisms of Mn-SOD step-by-step. Reaction 3 shows the net reaction of Mn-SOD. Mn 3+ -SOD + O 2 - O 2 + Mn 2+ -SOD Reaction 1 Mn 2+ -SOD + O 2 - + 2H + H 2 O 2 + Mn 3+ -SOD Reaction 2 Mn-SOD O - 2 + O - 2 + 2H + O 2 + H 2 O 2 Reaction 3 The catalytic reactions of Cu/Zn-SOD and ECSOD are parallel to the reactions above. 1.3 Glutathione Peroxidase GPx reduces H 2 O 2 to H 2 O in combination with oxidation of glutathione (GSH). Compared with Prdx and CAT, most GPxs contain selenium at their reaction center. The reaction is at the selenocysteine site of each monomer of GPx. GPx-1 is a tetramer. GPx initially reacts with H 2 O 2 at the selenocysteine site. Cys-SeOH is formed and then reduced by GSH. The overall reaction is shown by Reaction 4.

3 GPx 2GSH + H 2 O 2 GSSG + 2H 2 O Reaction 4 GPx-1 [8]. There are at least five major GPxs in mammalian cells. The most abundant is 1.4 Thioredoxin Reductase Thioredoxin reductase is a homodimeric selenoprotein. It functions mainly to catalyze the NADPH-dependent reduction of thioredoxins (Trx). In mammalian cells, TrxR basically has three isoforms: TrxR-1, the most dominant one, TrxR-2, mostly in mitochondria, and SpTrxR mostly in male sperm [9]. These three isoforms correspond to the three members of the Trx family, Trx1, Trx2, and Trx3. The overall reaction is shown by Reaction 5. TrxR Trx-(S-S) + NADPH + H + TRx-(SH) 2 + NADP + Reaction 5 1.5 Peroxiredoxin Peroxiredoxin (Prdx) is a novel peroxidase family, which formerly was called thioredoxin peroxidase. Prdxs can reduce H 2 O 2 and alkyl hydroperoxides to water and alcohol, respectively [10, 11]. There are six members of the mammalian Prdx family,

4 Prdx I VI. These members locate in different organelles in cells. Prdx I and II are in the cytosol; Prdx III specifically locates in mitochondria; Prdx IV contains a NH-terminal signal sequence for secretion and is found in the endoplasmic reticulum and extracellular space; Prdx V is found in mitochondria and peroxisomes; and Prdx VI is found in the cytosol [12]. Based on catalytic mechanisms, they can also be classified into 3 categories: typical 2-cysteine, atypical 2-cysteine and 1-cysteine. Typical 2-cysteine Prdxs form disulfide bonds between two molecules; atypical 2-cysteine Prdxs form a disulfide bond in the same molecule; 1-cysteine Prdxs has only 1 cysteine per molecule joining the reaction [13]. The overall reaction is shown by Reaction 6. Note that the electrons to reduce H 2 O 2 to H 2 O are from thiol group, parallel to Reaction 4 with GPx. -(SH)2 + H 2 O 2 2 H 2 O + -S-S- Reaction 6 1.6 Catalase Unlike Prdxs, catalase is a dismutase, which is localized in peroxisomes in mammalian cells. It reduces H 2 O 2 to H 2 O, but at the same time, it also oxidizes a second H 2 O 2 (or H 2 R) to O 2 (or R) [14]. H 2 R includes phenols, formic acid, formaldehyde and ethanol. CAT removes H 2 O 2 formed in peroxisomes as well as H 2 O 2 from the cytosol that diffuses into peroxisomes. Reactions 7 and 8 show the reactions of CAT.

5 CAT 2 H 2 O 2 2 H 2 O + O 2 Reaction 7 CAT H 2 O 2 + H 2 R 2 H 2 O + R Reaction 8 Reaction 7 is the catalytic reaction while Reaction 8 is a peroxidatic reaction. 1.7 Selenium Selenium is of fundamental importance to human health. Daily intake of 200 µg has considerable immunoenhancing effects. The average plasma concentration of selenium is around 1 µm [15]. Deficiency of selenium can cause disease, such as Keshan disease [15, 16]. In cell culture, the main selenium source is fetal bovine serum. In this thesis, sodium selenite (Na 2 SeO 3 ) and seleno-l-methionine (SeM) were used as additional sources of selenium (Se). In cells, selenium has been identified as a component in about 35 selenoproteins, GPx and TrxR are two of them [15, 16]. Selenium plays an essential role in both the glutathione (GSH) and the thioredoxin peroxide-removal systems. Each system has an essential Se-containing enzyme: GPx and TrxR, which contribute to H 2 O 2 removal by cells together with catalase. Therefore, Se could influence peroxide removal by cells indirectly. Ganther proposed a possible mechanism of the reaction between sodium selenite and GSH [23], which is evidence for Se directly influencing peroxide removal. Later, investigators found that superoxide (O - 2 ), one of the reactive oxygen species, was

6 detectable during this reaction [24, 25, 26, 27]. This idea comes to a totally opposite direction from the protective role of Se. Se compounds are able to react with GSH directly to convert O 2 to O - 2, which might be an explanatory mechanism Se cytotoxicity [26]. 1.8 Overview of Thesis Chapter I is an introduction to the peroxide-removal system of cells and tissues in Figure 1.1. Chapter II focuses on cellular responses to selenium exposure. Different cellular experiments are shown to demonstrate the different effects of selenium on redox states. Chapter III presents information on the reaction of selenium and GSH. Experiments were conducted to explore the underlying chemistry. Chapter IV brings forward a generalization and hypothesis that the redox states are all about balance and a proposal how to model this balance.

Figure 1.1 The hydrogen peroxide-removal system [3]. There are at least three nodes for the intracellular removal of H 2 O 2 ; with GPx (H 2 O 2 is reduced to H 2 O, and glutathione (GSH) is oxidized to GSSG), CAT (CAT firstly recruits one H 2 O 2 to form compound I and water, and then oxidizes a second H 2 O 2 to O 2 ); and Prdx (Prdx removes H 2 O 2 using reducing equivalents from thioredoxin (Trx)). 7

8 CHAPTER II CELLULAR RESPONSES TO SELENIUM EXPOSURE 2.1 Introduction The role of selenium in the peroxide-removal system has been shown to be paradoxical. As well known, selenium plays an essential role in both the GSH and the Trx nodes of peroxide-removal system, which contribute to H 2 O 2 removal by cells together with catalase. However, studies have shown that selenium compounds, such as sodium selenite (Na 2 SeO 3 ) and seleno-l-methionine (SeM), are able to react with GSH directly and convert O 2 to O - 2 and H 2 O 2, which might explain the cytotoxicity by Se at high doses. Selenius et al. also proposed similar selenium paradoxes (Figure 2.1) [28]. GPx and TrxR are very important for peroxide removal. Depletion of either one could cause adverse cell responses, such as apoptosis [29, 30, 31, 32, 33]. Data show that GPx and TrxR activities achieve their maximum at very different Se concentrations. In another words, GPx and TrxR might have quite different demands on Se, which is not as simple as the ratio of 2:1 that is the ratio of monomers to form tetramer or dimer. Investigations on this difference might lead us to further understanding of the different behaviors of GPx and TrxR in cells. Moreover, supplementation with 30 nm selenite has been suggested in the literatures. Studies on GPx and TrxR activities upon different Se supplementation might lead us to optimal amounts for Se supplementation in cell culture experiments based on different requirements. In this chapter, the effects of selenium on the peroxide-removal system will be studied and discussed. The changes of GPx and TrxR from dose-dependent Se supplementation are investigated.

9 2.2 Materials And Methods 2.2.1 Clonogenic Survival Assay Cells with or without treatment were plated in triplicate into 60-mm tissue culture dishes (400 cells per well) and were incubated for 2 weeks to allow colony formation. The colonies were then fixed in 70% ethanol and stained with 0.1% Coomassie blue (Amresco, Solon, OH). Colonies with a population of more than 50 cells were counted. 2.2.2 Glutathione Peroxidase Activity Assay GPx activity was determined with a spectrophotometer by coupling the oxidation of GSH and NADPH. Briefly, H 2 O 2 was prepared in 50 ml of water. NADPH and cell samples were then added to working buffer, which contained 1 mm of GSH and 100 U/mL of glutathione reductase. H 2 O 2 was then added into the mixture after 5 minutes, which was then monitored spectrophotometrically. 2.2.3 TrxR Activity Assay The protocol was revised from Sigma. TrxR was used as positive control. NADPH was prepared in working buffer (final concentration: 100 nm potassium phosphate with 10 mm EDTA and 0.24 mm NADPH). DTNB (39.6 mg/ml) was prepared in DMSO. TrxR inhibitor solution was used to exclude any background activity. Enzyme or sample, assay buffer, diluted inhibitor solution, working buffer, and DTNB

10 were added to a 96-well plate ordinally. Absorbance was measured at 412 nm with a microplate reader. 2.2.4 Western Blotting Assay Proteins were collected 48 h after cellular treatment and electrophoresed in SDSpolyacrylamide running gel and a 5% stacking gel. Total protein of 20 µg was loaded. The proteins were then electrotransferred to polyvinylidene fluoride (PVDF) membranes. After blocked in 5% non-fat milk TPBS buffer, the membranes were then treated with primary and secondary antibodies. Afterwards, they were washed 5 min for 3 times and treated with western blot detection solution and exposed to x-ray films. 2.2.5 H 2 O 2 -Removal Assay Cells at a density of forty thousand cells per well were plated to 96-well plates. Medium was removed and cells were washed with HBSS buffer 3 times. HBSS of 50 µl was added to each well. Another 100 µl of HBSS added with HEPES, PHPA, HRP and sodium bicarbonate were added to each well. H 2 O 2 (10 µm) was then added at different time points. Standards were also made at the same time in wells without cells. The plate was read at excitation wavelength of 345 nm and emission wavelength of 425 nm.

11 2.3 Results And Discussion 2.3.1 Cytotoxicity To avoid using toxic concentrations of selenium in our experiements, cytotoxicity assay (cell survival assay) was first carried out in PC3 and HL60 cells. Results can be seen in Figures 2.2 and 2.3. In Figure 2.2, PC3 cells were first treated with selenite (30 nm) for 48 h and then replated for clonogenic assay. Cells were separated into four groups (Figure 2.2 Group A, B, C and D). Group B, C and D shows similar colony number to Group A. Selenium of that concentration (30 nm) is not cytotoxic. In Figure 2.3, selenite and SeM were used with concentrations of 10 nm, 1 µm, 10 µm and 1 mm. HL60 cells exhibited growth inhibition with the treatment of 10 µm selenite, while concentrations of 1 µm and below didn t show any inhibition effects on HL60 growth. Surprisingly, HL60 cells could take SeM at very high doses up to 1 mm and higher without showing any growth inhibition consequences. One possible explanation is that SeM could scavenge H 2 O 2 present on amino acids, peptides, and proteins stoichiometrically and catalytically in the presence of GSH or TrxR system and hence SeM is critical for peptide and protein oxidation products removal [34]. This result also indicates that investigation with selenium treatment can be done in a wide concentration range.

12 2.3.2 Enzyme Expressions To determine the responses of protein expression to selenium exposure in cells, western blotting assay was carried out. MB231, PC3, MCF-7, ASPC, and Panc-1 cells were used. Results can be seen from Figures 2.4 to 2.7. After treatment with 30 nm of selenium, there were no significant changes of Prdx and CAT in MB231 (Figure 2.4), PC3 (Figures 2.5, 2.7), ASPC (Figure 2.6), Panc-1 (Figure 2.6), and MCF-7 (Figure 2.6) cells. Selenium (30 nm) increased the levels of GPx-1, TrxR-1 in both MB231 and PC3 cells (Figures 2.4, 2.5C and 2.5D). Comparison of Prdx-1 and Prdx-3 was also screened in PC3 cells with treatment of 30 nm selenite. PC3 cells and PC3 cells treated with selenium were then used to determine the difference on Prdx-1 and Prdx-3 (Figure 2.7). The amount of Prdx-3 in PC3 cells is apparently lower than that of Prdx-1, but there were no significant difference on Prdx-1 and Prdx-3 between PC-3 cells and PC-3 cells treated with selenium. For future studies, we need to do the same experiment on HL60 cells to show the enzyme expression response to selenium. We can also use selenium dose dependently to see if this experiment is consistent to activity assay results in doses. 2.3.3 GPx and TrxR Activities GPx and TrxR activity assays were then used to determine activities of GPx and TrxR after the treatment of selenium compounds. As expected, because GPx is selenium-dependent, addition of selenite increased GPx activity in PC3 cells (Figure 2.8A). Experiments on HL60 cells showed that the GPx activities went up dose-dependently with the treatment of selenite from control to 200 nm (Figure 2.8, Control: 6.2 10 12 active enzyme monomer per mg protein; 10 nm:

13 1.1 10 13 active enzyme monomer per mg protein; 20 nm: 1.6 10 13 active enzyme monomer per mg protein; 200 nm: 2.0 10 13 active enzyme monomer per mg protein; 2000 nm: 1.9 10 13 active enzyme monomer per mg protein). At lower doses of selenite treatment, GPx activity has a greater rate of change than that at higher doses of selenite treatment (Figure 2.8A). When treating cells with 2000 nm, the GPx activity decreased slightly, which might indicate that HL60 started experiencing apoptosis at the very high level [33]; but this requires further research. GPx activities in HL60 cells tend to achieve maxima with treatment of approximately 100 nm selenite. This trend in MB231 cells has been observed by Hu et al. in MCF-7 cells (Figure 2.9) [35] and by Makropoulos et al. in Jurkat T cells [36]. In both cell lines, maximum GPx activities were achieved with treatment of 100 nm selenite. TrxR activities increased in HL60 cells dose-dependently with treatment of SeM from control to 10 µm. However, TrxR activity tends to reach its maximum at a much higher Se concentration compared to GPx activity, and at low treatment doses (less than 200 nm), TrxR changed very little (Figure 2.10). Similar trends of TrxR activity can also be seen in MCF-7, A549, and HT-27 cells [36, 37] with treatment of selenite (Figure 2.11, adapted from [36]). In all these cells, TrxR activities can increase with treatment of up to 5 µm or even higher Se. These results show that although GPx and TrxR are both Se-dependent, their sensitivities to Se and demands for Se are indeed quite different. TrxR activity requires more Se, and GPx activity might reach its maximum at a much lower Se dose. For our future studies, we need to expand GPx activity assay to a broader range of Se concentrations in HL60 cells. SeM should be used as well. Besides, we need to do the same GPx activity assay in MB231 cells. For TrxR activity, we need to repeat

14 the experiment with treatment of selenite in HL60 cells. In addition, we would also like to measure TrxR activity in MB231 cells. If time permits, we can also do SOD and catalase activity assays to exclude any disturbing possibilities from SOD and catalase. Furthermore, GPx activities were increased in MCF-7, ASPC, MB231, MCF-7 and Panc-1 cells. MCF-7 cells increased GPx activities by 1.9-fold, MB231 3.7-fold, ASPC 2.1-fold, and Panc-1 3.9-fold (Figure 2.12). Besides, MCF-7 had the lowest GPx activity, compared to the other four cell lines. 2.3.4 H 2 O 2 Removal Results (Figures 2.4-2.12) above suggest that a higher level of GPx in PC3 cells should lead to greater potential to remove H 2 O 2 (either exogenous or endogenous) by supplementation with selenite. Interestingly, addition of Se did not increase the rate of exogenous H 2 O 2 removal by PC3 cells. There was no statistically significant difference between groups with or without Se treatment (Figure 2.13). The absolute value even went down slightly from 2.2 10-12 L s -1 cell -1 to 1.8 10-12 L s -1 cell -1. This is interesting in that the whole cell exogenous H 2 O 2 removing ability did not correspond to GPx activity and protein level increase and was not significantly increased, though Johnson et al. (2002, 2010) showed GPx-1 participated in exogenous H 2 O 2 removal. To see if other cells could still exhibit similar phenomena, we further examined MCF-7, MB231, ASPC and Panc-1 cells. Interestingly, total H 2 O 2 removal in these cells all showed no statistically significant changes with Se treatment (Figure 2.14), though different cells exhibited various H 2 O 2 removal abilities (MCF-7, 3.8 10-13 L s -1 cell -1 ; MB231, 4.5 10-13 L s -1 cell -1 ; ASPC, 5.0 10-13 L s -1 cell -1 ; Panc-1, 4.8 10-13 L s -1 cell -

15 1 ). Among these cells, MCF-7 cells eliminated H 2 O 2 relatively slowly, 17% less than that of MB231. After adding Se, GPx activities in these cells all went up. However, H 2 O 2 removal remains at the similar level. 2.4 Conclusions From the above studies, addition of selenium increases both GPx and TrxR protein expressions and enzymatic activities in cells, with CAT and Prdx expression not affected. The responses of GPx and TrxR activities to selenium supplement show different patterns. GPx is more likely to reach the threshold at low selenium concentrations (50 nm ~ 200 nm of selenium) in HL60 and MCF7 cells. TrxR is more likely to reach the threshold at much higher selenium concentrations (higher than 1 µm of selenium). However, peroxide-removal appeared no significant change with treatment of 30 nm selenite. Supplementation of selenite increased GPx and TrxR, which indicates H 2 O 2 should be removed at a greater rate. However, the rate of removal of bolus 10 µm H 2 O 2 in PC3 cells did not change. This might be because our method is not sensitive enough to detect the changes made by GPx. With treatment of 10 µm H 2 O 2, GSH can be depleted quickly and CAT dominates the peroxide-removal, which is not seleniumdependent. Therefore, there seem no significant changes on H 2 O 2 -removal when large amount of H 2 O 2 are presented to cells. This needs further investigation.

Figure 2.1 The paradox of selenium effects vs. concentration. Note the logarithmic scale for selenium concentrations from low to high. Se depletion and over exposure both could result in cell death. At moderate concentrations, Se exhibits antioxidant action, e.g. it helps to prevent cancer, but can be prooxidative either at very low as well as very high concentrations. (Figure adapted from [28]) 16

17 60 Survival Colony Number 50 40 30 20 10 0 Group A Group B Group C Group D Figure 2.2 Selenite of 30 nm shows neither cytotoxicity nor proliferative effect. Numbers of colonies from clonogenic survival assays on PC3 cells are shown. Group A: PC3 cells were grown 48 h and then re-plated with medium without Se; Group B: PC3 cells grew 48 h and then were re-plated with medium with selenite (30 nm); Group C: PC3 cells were treated with selenite (30 nm) and grew 48 h and then were re-plated with medium without selenite; Group D: PC3 cells were treated with selenite (30 nm) and grew 48 h and then were re-plated with medium with selenite (30 nm). No significant differences were observed, p > 0.05 (n = 3).

18 A B Figure 2.3 HL60 cell growth curves with treatment of different doses of selenite or SeM. A: HL60 growth curves with treatment of selenite (control, 10 nm, 1 µm, 10 µm and 1 mm). Treatment of 1 µm selenite didn t show growth inhibition while treatment with 10 µm started inhibiting HL60 growth from Day 2. Changes were significant (p < 0.01). HL60 cell growth was completely inhibited with treatment of 1 mm selenite. B: HL60 growth curves with treatment of SeM (control, 10 nm, 1 µm, 10 µm and 1 mm). HL60 cells show no apparent growth inhibition with up to exposure to as much as 1 mm SeM in the growth media.

Figure 2.4 Selenium increases protein expression of GPx-1 and TrxR, but not CAT or Prdx-1 in MB231 cells. Protein levels of redox enzymes, GPx-1, TrxR-1, catalase and Prdx-1, were studied in MB231 cells with or without supplement of selenite (30 nm). Selenite did not change Prdx-1 and catalase in MB231 cells. Selenite increased GPx-1 and TrxR-1 protein levels after treating cells with 30 nm selenite. (n = 3) 19

A B 20 C D Figure 2.5 Selenium increased TrxR-1 and GPx-1 in PC3 cells, but had no effect on Prdx-1, Prdx-3 or CAT. Protein levels of redox enzymes, GPx-1, TrxR-1, catalase, Prdx-1 and Prdx-3, were studied in PC3 cells with or without supplement of selenite (30 nm). A, B: Selenite did not change Prdx-1, Prdx-3 and catalase in PC3 cells. C, D: Selenite increased GPx-1 and TrxR-1 protein levels. (n = 3)

A 21 B Figure 2.6 Prdx-1 and CAT were not changed in ASPC, Panc-1, or MCF7 cells. Protein levels of redox enzymes, catalase and Prdx-1, were studied in other cells with or without supplement of selenite (30 nm). A: Selenite did not change Prdx-1 and catalase expression in ASPC and Panc-1 cells with the treatment of selenite. B: Selenite did not change Prdx-1 and catalase expression in MCF-7 cells with the treatment of selenite. (n = 3)

Figure 2.7 Prdx-1 and Prdx-3 were not increased by addition of selenium in PC3 cells. Comparison of Prdx-1 and Prdx-3 on PC3 cells and PC3 cells treated with selenium was performed. 22

A Number of Active GPx per mg protein 23 * 1.6E+13 1.4E+13 1.2E+13 1.0E+13 8.0E+12 6.0E+12 4.0E+12 2.0E+12 0.0E+00 [Selenium] / nm PC3 PC3 + Se Amount of Active GPx per mg protein B 2.0E+13 1.81E+13 1.8E+13 1.6E+13 1.58E+13 1.4E+13 1.2E+13 1.14E+13 1.0E+13 8.0E+12 5.96E+12 6.0E+12 4.0E+12 0 50 100 150 200 250 selenium [Selenium] / nm(nm) Figure 2.8 Number of effective active glutathione peroxidase enzymes after treatment of selenite in PC3 cells and HL60 cells. A: PC3 cell extracts with or without treatment of selenite (30 nm) were used for GPx activity assay. Numbers of active GPx per mg protein were calculated (n = 3). B: HL60 cell dose-response. The trend of selenite dose-dependent effects on GPx activity in HL60 cells is shown. C: Number of GPx molecules (monomers) per mg protein were calculated (Control: 6.2 1012 per mg protein; 10 nm: 1.1 1013 per mg protein; 20 nm: 1.6 1013 per mg protein; 200 nm: 2.0 1013 per mg protein; 2000 nm: 1.9 1013 per mg proteins). At lower doses, GPx activities were increased faster than at higher doses. GPx activities tended to achieve the maximum near treatment with 100 nm selenite. (*, p < 0.05)

24 C Amount of Active GPx 10 per mg protein 12 2.5E+13 2.0E+13 1.5E+13 1.0E+13 5.0E+12 0.0E+00 * * * * 0 10 nm 20 nm 200 nm 2000 nm Concentrations of Selenite [Selenium] / nm Figure 2.8 (continued)

Figure 2.9 Similar trends on GPx activities on other cells. Selenium induction of GPx activity in MCF-7 cells. The GPx activity is expressed as the relative increase in activity compared with the baseline value. From the figure, GPx activities tended to achieve the maximum near the treatment with100 nm selenite as well. (Figure adapted from [34]) 25

26 45 TrxR (μmol/min/ml) 40 35 TrxR / µmol min -1 ml -1 30 25 20 15 10 5 0.1 μm 1 μm 10% 5bs 10 μm 0 0 2 4 6 8 10 [SeM]/μM [SeM] / µm 10% fbs Figure 2.10 TrxR activities with treatment of different SeM doses in HL60 cells. Different concentrations (0.1 µm to 10 µm) of SeM were used to determine the dose-dependent TrxR activity responses. TrxR activity increases very little at low doses of SeM, but it rises considerably at very high doses of SeM. The TrxR activity for control is displayed separately. Concentration of 0 µm is used as simulation.

Figure 2.11 Similar trends on TrxR activities on other cells with treatment of selenite. TrxR activities with treatment of different concentrations of selenite were measured in MCF-7, A549 and HT-27 cells as well. Similar trends were seen as in HL60 cells and TrxR activity tends to reach its maximum at a much higher selenium concentration compared to GPx activity (Invert triangle, A549 lung cancer cells; Circle, MCF-7 breast cancer cells; Solid circle, HT-29 colon cancer cells). (Figure adapted from [36]) 27

28 Number of Active GPx Molecules per cell 1.2E+08 1.0E+08 8.0E+07 6.0E+07 4.0E+07 2.0E+07 0.0E+00 ASPC ASPC Se MB231 MB231 Se MCF7 MCF7 Se PANC PANC Se 1 Figure 2.12 GPx activities were increased in ASPC, MB231, MCF-7 and Panc-1 cells. Cells had different GPx activities. Treatment of Se (30 nm) increased GPx activity in ASPC, MB231, MCF-7 and Panc-1 cells. MCF-7 cells increased GPx activities by 1.9-fold, MB231 3.7-fold, ASPC 2.1-fold, and Panc-1 3.9-fold. MCF7 exhibited the lowest GPx activity among these cells.

29 3.0E- 12 k cell / L s - 1 cell - 1 2.5E- 12 2.0E- 12 1.5E- 12 1.0E- 12 5.0E- 13 0.0E+00 PC3 PC3+Se Figure 2.13 Capacity for removal of extracellular H 2 O 2 by PC3 cells is not affected significantly by increased media Se. H 2 O 2 removal was determined in PC3 cells with or without treatment of 30 nm selenite. There was no statistically significant difference between two groups (n = 3). The assay is based on the rate of removal of a 10 µm bolus addition of H 2 O 2 per cell.

30 7.0E- 13 6.0E- 13 k cell / L s - 1 cell - 1 5.0E- 13 4.0E- 13 3.0E- 13 2.0E- 13 1.0E- 13 0.0E+00 MCF7 MCF7+Se MB231 MB231+Se Panc- 1 Panc- 1+Se ASPC ASP+Se Figure 2.14 H 2 O 2 -removal in MCF7, MB231, Panc-1 and ASPC cells. H 2 O 2 (10 µm) was exposed to cells. Cells exhibited different abilities to remove H 2 O 2 (MCF-7, 3.8 10-13 L s -1 cell -1 ; MB231, 4.5 10-13 L s -1 cell -1 ; ASPC, 5.0 10-13 L s -1 cell -1 ; Panc-1, 4.8 10-13 L s -1 cell -1 ). However, after adding Se (30 nm), H 2 O 2 removal showed no statistically significant changes.

31 CHAPTER III SELENIUM AND GSH PRODUCE REACTIVE OXYGEN SPECIES 3.1 Introduction Selenium influences peroxide removal not only indirectly by regulating redox enzymes, GPx and TrxR, but also directly by reacting with GSH to produce O - 2 and H 2 O 2, which presents us the chemically paradoxical roles of Se. This paradox can be an important addition to the peroxide balancing system in cells and it surely results in better understanding of peroxide balance. A possible mechanism of GSH reacting with selenium has been studied [23, 24]. Figure 3.1 indicates that each selenite anion reacts with 6 molecules of GSH and generates peroxides. In this chapter, this mechanism is tested. The direct contributions of Se compounds (selenite and SeM) to H 2 O 2 formation and removal cycles are studied. Specifically, we have studied the formation of O - 2 and H 2 O 2 from the reaction of Se compounds with GSH. Selenite and SeM show different reaction patterns and rates. 3.2 Materials And Methods 3.2.1 Oxygen Consumption Assay

32 Oxygen uptake was monitored with an oxygen monitor (Clark electrode). GSH, Na 2 SeO 3 and SeM were prepared in phosphate buffer (ph 7.2). Different concentrations of GSH, Na 2 SeO 3, Se-Met, SOD or CAT was added into the chamber. Oxygen consumption rates (OCR) were calculated using the initial rate of change of O 2 (240 µm) being at room temperature. 3.2.2 GSSeSG Detection Assay UV-Vis photo-spectrometer (Hewlett Packard 8453) was used to detect GSSeSG. Both kinetic and standard methods were used. Solutions were prepared in phosphate buffer (ph 7.2).The absorbance of 263 nm was monitored to determine the changes in the concentrations of GSSeSG. 3.3 Results And Discussion 3.3.1 Selenite and GSH react to consume oxygen The ratio of GSH to selenite was important for oxygen consumption in the reaction. Different ratios were used during the performance of the oxygen consumption assay. Selenite and GSH are proposed to react with oxygen being reduced to produce O - 2 [23, 24, 25, 26, 27]. Studies on the reaction rates from different ratios of GSH to selenite were a good start to know more about the reaction kinetics (Figure 3.2). The concentration of selenite was first kept at 10 µm and GSH was changed to make different

33 ratios of GSH to selenite (mol:mol) from 25:1 to 225:1 (Figure 3.2A). As more GSH was presented, oxygen was obviously consumed faster and faster, from 0.6 µm/min to 2.3 µm/min, almost 4-fold increase. GSH was then kept at a constant concentration of 1 mm and selenite was adjusted to make different ratios of GSH to selenite (mol:mol) from 25:1 to 2000:1 (Figure 3.2B). As ratios went up, the reaction rate increased as seen by oxygen depletion. The maximum OCR was at the ratio of 500:1 (2.1 µm/min), 3 times greater than that at 25:1. 3.3.2 Seleno-L-methionine and GSH reacted to consume oxygen The ratio of GSH to SeM was also important for oxygen consumption by the reaction. Different ratios were used during the performance of oxygen consumption assay. Similar to selenite, SeM and GSH are thought to react with each other and oxygen is also consumed in this process [24, 27]. The reaction rates, which were observed as OCRs, were also studied on SeM and GSH (Figure 3.3). The concentration of SeM was first kept at 10 µm and GSH was changed to make different ratios of GSH to SeM (mol:mol) from 100:1 to 1600:1 (Figure 3.3A). As more GSH was presented, oxygen was depleted faster and faster, from 0.5 µm/min to 1.2 µm/min. GSH was then kept at the same constant concentration of 1 mm and SeM was adjusted to make different ratios of GSH to SeM (mol:mol) from 25:1 to 2000:1 (Figure 3.3B). As ratios went up, the reaction went faster and oxygen was consumed faster, from 1.2 µm/min to 1.9 µm/min. Here the difference between the OCRs at the same ratio in

34 Figure 3.3A and 3.3B might tell us SeM is more complicated than selenite, due to a more complicated structure, so the reaction rates might vary more. 3.3.3 GSSeSG was formed during the reactions of GSH with selenite or SeM Ganther [23] previously developed a method to detect GSSeSG and proposed that GSSeSG was one of the intermediate products of the reaction between GSH and selenite. The same method was used to detect the formation of GSSeSG during GSH reacting with selenite or SeM (Figure 3.4). From Figure 3.4A, an obvious increase of GSSeSG can be seen. However, in Figure 3.4B, the curve goes down and the values are much smaller than that in Figure 3.3A. This might be because a very low level of GSSeSG was produced initially and then dropped fast, or be because of disturbance of high background of GSH. The reason cannot be determined in this study and requires further study. 3.3.4 Peroxide generation during reaction of GSH and selenite and SeM. Superoxide, O 2 -, was detected by SOD and catalase slowed the OCR by converting H 2 O 2 to H 2 O and O 2 To confirm the formation of O - 2 and H 2 O 2, SOD and catalase was introduced during the reactions between GSH and selenite or SeM. Results are shown as in Figure 3.5. In the middle of the reactions between GSH and SeM or selenite (mol:mol: 100:1), SOD (1000 U/mL) was firstly added into the solution. Distinct increases of the slopes were seen in both Figure 3.5A and 3.5B, which meant SOD accelerated both reactions due to converting O - 2 into H 2 O 2. Catalase of 1200 U/mL was then introduced into the

35 solution to see if the system worked. OCR was lowered because more O 2 was present after catalase was added. 3.4 Conclusions From the studies in this chapter, both selenite and SeM can react with GSH directly to produce O - 2, and H 2 O 2. During the reactions, GSSeSG was detectable. Oxygen monitor was mainly used to study the reaction between GSH and Se compounds, SeM and selenite. The higher the ratio of GSH to selenite or SeM, the faster the reaction. One molecule of Se in selenite or SeM needs more than 1 molecule of GSH for the reactions, which makes the reactions complicated. Based on previous studies [23, 24], one Se needs six GSH for selenite. Possible reaction mechanism for selenite reacting with GSH can be seen in Figure 3.1. GSSeSG and the Se-carboxymethyl derivative of glutathione selenopersulfide (GSSeH) are produced as intermediates. GSSeSG was detected in our experiments (Figure 3.4). GSH is oxidized to GSSG and O 2 is converted into O - 2. We confirmed this by adding SOD and catalase. We also observed that during the reaction between GSH and SeM, O 2 was consumed as well, but the formation of GSSeSG was very different from that of selenite. Actually, it was totally opposite to our expectation and GSSeSG was decreasing from a higher level rather than accumulating as reaction was happening. SeM could react with GSH directly as well, and also consume O 2 and produce O - 2. But the mechanism could be very different from that of selenite. This needs further study.

36 Together with Chapter II, we can see that selenium can influence the H 2 O 2 - system: by reacting with GSH (Chapter III); or by regulating the protein levels and activities of GPx and TrxR (Chapter II).

Figure 3.1 Possible mechanism of the reaction between selenite and GSH. Selenite is able to react with GSH and consume O 2 to produce O 2 -. Hence oxygen consumption can be used to test this reaction. In the meanwhile, 6 molecules of GSH are involved with 1 Se, and GSSG is accumulated. H 2 O 2 might be one of the final products, but this remains for further investigation [23, 24]. 37

38 A 3.0E- 04 2.5E- 04 buffer 10 µm Selenite 25:1 50:1 75:1 [O 2 ] /μm 2.0E- 04 1.5E- 04 1.0E- 04 100:1 150:1 225:1 5.0E- 05 0 500 1000 1500 2000 2500 Time / s B [O 2 ] / μm 2.8E- 04 2.6E- 04 2.4E- 04 2.2E- 04 2.0E- 04 1.8E- 04 1.6E- 04 1.4E- 04 1.2E- 04 1.0E- 04 buffer 1 mm GSH 2000:1 1000:1 500:1 200:1 100:1 50:1 25:1 0 1000 2000 3000 4000 5000 Time / s Figure 3.2 Influence of ratios of GSH to selenite on their reaction. Oxygen uptake was observed with Clark electrode in phosphate buffer (ph 7.2). A: [selenite] was kept constant at 10 µm. Different concentrations of GSH were added to make different ratios (GSH:selenite, mol:mol) from 25:1 to 225:1. Oxygen consumption rates were 0.6, 0.8, 1.0, 1.3, 1.8, and 2.3 (µm/min). Controls (buffer and selenite) both had a rate of +0.3 µm/min. B: [GSH] was kept constant at 1 mm. Selenite at different concentrations was added to make ratios (GSH:selenite, mol:mol) from 25:1 to 2000:1. Oxygen consumption rates were 1.7, 1.9, 2.1, 2.0, 1.3, 1.1 and 0.7 (µm/min). Buffer had a rate of 0.1 µm/min and GSH had a rate of 0.6 µm/min.

39 A [O 2 ] / μm 3.0E- 04 2.5E- 04 2.0E- 04 1.5E- 04 1.0E- 04 buffer 10 µm SeM 100:1 200:1 400:1 800:1 1600:1 5.0E- 05 0.0E+00 0 500 1000 1500 2000 2500 3000 Time / s B [O 2 ] / μm 2.5E- 04 2.3E- 04 2.1E- 04 1.9E- 04 1.7E- 04 1.5E- 04 1.3E- 04 1.1E- 04 9.0E- 05 7.0E- 05 5.0E- 05 buffer 1 mm GSH 2000:1 1000:1 500:1 200:1 100:1 50:1 25:1 0 1000 2000 3000 4000 5000 Time / s Figure 3.3 Influence of ratios of GSH to SeM on their reaction. Oxygen uptake was observed by a Clark electrode in phosphate buffer (ph 7.2). A: [SeM] was constant at 10 µm. Different concentrations of GSH were added to make ratios (GSH:SeM, mol:mol) from 100:1 to 1600:1. Oxygen consumption rates were 0.5, 1.1, 1.9, 2.3, and 1.2, (µm/min). Buffer had a rate of +0.1 µm/min and SeM had a rate of 0.2 µm/min. B: [GSH] was constant at 1 mm. SeM at different concentrations was added to make ratios (GSH:SeM, mol:mol) from 25:1 to 2000:1. Oxygen consumption rates were 1.9, 1.8, 1.7, 1.7, 1.6, 1.4 and 1.2 (µm/min). Buffer had a rate of 0.3 µm/min and GSH has a rate of 0.9 µm/min.

40 A Concentration / μm 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 0.0 0 500 1000 1500 2000 Time / s B Concentration / μm 1.2 1.0 0.8 0.6 0.4 0.2 0.0 0 500 1000 1500 2000 Time / s Figure 3.4 Reactions of GSH and selenite or SeM show different patterns. Detection of GSSeSG was performed by using UV-Vis photo-spectrometer at absorbance of 263 nm. All solutions were prepared at ph 7.2. A: Formation of GSSeSG by the reaction of GSH and selenite. [Selenite] = 10 um, [GSH] = 1 mm. The ratio of GSH:Selenite (mol:mol) was 100:1. B: Formation of GSSeSG by the reaction of GSH and SeM. The concentration of SeM was 10 um and that of GSH was 1 mm. The ratio of GSH:SeM (mol:mol) was 100:1.

41 A 6.0E- 04 5.0E- 04 +H 2 O 2 1 mm [O2] / μm 4.0E- 04 3.0E- 04 2.0E- 04 buffer 3 mm GSH GSH:SeM 200:1 +SOD 1000 U/mL +Catalase 1200 U/mL 1.0E- 04 0.0E+00 0 200 400 600 800 1000 1200 Time / s B [O 2 ] / μm 1.4E- 03 1.2E- 03 1.0E- 03 8.0E- 04 6.0E- 04 4.0E- 04 2.0E- 04 0.0E+00 GSH:Selenite 100:1 +SOD 1000 U/mL +Catalase 1200 U/mL +H 2 O 2 0 5000 10000 15000 20000 25000 30000 35000 Time / s Figure 3.5 Formation of O 2 - and H 2 O 2 in the reaction of GSH and SeM or selenite. A: SeM was added to 3 mm GSH for reaction. Superoxide dismutase was then added to determine the formation of superoxide. Catalase was added afterwards to determine the generation of H 2 O 2. In the end, H 2 O 2 was added to show catalase was functional. The oxygen consumption rates were: 0 buffer; 0.6 GSH; 2.9 GSH+SeM; 4.1 +SOD; and 2.5 +catalase (µm/min). Addition of SOD accelerated the OCR, and addition of catalase decreased the OCR. B: Similar experiment was also done for the reaction of GSH and selenite. SOD was used to accumulate H 2 O 2 and was treated before CAT. After adding SOD, the slope of the curve went larger and O 2 was consumed faster. CAT slowed the total O 2 depletion in the solution by about a half.

42 CHAPTER IV THEORY OF ENZYMATIC COOPERATION 4.1 Introduction Three enzyme families have been identified in the redox system of cells to remove hydrogen peroxide (H 2 O 2 ) directly, glutathione peroxidase (GPx), catalase (CAT), and peroxiredoxin (Prdx) [2]. The role of each enzyme in relieving oxidative stress, such as that initiated by H 2 O 2, organic peroxides and disulfide bonds, could be important for understanding the cellular redox system, since it might give researchers a macroscopic depiction of how each enzyme contributes to oxidative stress-removal and hence help investigators to better understand redox balance. Johnson et al. [38] showed what each enzyme does in mouse red blood cells (RBCs). They demonstrated in mouse RBCs: (1) Prdx-2 plays a major role in removing endogenous H 2 O 2, but plays little or no role in eliminating exogenous H 2 O 2 or organic peroxides under their experimental conditions; and (2) CAT and GPx-1 both participate in removal of exogenous H 2 O 2 but CAT removes more exogenous H 2 O 2 compared to GPx-1. Another paper also by Johnson et al. [39] demonstrates that GPx-1 helps mouse erythrocytes protect against organic peroxides. Red blood cells are simpler than other mammalian cells, so it is often easier for investigators to start some projects with RBCs. However, RBCs exhibit very different behavior from other cells. Mammalian RBCs do not have a nucleus, mitochondria, golgi apparatus or endoplasmic reticulum. Therefore, events in RBCs might not be directly applicable in other mammalian cells. The role of each enzyme in other cells requires

43 study. In our laboratory, we mainly focus on human cells and work to establish quantitative information for cellular redox systems. We supplemented cell culture media with selenite (Na 2 SeO 3, 30 nm, Se) for our first step and observed changes in the intracellular redox system. After the supplementation, active GPx and TrxR went up, but H 2 O 2 -removal remained at a similar level under our experimental conditions. More detailed observations and discussion follows. 4.2 Results And Discussion In PC3 cells, GPx activity and protein levels were increased after addition of Se (Figures 2.5, 2.8), and more H 2 O 2 could be eliminated. However, the rate of removal of exogenous H 2 O 2 remains at a similar level (changes were not statistically significant, Figure 2.13). Because TrxR is also Se-dependent, as expected, the protein level of TrxR was also increased after Se addition (Figure 2.5). CAT, Prdx-1, and Prdx-3 kept a similar protein level with the same Se treatment (Figure 2.5). This was interesting in that the whole cell exogenous H 2 O 2 removing ability did not correspond to GPx activity and protein level increase and was not significantly increased, though Johnson et al. (2002, 2010) [38, 39] showed GPx-1 participated in exogenous H 2 O 2 removal. To see if other cells could still exhibit similar phenomena, we further examined MCF-7, MB231, ASPC and Panc-1 cells. After adding Se, GPx activities in these cells all went up. MCF-7 cells increased GPx activities by 1.9-fold, MB231 3.7-fold, ASPC 2.1- fold, and Panc-1 3.9-fold (Figure 2.12). MCF-7 had the lowest GPx activity, compared to other cells.

44 Similar to that in PC3 cells, catalase and Prdx-1 does not change in Panc-1 and ASPC cells (Figure 2.6). Interestingly, total H 2 O 2 removal in these cells all showed no statistically significant changes with Se treatment (Figure 2.14), though different cells exhibited various H 2 O 2 removal abilities (MCF-7, 3.9 10-13 L s -1 cell -1 ; MB231, 4.5 10-13 L s -1 cell -1 ; ASPC, 5.0 10-13 L s -1 cell -1 ; Panc-1, 4.8 10-13 L s -1 cell -1 ). Among these cells, MCF-7 cells eliminated H 2 O 2 relatively slowly, 17% less than MB231. Since we have only done exogenous H 2 O 2 removal on cells, one possible explanation to this is not like in mouse RBCs, GPx-1 plays little role in the removal of large amounts of exogenous H 2 O 2 in these cells. To make this conclusion, we still need further investigation to show how GPx-1 deficiency influence exogenous H 2 O 2 removal. We examined the enzyme levels in the 4 different cell lines. Results showed little difference between cells before and after Se treatment on CAT and Prdx-1 in all 4 kinds of cells, MCF-7, MB231, ASPC and Panc-1 cells (Figures 2.4, 2.6). We need to repeat western blotting assays on GPx-1 and TrxR due to antibody inefficiency. From review of the whole peroxide-removal system, there are three enzymes that have been discovered that directly remove H 2 O 2 : GPx, CAT, and Prdx (Figure 1.1) [3]. Further study of the contributions of these enzymes to peroxide removal can not only provide better understanding of the H 2 O 2 removal system, but also give some clinical inspirations as H 2 O 2 might be a critical mediator for clinical therapy [40]. In the future, we should also take organic peroxides, endogenous H 2 O 2, disulfide bonds and other oxidative stress into consideration, for these enzymes will likely behave distinctly in different cells and situations. For example, CAT and GPx-1 remove