Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Glutathione in fish : transport, influence of temperature and growth rate, and interaction with the stress… Leggatt, Rosalind Alexandra 2006

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_2006-200131.pdf [ 10.91MB ]
Metadata
JSON: 831-1.0092848.json
JSON-LD: 831-1.0092848-ld.json
RDF/XML (Pretty): 831-1.0092848-rdf.xml
RDF/JSON: 831-1.0092848-rdf.json
Turtle: 831-1.0092848-turtle.txt
N-Triples: 831-1.0092848-rdf-ntriples.txt
Original Record: 831-1.0092848-source.json
Full Text
831-1.0092848-fulltext.txt
Citation
831-1.0092848.ris

Full Text

G L U T A T H I O N E IN F I S H : T R A N S P O R T , I N F L U E N C E O F T E M P E R A T U R E A N D G R O W T H R A T E , A N D I N T E R A C T I O N S WITH T H E S T R E S S R E S P O N S E By R O S A L I N D A L E X A N D R A L E G G A T T B .Sc . University of Gue lph , 1997 A T H E S I S S U B M I T T E D IN P A R T I A L F U L F I L M E N T O F T H E R E Q U I R E M E N T S F O R T H E D E G R E E O F D O C T O R O F P H I L O S O P H Y in T H E F A C U L T Y O F G R A D U A T E S T U D I E S (Animal Sc ience) U N I V E R S I T Y O F BRIT ISH C O L U M B I A August 2006 © Rosal ind A lexandra Leggatt, 2006 Abstract Glutathione ( G S H ) is a ubiquitous antioxidant involved in many cellular p rocesses in mammals . In contrast, fish have relatively low G S H levels, as well as low turnover and activity of assoc ia ted enzymes . A s the metabol ism and function of G S H have been poorly studied in f ish, I examined the mechan isms by which fish t issues transport G S H , the influence accl imation temperature and growth rate have on G S H dynamics, and the importance of G S H during heat stress in fish to elucidate the differing roles of G S H in fish and mammals . In Chapter 2 I demonstrated that posterior kidney, and to a lesser extent liver and gill, of rainbow trout took up G S H by extracellular breakdown and intracellular synthesis. Direct uptake of exogenous G S H was observed in the liver as apparent gradient-dependent transport. In Chapter 3 I demonstrated that G S H turnover was proportional to incubation temperature in a rainbow trout hepatoma cell line in vitro. In killifish in vivo, t issue G S H levels and activity of GSH-assoc ia ted enzymes were proportional to accl imation temperature. Al though these experiments showed that temperature inf luences G S H dynamics in f ish, the effects of temperature could only partially explain the low G S H levels in fish compared with mammals . In Chapter 4 I showed that high G S H levels and activity of assoc ia ted enzymes in growth hormone transgenic coho sa lmon with high metabolic rates were due to increased feeding and growth rather than to direct effects of the t ransgene. Oxygen consumption corresponded to liver G S H levels in control and transgenic f ish, but not to other components of the G S H system. Although long-term changes in temperature influenced G S H dynamics in killifish, in Chapter 5 I demonstrated that acute temperature changes had little effect on G S H dynamics in rainbow trout in vivo or in vitro. However, altered G S H levels inconsistently altered the cellular response to acute heat stress in rainbow trout in vitro, and influenced the general ized and cellular responses to acute heat stress in vivo. Taken together, these studies demonstrate that fish can substantial ly modulate G S H levels in response to demand , and that altered G S H levels influence other physiological sys tems. ii Table of Contents Abstract i . . Table of Contents iii List of Tab les v List of Figures vi Abbreviat ions ix Preface xi Acknowledgments xii CHAPTER 1: General introduction and thesis overview 1 The Redox Environment 1 The Glutathione Antioxidant Sys tem 2 Glutathione and Fish 5 Glutathione and the Stress Response in F ish 7 Thes is Outl ine 8 Justif ication of Experimental Models 10 CHAPTER 2: Can rainbow trout directly take up glutathione into their tissues?.. 17 Introduction i ; 17 Materials and Methods 19 Experiment 2.1: Is GSH uptake independent of synthesis? 20 Experiment 2.2: Is GSH uptake independent of breakdown and synthesis? 20 Resul ts 22 Experiment 2.1: Is GSH uptake independent of synthesis? 22 Experiment 2.2: Is GSH uptake independent of breakdown and synthesis? 24 Discuss ion 38 References 41 CHAPTER 3: Effects of incubation and acclimation temperature on glutathione dynamics in fish in vitro and in vivo 43 Introduction 43 Materials and Methods 44 Experiment 3.1: Influence of incubation temperature on tGSH turnover in vitro .... 44 Experiment 3.2: Influence of acclimation temperature on GSH levels and enzyme activity in vivo 45 Resul ts 48 Experiment 3.1: Influence of incubation temperature on tGSH turnover in vitro .... 48 Experiment 3.2: Influence of acclimation temperature on GSH levels and enzyme activity in vivo :: 51 Discuss ion 58 References 62 CHAPTER 4: Glutathione dynamics and metabolism in growth hormone transgenic salmon during feeding and starvation 66 Introduction 66 Materials and Methods 67 Experiment 4.1: Oxygen consumption during feeding and starvation 67 Experiment 4.2: Glutathione dynamics during feeding and starvation 68 iii Resul ts : 70 Experiment 4.1: Oxygen consumption during feeding and starvation 70 Experiment 4.2: Glutathione dynamics during feeding and starvation 75 Discuss ion 81 References 87 CHAPTER 5: Glutathione levels, oxidation and activity of associated enzymes during an acute heat stress in rainbow trout in vivo and in vitro 90 Introduction 90 Materials and Methods 91 Experimental Series 5.1: GSH dynamics in response to a heat stress in vivo and in vitro 91 Experiment 5.2: GSH dynamics in response to heat and oxidative stresses in vitro 92 Resul ts 95 Experimental Series 5.1: GSH dynamics in response to a heat stress in vivo and in vitro 95 Experiment 5.2: GSH dynamics in response to heat and oxidative stress in vitro . 95 Discuss ion i 101 References 105 CHAPTER 6: The effects of glutathione levels on components of the generalized and cellular stress responses in rainbow trout in vivo and in vitro 108 Introduction 108 Materials and Methods 110 Experimental Series 6.1: The effects of altered GSH levels on the generalized and cellular stress responses in vivo 110 Experimental Series 6.2: The effects of altered GSH levels on the cellular stress response in vitro 111 Resul ts : 114 tGSH levels in rainbow trout models after injection or incubation with GSH or BSO ,....„ 114 Experimental Series 6.1: The effects of altered GSH levels on the generalized and cellular stress responses in vivo 117 Experimental Series 6.2: The effects of altered GSH levels on the cellular stress response in vitro 123 Discuss ion 126 References 132 CHAPTER 7: General discussion and future directions 135 Future Directions 139 References 140 iv List of Tables Table 2.1: Plasma total.GSH, plasma Cortisol, as well as oxidized glutathione in liver and posterior kidney from rainbow trout injected with combinations of G S H and a G S H synthesis blocker daily for 3 days 25 Table 2.2: Percent oxidized glutathione in liver, brain, and posterior kidney from rainbow trout injected with combinations of G S H , a G S H synthesis blocker and a G S H breakdown blocker 33 Table 2.3: Estimation of intracellular total G S H in rainbow trout after removing the plasma component for Experiment 2.1 36 Table 2.4: Estimation of intracellular total G S H in rainbow trout after removing the plasma component for Experiment 2.2 36 Table 3.1: Total protein of RTH-149 cells over time, held at different temperatures with or without the G S H synthesis blocker B S O 52 Table 3.2: Lactate dehydrogenase released from RTH-149 cells over time, held at different temperatures with or without the G S H synthesis blocker BSO 53 Table 3.3: Average acclimation temperature and oxygen content of the water, amount eaten, and hematocrit of killifish acclimated to different temperatures for three weeks 53 Table 4.1: Weight and length of growth hormone transgenic coho salmon fed to satiation, fed a fixed ration or starved for one month, and control coho salmon fed to satiation or starved for one month 76 Table 5.1: Tissue total G S H levels in rainbow trout exposed to a 2 h, 23°C heat stress (+10°C, control 13°C) 96 Table 5.2: Reduced G S H and viability in a primary culture of rainbow trout hepatocytes after a 1 h, 30°C heat stress (+15°C, control 15°C) 99 Table 6.1: Oxidized glutathione in rainbow trout injected once a day for three days with saline, G S H or BSO 118 v List of Figures Figure 1.1: Cellular dynamics and metabolism of glutathione 3 Figure 1.2: Integration between glutathione and other antioxidants 4 Figure 2.1: The transport of glutathione across the cellular membrane 18 Figure 2.2: Liver (A) and posterior kidney (B) total G S H levels in rainbow trout injected daily for three days with combinations of saline, G S H , and a G S H synthesis blocker •. 23 Figure 2.3: y-Glutamyltranspeptidase activity in various tissues of 250g rainbow trout 25 Figure 2.4: y-Glutamyltranspeptidase (yGT) activity in posterior kidney of rainbow trout injected with saline or a yGT activity blocker 26 Figure 2.5: Tissue total G S H levels in rainbow trout after a single injection with 0.8mmol GSH/kg fish.... 26 Figure 2.6: Liver total G S H levels in rainbow trout injected with combinations of enzyme blockers: saline, a G S H synthesis blocker (BSO), a G S H breakdown blocker (acivicin), or B S O and acivicin, followed by saline or G S H injection 27 Figure 2.7: Posterior kidney total G S H levels in rainbow trout injected with combinations of enzyme blockers: saline, a G S H synthesis blocker (BSO), a G S H breakdown blocker (acivicin), or BSO and acivicin, followed by saline or G S H injection 28 Figure 2.8: Gill total G S H levels in rainbow trout injected with combinations of enzyme blockers: saline, a G S H synthesis blocker (BSO), a G S H breakdown blocker (acivicin), or B S O and acivicin, followed by a saline or G S H injection 30 Figure 2.9: Brain total G S H levels in rainbow trout injected with combinations of enzyme blockers: saline, a G S H synthesis blocker (BSO), a G S H breakdown blocker (acivicin), or B S O and acivicin, followed by a saline or G S H injection 31 Figure 2.10: Plasma total G S H levels in rainbow trout injected with combinations of enzyme blockers: saline, a G S H synthesis blocker (BSO), a G S H breakdown blocker (acivicin), or B S O and acivicin, followed by a saline or G S H injection 32 Figure 2.11: y-Glutamyltranspeptidase activity in posterior kidney of rainbow trout injected with combinations of enzyme blockers: saline, a G S H synthesis blocker (BSO), a G S H breakdown blocker (acivicin), or B S O and acivicin, followed by a saline or G S H injection 34 Figure 2.12: Initial and final liver to plasma total G S H concentration gradients as set up by G S H injection in rainbow trout injected concurrently with saline or B S O (Experiment 2.1) or with saline or combinations of B S O and acivicin (Experiment 2.2) 37 Figure 2.13: Initial and final posterior kidney to plasma total G S H concentration gradients as set up by G S H injection in rainbow trout injected concurrently with saline or B S O (Experiment 2.1) or with saline or combinations of BSO and acivicin (Experiment 2.2) 37 Figure 3.1: Percent of total G S H remaining in RTH-149 cells after 24 h incubation with various concentrations of a G S H synthesis inhibitor 46 Figure 3.2: Total G S H levels in (A) G S H synthesis inhibitor incubated and (B) control RTH-149 cells held at different temperatures 49 Figure 3.3: The effect of incubation with a G S H synthesis inhibitor on total G S H levels over time in RTH-149 cells held at different temperatures. A) The relationship between tGSH levels (as nmol/mg protein) and time as expressed by exponential decay. B) The relationship between tGSH levels (as % of control levels at equal time points) and time as expressed by exponential decay 50 Figure 3.4: Decay coefficients of total G S H decrease during incubation with a G S H synthesis inhibitor in cultured RTH-149 cells as a function of incubation temperature 52 Figure 3.5: Tissue total G S H levels of killifish acclimated to 6, 17, 25, 30 or 33°C for three weeks 55 Figure 3.6: Total G S H levels of killifish (A) liver, (B) brain, and (C) muscle expressed as a function of acclimation temperature 56 Figure 3.7: (A) Oxidized and (B) reduced G S H levels in killifish liver acclimated to different temperatures. 57 vi Figure 3.8: Glutathione peroxidase (GPx) and glutathione reductase (GR) activity in the liver of killifish acclimated to different temperatures. (A) GPx activity measured at 28°C; (B) G R activity measured at 28°C; (C) G P x activity measured at respective acclimation temperatures; (D) G R activity measured at respective acclimation temperatures 57 Figure 4.1: Oxygen consumption rates of representative coho salmon over 4 days starvation and 4 days feeding, measured every 42 min. A) A growth hormone transgenic fish fed 1% body weight/day, B) a growth hormone transgenic fish fed to satiation, and C) a control fish fed 1% body weight/day 72 Figure 4.2: Frequency histogram of oxygen consumption rates of a representative coho salmon over one day/night or feeding period 73 Figure 4.3: Resting oxygen consumption rate of growth hormone transgenic and control salmon during a four day acclimation and starvation period, calculated every day/night period 73 Figure 4.4: Oxygen consumption rates of growth hormone transgenic and control salmon 74 Figure 4.5: Resting oxygen consumption rates after feeding of growth hormone transgenic and control coho salmon at different feeding levels 76 Figure 4.6: Total G S H levels in A) liver, B) posterior kidney, C) intestinal mucosa, D) plasma, and E) muscle in the following groups of control and growth hormone transgenic coho salmon: transgenic salmon fed to satiation, control salmon fed to satiation, transgenic salmon fed an equal ration as control salmon, transgenic salmon starved for one month, and control salmon starved for one month 78 Figure 4.7: y-Glutamylcysteine synthetase activity in the liver of the following groups of control and growth hormone transgenic coho salmon: transgenic salmon fed to satiation, control salmon fed to satiation, transgenic salmon fed an equal ration as control salmon, transgenic salmon starved for one month, and control salmon starved for one month 79 Figure 4.8: A) Oxidized and B) reduced glutathione in liver of the following groups of control and growth hormone transgenic coho salmon: transgenic salmon fed to satiation, control salmon fed to satiation, transgenic salmon fed an equal ration as control salmon, transgenic salmon starved for one month, and control salmon starved for one month 79 Figure 4.9: A) Glutathione reductase and B) glutathione peroxidase activities in the liver of the following groups of control and growth hormone transgenic coho salmon: transgenic salmon fed to satiation, control salmon fed to satiation, transgenic salmon fed an equal ration as control salmon, transgenic salmon starved for one month, and control salmon starved for one month 80 Figure 4.10: y-Glutamyltranspeptidase activity in A) intestinal mucosa and B) posterior kidney of the following groups of control and growth hormone transgenic coho salmon: transgenic salmon fed to satiation, control salmon fed to satiation, transgenic salmon fed an equal ration as control salmon, transgenic salmon starved for one month, and control salmon starved for one month 80 Figure 4.11: Correlation between oxygen consumption rate and A) liver total G S H , B) intestinal y-glutamyltranspeptidase, and C) intestinal total G S H in growth hormone transgenic and control coho salmon during feeding and starvation 82 Figure 5.1: Total G S H levels in the liver of rainbow trout after a 2 h, 23°C heat stress (+10°C, control 13°C) 96 Figure 5.2: A) Glutathione peroxidase and B) glutathione reductase activity in the liver of rainbow trout after a 2 h, 23°C heat stress (+10°C, control 13°C) 97 Figure 5.3: A) Total G S H and B) oxidized G S H in a primary culture of rainbow trout hepatocytes after a 1 h, 30°C heat stress (+15°C, control 15°C) 98 Figure 5.4: A) Total G S H and B) Hsp70 levels in RTH-149 cells exposed to a 1 h heat (33°C, +13°C) or oxidative (10mM hydrogen peroxide) stress (control 20°C), with or without fetal calf serum in the media.99 Figure 5.5: A) Total G S H , B) oxidized G S H , and C) reduced G S H in RTH-149 cells exposed to a 1 h heat (33°C, +13°C), oxidative (10mM hydrogen peroxide), or control (20°C) treatment 100 Figure 5.6: Glutathione reductase activity in RTH-149 cells exposed to a 1 h heat (33°C, +13°C), oxidative (10mM hydrogen peroxide), or control (20°C) treatment 102 Figure 5.7: Lactate dehydrogenase released from RTH-149 cells exposed to a 1 h heat (33°C, +13°C), oxidative (10mM hydrogen peroxide), or control (20°C) treatment 102 vii Figure 6.1: Hsp70 levels in RTH-149 cells following exposure to a A) 28°C (+8°C) or B) 33°C (+13°C) 1 h heat stress (ambient temperature 20°C) 112 Figure 6.2: Total G S H in A) liver and B) posterior kidney of rainbow trout injected once a day for three days with saline, G S H or B S O , and then exposed to a 2 h, 20.5°C heat stress (+12°C, ambient temperature 8.5°C) 115 Figure 6.3: Total G S H in A) liver and B) posterior kidney of rainbow trout injected once a day for three days with saline, G S H or B S O , and then exposed to a 2 h, 25.5°C heat stress (+10.5°C, ambient temperature 15°C) 116 Figure 6.4: Total G S H levels in RTH-149 cells incubated with normal media, media containing G S H for 6 h, or media containing B S O for 48 h, and then exposed to a 1 h heat stress of varying temperatures (ambient temperature 20°C) 118 Figure 6.5: Total G S H levels in RTH-149 cells incubated with normal media, media containing G S H for 6 h, media containing B S O for 48 h, or media containing BSO for 6 h. tGSH levels were measured immediately prior to a 1 h, 33°C heat stress (+13°C, ambient temperature 20°C) 119 Figure 6.6: Plasma Cortisol levels in rainbow trout injected once a day for three days with saline, G S H or BSO, and then exposed to a 2 h, 20.5°C (+12°C) heat stress or maintained at ambient temperature (8.5°C) 119 Figure 6.7: Plasma glucose levels in rainbow trout injected once a day for three days with saline, G S H or BSO, and then exposed to a 2 h, 20.5°C (+12°C) heat stress or maintained at ambient temperature (8.5°C) 120 Figure 6.8: Plasma Cortisol levels in rainbow trout injected once a day for three days with saline, G S H or BSO, and then exposed to a 3 min chasing stress, or 2 h, 25.5°C heat stress (+10.5°C, ambient temperature 15°C) 120 Figure 6.9: Liver and posterior kidney Hsp70 levels in rainbow trout injected once a day for three days with saline, G S H or B S O , and then exposed to a 2 h, 20.5°C (+12°C) heat stress, or maintained at ambient temperature (8.5°C) 121 Figure 6.10: Liver and posterior kidney A) Hsp70 and B) Hsp30 levels in rainbow trout injected once a day for three days with saline, G S H or BSO, and then exposed to a 2 h, 25.5°C heat stress (+10.5°C, ambient temperature 15°C) 122 Figure 6.11: Hsp70 levels in RTH-149 cells incubated with normal media, media containing G S H for 6 h, or media containing B S O for 48 h, and then exposed to a 24.5 (+4.5°C) or a 27°C (+7°C) 1 h heat stress, or maintained at ambient temperature (20°C) 124 Figure 6.12: A) Hsp70 and B) Hsp30 levels in RTH-149 cells incubated with normal media, media containing G S H for 6 h, or media containing BSO for 48 h, and then exposed to a 1 h heat stress of varying temperatures, or maintained at ambient temperature (20°C) 124 Figure 6.13: Percent lactate dehydrogenase released from RTH-149 cells incubated with normal media, media containing G S H for 6 h, or media containing B S O for 48 h, and then exposed to a 1 h heat stress of varying temperatures (ambient temperature 20°C) 125 Figure 6.14: Hsp70 levels in RTH-149 cells incubated with normal media, media containing G S H for 6 h, media containing BSO for 48 h, or media containing B S O for 6 h, arid then exposed to a 1 h, 33°C heat stress (+13°C, ambient temperature 20°C) 127 Figure 6.15: Hsp30 levels in RTH-149 cells incubated with normal media, media containing G S H for 6 h, media containing B S O for 48 h, or media containing BSO for 6 h, and then exposed to a 1 h, 33°C heat stress (+13°C, ambient temperature 20°C) 127 Figure 6.16: Percent lactate dehydrogenase released from RTH-149 cells incubated with normal media, media containing G S H for 6 h, media containing BSO for 48 h, or media containing B S O for 6 h, and then exposed to a 1 h, 33°C heat stress (+13°C, ambient temperature 20°C) 128 viii Abbreviations A 0 2 change in oxygen level At change in time a.a. amino acid A O antioxidant BSO L-buthionine-sulfoximine BW body weight CAT catalase cys cysteine D M E M Dulbecco's Modified Eagle's Medium DTNB 5,5'-dithiobis(2-nitrobenzoic acid) E C V extracellular volume EDTA ethylenediaminetetraacetic acid ELISA enzyme-linked immunosorbant assay F C S fetal calf serum FR free radical g gravitational force G-6-PDH glucose-6-phosphate dehydrogenase G C S y-glutamylcysteine synthetase y-glu Y - 9 l u t a m i c a c i d yGT v-glutamyltranspeptidase gly glycine GPx glutathione peroxidase G R glutathione reductase G S glutathione synthetase G S H glutathione G S S G oxidized glutathione G S T glutathione-S-transferase H 2 0 2 hydrogen peroxide Hsc heat shock cognate HSF1 heat shock factor 1 Hsp's heat shock proteins Hsp30 heat shock protein molecular weight 30 Hsp70 heat shock protein molecular weight 70 LDH lactate dehydrogenase M R P multidrug resistance-associated protein MS-222 tricaine methanesulfonate NADH B-nicotinamide adenine dinucleotide, reduced NADPH B-nicotinamide adenine dinucleotide phosphate, reduced 0 2 ' " superoxide radical Oatp organic anion transport protein ix P B S phosphate-buffered saline P-SH protein thiols P S - S P protein-mixed disulfides P U F A polyunsaturated fatty acids Q 1 0 ratio of increase in activity with a temperature increase of 10°C R B C red blood cell RMR routine metabolic rate R O S reactive oxygen species RTH-149 rainbow trout hepatoma cell line ATCC# CRL-1710 S D S - P A G E sodium dodecyl sulphate polyacrylamide gel electrophoresis S M R standard metabolic rate S O D superoxide dismutase tGSH total glutathione (reduced glutathione + 2><oxidized glutathione) tGSHj intracellular total glutathione = {measured tissue total glutathione - (extracellular volume x measured plasma total glutathione)}/ (1 - extracellular volume) V volume (L) Vit vitamin x Preface This thesis has been written in manuscript format according to the guidelines of the Faculty of Graduate Studies, University of British Columbia. Oxygen consumption data from Chapter 4 (Experiment 4.1) have been previously published as "Leggatt, R.A., Devlin, R.H., Farrell, A .P . , and Randall, D.J. 2003. Oxygen uptake of growth hormone transgenic coho salmon during starvation and feeding. Journal of Fish Biology. 62: 1053-1066." The glutathione section of Chapter 4 (Experiment 4.2) was a joint project between myself and Dr. Robert Devlin, Dionne Sakrani and Carlo Biagi (Fisheries and Oceans Canada). Feeding and starvation trials in the second part of this chapter were set up and performed by Dionne Sakrani and Carlo Biagi, and all data were analyzed by me. The in vivo heat stress experiment of Chapter 5 (Experimental Series 5.1 in vivo) was a joint project between myself and Dr. Luis Afonso (NRC Institute for Marine Biology). Dr. Afonso set up this experiment, and all data were analyzed by me. All other experimental set-ups and analyses were performed by me. xi A c k n o w l e d g m e n t s Many years later, it is wonderful to be standing at the end point of my thesis. When I began this journey I never expected it to take me where it did or to lead me as far down the path. I would like to thank my supervisor, Dr. George Iwarria, for the inspiration and for fostering an excitement of science within me. To my co-supervisor, Dr. Colin Brauner, thank you for picking me up and giving me the encouragement to keep going. To Dr. Patricia Schulte, thank you for pushing me to improve, and for making me look forward to criticism, knowing the improvements it can bring. And to Dr. Robert Devlin, thank you for sticking with me, and for always giving me a new way of looking at things. Many people have helped me through my thesis, in its many forms. Thank you to Dr. Kim Cheng and Dr. Tilmann Benfey for your support and encouragement. Thank you to Dr. David Randall and Dr. Tony Farrell for leading me through my first experiment, first paper, first rejection and finally first triumph. As well, thank you to Ken Scheer and Dr. Jeff Richards for giving me new things to think about. I have had the great fortune to be part of a number of labs through my thesis. From the Iwama lab, thank you so much to Dr. Anne Todgham and Dr. Paige Ackerman. You two have led the way for me, given me role models and reminded me that this is a tough road, but that's what makes it worthwhile. Thank you as well to Dr. Kazumi Nakano, Dr. Luis Afonso, Ellen Teng, Dom Toa, Nil Basu, Alberto Schlict and Dr. Robert Forsyth. Thanks to the Brauner lab for making me a new home, and sticking my picture up so even if I wasn't there in person, I always felt included. Thanks to the Schulte lab for taking me in when things were quiet. And last but not least thank you to the Farrell lab for taking over the lonely MacMillian aquaculture lab and filling it with laughter, beer, and lots of great science. I would not have gotten as far as I did without the intellectual and emotional support of everyone above. I would also still be dissecting fish from my first experiment if it weren't for the many hands helping me along the way. Thanks to Paige Ackerman, Luis Afonso, Dan Baker, Carlo Biagi, Erika Eliason, Nann Fangue, Bruce Gillespe, Amelia Grant, Linda Hanson, Licia Lundstedt, Vicki Marlott, Kazumi Nakano, Jonathan Ritchie, Dionne Sakrani, Danielle Simonot, Dom Toa, Anne Todgham, Marianna Veiga, and Natalie West. My thesis was funded with the generous support of AquaNet (Dr. George Iwama), N S E R C (Dr. George Iwama, Dr. Colin Brauner and Dr. Patricia Schulte), the Canadian Regulatory System for Biotechnology (Dr. Robert Devlin), the Canadian Biotechnology Strategy (Dr. Robert Devlin), the National Research Council's Institute for Marine Biology, and the Fraser Valley Trout Hatchery. Personal support was from N S E R C (PGS A and B), UBC (Leonard S. Klinck Memorial Fellowship, University Graduate Fellowship), the Faculty of Land and Food Systems (Elizabeth Howland Fellowship, travel award), AquaNet (salary and travel award), Fisheries and Oceans Canada ( N S E R C DFO Supplemental Scholarship), the National Research Council's Institute for Marine Biology (salary), Aquaculture Association of Canada (student scholarship and travel awards), the Fish Biology Congress (travel award), the Faculty of Graduate Studies (travel award), and the Canadian Society of Zoologists (travel awards). Thank you to my husband Jonathan Ritchie. Although you may not have known what I was doing all this time, you've been with me the entire way and given me the support, encouragement and occasional kick in the pants to get me finished. Finally I'd like to acknowledge the unwilling support of the many fish that showed me everything I know about glutathione. xii CHAPTER 1 General Introduction and Thesis Overview Fish live in a constantly changing environment where temperature, oxygen availability, light and compounds in the water can fluctuate on an hourly, daily, and seasonal basis. To maintain internal homeostasis in the face of such fluctuations, fish are equipped with adaptable, integrated physiological control systems. One such system is the antioxidant glutathione (GSH) and its associated enzymes, which act to maintain the redox balance of many cellular systems, including those of fish. In this thesis I describe the transport of G S H within fish, the importance of G S H during changing temperature and metabolism, and the role of G S H in the stress response off ish. The Redox Environment The cellular redox environment is a balance between oxidative and reductive reactions that involve the transfer of electrons between or within molecules (Ziegler, 1991). The main threat to the redox balance within a cell is the presence or production of molecules with one or more unpaired electrons, termed free radicals (FR). Reactive oxygen species (ROS) are FR forms of molecular oxygen and represent a large biological threat to the redox balance of cells. There are many sources of R O S and other FR within an organism. One of the most significant sources of R O S is from mitochondrial respiration, where approximately 0.1% of oxygen entering the electron transport chain is released as R O S - mainly as superoxide (0 2 " ) and hydrogen peroxide (H 2 0 2 ) (reviewed by Fridovich, 2004). Other sources of R O S include transition metal reactions, action of oxidase enzymes, redox cycling of quinones, respiratory burst activity, and environmental sources such as UV, radiation, pollution and xenobiotics (reviewed by Stohs, 1995; Berry and Kohen, 1999; Hayes and McLellan, 1999). Moderate levels of R O S are important in cell function as they can act as signals for DNA synthesis, enzyme activation, gene expression and the cell cycle (Deplancke and Gaskins, 2002). However, high levels of R O S can result in damage to key biological molecules such as lipids, proteins, and DNA (reviewed by Kidd, 1997; Hayes and McLellan, 1999). The cellular redox balance is maintained by reduction of FR, R O S and their reactive oxygen products by a suite of coordinated, integrated antioxidants (AO) that include the A O enzymes (superoxide dismutase (SOD), glutathione peroxidase (GPx), and catalase (CAT)), as well as small water soluble (such as G S H , Vit C, and uric acid) and lipid soluble (such as Vit E, Vit K, and (3-carotene) molecules (see Ternay and Sorokin, 1997; Hayes and McLellan, 1999). If R O S production increases beyond the capacity of the AO system, this can result in membrane damage, enzyme malfunction, DNA strand breaks and mutation, and cell death (reviewed by Acworth et al., 1997; Kidd, 1997). My thesis focuses on G S H and its associated enzymes as this antioxidant system is particularly important in maintaining the redox balance in the cell. G S H is present in high intracellular concentrations relative to other antioxidants (up to 10mM), and is consequently considered part of the first line of defence against exogenous or metabblically produced R O S and electrophiles (Ziegler, 1991; Wilhelm-Filho et al., 2000; Martinez-Alvarez et al., 2005). As well, G S H is the major thiol buffer in the cell (Deplancke and Gaskins, 2002). As such, G S H influences the redox status of protein thiols, and hence 1 protein function and regulation, as well as cell signaling (Arrigo, 1999; Schaferand Buettner, 2001; Lind et al., 2002; Ghezzi and Bonetto, 2003). The Glutathione Antioxidant System The antioxidant G S H and its associated enzymes constitute one of the main systems involved in the redox balance of cells. G S H is a tripeptide, composed of y-glutamic acid (y-glu), cysteine (cys) and glycine (gly). In mammals, the metabolism of G S H and its role in toxicity and disease have been extensively studied and well reviewed (including, but not limited to Meister and Anderson, 1983; Meister, 1988; Kidd, 1997; Anderson, 1998; Griffith, 1999; Hayes and McLellan, 1999; Pastore etal . , 2003; Pompella et al., 2003). In the 118 years since the discovery of G S H by de Ray-Pailhade in 1888 (Meister, 1988), there have been tens of thousands of research articles examining G S H metabolism and cellular roles. The basic components of G S H metabolism and function are similar among vertebrates, although some tissue- and species-specific differences exist. G S H has three main functions within a cell (Figure 1.1). First, G S H is an important antioxidant, reducing H 2 0 2 and lipid hydroperoxides with the GPx enzymes. In the process, G S H is oxidized (to oxidized glutathione - G S S G ) . The A O power of G S H relies on the ability of G S S G to be readily reduced back to G S H by glutathione reductase (GR), which maintains low levels of G S S G . High G S S G levels (10-50% of total G S H ) can indicate oxidative stress (Pastore et al., 2003). The second cellular function of G S H is as a cofactor of the glutathione-S-transferase (GST) enzymes. These enzymes conjugate G S H to a variety of exogenous toxicants or endogenously produced metabolites, which are then more easily excreted out of the cells via multidrug resistance-associated transport proteins (MRP, see Hayes and McLellan, 1999). Finally, G S H acts to maintain protein thiols in a reduced state and hence maintain protein structure and enzyme function (reviewed by Pastore et al., 2003). Through these three functions, G S H has been implicated in numerous cellular processes, including protein and DNA synthesis (Carelli et al., 1997; Kidd, 1997), cell growth and division (Kidd, 1997), the ubiquitin pathway of protein degradation (Jahngen-Hodge et al., 1997), and the process of apoptosis (Coppola and Ghibelli, 2000). G S H is an important part of the integrated A O system, and maintains other non-protein antioxidants in their reduced and biologically significant state (e.g. Figure 1.2, Sciuto, 1997; Berry and Kohen, 1999). Glutathione dynamics and metabolism is outlined in Figure 1.1. Glutathione dynamics are defined here, and throughout the thesis, as the sum of cellular reactions taking place of which G S H is the primary component (approximated by G S H levels, oxidation, and activity of associated enzymes). The rate limiting enzyme in G S H synthesis is y-glutamylcysteine synthetase (GCS). This enzyme is inhibited by high levels of G S H , and consequently regulates G S H levels within the cell (Richman and Meister, 1975). Amino acids for G S H synthesis are obtained from the diet, from protein breakdown, from conversion of other amino acids, and from breakdown of extracellular G S H (see Forman et al., 1995; Kidd, 1997). The only known catabolic enzyme of G S H is the extracellular peripheral protein y-glutamyltranspeptidase (yGT). yGT breaks down G S H on the surface of the cell membrane for import as amino acids (reviewed by Forman et al., 1995). Although G S H can be directly transported out of various cell types, the presence of a biologically significant, intact G S H import is a matter of some debate 2 Figure 1.1: Cellular dynamics and metabolism of glutathione (GSH). 1) G S H , along with the G S H peroxidase enzymes (GPx), reduces hydrogen peroxide (H 2 0 2 ) or lipid hydroperoxides to less harmful substances. 2) In the process, G S H is oxidized to G S S G , which is then reduced by G S S G reductase (GR), using N A D P H . N A D P H is maintained in the reduced form by glucose-6-phosphate dehydrogenase (G-6-PDH). 3) If G S S G levels rise due to oxidative stress, G S S G may be excreted from the cell. 4) G S H can be conjugated to reactive oxygen products or exogenous or endogenously produced toxicants by the GSH-S-transferase enzymes (GST), which are then excreted from the cell via the multidrug resistance-associated proteins (A). 5) G S H maintains protein thiols (P-SH) in a reduced state and 6) G S S G can result in the formation of protein-mixed disulfides (PS-SP) . These can then be reduced by glutaredoxin, or may be excreted from the cell. 7) In some tissues G S H may be transported directly out of the cell, through uni- or bi-directional transporters, although whether G S H import by this means takes place is not clear. 8) G S H is synthesized from its component amino acids (a.a.) by y-glutamylcysteine synthetase (GCS) and G S H synthetase (GS). G C S is regulated by G S H levels which inhibit its activity. 9) Extracellular G S H is broken down by the surface enzymes y-glutamyltranspeptidase (yGT) and dipeptidase, transported into the cell as a.a. or dipeptides, and resynthesized intracellularly. (Modified from Forman et al., 1995; Hayes and McLellan, 1999; Landino et al., 2004). 3 reactive oxygen products reduced products excreted out of cell GSH Figure 1.2: Integration between glutathione (GSH) and other antioxidants. 0.1% of oxygen entering the mitochondria is released as superoxide (02"~) or hydrogen peroxide (H 2 0 2 ) . 0 2 ' " is reduced to H 2 0 2 by the superoxide dismutase enzymes (SOD). H 2 0 2 is then reduced to H 2 0 by catalase (CAT), or G S H and glutathione peroxidase (GPx). Reactive oxygen species may react with biological molecules, resulting in reactive oxygen product formation. These may be conjugated to G S H by the glutathione-S-transferase enzymes (GST) and excreted out of the cell. They may also be reduced by a number of antioxidants including G S H and GPx, ascorbic acid (VitC), and a-tocopherol (VitE). The consequent VitC and VitE radicals formed are then returned to reduced form directly or indirectly by G S H . (Acworth et al., 1997; Kidd, 1997; Sciuto, 1997; Berry and Kohen, 1999; Dalton et al., 1999; Hayes and McLellan, 1999). 4 (reviewed by Ballatori et al., 2005). Cellular turnover of G S H is a balance between its synthesis from amino acids, and its export as G S H , G S S G , or glutathione-S-conjugates (Lunn et al., 1979; Lautenberg et al., 1984, Scott eta l . , 1993). In mammals, there are organ-specific differences in G S H metabolism and enzyme activity. The liver has the highest concentration of G S H (up to 10mM, Kretzschmar and Muller, 1993; Kidd, 1997). G S H is important in this organ, as the liver is the main site of exogenous and endogenous metabolite detoxification for excretion into the bile. As well, the liver is the main source of plasma G S H , and maintains plasma G S H in pM concentrations (Anderson et al., 1980; Lauterburg et al., 1984). In the plasma, G S H functions as an extracellular antioxidant with plasma GPx (see Hayes and McLellan, 1999), and is transported from the liver to other tissues (Anderson et al., 1980). The main site of plasma G S H uptake is the kidney, as it is the organ with the highest concentration of yGT (Anderson et al., 1980; Puri and Meister, 1983; Scott et al., 1993). The kidney has a large role in GSH-S-conjugate excretion (see Uchida and Osawa, 1999). Both the lungs and lens take up plasma G S H as well, likely to protect against air-borne toxicants and UV-produced R O S , respectively (Hahn et al., 1978; Hagen and Jones, 1989; Favilli et al., 1997; Kannan et al., 1999). G S H levels are also high in the brain, possibly to protect against metabolically produced R O S , although the function of G S H in this tissue is not clear (Uchida and Osawa, 1999). The roles of G S H in relation to toxicant and oxidative stressors have been well examined in fish. However, relatively little work has examined G S H metabolism and other cellular roles of G S H in fish. In contrast, G S H metabolism and function have been extensively studied in mammals. Consequently, this body of knowledge provides an excellent framework with which to compare and contrast the dynamics of G S H in the much less studied fish. This can help elucidate unique features of G S H within fish or mammals, as well as universal functions of G S H within the vertebrates and possible evolutionary processes of G S H metabolism. Glutathione and Fish Fish differ from mammals in several key aspects that may result in unique characteristics of G S H function and regulation in fish. Fish can experience daily and seasonal fluctuations in body temperature and oxygen availability, both of which could potentially affect the rate of R O S production (Wilhelm-Filho et al., 2000; Davidson and Schiestl, 2001), and consequently the necessity of G S H . In addition, extensive gill and water contact make fish particularly susceptible to uptake of water-borne toxicants, suggesting fish may have a relatively large demand on their G S H system. However, G S H levels, turnover, and activity of associated enzymes are noted as being relatively low in fish compared to mammals (Wallace, 1989; Gallagher et al., 1992; Wilhelm-Filho et al., 2000). There may a number of reasons for this. Most notably, fish are able to excrete H 2 0 2 in substantial quantities across their gills (approximately 0.8nmol/min/g fish), and consequently the AO activity of G S H and G P x may be in lower demand (Wilhelm-Filho et al., 1994). As well, fish generally have lower mass-specific metabolic rates than mammals, even when corrected for temperature (Else and Hulbert, 1981), and consequently may produce less R O S from the mitochondria. Work by Janssens et al. (2000) suggests A O levels in fish are 5 related to metabolic rate, as they found G P x and SOD, but not CAT, activities of deep sea fish species were proportional to the activity of citrate synthetase. As well, Wilhelm-Filho (1996) found active fish had higher AO levels than sluggish fish in the marine environment, but not in the freshwater environment. In contrast, Speers-Roesch and Ballantyne (2005) found both G R and CAT activity were inversely proportional to cytochrome c oxidase activity in Arctic and temperate fish species, bringing into question the relationship between metabolic rate and antioxidant levels in fish. Wilhelm-Filho et al. (2000) proposed AO status would be proportional to water temperature in thermoconformers such as fish, as oxygen consumption is generally proportional to temperature, and consequently R O S production by the mitochondria. However, whether temperature or other influences of metabolism directly affect G S H dynamics in fish, or can explain the difference in G S H dynamics between fish and mammals have not previously been investigated. Fish have other note-worthy differences in G S H metabolism compared to mammals. In mammals, G S H levels and activity of most associated enzymes are several fold (4-20) higher in the liver than in other tissues such as the kidney (Sen et al., 1994; Leeu'wenburgh and J i , 1995; Rebrin and Sohal, 2004). However, in fish, G S H levels and enzyme activity in extra-hepatic tissues such as the kidney and gill are closer to, equal to, or in some cases greater than levels in the liver (Gallagher and Di Giulio, 1992; Otto et al., 1997a, 1997b; Lushchak et al., 2001). This suggests the kidney and gill off ish may have similar functions in antioxidant defence and detoxification as the liver (as hypothesized by Gallagher and Di Giulio, 1992). Wilhelm-Filho et al. (2000) postulated that fish can tolerate high levels of G S S G , as several studies report particularly high resting G S S G levels (approximately 30% of total GSH) . However, other studies report resting G S S G levels in fish as similar to those of mammals (approximately 5% of total G S H , Bell and Cowey, 1990; Gallagher and Di Giulio, 1992; Otto et al., 1997a; Ritola et al., 1999; Pena-Llopis et al., 2001; Stephensen et al., 2002). The differences between studies may be due to species differences in G S S G levels (as noted between the channel catfish - Ictalurus punctatus and brown bullhead - Ameriurus nebulosus, Hasspieler et al., 1994), or to falsely high G S S G levels in some studies due to auto-oxidation of G S H during sample preparation. Cellular transport of G S H may differ between fish and mammals. In mammals, G S H is not directly transported into cells but is broken down extracellularly and resynthesized within the cell. Given that G C S activity is inhibited by G S H , endogenous G S H prevents further synthesis of G S H with increased amino acid supply (Meister, 1995). Thus, attempting to increase tissue levels with exogenous G S H in mammals has little effect on G S H in tissues such as the liver, kidney, muscle, lung and lymphoid cells (Puri and Meister, 1983; Wellner et al., 1984; Martensson et al., 1989; Martensson and Meister, 1989). However, Otto et al. (1997b) found intraperitoneal injections of G S H in rainbow trout (Oncorhynchus mykiss) and American eel (Anguilla rostrata) increased G S H levels in the liver, kidney and gills, and postulated that direct G S H uptake may be a distinct feature of fish tissues. Although the mechanisms by which fish tissues take up exogenous G S H are not known, the ability to alter G S H levels with exogenous G S H make fish an interesting model to examine cellular roles of G S H . 6 Glutathione and the Stress Response in Fish The majority of studies examining G S H in fish have focused on the role of G S H during exposure to oxidative or toxicant stressors. Although there are species (Ploch et al., 1999), tissue- (Otto and Moon, 1995; Feng et al., 2001; Lushchak et al., 2001), AO type- (Almar et al., 1998), and stressor- (Bell and Cowey, 1990; Ritola et al., 2002; Stephensen et al., 2002; Lima et al., 2006) specific differences, in general, exposure to toxicant and oxidative stressors transiently decreases and eventually increases G S H levels in fish, and alters G S H enzyme activity (as above and Lindstrom-Seppa et al., 1996; Meyer et al., 2003; Hughes and Gallagher, 2004). This suggests G S H and its associated enzymes are generally increased in response to stressors that have toxicological effects or threaten redox homeostasis. In addition, G S H status appears to be directly related to resistance to oxidative or toxicological stressors in fish. Increased G S H levels have been associated with decreased pollution-related neoplasms (Hasspieler et al., 1994), and resistance to herbicides (Pena-Llopis et al., 2001), H 2 0 2 (Wright et al., 2000), and tert-butyl hydroperoxide (Meyer et al., 2003), and have been associated with increased resistance to arsenic in arsenic-sensitive, but not arsenic-resistant, cell lines (Wang et al., 2004). As well, decreased G S H levels have been associated with increased lipid peroxide formation with or without exposure to tert-butyl hydroperoxide (Ploch et al., 1999), decreased survival during pesticide exposure (Dorval and Hontela, 2003), and increased cytotoxicity to "soft" (Hg, Cu , Cd) but not "hard" (Zn, Ni, Pb) metals (Maracine and Segner, 1998). Adequate levels of G S H likely maintain cell and organism health during potentially harmful chemical exposure by providing AO activity and as a substrate for GPx and G S T enzymes. Although the importance of G S H during toxicant and oxidative stressors in fish is well documented, there is increasing evidence that G S H may have a more general role in the response to stress. When environmental conditions fluctuate beyond the normal range experienced by an organism, this can compromise internal homeostasis and result in the organism entering a stressed state. In order to maintain homeostasis during such fluctuations, fish mount an integrated, generalized stress response. The primary component of the generalized stress response is increased plasma levels of stress hormones such as catecholamines and corticosteroids (see Barton and Iwama, 1991). Increased plasma catecholamine levels result in the mobilization of energy stores, increased cardiac output, ventilation, gill perfusion and oxygen carrying capacity of the blood, as well as altered osmotic and acid/base regulation (see Wendelaar Bonga, 1997; Reid et al., 1998; Perry and Bernier, 1999). Increased plasma Cortisol levels result in a mobilization of energy stores from increased carbohydrate, protein arid lipid metabolism, as well as alterations in ion regulation, and increased cytochrome P450 induction (see Wendelaar Bonga, 1997; Mommsen et al., 1999). These secondary changes allow for adequate energy and oxygen supply, as well as maintaining osmotic and acid base regulation during stress (see Barton and Iwama, 1991; Wendelaar Bonga, 1997; Re ide ta l . , 1998; Mommsen et al., 1999; Perry and Bernier, 1999). If the stress persists, or is of sufficient severity, this can result in osmotic or acid-base imbalances, decreased immune function, reproductive capacity, and growth, and finally death (see Barton and Iwama, 1991; Wendelaar Bonga, 1997). 7 Fish also respond to stressors on a cellular level by inducing a number of stress proteins, termed heat shock proteins (Hsp's, Iwama et al., 1998; Iwama et al., 1999; Basu et al., 2002). Hsp's are upregulated in response to a number of stressors that compromise protein integrity, such as heat and chemical stress, energy depletion, ion imbalance and pathogens (Macario, 1995; Forsyth et al., 1997; Feder and Hofmann, 1999; Ackerman and Iwama, 2001). Hsp's act as protein chaperones, assisting in the proper folding and translocation of cellular proteins, as well as maintaining protein integrity and repairing damaged proteins during stress (see Macario, 1995; Feder and Hofmann, 1999; Kregel, 2002). Some Hsp's are constitutively expressed (e.g. heat shock cognate 70 - Hsc70), some constitutively expressed and upregulated during stress (e.g. Hsp70), and others are only synthesized in response to stress (e.g. Hsp30, see Basu et al., 2002). There is mounting evidence that the generalized and cellular stress responses are integrated in fish. Increased plasma Cortisol levels are reported to reduce the increase in Hsp70 levels after heat stress (Ackerman et al., 2000; Basu et al., 2001; Boone et al., 2002), possibly through interactions with the proteasome and the glucocorticoid receptor (Boone and Vijayan, 2002; Basu et al., 2003). G S H may be part of the generalized response to stress, and not specific to toxicant and oxidant stressors, through a number of ways. Stress can increase oxygen consumption (see Wendelaar Bonga, 1997) which in turn can increase reactive oxygen species production by the mitochondria (Muradian et al., 2002). As such, stress in general may compromise the cellular redox balance, and consequently increase demand on the G S H AO system. Gasch et al. (2000) found several genes in yeast involved in antioxidant processes were consistently induced by a wide variety of environmental stressors, suggesting maintenance of the redox balance is a key function of the response to stress. Stress could also increase levels of endogenous toxicants requiring G S T and G S H for detoxification and excretion (e.g. catecholamine metabolites, Baez et al., 1997). In mammals, G S H reportedly interacts with the thiol groups of a number of components of the stress response, such as the glucocorticoid receptor (Esposito eta l . , 1995;Makino etal . , 1999), Hsc70 (Hoppe et al., 2004) and the small heat shock proteins Hsp25-27 (Zavialov et al., 1998; Papp et al., 2003). Consequently, adequate G S H levels may be necessary to maintain cellular functions of these components. As well, catecholamines and corticosteroids are reported to increase G S H synthesis and levels of A O enzymes in various vertebrates, including fish (Ignatius and Oommen, 1990; Lu et a|., 1992; Speck et al., 1993; Walther et al., 1996). G S H dynamics are altered by stressors that do not directly comprise oxidized or toxicant stress, such as heat (Parihar et al., 1997; Kaur et al., 2005), and salinity stress (Martinez-Alvarez et al., 2002), suggesting G S H is important during stress in general. However, the function of G S H as a significant component of the generalized and cellular stress responses in fish requires further examination. Thesis Outline I addressed three primary objectives in my thesis: Objective I: To determine the mechanisms by which G S H is transported in fish (Chapter 2). Objective II: To examine the importance of G S H during conditions that alter metabolism (Chapters 3 and 4). 8 Objective III: To examine the role of G S H as a component of the generalized and cellular stress responses in fish (Chapters 5 and 6). Through these objectives I hoped to gain insights into the roles of G S H in vertebrates in general, and fish in particular, and identify how G S H integrates with other systems to enable fish to reside in a fluctuating environment. Objective I: In Chapter 2, I examined how exogenous G S H is transported into fish tissues. Otto et al. (1997b) postulated fish may be unique in the vertebrate world by being able to transport G S H directly into their tissues. Using rainbow trout as a model, I determined the validity of this hypothesis, and gained insights into unique mechanisms of tissue-specific G S H metabolism in rainbow trout. Objective II: In Chapters 3 and 4, I examined the effects of factors that influence metabolism (i.e. acclimation temperature and growth rate) on G S H dynamics in fish. G S H levels, turnover and activity of associated enzymes are relatively low in fish compared with mammals. In Chapter 3, I examined the effects of incubation or acclimation temperature on G S H turnover in a rainbow trout hepatoma cell line (RTH-149), and G S H levels and activity of associated enzymes in tissues of the eurythermal killifish (Fundulus heteroclitus macrolepidotus). Using these models I determined whether G S H utilization, levels, and corresponding activity of associated enzymes are altered to cope with changes in environmental temperature, and determined to what extent body temperature can explain the differences in G S H dynamics between fish and mammals. In Chapter 4, I further examined the effects of metabolic rate on G S H dynamics using growth hormone transgenic coho salmon (Oncorhynchus kisutch) as a fish with accelerated growth. I examined the effects of a growth hormone transgene on metabolic rate and G S H dynamics during feeding and starvation to determine if G S H dynamics could be correlated to metabolic rates in fish. These chapters addressed whether G S H utilization, levels, and activity of associated enzymes are altered in response to changing environmental and physiological conditions in various fish species. Objective III: In Chapters 5 and 6, I examined whether G S H has a role in the response to stress in general, or if its importance is limited to stressors that directly cause oxidative stress. For this objective I used heat stress as my model stressor. Heat stress is an interesting model as it does not directly compromise the redox status of cells, but may indirectly induce oxidative stress by increasing metabolic rate. As well, heat stress is known to induce components of the generalized stress response (e.g. plasma Cortisol and glucose) as well as the cellular stress response (various heat shock protein families), and hence allows for determination of interactions between these components and the G S H system in fish. In Chapter 5,1 examined whether heat stress alters G S H dynamics in three levels of organization of rainbow trout: in vivo, in primary cultured hepatocytes in vitro, and RTH-149 cells in vitro. In addition, G S H dynamics after heat stress were compared with those after an oxidative stress in vitro, to determine if heat stress results in oxidative modulation of the G S H system. In Chapter 6, I examined whether altered G S H levels affect the generalized or cellular stress responses to heat and chasing stress in rainbow trout, using both in vivo, and RTH-149 in vitro models. These chapters examined the role of G S H in the generalized and cellular responses to stress, and expand the understanding of how rainbow trout and other fish may respond to acute environmental fluctuations. 9 In Chapter 7, I integrated the findings addressed in Objectives l-lll and addressed future research needed to further the understanding of G S H function and metabolism in fish. Justification of Experimental Models I used various experimental models to address the above objectives. In Objective I, Chapter 2, rainbow trout were used to examine G S H transport, as this fish has previously demonstrated uptake of exogenous G S H (Otto et al., 1997b). As G S H levels are easily manipulated in these fish, rainbow trout were also used to determine the effects of altered G S H levels on components of the primary and cellular stress response (Objective III, Chapter 6). To be consistent, rainbow trout were also used to determine the effects of heat shock on G S H dynamics in fish (Objective III, Chapter 5). In Objective II, I examined the effects of factors that influence metabolic rate on G S H dynamics in fish. To determine the effects of acclimation temperature on G S H dynamics (Chapter 3) I used the killifish as this fish is extremely eurythermal and therefore allowed for examination of temperature effects over a wide range. To more closely examine the relationship between metabolism and G S H dynamics (Chapter 4), I used growth hormone transgenic coho salmon as my model. The rapid growth rate of this fish makes it a useful model as growth rate, and hence metabolic rate, can be easily controlled by feeding levels. This allows for comparisons of G S H dynamics and metabolic rates in fish. I was interested in cellular aspects of G S H metabolism and function in several of the above objectives - specifically cellular G S H turnover (Objective II, Chapter 3), and interactions between G S H and the cellular stress response (Objective III, Chapters 5-6). My original plan was to use a primary culture of hepatocytes for these experiments, as this is a reduced model that maintained similar traits to in vivo hepatocytes. However, G S H levels in primary cultured hepatocytes increased steadily over time (see Chapter 5, Figure 5.3). As such, this may be a poor model for examining cellular G S H dynamics in fish, as the effects of culture on G S H dynamics may influence cellular functions of G S H . Consequently, for most in vitro studies I used a rainbow trout hepatoma cell line, as G S H levels were more stable in this model (Chapter 5, Figure 5.5). Reduced preparations, such as the RTH-149 cell line, allow for determination of cellular pathways, without the influence of whole animal controlling factors such as hormones. 10 References Ackerman, P.A., Iwama, G.K., 2001. Physiological and cellular stress responses of juvenile rainbow trout to vibriosis. Journal of Aquatic Animal Health 13: 173-180. Ackerman, P.A., Forsyth, R.B., Mazur, C.F. , Iwama, G.K., 2000. Stress hormones and the cellular stress response in salmonids. Fish Physiology and Biochemistry 23: 327-336. Acworth, I.N., McCabe, D.R., Maher, T.J . , 1997. The analysis of free radicals, their reaction products, and antioxidants. In: Baskin, S.I. Salem, H. (Eds.), Oxidants, Antioxidants, and Free Radicals. Taylor and Francis, Washington, pp. 23-78. Almar, M., Otero, L , Santos, C , Gallego, J .G . , 1998. Liver glutathione content and glutathione-dependent enzymes of two species of freshwater fish as bioindicators of chemical pollution. Journal of Environmental Science and Health. Part B: Pesticides, Food Contaminants, and Agricultural Wastes 33: 769-783. Anderson, M.E., 1998. Glutathione: an overview of biosynthesis and modulation. Chemico-Biological Interactions 111-112: 1-14. Anderson, M.E., Bridges, R.J. , Meister, A., 1980. Direct evidence for inter-organ transport of glutathione and that the non-filtration renal mechanism for glutathione utilization involves Cl-glutamyl transpeptidase. Biochemical and Biophysical Research Communications 96: 848-853. Arrigo, A . -P . , 1999. Gene expression and the thiol redox state. Free Radical Biology & Medicine 27: 936-944. Baez, S., Segura-Aguilar, J . , Widersten, M., Johansson, A . -S . , Mannervik, B., 1997. Glutathione transferases catalyse the detoxication of oxidized metabolites (o-quinones) of catecholamines and may serve as an antioxidant system preventing degenerative cellular processes. Biochemical Journal 324: 25-28. Ballatori, N., Hammond, C.L., Cunningham, J.B., Krance, S.M. , Marchan, R., 2005. Molecular mechanisms of reduced glutathione transport: role of the M R P / C F T R / A B C C and OATP/SLC21A families of membrane proteins. Toxicology and Applied Pharmacology 204: 238-255. Barton, B.A., Iwama, G.K., 1991. Physiological changes in fish from stress in aquaculture with emphasis on the response and effects of corticosteroids. Annual Reviews of Fish Diseases 1: 3-26. Basu, N., Kennedy, C . J . , Iwama, G.K., 2003. The effects of stress on the association between hsp70 and the glucocorticoid receptor in rainbow trout. Comparative Biochemistry and Physiology. Part A, Physiology 134: 655-663. Basu, N., Nakano, T., Grau, E.G., Iwama, G.K., 2001. The effects of Cortisol on heat shock protein 70 levels in two fish species. General & Comparative Endocrinology 124: 97-105. Basu, N., Todgham, A.E. , Ackerman, P.A., Bibeau, M.R., Nakano, K., Schulte, P.M., Iwama, G.K., 2002. Heat shock protein genes and their functional significance in fish. Gene 295: 173-183. Bell, G. , Cowey, C.B. , 1990. Oxidation of glutathione during hydroperoxide metabolism in isolated hepatocytes of rainbow trout (Salmo gairdneh). Fish Physiology and Biochemistry 8: 11-17. Berry, E.M., Kohen, R., 1999. Is the biological antioxidant system integrated and regulated? Medical Hypothesis 53: 397-401. Boone, A .N . , Vijayan, M.M., 2002. Glucocorticoid-mediated attenuation of the hsp70 response in trout hepatocytes involves the proteasome. American Journal of Physiology - Regulated and Integrated Comparative Physiology 283: R680-R687. Boone, A .N . , Ducouret, B., Vijayan, M.M., 2002. Glucocorticoid-induced glucose release is abolished in trout hepatocytes with elevated hsp70 content. Journal of Endocrinology 172: R1-R5. Carelli, S., Ceriotti, A., Cabibbo, A., Fassina, G. , Ruvo, M., Sitia, R., 1997. Cysteine and glutathione secretion in response to protein disulfide bond formation in the ER. Science 277: 1681-1684. Coppola, S., Ghibelli, L., 2000. G S H extrusion and the mitochondrial pathway of apoptotic signalling. Biochemical Society Transactions 28: 56-61. Dalton, T.P., Shertzer, H.G., Puga, A., 1999. Regulation of gene expression by reactive oxygen. Annual Reviews in Pharmacology and Toxicology 39: 67-101. 11 Davidson, J .F. , Schiestl, R.H., 2001. Mitochondrial respiratory electron carriers are involved in oxidative stress during heat stress in Saccharomyces cerevisiae. Molecular and Cellular Biology 21: 8483-8489. Deplancke, B., Gaskins, H.R., 2002. Redox control of the transsulfuration and glutathione biosynthesis pathways. Current Opinion in Clinical Nutritional and Metabolic Care 5: 85-92. Dorval, J . , Hontela, A., 2003. Role of glutathione redox cycle and catalase in defense against oxidative stress induced by endosulfan in adrenocortical cells of rainbow trout (Oncorhynchus mykiss). Toxicology and Applied Pharmacology 192: 191-200. Else, P.L., Hulbert, A . J . , 1981. Comparison of the "mammalian machine" and the "reptile machine": energy production. American Journal of Physiology 240: R3-R9. Esposito, F., Cuccovillo, F., Morra, F., Russo, T., Cimino, R., 1995. DNA binding activity of the glucocorticoid receptor is sensitive to redox changes in intact cells. Biochimica et Biophysica Acta 1260: 308-314. Favilli, F., Marraccini, P., lantomasi, T., Vincenzini, M.T., 1997. Effect of orally administered glutathione on glutathione levels in some organs of rats: role of specific transporters. British Journal of Nutrition 78: 293-300. Feder, M.E., Hofmann, G.E. , 1999. Heat-shock proteins, molecular chaperones, and the stress response: evolutionary and ecological physiology. Annual Reviews in Physiology 61: 243-282. Feng, T., Zheng, W.Y., Guo, X.-Q. , Zhang, Y., 2001. Reduced-glutathione concentrations in Boleophthalmus pectinirostris tissues exposed to benzo[a]pyrene. Journal of Environmental Sciences 13: 308-310. Forman, H.F., Liu, R.-M., Shi, M.M., 1995. Glutathione synthesis in oxidative stress. In: Packer, L. Cadenas, E. (Eds.), Antioxidants in Health and Disease, Biothiols in Health and Disease. M. Dekker, New York, pp. 189-212. Forsyth, R.B., Candido, E.P.M. , Babich, S.L., Iwama, G.K., 1997. Stress protein expression in coho salmon with bacterial kidney disease. Journal of Aquatic Animal Health 9: 18-25. Fridovich, I., 2004. Mitochondria: are they the seat of senescence? Aging Cell 3: 13-16. Gallagher, E.P., Di Giulio, R.T., 1992. A comparison of glutathione-dependent enzymes in liver, gills and posterior kidney of channel catfish {Ictalurus punctatus). Comparative Biochemistry and Physiology. Part C, Pharmacology, Toxicology and Endocrinology 102: 543-547. Gallagher, E.P., Hasspieler, B.M., Di Giulio, R.T., 1992. Effects of buthionine sulfoximine and diethyl maleate on glutathione turnover in the channel catfish. Biochemical Pharmacology 43: 2209-2215. Gasch, A .P . , Spellman, P.T., Kao, C M . , Carmel-Harel, O., Eisen, M.B., Storz, G. , Botstein, D., Brown, P.O., 2000. Genomic expression programs in the response of yeast cells to environmental changes. Molecular Biology of the Cell 11:4241-4257. Ghezzi , P., Bonetto, V., 2003. Redox proteomics: Identification of oxidatively, modified proteins. Proteomics 3: 1145-1153. Griffith, O.W., 1999. Biological and pharmacologic regulation of mammalian glutathione synthesis. Free Radical Biology and Medicine 27: 922-935. Hagen, T.M., Jones, D.P., 1989. Role of glutathione transport in extrahepatic detoxication. In: Taniguchi, N., Higashi, T., Sakamoto, Y. Meister, A. (Eds.), Glutathione Centennial: Molecular Perspectives and Clinical Implications. Academic Press, Inc., Tokyo, pp. 423-433. Hahn, R., Wendel, A., Flohe, L., 1978. Fate of extracellular glutathione in rat. Biochimica et Biophysica Acta 539: 324-337. Hasspieler, B.M., Behar, J.V., Di Giulio, R.T., 1994. Glutathione-dependent defense in channel catfish (Ictalurus punctatus) and brown bullhead (Ameriurus nebulosus). Ecotoxicology and Environmental Safety 28: 82-90. Hayes, J.D., McLellan, L.I., 1999. Glutathione and glutathione-dependent enzymes represent a co-ordinately regulated defense against oxidative stress. Free Radical Research 31: 273-300. 12 Hoppe, G. , Chai, Y .C . , Crabb, J.W., Sears, J . , 2004. Protein s-glutathionylation in retinal pigment epithelium converts heat shock protein 70 to an active chaperone. Experimental Eye Research 78: 1085-1092. Hughes, E., Gallagher, E., 2004. Effects of D-naphthoflavone on hepatic biotransformation and glutathione biosynthesis in largemouth bass (Micropterus salmoides). Marine Environmental Research 58: 675-679. Ignatius, J . , Oommen, O.V., 1990. Effects of corticosteroids and protein synthesis inhibitors on activities of oxidative enzymes in a bony fish, Anabas testudineus (Bloch). General & Comparative Endocrinology 78: 303-310. Iwama, G.K., Thomas, P.T., Forsyth, R.B., Vijayan, M.M., 1998. Heat shock protein expression in fish. Reviews in Fish Biology and Fisheries 8: 35-56. Iwama, G.K., Vijayan, M.M., Forsyth, R.B., Ackerman, P., 1999. Heat shock proteins and physiological stress in fish. American Zoology 39: 901-909. Jahngen-Hodge, J . , Obin, M.S., Gong, X. , Shang, F., Nowell, T .R.J . , Gong, J . , Abasi , H., Blumberg, J . , Taylor, A., 1997. Regulation of ubiquitin-conjugating enzymes by glutathione following oxidative stress. The Journal of Biological Chemistry 45: 28218-28226. Janssens, B.J. , Childress, J . J . , Baguet, F., Rees, J. -F. , 2000. Reduced enzymatic antioxidative defense in deep-sea fish. The Journal of Experimental Biology 203: 3717-3725. Kannan, R., Bao, Y.Z. , Wang, Y., Sarthy, V .P . , Kaplowitz, N., 1999. Protection from oxidant injury by sodium-dependent G S H uptake in retinal Muller cells. Experimental Eye Research 68: 609-616. Kaur, M., Atif, F., AN, M., Rehman, H., Raisuddin, S., 2005. Heat stress-induced alterations of antioxidants in the freshwater fish Channa punctata Bloch. Journal of Fish Biology 67: 1653-1665. Kidd, P.M., 1997. Glutathione: systemic protectant against oxidative and free radical damage. Alternative Medicine Review 2: 155-176. Kregel, K.C., 2002. Heat shock proteins: modifying factors in physiological stress responses and acquired thermotolerance. Journal of Applied Physiology 92: 2177-2186. Kretzschmar, M., Muller, D., 1993. Aging, training and exercise - a review of effects on plasma glutathione and lipid peroxides. Sports Medicine 15: 196-209. Landino, L.M., Moynihan, K.L., Todd, J.V., Kennett, K.L., 2004. Modulation of the redox state of tubulin by the glutathione/glutaredoxin reductase system. Biochemical and Biophysical Research Communications 314: 555-560. Lauterburg, B.H., Adams, J.D., Mitchell, J.R., 1984. Hepatic glutathione homeostasis in the rat: efflux accounts for glutathione turnover. Hepatology 4: 586-590. Leeuwenburgh, C , J i , L , 1995. Glutathione depletion in rested and exercised mice - biochemical consequence and adaptation. Archives of Biochemistry and Biophysics 316: 941-949. Lima, P.L., Benassi, J .C . , Pedrosa, R . C , Dal Magro, J . , Oliveira, T.B., Wilhelm-Filho, D., 2006. Time-course variations of DNA damage and biomarkers of oxidative stress in tilapia (Oreochromis niloticus) exposed to effluents from a swine industry. Archives of Environmental Contamination and Toxicology 50: 23-30. Lind, C , Gerdes, R., Hamnell, Y., Schuppe-Koistinen, I., von Lowenhielm, H.B., Holmgren, A., Cotgreave, I.A., 2002. Identification of S-glutathionylated cellular proteins during oxidative stress and constitutive metabolism by affinity purification and proteomic analysis. Archives of Biochemistry and Biophysics 406: 229-240. Lindstrom-Seppa, P., Roy, S., Huuskonen, S., Tossavainen, K., Ritola, O., Marin, E., 1996. Biotransformation and glutathione homeostasis in rainbow trout exposed to chemical and physical stress. Marine Environmental Research 42: 323-327. Lu, S.C. , J . , G., Kuhlenkamp, J . , Kaplowitz, N., 1992. Insulin and glucocorticoid dependence of hepatic y-glutamylcysteine synthetase and glutathione synthesis in the rat. Journal of Clinical Investigation 90: 524-532. Lunn, G. , Dale, G.L., Beutler, E., 1979. Transport accounts for glutathione turnover in human erythrocytes. Blood 54: 238-244. 13 Lushchak, V.I., Lushchak, L.P., Mota, A.A., Hermes-Lima, M., 2001. Oxidative stress and antioxidant defenses in goldfish Carassius auratus during anoxia and reoxygenation. American Journal of Physiology 280: R100-R107. Macario, A.J.L. , 1995. Heat-shock proteins and molecular chaperone: implications for pathogenesis, diagnostics, and therapeutics. International Journal of Clinical and Laboratory Research 25: 59-70. Makino, Y., Yoshikawa, N., Okamoto, K., Hirota, K., Yodoi, J . , Makino, I., Tanaka, H., 1999. Direct association with thioredoxin allows redox regulation of glucocorticoid receptor function. The Journal of Biological Chemistry 274: 3182-3188. Maracine, M., Segner, H., 1998. Cytotoxicity of metals in isolated fish cells: Importance of the cellular glutathione status. Comparative Biochemistry and Physiology. Part A, Physiology 120: 83-88. Martensson, J . , Meister, A., 1989. Mitochondrial damage in muscle occurs after marked depletion of glutathione and is prevented by giving glutathione monoester. Proceedings of the National Academy of Sciences of the United States of America 86: 471-475. Martensson, J . , Jain, A., Frayer, W., Meister, A., 1989. Glutathione metabolism in the lung - inhibition of its synthesis leads to lamellar body and mitochondrial defects. Proceedings of the National Academy of Sciences of the United States of America 86: 5296-5300. Martinez-Alvarez, R.M., Morales, A .E . , Sanz, A., 2005. Antioxidant defenses in fish: Biotic and abiotic factors. Reviews in Fish Biology and Fisheries 15: 75-88. Martinez-Alvarez, R.M., Hidalgo, M.C., Domezain, A., Morales, A .E . , Garcia-Gallego, M., Sanz, A., 2002. Physiological changes of sturgeon Acipenser naccarii caused by increasing environmental salinity. Journal of Experimental Biology 205: 3699-3706. Meister, A., 1988. On the discovery of glutathione. Trends in Biochemical Sciences 13: 185-188. Meister, A., 1995. Strategies for increasing cellular glutathione. In: Packer, L. Cadenas, E. (Eds.), Biothiols in Health and Disease. M. Dekker, New York, pp. 165-188. Meister, A., Anderson, M.E., 1983. Glutathione. Annual Review of Biochemistry 52: 711-760. Meyer, J . , Smith, J . , Winston, G. , Di Giulio, R., 2003. Antioxidant defenses in killifish (Fundulus heteroclitus) exposed to contaminated sediments and model prooxidants: short-term and heritable responses. Aquatic Toxicology 65: 377-395. Mommsen, T.P., Vijayan, M.M., Moon, T.W., 1999. Cortisol in teleosts: dynamics, mechanisms of action, and metabolic regulation. Reviews in Fish Biology and Fisheries 9: 211-268. Muradian, K.K., Utko, N.A., Mozzhukhina, T.G. , Litoshenko, A.Y., Pishel, I.N., Bezrukov, V.V., Fraifield, V .E . , 2002. Pair-wise linear and 3D nonlinear relationships between the liver antioxidant enzyme activities and the rate of body oxygen consumption in mice. Free Radical Biology and Medicine 33: 1736-1739. Otto, D.M.E., Moon, T.W., 1995. 3,3',4,4'-Tetrachlorobiphenyl effects on antioxidant enzymes and glutathione status in different tissues of rainbow trout. Pharmacology and Toxicology 77: 281-287. Otto, D.M.E., Sen, C.K., Casley, W.L., Moon, T.W., 1997a. Regulation of 3,3',4,4'-tetrachlorobiphenyl induced cytochrome p450 metabolism by thiols in tissues of rainbow trout. Comparative Biochemistry and Physiology. Part C, Pharmacology, Toxicology and Endocrinology 117: 299-309. Otto, D.M.E., Sen, C.K., Hidiroglou, N., Madere, R., Moon, T.W., 1997b. Role of exogenous glutathione in teleost fish and its effects on antioxidant defense responses in rainbow trout exposed to 3,3',4,4'-tetrachlorobiphenyl. Fish Physiology and Biochemistry 16: 449-457. Papp, E., Nardai, G. , Soti, C , Csermely, P., 2003. Molecular chaperones, stress proteins and redox homeostasis (Reprinted from Thiol metabolism and Redox Regulation of Cellular Functions). Biofactors 17: 249-257. Parihar, M.S., Javeri, T., Hemnani, T., Dubey, A.K., Prakash, P., 1997. Responses of superoxide dismutase, glutathione peroxidase and reduced glutathione antioxidant defenses in gills of the freshwater catfish (Heteropneustes fossilis) to short-term elevated temperature. Journal of Thermal Biology 22: 151-156. Pastore, A., Federici, G. , Bertini, E., Piemonte, F., 2003. Analysis of glutathione: implication in redox and detoxification. Clinica Chimica Acta 333: 19-39. 14 Pena-Llopis, S., Pena, J.B., Sancho, E., Fernandez-Vega, C , 2001. Glutathione-dependent resistance of the European eel Anguilla anguilla to the herbicide molinate. Chemosphere 45: 671-681. Perry, S.F. , Bernier, N.J., 1999. The acute humoral adrenergic stress response in fish: facts and fiction. Aquaculture 177: 285-295. Ploch, S.A., Lee, Y . -P . , MacLean, E., Di Giulio, R.T., 1999. Oxidative stress in liver of brown bullhead and channel catfish following exposure to ferf-butyl hydroperoxide. Aquatic Toxicology 46: 231-240. Pompella, A., Visvikis, A., Paolicchi, A., De Tata, V., Casini, A.F. , 2003. The changing faces of glutathione, a cellular protagonist. Biochemical Pharmacology 66: 1499-1503. Puri, R.N., Meister, A., 1983. Transport of glutathione, as gamma-glutamylcysteinylglycyl ester, into liver and kidney. Proceedings of the National Academy of Sciences of the United States of America 80: 5258-5260. Rebrin, I., Sohal, R., 2004. Comparison of thiol redox state of mitochondria and homogenates of various tissues between two strains of mice with different longevities. Experimental Gerontology 39: 1513-1519. Reid, S .G . , Bernier, N.J., Perry, S.F. , 1998. The adrenergic stress response in fish: control of catecholamine storage and release. Comparative Biochemistry and Physiology. Part C, Pharmacology, Toxicology and Endocrinology 120: 1-27. Richman, P., Meister, A., 1975. Regulation of gamma-glutamyl-cysteine synthetase by non-allosteric feedback inhibition by glutathione. Journal of Biological Chemistry 250: 1422-1426. Ritola, O., Livingstone, D.R., Peters, L.D., Lindstrom-Seppa, P., 2002. Antioxidant processes are affected in juvenile rainbow trout (Oncorhynchus mykiss) exposed to ozone and oxygen-supersaturated water. Aquaculture 210: 1-19. Ritola, O., Kiuru, T., Koponen, K., Molsa, H., Lindstr6m-Seppa, P., 1999. Rainbow trout (Oncorhynchus mykiss) exposed to oxygen supersaturation and handling stress: plasma Cortisol and hepatic glutathione status. Acta Biologica Hungarica 50: 215-227. Schafer, R.Z., Buettner, G.R., 2001. Redox environment of the cell as viewed through the redox state of the glutathione disulfide/glutathione couple. Free Radical Biology & Medicine 30: 1191-1212. Sciuto, A .M. , 1997. Antioxidant properties of glutathione and its role in tissue protection. In: Baskin, S.I. Salem, H. (Eds.), Oxidants, Antioxidants, and Free Radicals. Taylor and Francis, Washington, pp. 171-192. Scott, R.D., Hughey, R.P., Curthoys, N.P., 1993. Role of apical and basolateral secretion in turnover of glutathione in Llc-Pk(1) cells. American Journal of Physiology 265: F723-F728. Sen, C.K., Atalay, M., Hanninen, O., 1994. Exercise-induced oxidative stress: glutathione supplementation and deficiency. Journal of Applied Physiology 77: 2177-2187. Speck, R.F., Schranz, C , Lauterburg, B.H., 1993. Prednisolone stimulates hepatic glutathione synthesis in mice: protection by prednisolone against acetaminophen hepatotoxicity in vivo. Journal of Hepatology 18: 67. Speers-Roesch, B., Ballantyne, J .S. , 2005. Activities of antioxidant enzymes and cytochrome c oxidase in liver of Arctic and temperate teleosts. Comparative Biochemistry and Physiology. Part A, Physiology 140: 487-494. Stephensen, E., Sturve, J . , Forlin, L., 2002. Effects of redox cycling compounds on glutathione content and activity of glutathione-related enzymes in rainbow trout liver. Comparative Biochemistry and Physiology. Part C, Pharmacology, Toxicology and Endocrinology 133: 435-442. Stohs, S .J . , 1995. The role of free radicals in toxicity and disease. Journal of Basic and Clinical Physiology and Pharmacology 6: 205-228. Ternay, A .L .J . , Sorokin, V., 1997. Redox, radicals, and antioxidants. In: Baskin, S.I. Salem, H. (Eds.), Oxidants, Antioxidants, and Free Radicals. Taylor and Francis, Washington, pp. 1-22. Uchida, K., Osawa, T., 1999. Cellular response to lipid peroxidation products. In: Packer, L. Yodoi, J . (Eds.), Redox regulation of cell signalling and its clinical application. Marcel Dekker Inc., New York, pp. 105-115. 15 Wallace, K.B., 1989. Glutathione-dependent metabolism in fish and rodents. Environmental Toxicology and Chemistry 8: 1049-1055. Walther, F.J. , David-Cu, R., Mehta, E.I., Polk, D.H., Jobe, A.H. , Ikegami, M., 1996. Higher lung antioxidant enzyme activity persists after single dose of corticosteroids in preterm lambs. American Journal of Physiology 15: L187-L191. Wang, Y., Chaung, R., Tung, L , 2004. Comparison of the cytotoxicity induced by different exposure to sodium arsenite in two fish cell lines. Aquatic Toxicology 69: 67-79. Wellner, V .P . , Anderson, M.E., Puri, R.N., Jensen, G.L., Meister, A., 1984. Radioprotection by glutathione ester - transport of glutathione ester into human lymphoid-cells and fibroblasts. Proceedings of the National Academy of Sciences of the United States of America 81: 4732-4735. Wendelaar Bonga, S.E. , 1997. The stress response in fish. Physiological Reviews 77: 591-625. Wilhelm-Filho, D., 1996. Fish antioxidant defenses - A comparative approach. Brazilian Journal of Medical and Biological Research 29: 1735-1742. Wilhelm-Filho, D., Gonzalez-Flecha, B., Boveris, A., 1994. Gill diffusion as a physiological mechanism for hydrogen peroxide elimination by fish. Brazilian Journal of Medical and Biological Research 27: 2879-2882. Wilhelm-Filho, D., Torres, M.A., Marcon, J.L., Fraga, C .G. , Boveris, A., 2000. Comparative antioxidant defences in vertebrates - emphasis on fish and mammals. Trends in Comparative Biochemistry And Physiology 7: 33-45. Wright, J . , George, S., Martinez-Lara, E., Carpene, E., Kindt, M., 2000. Levels of cellular glutathione and metallothionein affect the toxicity of oxidative stressors in an established carp cell line. Marine Environmental Research 50: 503-508. Zavialov, A.V., Gaestel, M., Korpela, T., Zav'yalov, V .P . , 1998. Thiol/disulfide exchange between small heat shock protein 25 and glutathione. Biochimica et Biophysica Acta 1388: 123-132. Ziegler, D.M., 1991. Non-radical mechanisms for the oxidation of glutathione. In: Sies, H. (Eds.), Oxidative Stress: Oxidants and Antioxidants. Academic Press Ltd., London, pp. 85-97. 16 CHAPTER 2 Can rainbow trout directly take up glutathione into their tissues? Introduction Glutathione (GSH) is an important intracellular antioxidant and thiol and is involved in cellular protection from oxidative stress, toxicants and irradiation. Although its metabolism,has been extensively studied in mammals, little is known of G S H dynamics in fish. In mammals, G S H is involved in a number of cellular processes such as protein and DNA synthesis, cell growth and division, and apoptosis, and its decrease is associated with aging and many human diseases (reviewed by Kidd, 1997; Coppola and Ghibelli, 2000). Consequently, there is much interest in increasing intracellular G S H levels with exogenous G S H for both medical applications and to elucidate cellular roles of G S H (Meister, 1995). However, in mammals this has proved problematic as most mammalian tissues do not take up G S H directly, but break it down into its component amino acids (y-glutamic acid - y-glu, cysteine and glycine), transport it across the membrane and resynthesize it intracellular^ (Figure 2.1, Meister, 1995). The first and rate-limiting enzyme in G S H synthesis (y-glutamylcysteine synthetase - G C S ) is inhibited by the presence of G S H (Richman and Meister, 1975). Normal G S H levels in mammalian tissues are sufficient to inhibit the activity of G C S , and consequently increasing intracellular amino acid supply with exogenous G S H does not increase synthesis of intracellular G S H . Intraperitoneal injection of G S H in mammals has little effect on G S H levels in liver, kidney, muscle, lung and lymphoid cells (Puri and Meister, 1983; Wellner et al., 1984; Martensson and Meister, 1989; Martensson et al., 1989). In fish, however, Otto et al. (1997) found daily injections of G S H increased G S H levels after 3 d in the liver, kidney and gill of rainbow trout (Oncorhynchus mykiss) and the liver and kidney of American eel (Anguilla rostrata). The authors speculated that, unlike mammals, fish tissues may be able to directly transport G S H from the plasma. However, the mechanisms by which exogenous G S H increases G S H levels in rainbow trout and American eel tissues, and whether these fish are able to directly transport G S H , have yet to be determined. There are three known pathways by which exogenous G S H could potentially increase intracellular G S H levels in fish tissues (Figure 2.1): Pathway 1) by G S H uptake via complete extracellular breakdown and intracellular synthesis - where existing G S H levels are not high enough to fully inhibit G C S activity, Pathway 2) by G S H uptake via partial G S H breakdown and uptake of y-glutamylcysteine, which then enters G S H synthesis after the inhibited G C S step, and Pathway 3) by direct transport of G S H across the cell membrane (Forman et al., 1995; Anderson and Meister, 1983). No import transporters have been identified in vertebrates as yet, although some members of the mammalian multidrug resistance-associated protein (MRP) and organic anion transport protein (Oatp) families are found to have G S H transporting capabilities (reviewed by Ballatori et al., 2005). The majority of these are export only, and while some are bi-directional, G S H levels are generally several magnitudes higher in the tissues than plasma, and consequently these proteins function as exporters only (Garciaruiz et al., 1992; Lu et al., 1994; FernandezCheca et al., 1996; Mittur et al., 2002; Ballatori et al., 2005). An A T P -dependent G S H and GSH-S-conjugate transporter has been identified and classified as an M R P 17 Extracellular ,y-glu-a.a.T^ (2) cys-gly anspo > O 00 yGT dipeptidase s - • G S H * Intracellular Figure 2.1: The transport of glutathione (GSH) across the cellular membrane. In most mammalian cells, G S H is first broken down into its component amino acids by y-glutamyltranspeptidase (yGT) which removes y-glutamate (y-glu) by addition of an amino acid (a.a.), then dipeptidase which separates cysteine (cys) and glycine (gly). The a.a. and dipeptides are transported across the membrane via a.a. transporters. The y-glu-a.a. complex is transformed to y-glu, using one ATP . Cys is added to y-glu via y-glutamylcysteine synthetase (GCS), and then gly is added via glutathione synthetase (GS), forming G S H . Each synthesis step requires one ATP . Excess G S H binds to G C S and prevents further G S H synthesis. (Modified from Forman et al., 1995). Exogenous G S H may enter a fish cell by: Pathway 1) a lack of G C S inhibition due to low G S H levels; Pathway 2) by the addition of cys to y-glu in the extracellular space, thereby skipping the blocked G C S synthesis step; or Pathway 3) by direct transport via a membrane transporter. Methods of transport can be determined by inhibiting yGT with acivicin and y G C S with buthionine sulfoximine (BSO). 18 homologue in the skate liver (Raja erinacea, Rebbeor et al., 2000), although its function as a G S H importer has not been examined. The pathways by which fish tissues take up G S H may not be universal among tissues. In mammals, G S H has tissue-specific roles and consequently tissue-specific metabolism and transport. The liver has the highest synthesis rate and levels of G S H in mammals, likely due to its role in glutathione-S-transferase-associated detoxification in this tissue. As well, the liver exports G S H to the plasma for transport to other tissues (Anderson et al., 1980; Lauterburg et al., 1984). The kidney contains the highest quantity of the extracellular G S H breakdown enzyme y-glutamyltranspeptidase (yGT) and removes 67-80% of plasma G S H (Puri and Meister, 1983; Anderson et al., 1980; Scott et al., 1993), potentially due to detoxification functions within the kidney. Normal G S H levels inhibit G C S activity, and prevent resynthesis of the majority of G S H precursors taken up by the kidney, although minor quantities of G S H enter by partial breakdown and resynthesis (Pathway 2; Anderson et al., 1980). Direct transport of plasma G S H has been identified in a few mammalian tissues that are exposed to toxicants or irradiation (lung and retinal cells; Hagen and Jones, 1989; Favilli et al., 1997; Kannan et al., 1999). As well, the intestine can directly import lumen G S H originating from the diet or bile, but not from the plasma (Hagen and Jones, 1989; Vincenzini et al., 1992; Favilli et al., 1997). In the brain, G S H has important roles as a neuroprotector, antioxidant and possibly as a neurotransmitter, and may be transported directly from the plasma (Kannan et al., 2000). Clearly, large differences exist in tissue-specific transport of G S H in mammals. Little work has been done on the transport and metabolism of G S H within fish, although fish are noted as having lower levels and turnover of G S H , and lower activity of associated enzymes than mammals (Wallace, 1989; Gallagher et al., 1992). By determining how G S H is transported in fish tissues, it may be possible to identify unique features of G S H metabolism within fish, gain insights into specific roles of G S H within fish tissues, and obtain an understanding of how G S H is metabolized on a whole animal level. I investigated G S H transport in various' rainbow trout tissues to address the following 2 objectives: Objective 2.1: To determine the potential offish tissues to directly take up G S H across the cell membrane by determining if uptake was independent of synthesis in various tissues. Objective 2.2: To determine if G S H uptake in fish tissues was by Pathway 1: extracellular breakdown and intracellular resynthesis (dependent on both G S H breakdown and synthesis), Pathway 2: partial extracellular breakdown and intracellular synthesis (dependent on yGT breakdown and independent of synthesis) or Pathway 3: direct transport (independent of both breakdown and synthesis, Figure 2.1). Materials and Methods G S H synthesis was prevented by inhibition of G C S using the specific inhibitor buthionine sulfoximine (BSO; Griffith and Meister, 1979ba), and G S H breakdown was prevented by inhibition of yGT activity using acivicin - a chemical that binds to the y-glu binding site of yGT and other enzymes (Stole et al., 1994). The above objectives were addressed in two main experiments: Experiment 2.1: To determine if G S H uptake is independent of G S H synthesis, rainbow trout were injected once a day for 19 three days with combinations of G S H and BSO. Experiment 2.2: To determine if G S H uptake is independent of G S H breakdown and synthesis, rainbow trout were injected with combinations of G S H , BSO, and acivicin. The effects of G S H and blocker injection on the total G S H pool (tGSH = reduced G S H + 2*oxidized GSH) were measured in a variety of tissues, as well as oxidized G S H (GSSG) levels and yGT activity in selected tissues. Fish were held in dechlorinated Vancouver municipal water at approximately 8.5°C. All experiments took place in the spring (March-April). Injection Protocol All injection mixtures were made in physiological saline (100mM NaCl , 2.5mM KCI, 1.5mM CaCI 2 , 1 .OmM MgCI 2 , 0.5mM N a H 2 P 0 4 , 5mM N a H C 0 3 , pH 6.8). To inject for the following experiments, fish were removed from the tanks and lightly anaesthetized with 50mg/L tricaine methanesulfonate and 100mg/L sodium bicarbonate. Fish were then weighed and injected intraperitoneally with 10(iL of injection mixture per g off ish. Fish were allowed to recover in a 100L aerated holding tank and then returned to their respective tanks. Experiment 2.1: Is GSH uptake independent of synthesis? Rainbow trout (32.9±7.4g) were obtained from Spring Valley Trout Hatchery (Mission, BC , Canada), and divided into four 50L tanks, with one tank per injection treatment. Fish were fed to satiation with a commercially available trout diet (Skretting, Vancouver, B C , Canada) once a day until the start of the experiment. To determine if rainbow trout could take up G S H independent of synthesis, fish were injected once a day for three days with 0.8mmol GSH/kg fish, 1.8mmol BSO/kg fish (GSH synthesis blocker), G S H and B S O , or saline as control as per Otto et al. (1997). 6 h after the final injection, eight fish per treatment were sacrificed by an overdose of anesthetic (2g/L tricaine methanesulfonate, 4g/L sodium bicarbonate, for this and all future experiments). Blood was taken by caudal puncture, centrifuged at 6800*g for 3 min and plasma removed and frozen in liquid nitrogen. Liver and posterior kidney tissues were then excised, rinsed in ice-cold phosphate-buffered saline (PBS), blotted dry and frozen on liquid nitrogen. Samples were stored at -80°C and analyzed within one month for tissue protein and tGSH, and plasma tGSH and Cortisol. In addition, a subset of tissue was analyzed for G S S G . Experiment 2.2: Is GSH uptake independent of breakdown and synthesis? In order to determine if extracellular G S H breakdown could be a factor in uptake of G S H in fish tissues, it was first necessary to determine the presence of yGT activity in various fish tissues, as well as the ability of acivicin to inhibit yGT activity in fish. In addition, a protocol of injecting fish with a single injection of G S H was established. To determine the presence of yGT activity in a variety of rainbow trout tissues, eight 250g rainbow trout were sacrificed by an overdose of anesthetic. Blood was taken by caudal puncture and the following tissues excised, rinsed in ice-cold P B S and blotted dry: liver, heart, posterior and anterior kidney, spleen, intestine, pyloric caeca, white muscle, brain, and gills. All tissues were frozen in liquid nitrogen and stored at -80°C until analysis for yGT. Prior to tissue preparation, the intestine was thawed and intestinal mucosa scraped off musculature. To determine if acivicin blocks yGT activity in rainbow trout in vivo, 25g rainbow trout were divided into two 70L tanks and injected with either O.lmmol acivicin/kg fish or saline as control, as modified from Lantum et al. (2004). Four fish per injection treatment were then sacrificed at each of the 20 following time points: 20 min and 1, 2, 4, 8, 24 and 48 h post-injection. Fish were killed by an overdose of anaesthetic and posterior kidney removed, rinsed in ice cold P B S and frozen on liquid nitrogen for yGT activity analysis. To avoid potential complications of injecting fish daily over three days, it was determined if a single injection of G S H would increase tissue G S H levels. 25g rainbow trout were injected with 0.8mmol GSH/kg fish, and four fish were sacrificed at each of the following time points: pre-injection and 1, 2, 4, 8, and 24 h post-GSH injection. Liver and posterior kidney were sampled as tissues previously demonstrating G S H uptake. In addition, white muscle and intestine were sampled as the ability to transport G S H was unknown in both tissues, but some yGT activity was detected. Liver and kidney were excised from fish at all time points and muscle and intestine at pre-injection, and 4, 8 and 24 h post-injection. All tissues were rinsed in P B S , frozen in liquid nitrogen and stored at -80°C until G S H analysis. To address if G S H uptake in rainbow trout tissues was independent of extracellular breakdown and synthesis, the following experiment was performed. Rainbow trout (25.9±5.1g) were obtained from Richard Henley farm (Langley, BC , Canada) and divided into eight 50L tanks, 24 fish per tank, 1 tank per injection treatment. Fish were fed once a day as above until the start of the experiment. Fish were divided into four blocker injection groups (two tanks per group) and injected with one of four enzyme blocker treatments: 4mmol BSO/kg fish, 0.1 mmol acivicin/kg fish, B S O and acivicin, or saline as a control. 24 h later one tank of each blocker group was injected with 0.8mmol GSH/kg fish and the other with saline as a control. Fish originally injected with BSO were given an additional injection of B S O . Four and 8 h after G S H injection, fish were sacrificed. Blood was taken by caudal puncture, centrifuged at 6800*g for 3 min and plasma removed and frozen in liquid nitrogen. Liver, posterior kidney, brain and gill tissues were excised, rinsed in ice-cold P B S , blotted dry and frozen in liquid nitrogen. Samples were stored at -80°C and analyzed for tissue total protein, t G S H , kidney yGT, plasma tGSH, and a subset of liver, kidney and brain tissues were analyzed for G S S G . Tissue Preparation and Analysis Glutathione reductase was purchased from Roche Diagnostics (Laval, Q C , Canada), and tricaine methanesulfonate was purchased from Syndel, Canada (Vancouver, BC , Canada). All other chemicals were purchased from Sigma-Aldrich Company (Oakville, ON, Canada). Tissue samples were homogenized (muscle tissue) or sonicated (all other tissues) in 125mM sodium phosphate buffer (pH 7.5) with 6mM ethylenediaminetetraacetic acid (EDTA) (tGSH, G S S G analyses), or 100mM Tris-HCI (pH 8.0) with 1mM EDTA, 1pM pepstatin, 1pM leupeptin, 0.15uM aprotinin, and 0.5mM phenylmethylsulfonyl fluoride protease inhibitors (yGT analysis). Tissues were sonicated at a ratio of 0.1g/mL and supernatant obtained by centrifugation at 11 600*g at 4°C for 5 min. A portion of supernatant was removed for protein and yGT analysis. For tGSH and G S S G analysis, an equal volume of supernatant or plasma was added to 10% sulfosalicylic acid. The sulfosalicylic acid mix was then centrifuged at 10 000*g at 4°C for 10 min and the supernatant retained. Preliminary studies found no effect on tGSH or G S S G values of sonicating tissues in sodium phosphate buffer before addition of sulfosalicylic acid, compared with sonicating in sulfosalicylic acid alone. 21 All analyses were performed in triplicate on 96-well microplates and measured using a SpectraMax spectrophotometer equipped with SoftmaxPro Software (Molecular Devices Corporation). Protein content was analyzed using the bicinchoninic acid method as per Smith et al. (1985), using bovine serum albumin as standards and analyzed at 516nm. tGSH and G S S G analyses were modified from Griffith (1980) as follows. For tGSH analyses, 6|^L of triethanolamine was added per 10O^L of tissue or cell supernatant and G S S G standards. 10p.L of samples or G S S G standards and 200(iL of reaction mixture were added to a 96 well plate. The reaction mixture contained 0.22mM reduced B-nicotinamide adenine dinucleotide phosphate, 0.62mM 5,5'-dithiobis(2-nitrobenzoic acid), 0.56% triethanolamine and 0.5U7mL glutathione reductase in 125mM sodium phosphate, 6mM EDTA buffer, pH 7.4. The change in absorbance was monitored at 412nm for 5 min. For G S S G analyses, 6pL of 2-vinylpyridine (to bind reduced GSH) and 20pL of 0.1% triethanolamine in P B S was added per 100pL of sample or G S S G standard. Samples and standards were mixed vigorously for 1 min and then incubated for 50 min. G S S G analyses were then carried out as per tGSH above, except dithiobis(2-nitrobenzoic acid) concentration in the reaction mixture was 1pM. G S S G data are presented as % of tGSH. yGT activity analysis was modified from Meister e ta l . (1981). In brief, 10pL of sample or Tris-HCI blank was added to a 96-well plate, followed by 200pL of a reaction mixture containing 25mM glycyl-glycine and 1.25mM L-y-glutamic acid-p-nitroanilide. The change in absorbance was monitored at 410nm for 5 min. The activity of yGT (pmol p-nitroaniline formed/min/mg protein) was determined using an extinction coefficient calculated for p-nitroaniline. Statistical Analyses The effects of blocker and G S H injection were analyzed using 1-way or 2-way A N O V A as appropriate, using SigmaStat ( S P S S Inc.). Differences between individual factors were analyzed using Tukey's post-hoc comparison test. If normality or equal variance failed, data were transformed and reanalyzed. If transformation failed to normalize data or bring about equal variance, data were analyzed by Kruskal-Wallis non-parametric A N O V A on ranks, followed by Dunn's post-hoc comparison test. Differences were considered significant if p<0.05. All data are presented as mean ± standard error of the mean. Results Experiment 2.1: Is GSH uptake independent of synthesis? tGSH levels in rainbow trout injected with combinations of G S H and B S O once a day for three days are reported in Figure 2.2. In the liver there was no interaction between G S H and B S O injections on tGSH levels (p=0.313). G S H injection increased tGSH levels above those of non-GSH-injected fish (p<0.001), and B S O injection decreased tGSH levels below those of non-BSO-injected fish (p<0.001). In the kidney, although G S H injection increased tGSH levels overall (p<0.001) and B S O injection decreased tGSH levels overall (p<0.001), there was a significant interaction between G S H and B S O injection on tGSH levels (p=0.014). Concurrent injection with BSO diminished the increase in tGSH levels with G S H injection to 14% of G S H injection alone. Injection of G S H without BSO resulted in a greater increase in 22 X o A 3 3 0 o cx e ^ 2 0 1 0 1 1 . 0 b 9.5 I Pre- Saline BSO injection 4 0 3 0 2 0 1 0 B 1 8 . 7 I I I Saline f ^ ^ l GSH 2 . 6 Pre- Saline BSO injection Figure 2.2: Liver (A) and posterior kidney (B) total G S H levels in rainbow trout injected daily for three days with combinations of saline, G S H , and a G S H synthesis blocker (BSO). Tissue tGSH levels were measured 6 h after last injection. Significant differences among injection groups, within tissues, are indicated by differing letters (a,b,c,d). Numbers above bars are mean differences in G S H levels between G S H - and non-GSH-injected fish within blocker group. n=8. 23 tGSH levels in the kidney than in the liver (18.7 and 11.0 nmol/mg protein respectively). Plasma tGSH levels are given in Table 2.1. G S H injection greatly increased plasma tGSH levels above pre-injection and non-GSH-injected levels (p<0.001), regardless of BSO injection. A subset (n=4) of liver and kidney tissues were measured for G S S G (as percent of tGSH, Table 2.1). There was no difference in G S S G among injection groups in either the liver (p=0.327) or kidney (p=0.089). Fish injected with combinations of G S H and BSO had higher plasma Cortisol levels after 3 d injection than saline-injected fish (Table 2.1), although this was only significant in BSO-injected fish (p=0.014). The plasma C o r t i s o l levels of B S O - and GSH-injected fish did not differ from pre-stress levels and were within the range reported as unstressed levels in salmonids (Barton and Iwama, 1991), indicating G S H and B S O injection did not cause significant stress in the fish. Experiment 2.2: Preliminary experiments yGT activity was detectable in the posterior kidney, intestine, pyloric caeca and, to some extent, the brain and white muscle of rainbow trout (Figure 2.3). The highest activity was measured in the posterior kidney, where activity levels were over 10 times that of the pyloric caeca, and over 15 times that of the intestine. No yGT activity was detectable in the blood, heart, liver, anterior kidney, spleen or gills of rainbow trout. A single injection of O.lmmol acivicin/kg fish decreased detectable yGT activity in rainbow trout posterior kidney at every measured time point (p<0.001, Figure 2.4). yGT activity decreased to 6% of saline-injected fish 20 min after injection. By 8 h, yGT activity was less than 1% of saline-injected fish and remained at this level for the remaining sampling times (up to 48 h). A single injection of 0.8mmol GSH/kg fish increased tGSH significantly at 8 h post-injection in the liver (p=0.005), and 1, 2, 4 and 8 h post-injection in kidney (p<0.001, Figure 2.5). However, there was no effect of G S H injection on muscle (p=0.487) or intestine mucosa (p=0.749) t G S H . Consequently, muscle and intestine were not sampled in Experiment 2.2. Experiment 2.2: Is GSH uptake independent of breakdown and synthesis? Liver tGSH levels in fish injected with combinations of G S H and G S H synthesis (BSO) and breakdown (acivicin) blockers are given in Figure 2.6. At 4 h post-GSH injection, there was a significant interaction between G S H and blocker injection on tGSH levels (p=0.03). G S H injection had no effect on tGSH levels compared with non-GSH-injected fish in saline and B S O injection groups, but resulted in significant increases in tGSH levels in acivicin and BSO/acivicin injection groups. At 8 h post-GSH injection, there was no significant interaction between G S H and blocker injection on tGSH levels in the liver (p=0.30). Overall, G S H injection increased tGSH levels in the liver above non-GSH-injected groups (p<0.001). However, this was only significant in saline- and BSO/acivicin-pre-injected fish. The rise in tGSH levels with G S H injection above those of non-GSH-injected fish was 2 times higher in saline- than acivicin- or BSO/acivicin-injected fish, and 3 times higher in saline- than BSO-injected fish. Injection with BSO/acivicin decreased tGSH levels in the liver at 4 and 8 h, and B S O injection alone decreased tGSH levels at 8 h compared to saline-injected fish. Posterior kidney tGSH levels in fish injected with combinations of G S H and G S H synthesis and breakdown blockers are given in Figure 2.7. At 4 h post-GSH injection, there was a significant interaction between G S H and blocker injection on kidney tGSH levels (p<0.001). G S H injection significantly 24 Table 2.1: Plasma total G S H (uM, n=8), plasma Cortisol (ng/mL, n=8), as well as oxidized glutathione ( G S S G as % of tGSH) in liver and posterior kidney (n=4) from rainbow trout injected with combinations of G S H and a G S H synthesis blocker (BSO) daily for 3 days. Fish were sampled 6 h after final G S H / B S O injection. Significant differences among injection groups are indicated by differing letters (a,b). No significant differences were observed in liver or kidney G S S G . Injection Plasma tGSH Plasma Cortisol Liver % G S S G Kidney % G S S G Pre-injection 23.4±1.6 a 7.6±2.1 a 2.5±1.1 1.2±0.1 Saline 23.0±0.9 a 2.2±0.4 b 1.2±0.2 0.9±0.2 G S H 1801.0±139.7 b 5.4±0.8 a b 0.9±0.3 0.7±0.2 B S O 14.8±1.0 a 6.5±0.8 a 1.4±0.3 1.1±0.1 G S H / B S O 1265.1±86.1 b 6.5±1.9 a b 1.1±0.3 0.6+0.2 a o l-H OH 14 H I 12 ~ 10 3, o cd H a V o o > T 3 i -O D . O (D 'EH 1 ^ 1 G ii o o o DH o 1/3 13 C 33 OX) Figure 2.3: y-Glutamyltranspeptidase (yGT) activity in various tissues of 250g rainbow trout. n=8. 25 g 0 10 20 30 40 50 time (hours from acivicin injection) Figure 2.4: Y-Glutamyltranspeptidase (yGT) activity in posterior kidney of rainbow trout injected with saline or a yGT activity blocker (0.1 mM acivicin/kg fish). n=4. 50 * 40 H o >-< s 6 m O 3 30 20 H 10 - • — Liver -O— Kidney - T — - Muscle Intestine 5 10 15 time (hours from GSH injection) 20 25 Figure 2.5: Tissue total G S H levels in rainbow trout after a single injection with 0.8mmol GSH/kg fish, indicates significant difference from pre-injection, within tissue. n=4. 26 30 25 A 4 h I I Saline GSH c +-< o s 'o s o > 20 H 15 10 0.5 a a 8 h 25 20 15 10 H 8.1 a X 4.9 1.0 ab ab 4.5 2.5 ae ce cd ab X 2.4 3.6 Saline BSO Acivicin B/A Blocker injection Figure 2.6: Liver total G S H levels in rainbow trout injected with combinations of enzyme blockers: saline, a G S H synthesis blocker (BSO), a G S H breakdown blocker (acivicin), or BSO and acivicin (B/A). Fish were then injected with saline or G S H and tissue tGSH levels measured 4 h and 8 h after G S H injection. Significant differences among injection groups, within time, are indicated by differing letters (a,b,c,d,e). Numbers above bars are mean differences in G S H levels between G S H and non-GSH-injected fish within blocker group. n=12. 27 '53 H—> o &, bo s "o s X 0 0 o c T3 40 30 A 20 10 30 -\ 20 10 4 h 14.5 b X 15.4 f de 1.0 c _=r=_ 3.8 ad 8 h 17.2 c X 12.1 cd ab cde 1.3 b 2.2 ade be n Saline BSO Acivicin Blocker injection B/A Figure 2.7: Posterior kidney total G S H levels in rainbow trout injected with combinations of enzyme blockers: saline, a G S H synthesis blocker (BSO), a G S H breakdown blocker (acivicin), or B S O and acivicin (B/A). Fish were then injected with saline or G S H and tissue tGSH levels measured 4 h and 8 h after G S H injection. Significant differences among injection groups, within time, are indicated by differing letters (a,b,c,d,e,f). Numbers above bars are mean differences in G S H levels between G S H - and non-GSH-injected fish within blocker group. n=12. 28 increased tGSH levels in all groups except BSO-pre-injected fish. G S H injection in saline-pre-injected fish was 3 times that of non-GSH-injected fish and the magnitude of increase was almost 2 times that of the maximum increase in liver tGSH. Fish pre-injected with BSO/acivicin had a greatly diminished increase in tGSH with G S H injection to approximately 25% of saline- and acivicin-pre-injected fish. The pattern of tGSH levels in the posterior kidney at 8 h was similar to 4 h post-GSH injection. There was a significant interaction between G S H and blocker injection (p<0.001) on tGSH levels at 8 h. G S H injection greatly increased tGSH levels in saline- and acivicin-pre-injected fish but not in B S O - or BSO/acivicin-pre-injected fish. B S O injection alone decreased tGSH levels and acivicin injection alone increased tGSH levels compared with saline-injected fish at both 4 and 8 h. Gill tGSH levels in fish injected with combinations of G S H and G S H synthesis and breakdown blockers are given in Figure 2.8. At 4 h post-GSH injection, there was no significant interaction between G S H and blocker injection on tGSH levels (p=0.17), but there were significant factor effects for both G S H (p<0.001) and blocker injection (p<0.001). G S H injection increased tGSH levels in all blocker groups, although this was not significant in BSO-pre-injected fish (p=0.07). G S H injection resulted in a similar magnitude of tGSH increase in saline-, acivicin- and BSO/acivicin-injected fish, and this increase was approximately 3 times greater than that of BSO-pre-injected fish. At 8 h post-GSH injection, there was a significant interaction between blocker and G S H injection (p=0.004), but few individual differences. G S H injection resulted in minor, but significant increases in tGSH levels in saline- and acivicin-pre-injected fish only. Both acivicin and BSO/acivicin injection alone increased gill tGSH levels at 4 h, but there were no effects of blocker injection on gill tGSH levels at 8 h. Brain tGSH levels in fish injected with combinations of G S H and G S H synthesis and breakdown blockers are given in Figure 2.9. At both 4 h and 8 h post-GSH injection, there was a significant interaction between G S H and blocker injection (p=0.04, p<0.001 respectively), but few individual differences. At 4 h, brain tGSH levels increased slightly, but significantly with G S H injection in acivicin-pre-injected fish only. At 8 h, G S H injection did not significantly increase tGSH levels in the brain in any blocker group. There were no effects of blocker injection on brain tGSH levels at 4 h, but injection with B S O or BSO/acivicin alone decreased tGSH levels compared with saline-injected fish at 8 h. G S H injection greatly increased tGSH levels in the plasma at 4 h (p<0.001) and 8 h (p<0.001), although to a lesser extent at 8 h post-GSH injection (Figure 2.10). There were no significant effects of blocker injection on plasma tGSH levels at 4 or 8 h. A subsample of liver, kidney and brain tissue was analyzed for G S S G (as % of tGSH, Table 2.2). There were no effects of injection on % G S S G in liver (p=0.155) or brain (p=0.525), although there was a significant effect of injection on % G S S G in the posterior kidney (p=0.004). However, only fish injected with GSH/acivicin and those injected with BSO/acivicin differed in % G S S G . G S H injection did not affect % G S S G in any blocker group or tissue. Acivicin injection decreased yGT activity in the posterior kidney by greater than 99% in all groups and times (p<0.001, Figure 2.11). G S H injection alone decreased yGT activity at 4 h only. There were no other effects of G S H or BSO injection on yGT activity within the kidney. Although tissues examined were rinsed in P B S before tGSH analysis, tGSH concentrations of any residual plasma within the tissues would result in underestimation of tissue tGSH where plasma 29 30 n 25 20 15 H 10 5 H 0 ' S -4-» o "o a X 25 0 0 O 3 20 15 10 4h 8h 5.3 be 2.7 cd ab 2.0 -0.6 ab b 6.2 2.3 ad 7.0 b 0.8 ab ad Saline BSO Acivicin Blocker injection B/A Figure 2.8: Gill total G S H levels in rainbow trout injected with combinations of enzyme blockers: saline, a G S H synthesis blocker (BSO), a G S H breakdown blocker (acivicin), or BSO and acivicin (B/A). Fish were then injected with saline or G S H and tissue tGSH levels measured 4 h and 8 h after G S H injection. Significant differences among injection groups, within time, are indicated by differing letters (a,b,c,d). Numbers above bars are mean differences in G S H levels between G S H - and non-GSH-injected fish within blocker group. n=12. 30 20 15 10 4 h "S 5 OH (DO s 1 0 OO 2 15 g '3 10 1.8 abc 8 h 0.1 a abc I I Saline I ^ GSH -0.4 2.8 ab J L ab 2.4 2.0 ab Saline BSO Acivicin Blocker injection 2.2 abc -0.3 c c B/A Figure 2.9: Brain total G S H levels in rainbow trout injected with combinations of enzyme blockers: saline, a G S H synthesis blocker (BSO), a G S H breakdown blocker (acivicin), or B S O and acivicin (B/A). Fish were then injected with saline or G S H and tissue tGSH levels measured 4 h and 8 h after G S H injection. Significant differences among injection groups, within time, are indicated by differing letters (a,b,c). Numbers above bars are mean differences in G S H levels between G S H - and non-GSH-injected fish within blocker group. n=12. 31 in O 1600 1400 1200 1000 800 600 400 200 40 A 20 0 4 h 1 1400 ^ 1200 1000 H 800 600 400 200 ^ 40 20 H 0 8 h 961.8 c X ab n 444.1 cd ab n 1185.1 478.3 be I abd 832.3 c X 180.9 be a n abd Saline BSO Acivicin Blocker injection 741.4 cd X 611.7 ab B/A Figure 2.10: Plasma total G S H levels in rainbow trout injected with combinations of enzyme blockers: saline, a G S H synthesis blocker (BSO), a G S H breakdown blocker (acivicin), or B S O and acivicin (B/A). Fish were then injected with saline or G S H and plasma tGSH levels measured 4 h and 8 h after G S H injection. Significant differences among injection groups, within time, are indicated by differing letters (a,b,c,d). Numbers above bars are mean differences in G S H levels between G S H - and non-GSH-injected fish within blocker group. n=12. 32 Table 2.2: Percent oxidized glutathione ( G S S G as % of total GSH) in liver (n=4), brain (n=4) and posterior kidney (n=8) from rainbow trout injected with combinations of G S H , a G S H synthesis blocker (BSO) and a G S H breakdown blocker (acivicin). Tissues were sampled 4 h (kidney) and 8 h (liver and brain) after G S H injection. Significant differences among injection groups within the kidney are indicated by differing letters (a,b). There were no significant differences in liver or brain G S S G among injection groups. Injection Liver Brain Injection (kidney) Kidney Saline 4.2±0.9 2.3±1.5 Saline 6.7±1.9 a D G S H 3.2±0.7 4.1+2.1 G S H 6.3±1.3 a b B S O 5.9±1.5 2.6±0.9 Acivicin 11.6±1.0 a b G S H / B S O 5.5±1.2 4.0±0.9 GSH/acivicin 4.9±1.4 a Acivicin 3.8±0.7 3.3±1.6 BSO/acivicin 13.1±1.4b GSH/acivicin 3.5+0.1 2.1±0.6 GSH/BSO/aciv ic in 10.1±2.4 a b 33 c '53 -4—» o 1-1 p . s o B o ca H O c H ^ en O OH 14.0 A 12.0 H IO.O A B 8.0 S O . 0.2, 0.1 0.0 14.0 12.0 H 10.0 8.0 A / 4 h c JL, c c JL J J J 1 8h I 0.2 0.1 0.0 b b J L J _ l Saline BSO Acivicin Blocker injection B/A Figure 2.11: y-Glutamyltranspeptidase (yGT) activity in posterior kidney of rainbow trout injected with combinations of enzyme blockers: saline, a G S H synthesis blocker (BSO), a G S H breakdown blocker (acivicin), or BSO and acivicin (B/A). Fish were then injected with saline or G S H and kidney yGT activity measured 4 h and 8 h after G S H injection. Significant differences among injection groups, within time, are indicated by differing letters (a,b,c). n=12. 34 levels were less than tissue levels, and overestimate tissue levels where plasma levels were greater than tissue levels. This could artificially increase the difference in tissue tGSH between saline- and G S H -injected fish, and result in a false significant uptake of G S H in different groups. To verify that this was not a factor in G S H uptake, intracellular concentrations of tGSH (tGSH,) were estimated by accounting for plasma concentration in combination with percent extracellular volume (ECV) using the following equation: tGSH; = {measured tissue tGSH - (ECV * measured plasma tGSH)}/ (1- ECV) where all tGSH levels were in mM (assuming a tissue density of 1g/ml_) and E C V used were from Bushnell et al. (1998) (28% in liver, 72% in posterior kidney and 11% in brain). The weight of the gill cartilage prevented calculation of tGSH levels as mM, and consequently this tissue was excluded from the above analysis. tGSHi levels are given in Table 2.3 (Experiment 2.1) and Table 2.4 (Experiment 2.2). The relationships between G S H injection and tGSHi levels were similar to original tGSH calculations (Figure 2.2, 2.6, 2.7 and 2.8) in all cases except kidney tGSH levels in Experiment 2.1. In this experiment, tGSH, levels did not significantly increase with G S H injection, and resulted in negative values in GSH/BSO-injected fish (Table 2.3). This may be due to an overestimation of % E C V in the kidney (as suggested by Bushnell et al., 1998), or to a lack of plasma remaining in the kidney after P B S rinse. As tGSH, levels did not greatly differ from kidney tGSH in Experiment 2.2, the original tGSH levels in posterior kidney are likely accurate. In the liver, minor significant increases in tGSH with G S H injection in acivicin and BSO/acivicin injection groups (Experiment 2.2) were not significant in tGSHi, indicating G S H uptake in this experiment was dependent on G S H breakdown and resynthesis. In addition, minor significant increases in tGSH with G S H injection in the BSO/acivicin group in kidney, and the acivicin group in brain (Experiment 2.2) were not significant with tGSHi. However, large significant increases with G S H injection remained so with tGSH, in both the liver and kidney. Plasma to tissue GSH concentration gradient Under normal physiological conditions, tissue G S H concentrations (average of saline-injected fish from Experiments 2.1 and 2.2: 1.12mM in liver, 0.58mM in kidney) are 25-50 times that of plasma (0.02mM). However, G S H injection resulted in similar or higher plasma tGSH levels (Table 2.1, Experiment 2.1; Figure 2.10, Experiment 2.2) than average tissue levels in several groups. To determine if apparent tissue to plasma concentration gradients were a factor in G S H tissue uptake, the effects of G S H injection on the concentration gradients were examined. The relationship between initial and final tissue to plasma tGSH concentration gradients as set up by G S H injection for Experiments 2.1 and 2.2 are given in Figure 2.12 (liver) and Figure 2.13 (posterior kidney). Initial G S H concentration gradients were estimated by the ratio of tissue tGSH of saline-injected fish to plasma tGSH of GSH-injected fish, and final concentration gradients were determined from the ratio of tissue G S H of GSH-injected fish to plasma tGSH levels of GSH-injected fish. Initial tissue to plasma tGSH concentration gradients less than 1 indicate that the concentration gradient could promote the import of G S H to tissues from plasma, and gradients greater than 1 could indicate the promotion of an outward movement of G S H from tissues to plasma (dotted line Figures 2.12 and 2.13). If tissue G S H levels were influenced by these concentration gradients, final tGSH concentration gradients between plasma and tissues should be equilibrated (dashed line Figures 2.12 and 2.13). 35 Table 2.3: Estimation of intracellular total G S H (tGSHj in mM) in rainbow trout after removing the plasma component for Experiment 2.1. Significant differences among injection groups within tissue are indicated by differing letters (a,b,c,d,e). n=8. Injection Liver Kidney Pre-injection 1.69±0.11 a 2.14±0.13 a b Saline 1.72±0.05a 2.97±0.21 a c d G S H 2.37±0.13 b 3.52±0.51 c BSO 0.88±0.07 c 1.19±0.11 b G S H / B S O 1.65±0.18a -1.24±0.33 e Table 2.4: Estimation of intracellular total G S H (tGSHj in mM) in rainbow trout after removing the plasma component for Experiment 2.2. Significant differences among injection groups within tissue are indicated by differing letters (a,b,c,d,e). n=12. Time/injection protocol Liver Kidney Brain 4 h Saline 1.73±0.15a D 2.82±0.22 a 1.12±0.12 a b G S H 1.53±0.15b 5.79±0.54 b c 1.23±0.16 a b B S O 1.38±0.09b 0.88±0.07 d 1.56±0.15a G S H / B S O 1.35±0.11 b 0.25±0.09 d 1.22±0.18 a b Acivicin 1.76±0.15a b 4.56±0.12 b e 1.05±0.11 a b GSH/Acivicin 2.13±0.17 a 7.07±0.52° 1.14±0.13 a b BSO/Acivicin 0.66±0.05 c 3.55±0.18 a e 0.78±0.08 b GSH/BSO/Aciv ic in 0.78±0.07 c 3.06±0.23 a 1.15±0.14 a b 8 h Saline 2.56±0.22 a 2.62±0.16 a 1.62±0.13 a b G S H 3.75±0.29b 6.12±0.30 b 1.26±0.09 b c d B S O 1.26±0.12 c d 0.66±0.17 c 1.27±0.08 b c d G S H / B S O 1.61±0.11 c 0.72±0.12 c 1.53±0.11 a b Acivicin 2.30±0.11 a 5.34±0.20 b 1.80±0. 1 2 a c GSH/Acivicin 2.73±0.21 a 10.50±0.65 d 2.09±0.11 c BSO/Acivicin 0.84±0.05 d 2.38±0.05 a 1.01±0.05d GSH/BSO/Aciv ic in 1.21±0.16 c d 1.69±0.14 a c 1.02±0.10 b c d 36 7 n T3 <D C J <u w CO o o GO O <u > « o GO O o GO a 4 H 3 H C/3 C import export • Experiment: 0 2.-A • 2-B 4h A 2-B 8h Blocker injection: O Saline O BSO O Acivicin • BSO/Acivicin —C-rf^- — i -• fiimLcfiacerrJralLorijraalJ enl e_qu aj 1 initial tGSH gradient liver GSH of saline injected fish plasma GSH of GSH injected fish Figure 2.12: Initial and final liver to plasma total G S H concentration gradients as set up by G S H injection in rainbow trout injected concurrently with saline or B S O (Experiment 2.1) or with saline or combinations of B S O and acivicin (Experiment 2.2). Dotted line indicates boundary between initial gradient of G S H import from plasma to liver and G S H export from liver to plasma. Dashed line indicates final concentration equal between plasma and liver. i - H » 4 H I GO O x GO O > T3 X GO a o X GO a O •£ -i-> T3 e so 0 Experiment: O 2-A • 2-B 4h A 2-B 8h Blocker injection: O Saline O B S O © Acivicin • BSO/Acivicin import • O O export A • final concentration gradient equal 0.0 0.2 0.4 0.6 — i 1 1 1—•• 1 1 0.8 1.0 1.2 1.4 1.6 1.8 initial tGSH gradient = kidney tGSH of saline injected fish plasma tGSH of GSH injected fish Figure 2.13: Initial and final posterior kidney to plasma total G S H concentration gradients as set up by G S H injection in rainbow trout injected concurrently with saline pr B S O (Experiment 2.1) or with saline or combinations of B S O and acivicin (Experiment 2.2). Dotted line indicates boundary between initial gradient of G S H import from plasma to kidney and G S H export from kidney to plasma. Dashed line indicates final concentration equal between plasma and kidney. 37 In the liver, plasma tGSH levels resulted in import concentration gradients in saline- and BSO-injected fish from Experiment 2.1, and BSO/acivicin-injected fish at 4 h in Experiment 2.2 (Figure 2.12). In these cases, this resulted in an equal final tissue concentration gradient between tissues and plasma. In contrast, where plasma tGSH levels set up initial import concentration gradients in the kidney, there was no clear pattern of final concentration gradients of G S H (Figure 2.13). Final G S H concentrations were lower in the kidney than the plasma in some groups (BSO-injected fish of Experiment 2.1 and 4 h Experiment 2.2) and higher in other groups (saline-injected fish of Experiment 2.1 and 4 h Experiment 2.2, and acivicin-injected fish at 8 h Experiment 2.2). These data imply liver tGSH levels were influenced by plasma to tissue concentration gradients, whereas posterior kidney tGSH levels were not. Discussion I have examined for the first time the mechanisms by which tissues of rainbow trout take up exogenous G S H from the plasma. The pathways by which rainbow trout tissues take up exogenous G S H , whether intact, or by complete breakdown and resynthesis, were dependent on the tissue, and to some extent the injection protocol and length of time after G S H injection. Evidence of direct transport of exogenous G S H into examined tissues (Pathway 3, Figure 2.1) was only detected in the liver as apparent gradient-driven uptake. When plasma levels of G S H were greater than liver levels, suppression of G S H synthesis did not affect apparent gradient-driven uptake of G S H . Although gradient-driven direct uptake of G S H in the liver of rainbow trout was not unequivocal, several gradient-driven transporters have been identified in mammals. In particular, the sinusoidal membrane of mammalian liver contains M R P and Oatp bi-directional transporters that export G S H from the liver to the plasma under physiological concentrations of plasma and liver G S H (Garciaruiz et al., 1992; Lu et al., 1994; Mittur et al., 2002; Ballatori et al., 2005). Although little work has been done on G S H transporters in fish, a MRP-l ike protein has been identified as a G S H and GSH-S-conjugate transporter in skate liver (Rebbeor et al., 2000), suggesting that functionally similar transporters exist in fish. This indicates the liver of fish likely has a similar role in G S H export to the plasma as in mammals. Gradient-driven uptake of exogenous G S H has not been observed in mammalian studies, likely due to high levels of G S H in the liver (approximately 10mM, Kidd, 1997), which would require a greater increase in plasma levels than possible with injection protocols used thus far. The posterior kidney, and to a lesser extent, the liver and gill of rainbow trout took up exogenous G S H by extracellular breakdown and intracellular resynthesis (Pathway 1, Figure 2.1), as suppression of G S H synthesis prevented or diminished the increase in tissue tGSH levels with exogenous G S H . In the case of the liver, this was only observed when plasma levels were too low to allow for apparent gradient-driven direct uptake. Neither the liver nor gill exhibited yGT activity. Consequently, plasma G S H was likely broken down by yGT in the kidney and transported to these tissues as amino acids or dipeptides. The lack of inhibition of G C S activity by resting levels of G S H in these tissues indicates the capacity for G S H synthesis is greater than can be supplied by normal concentrations of amino acid precursors. These tissues are responsible for neutralizing blood (posterior kidney and liver), food (liver), and water-borne (gill) contaminants. G S H is a cofactor in glutathione-S-transferase-dependent detoxification, and 38 consequently increased ability to synthesize G S H in these tissues may allow for rapid increase in G S H levels during times of contaminant stress. This supports a similar hypothesis by Gallagher et al. (1992), who found liver and gills of channel catfish (Ictalurus punctatus) had high G S H synthesis capacity relative to the tissue levels. In contrast, the brain, muscle and intestine of rainbow trout appeared to have little capacity for G S H uptake either directly or indirectly. The lack of G S H uptake indicates G S H synthesis is in equilibrium with amino acid precursor supply in these tissues, and may reflect the inability of G S H to cross the blood-brain barrier in rainbow trout. No tissue examined took up exogenous G S H by partial extracellular breakdown and intracellular synthesis (Pathway 2, Figure 2.1). Surprisingly, acivicin, the inhibitor of yGT, did not prevent G S H uptake in some tissues. As this is the first and pivotal step in G S H uptake by breakdown and resynthesis (Pathway 1, Figure 2.1), it was expected to prevent G S H uptake in a similar manner as the G S H synthesis inhibitor. Although this was observed in the liver, acivicin did not prevent uptake in the kidney and gill. This could be for a number of reasons. yGT activity was found to be over 99% inhibited by acivicin in the posterior kidney, but the small activity remaining could break down G S H to its component amino acids for resynthesis in the cell. However, a partial decrease in uptake would be expected to account for the greatly inhibited activity of yGT, which was not observed. Another possibility is that acivicin was affecting other systems in the cell that resulted in G S H uptake. Acivicin is not a specific inhibitor of yGT but binds to the y-glu site of many enzymes, and consequently acivicin could indirectly increase G S H uptake but an unknown pathway. This is unlikely in the posterior kidney, as injection of both B S O and acivicin prevented G S H uptake, demonstrating that uptake in the presence of acivicin is reliant on G S H synthesis, and consequently breakdown of G S H . In the gill, BSO/acivicin injection prevented G S H uptake at 8 h, but not 4 h post-injection, suggesting increased uptake by indirect effects of acivicin. A third possibility is that rainbow trout possess an alternate, and as yet unidentified, means of G S H breakdown. Such a system has not been alluded to in other study organisms, and further work is required to confirm the existence of an alternate pathway for G S H degradation. Of the tissues examined, the posterior kidney of rainbow trout was able to take up exogenous G S H to the largest extent (Figures 2.2, 2.7). This is consistent with mammalian studies that found the majority of plasma G S H is broken down and taken up by the kidney, although not necessarily resynthesized as G S H intracellularly (Anderson et al., 1980; Scott et al., 1993). When the supply of amino acid precursors increased in the form of exogenous G S H , kidney G S H levels (2.3mM, Experiment 2.2 4 h) increased to similar levels reported for mammals (average 2.8mM; Puri and Meister, 1983; Griffith and Meister, 1979ab; Leeuwenburgh and J i , 1995; Leeuwenburgh and J i , 1998). In this tissue, low G S H levels relative to mammals may not be a function of low synthesis rates but of low amino acid supply. The time frame of G S H uptake differed between tissues in the rainbow trout (Figures 2.5 through 2.9). G S H injection increased G S H levels in the kidney within 1 h of injection, and these levels remained elevated at 8 h. G S H injection increased tGSH levels in the gill at 4 h only and in the liver at 8 h only. This is consistent with the sequence of GSH-derived amino acid uptake in mammals, where uptake is first and most consistently detected in kidney, and is detected earlier in the lung than in the liver (Hahn et al., 1978). Rapid uptake in the kidney is likely due to the presence of high concentrations of yGT. In the 39 liver, no yGT activity was detected and consequently an adequate supply of amino acids for increased G S H synthesis must be transported from the kidney. However, gill tissue did not contain detectable levels of yGT either. Earlier increase in G S H levels in this tissue may be due to higher blood flow in the gill relative to the liver. In most tissues, G S H or blocker injection did not increase the percent of tGSH that was oxidized. This indicates that increased tGSH levels after G S H injection were due to increased reduced G S H and not the oxidized form. However, in the kidney of Experiment 2.2, % G S S G levels were higher than expected, particularly those off ish injected with acivicin, with or without B S O . This indicates acivicin may have an oxidative effect on the kidney, although alterations in tGSH with different injection groups did not correspond to alterations in G S S G . In particular, G S H injection did not increase % G S S G in any blocker group, indicating that the observed uptake of tGSH was due to the reduced and not oxidized form of G S H . Injection of acivicin alone was expected to decrease G S H levels in tissues that depend on G S H breakdown by yGT for synthesis of intracellular G S H . However, acivicin injection increased tGSH levels in the kidney and transiently increased levels in the gill of rainbow trout. Although the majority of mammalian studies have found the opposite, a few studies have reported acivicin increases G S H levels in the kidney and brain but not the liver (Wolff et al., 1998; Lantum et al., 2004). An adequate explanation for this has not yet been presented. However, this may be due to acivicin binding to the y-glu site of an unidentified enzyme(s), resulting in indirect effects on G S H metabolism. Inhibition of G S H synthesis with BSO decreased tGSH levels in the liver, kidney, and brain but not the gill of rainbow trout. This is likely due to G S H export from the tissues no longer being balanced by G S H synthesis within the cell. After 78 h of BSO exposure the levels of G S H in rainbow trout tissues remained at 50% of saline-injected fish. The decline in G S H is considerably slower than those reported in mammals, where G S H levels dropped to 20% of controls within 4 h of injection with 2mmol BSO/kg animal (Puri and Meister, 1983). This indicates that G S H turnover in the form of G S H export is much slower in fish than mammals, as postulated by Gallagher et al. (1992). Summary • Rainbow trout liver could directly transport exogenous G S H (Pathway 3), potentially by bi-directional physiological export transporters. This transport was not evident in other tissues. • The G S H synthesis capacity of several rainbow trout tissues was greater than endogenous amino acid supply, as indicated by the ability of kidney, liver and gill to take up G S H by extracellular breakdown and intracellular resynthesis (Pathway 1). This may lend flexibility to the G S H system in fish, where levels of G S H may rapidly increase by increased supply of amino acids, or during times of high demand, without increasing synthesis enzymes. 40 References Anderson, M.E., Meister, A., 1983. Transport and direct utilization of Y-glutamylcyst(e)ine for glutathione synthesis. Proceedings of the National Academy of Science of the United States of America 80: 707-711. Anderson, M.E., Bridges, R.J. , Meister, A., 1980. Direct evidence for inter-organ transport of glutathione and that the non-filtration renal mechanism for glutathione utilization involves g-glutamyl transpeptidase. Biochemical and Biophysical Research Communications 96: 848-853. Ballatori, N., Hammond, C.L., Cunningham, J.B., Krance, S.M. , Marchan, R., 2005. Molecular mechanisms of reduced glutathione transport: role of the M R P / C F T R / A B C C and OATP/SLC21A families of membrane proteins. Toxicology and Applied Pharmacology 204: 238-255. Barton, B.A., Iwama, G.K., 1991. Physiological changes in fish from stress in aquaculture with emphasis on the response and effects of corticosteroids. Annual Reviews of Fish Diseases 1: 3-26. Bushnell, P.G. , Conklin, D.J., Duff, D.W., Olson, K.R., 1998. Tissue and whole-body extracellular, red blood cell and albumin spaces in the rainbow trout as a function of time: a reappraisal of the volume of the secondary circulation. Journal of Experimental Biology 201: 1381-1391. Coppola, S., Ghibelli, L , 2000. G S H extrusion and the mitochondrial pathway of apoptotic signalling. Biochemical Society Transactions 28: 56-61. Favilli, F., Marraccini, P., lantomasi, T., Vincenzini, M.T., 1997. Effect of orally administered glutathione on glutathione levels in some organs of rats: role of specific transporters. British Journal of Nutrition 78: 293-300. FernandezCheca, J .C. , Y i , J.R., Ruiz, C .G . , Ookhtens, M., Kaplowitz, N., 1996. Plasma membrane and mitochondrial transport of hepatic reduced glutathione. Seminars in Liver Disease 16: 147-158. Forman, H.F., Liu, R.-M., Shi, M.M., 1995. Glutathione synthesis in oxidative stress. In: Packer, L. Cadenas, E. (Eds.), Antioxidants in Health and Disease, Biothiols in Health and Disease. M. Dekker, New York, pp. 189-212. Gallagher, E.P., Hasspieler, B.M., Digiulio, R.T., 1992. Effects of buthionine sulfoximine and diethyl maleate on glutathione turnover in the channel catfish. Biochemical Pharmacology 43: 2209-2215. Garciaruiz, C , Fernandezcheca, J .C. , Kaplowitz, N., 1992. Bidirectional mechanism of plasma membrane transport of reduced glutathione in intact rat hepatocytes and membrane vesicles. Journal of Biological Chemistry 267: 22256-22264. Griffith, O.W., 1980. Determination of glutathione and glutathione disulfide using glutathione reductase and 2-vinylpyridine. Analytical Biochemistry 106: 207-212. Griffith, O.W., Meister, A., 1979a. Glutathione: Interorgan translocation, turnover, and metabolism. Proceedings of the National Academy of Science of the United States of America 76: 5606-5610. Griffith, O.W., Meister, A., 1979b. Potent and specific inhibition of glutathione synthesis by buthionine sulfoximine (S-n-butyl homocysteine sulfoximine). Journal of Biological Chemistry 254: 7558-7560. Hagen, T.M., Jones, D.P., 1989. Role of glutathione transport in extrahepatic detoxication. In: Taniguchi, N., Higashi, T., Sakamoto, Y. Meister, A. (Eds.), Glutathione Centennial: Molecular Perspectives and Clinical Implications. Academic Press, Inc., Tokyo, pp. 423-433. Hahn, R., Wendel, A., Flohe, L., 1978. Fate of extracellular glutathione in rat. Biochimica et Biophysica Acta 539: 324-337. Kannan, R., Bao, Y.Z. , Wang, Y., Sarthy, V .P . , Kaplowitz, N., 1999. Protection from oxidant injury by sodium-dependent G S H uptake in retinal Muller cells. Experimental Eye Research 68: 609-616. Kannan, R., Chakrabarti, R., Tang, D., Kim, K., Kaplowitz, N., 2000. G S H transport in human cerebrovascular endothelial cells and human astrocytes: evidence for luminal localization of Na+-dependent G S H transport in H C E C . Brain Research 852: 374-382. Kidd, P.M. , 1997. Glutathione: systemic protectant against oxidative and free radical damage. Alternative Medicine Review 2: 155-176. Lantum, H., Iyer, R., Anders, M., 2004. Acivicin-induced alterations in renal and hepatic glutathione concentrations and in g-glutamyltransferase activities. Biochemical Pharmacology 67: 1421-1426. 41 Lauterburg, B.H., Adams, J.D., Mitchell, J.R., 1984. Hepatic glutathione homeostasis in the rat: efflux accounts for glutathione turnover. Hepatology 4: 586-590. Leeuwenburgh, C , J i , L., 1995. Glutathione depletion in rested and exercised mice - biochemical consequence and adaptation. Archives of Biochemistry and Biophysics 316: 941-949. Leeuwenburgh, C., J i , L., 1998. Glutathione and glutathione ethyl ester supplementation of mice alter glutathione homeostasis during exercise. Journal of Nutrition 128: 2420-2426. Lu, S.C. , Kuhlenkamp, J . , Ge, J.L., Sun, W.M. , Kaplowitz, N., 1994. Specificity and directionality of thiol effects on sinusoidal glutathione transport in rat liver. Molecular Pharmacology 46: 578-585. Martensson, J . , Meister, A., 1989. Mitochondrial damage in muscle occurs after marked depletion of glutathione and is prevented by giving glutathione monoester. Proceedings of the National Academy of Sciences of the United States of America 86: 471-475. Martensson, J . , Jain, A., Frayer, W., Meister, A., 1989. Glutathione metabolism in the lung - inhibition of its synthesis leads to lamellar body and mitochondrial defects. Proceedings of the National Academy of Sciences of the United States of America 86: 5296-5300. Meister, A., 1995. Strategies for increasing cellular glutathione. In: Packer, L. Cadenas, E. (Eds.), Biothiols in Health and Disease. M. Dekker, New York, pp. 165-188. Meister, A., Tate, S .S. , Griffith, O.W., 1981. g-Glutamyl transpeptidase. Methods in Enzymology 77: 237-253. Mittur, A., Wolkoff, A.W., Kaplowitz, N., 2002. The thiol sensitivity of glutathione transport in sidedness-sorted basolateral liver plasma membrane and in Oatpl-expressing HeLa cell membrane. Molecular Pharmacology 61: 425-435. Otto, D.M.E., Sen, C.K., Hidiroglou, N., Madere, R., Moon, T.W., 1997. Role of exogenous glutathione in teleost fish and its effects on antioxidant defense responses in rainbow trout exposed to 3,3',4,4'-tetrachlorobiphenyl. Fish Physiology and Biochemistry 16: 449-457. Puri, R.N., Meister, A., 1983. Transport of glutathione, as gamma-glutamylcysteinylglycyl ester, into liver and kidney. Proceedings of the National Academy of Sciences of the United States of America 80: 5258-5260. Rebbeor, J .F. , Connolly, G.C. , Henson, J .H. , Boyer, J.L., Ballatori, N., 2000. ATP-dependent G S H and glutathione S-conjugate transport in skate liver: role of an Mrp functional homologue. American Journal of Physiology-Gastrointestinal and Liver Physiology 279: G417-G425. Richman, P., Meister, A., 1975. Regulation of gamma-glutamyl-cysteine synthetase by non-allosteric feedback inhibition by glutathione. Journal of Biological Chemistry 250: 1422-1426. Scott, R.D., Hughey, R.P., Curthoys, N.P., 1993. Role of apical and basolateral secretion in turnover of glutathione in Llc-Pk(1) cells. American Journal of Physiology 265: F723-F728. Smith, P.K., Krohn, R.I., Hemanson, G.T., Mallia, A.K., Gartner, F.H., Provenzano, M.D., Fujimoto, E.K., Goeke, N.M., Olson, B.J. , Klenk, D.C., 1985. Measurement of protein using bicinchoninic acid. Analytical Biochemistry 150: 76-85. Stole, E., Smith, T.K., Manning, J .M. , Meister, A., 1994. Interaction of g-glutamyl transpeptidase with acivicin. The Journal of Biological Chemistry 269: 21435-21439. Vincenzini, M., Favilli, F., lantomasi, T., 1992. Intestinal uptake and transmembrane transport-systems of intact G S H - characteristics and possible biological role. Biochimica et Biophysica Acta 1113: 13-23. Wallace, K.B., 1989. Glutathione-dependent metabolism in fish and rodents. Environmental Toxicology and Chemistry 8: 1049-1055. Wellner, V .P . , Anderson, M.E., Puri, R.N., Jensen, G.L., Meister, A., 1984. Radioprotection by glutathione ester - transport of glutathione ester into human lymphoid-cells and fibroblasts. Proceedings of the National Academy of Sciences of the United States of America 81: 4732-4735. Wolff, J .E.A. , Munstermann, G. , Grebenkamper, K., Erben, M., 1998. Gamma-glutamyl transpeptidase does not act as a cystine transporter in brain microvessels. Neurochemical Research 23: 1175-1178. 42 CHAPTER 3 Effects of incubation and acclimation temperature on glutathione dynamics in fish in vitro and in vivo Introduction Although most animals rely heavily on aerobic metabolism to sustain life, reliance on oxidative metabolism for energy can be detrimental. Approximately 0.1% of all oxygen entering the mitochondrial electron transport chain is released as reactive oxygen species, which can cause damage to lipids, proteins, and DNA (reviewed by Fridovich, 2004). To combat this potential damage, cells are equipped with a suite of antioxidants, including the tripeptide glutathione (GSH). G S H , along with the glutathione peroxidase enzymes (GPx), reduces various oxidized products within a cell, forming oxidized G S H (GSSG) . The antioxidant activity of G S H relies on G S S G being readily reduced back to G S H by glutathione reductase (GR). The total cellular pool of G S H (tGSH = G S H + 2 G S S G ) is the result of a balance between its synthesis within the cell (via glutamylcysteine synthetase (GCS) and glutathione synthetase enzymes), and its transport from the cell (for the above see Meister, 1995; Kidd, 1997; Anderson, 1998; Hayes and McLellan, 1999). Although some free G S H may be lost by reversible interactions with protein thiols (Lind et al., 2002), cellular turnover of G S H can be accounted for by excretion of G S H , G S S G , and GSH-S-conjugates from cells in hepatocytes, kidney cells and erythrocytes in mammals (Lunn et al., 1979; Lauterburg et al., 1984; Scott et al., 1993). In mammals, G S H levels are high relative to other antioxidant and thiol molecules (up to 10mM, reviewed by Kidd, 1997; Hayes and McLellan, 1999). However, in fish G S H levels and the functional activity of GSH-associated enzymes are relatively low (Wallace, 1989). In addition, G S H turnover appears to be lower in fish than mammals (Gallagher et al., 1992, Chapter 2). The basis for lower G S H levels, turnover and activity of associated enzymes in fish relative to mammals is not fully understood, but may be due to one of two factors. Fish are able to excrete hydrogen peroxide from their gills (Wilhelm-Filho et al., 1994), thereby decreasing the need for antioxidants such as G S H . Secondly, lower metabolic rate in fish, from either or both lower body temperature and reduced metabolic cost of ectothermy versus endothermy, may decrease the necessity for G S H in fish compared to mammals. Increased metabolism results in increased oxygen uptake by mitochondria, and consequently increased production of reactive oxygen species such as hydrogen peroxide (Davidson and Schiestl, 2001). The lower metabolic rates of fish should result in decreased demand for G S H , and may account for the observed low G S H levels, turnover and activity of associated enzymes relative to mammals. The effect of temperature on G S H dynamics has been examined to some extent in fish and mammals. In mammals, hyperthermia results in a transient decrease followed by an increase in G S H levels in the blood, an increase in excretion of hepatic G S H , and an increase in lipid peroxidation (Skibba et al., 1991; Ohtsuka et al., 1994). Short-term heat stress was found to decrease G S H levels and increase lipid peroxidation in catfish (Heteropneustes fossilis, 'Parihar et al., 1996; Parihar et al., 1997), and snakeheads (Channa punctata, Kaur et al., 2005). Although no studies have directly addressed the 43 effects of long-term temperature changes on G S H dynamics, seasonal changes in G S H levels have been associated in part with changes in temperature in various ectothermic animals (Dziubek, 1987; Perez-Pinzon and Rice, 1995; Wilhelm-Filho et al., 2001; Gorbi et al., 2005). In Chapter 2, I demonstrated that G S H levels were influenced by precursor supply, and not necessarily synthesis rate. Consequently, G S H levels may be rapidly increased in fish tissues during times of high demand, such as altered temperature. Based on the above data, I hypothesized that G S H turnover, levels and activity of associated enzymes in fish would increase with increasing temperature, and that temperature would explain the differences in G S H dynamics between fish and mammals. To examine the effect of temperature on the turnover of the total G S H pool I incubated a rainbow trout (Oncorhynchus mykiss) hepatoma line (RTH-149) with an inhibitor of the G S H synthesis enzyme G C S (buthionine sulfoximine - B S O , Griffith and Meister, 1979). This cell line was chosen as it is able to grow over a wide range of temperature (at least 12-24°C, although growth decreases at 27°C, Lannan et al., 1984). I also examined the effects of acclimation temperature on G S H levels, oxidation and activity of GPx and G R in vivo, using the common killifish (Fundulus heteroclitus macrolepidotus), as this fish is known to tolerate a wide range of acclimation temperatures (Targett, 1978; Fangue et al., unpublished data). Using these models, I addressed the following objectives: Objective 3.1: To determine if incubation temperature alters G S H turnover of RTH-149 cells in vitro. Objective 3.2: To determine if acclimation temperature alters G S H levels, oxidation and activity of associated enzymes in killifish in vivo. Objective 3.3: To determine to what extent temperature can account for differences in G S H levels, turnover, and activity of associated enzymes between fish and mammals. Materials and Methods RTH-149 cells were purchased from the American Type Culture Collection (ATCC#: CRL-1710, Manassas, VA, USA). All cell culture media were from Invitrogen Corporation (Burlington, ON, Canada). G R was purchased from Roche Diagnostics (Laval, Q C , Canada), and tricaine methanesulfonate was purchased from Syndel, Canada (Vancouver, BC, Canada). All other chemicals were purchased from Sigma-Aldrich (Oakville, O N , Canada). Experiment 3.1: Influence of incubation temperature on tGSH turnover in vitro RTH-149 cells were cultured in 18.3mM H E P E S and 0.75% sodium bicarbonate buffered Dulbecco's Modified Eagle's Medium (DMEM) with 10% fetal calf serum, 0.1mM non-essential amino acids and 1% antibiotic/antimycotic (penicillin, streptomycin, amphotericin B), pH 7.4. During growth, cells were held at 20°C in air. In a preliminary experiment, the minimum BSO concentration required to produce maximum tGSH decrease after 24 h was determined. RTH-149 cells were grown on 6-well plates with 2mL of the above media until approximately 80% confluent. Media were changed and replaced with media as above but lacking fetal calf serum and containing varying concentrations of B S O (0 to 50mM). 24 h after the start of BSO incubation, 6 wells were sampled per BSO concentration. To sample, cells and media were scraped into a 2mL tube and centrifuged at 10 000*g for 20 sec. The cell pellet was washed twice in ice-44 cold phosphate-buffered saline and frozen on dry ice for tGSH levels as described below. The percentages of tGSH remaining after 24 h incubation with increasing concentrations of BSO are given in Figure 3.1. tGSH levels declined with increasing BSO concentration, to a minimum at 25-50mM BSO. Based on the results of this experiment, a concentration of 25mM B S O was selected for the subsequent experiment to test the effects of temperature on tGSH turnover. RTH-149 cells were plated on 6-well plates at 1.6*10 5 cells per 2mL media per well. Cells were grown at 18°C for 6 d until approximately 80% confluent. Cells were then divided into three groups and incubated at three different temperatures (10°C, 18°C or 26°C) for an incubation period of 16 h. All temperatures used were within the known proliferation zone of rainbow trout cell lines, with 18°C close to the reported optimum temperature of 20°C (Bols et al., 1992). At the end of the incubation period, the media were changed, further dividing cells into two groups: a control group receiving normal media and a B S O group that received media containing 25mM B S O . Because B S O can prevent cell replication, the media for both control and BSO-incubated cells contained no fetal calf serum to similarly prevent growth of control cells. While removal of fetal calf serum can alter G S H levels (see Chapter 5), it was necessary to remove fetal calf serum to prevent growth-related differences between control and BSO-incubated cells. All cells were sampled at the end of the incubation period (0 h). In addition, B S O cells were sampled 4, 8, 24, 48 and 72 h into the B S O incubation and control cells were sampled 24 and 72 h into the incubation period. Six wells were sampled per time, temperature, and media group. To sample each well, an aliquot of medium was removed and frozen on dry ice for analysis of lactate dehydrogenase (LDH). Cells were then scraped into a 2mL tube, centrifuged at 10 000*g for 20 sec, washed twice in ice-cold phosphate-buffered saline and frozen on dry ice for later measurements of cell protein, LDH and tGSH levels. All samples were stored at -80°C. Experiment 3.2: Influence of acclimation temperature on GSH levels and enzyme activity in v ivo Killifish (8.7±4.0g) were captured from Hampton, NH by Aquatic Research Organisms, Inc. (Hampton, NH, USA). The following experiment took place in the summer (August-September). Fish were maintained in 20L glass aquaria in 12ppt seawater (Kent Marine Salts (Acworth, GA, USA) dissolved in dechlorinated Vancouver municipal fresh water) and 12:12 lightdark photoperiod at a density of eight fish per aquarium. Fish were fed to satiation daily with trout food (Skretting, Vancouver, BC, Canada) and blood worms until the final day of temperature acclimation. From the starting temperature of 28°C, temperature in the aquaria was adjusted by 1°C per day until the tanks were at 6, 17, 25, 30 and 33°C, 1 tank per temperature. Once temperatures stabilized, fish were acclimated to their respective temperatures for three weeks. At the end of the three weeks, fish were sacrificed by an overdose of anesthetic (2g/L tricaine methanesulfonate, 4g/L sodium bicarbonate), and blood collected in a hematocrit tube by caudal severance. The blood was then centrifuged for 3 min, hematocrit recorded and plasma and R B C frozen separately in liquid nitrogen. Liver, brain, gill, muscle, and heart were then excised, rinsed in ice-cold phosphate-buffered saline, blotted dry and frozen in liquid nitrogen. All blood and tissue samples were stored at -80°C and analyzed within one month. tGSH levels in all blood and tissue samples were analyzed, as well as liver G S S G , GPx , and G R activity. 45 Figure 3.1: Percent of total G S H remaining in RTH-149 cells after 24 fi incubation with various concentrations of a G S H synthesis inhibitor (BSO). Where letters differ (a,b) significant differences in % tGSH remaining exist between BSO groups. n=6. 46 Sample Preparation and Analyses All sample preparation was performed on ice and analyses were performed at room temperature unless otherwise stated. RTH-149 cells were sonicated in 60(iL of 100mM H E P E S containing 1mM ethylenediaminetetraacetic acid (EDTA), 1pM pepstatin, 1uM leupeptin, 0.15uM aprotinin, and 0.5mM phenylmethylsulfonyl fluoride protease inhibitors. Portions of homogenate were removed for protein and LDH analyses, and the remaining homogenate added to an equal volume of 10% sulfosalicylic acid, centrifuged at 10 000*g for 10 min and the supernatant removed for tGSH analysis. Killifish tissues were sonicated in either 125mM sodium phosphate buffer (pH 7.5) with protease inhibitors (tGSH, G S S G , GR analyses) or the above supplemented with 1mM dithiothreitol (GPx analysis) at 100pL per 10mg tissue and centrifuged at 11 600*g at 4°C for 5 min. Portions of supernatant were removed for subsequent analysis of total protein content, and GPx and G R activity. The remaining supernatant was added to an equal volume of 10% sulfosalicylic acid, centrifuged at 10 000*g at 4°C for 10 min and supernatant removed for tGSH and G S S G analysis. All analyses were performed in triplicate, under saturating conditions, on 96-well microplates and measured using a SpectraMax spectrophotometer equipped with SoftmaxPro Software (Molecular Devices Corporation). Protein content was analyzed using the bicinchoninic acid method as per Smith et al. (1985), using bovine serum albumin as standards and analyzed at 516nm. LDH activities in RTH-149 cell homogenate and media were analyzed as previously described (Bergmeyer, 1985). In brief, cell homogenate was diluted 11 times in 100mM H E P E S , pH 7.4. 10uL of cell homogenate or 20pL of media were added to a 96-well plate, with appropriate H E P E S or media blanks. 0.4mM reduced B-nicotinamide adenine dinucleotide (NADH) in H E P E S buffer was added to each well for a final volume of 190uL. The plates were covered and incubated at 28°C for 15 min for cell homogenate and 30 min for media. Background absorbance decline was then read for 5 min at 340nm at 28°C. The reaction was initiated by the addition of 10uL of 28mM pyruvate and oxidation of NADH to N A D + was measured by absorbance decline at 340nm for 5 min at 28°C. tGSH and G S S G analyses were modified from a previous study (Griffith, 1980), as follows. For tGSH analyses, 6uL of triethanolamine was added per 10OpL of tissue or cell supernatant and G S S G standards. 10uL of samples or G S S G standards and 200uL of reaction mixture were added to a 96 well plate. The reaction mixture contained 0.22mM reduced 6-nicotinamide adenine dinucleotide phosphate (NADPH), 0.62mM 5,5'-dithiobis(2-nitrobenzoic acid) (DTNB), 0.56% triethanolamine and 0.5 U/mL GR in 125mM sodium phosphate, 6mM EDTA buffer, pH 7.4. The change in absorbance was monitored at 412nm for 5 min. For G S S G analyses, 6pL of 2-vinylpyridine (to bind reduced GSH) and 20uL of 0.1% triethanolamine in phosphate-buffered saline was added per 100uL of sample or G S S G standard. Samples and standards were mixed vigorously for 1 min and then incubated for 50 min. G S S G analyses were then as per tGSH above, except DTNB concentration in the reaction mixture was 1uM. G R and G P x activity of liver homogenates were analyzed by procedures modified from Stephensen et al. (2002). Homogenates were diluted to 5mg protein/mL with 125mM sodium phosphate, 6mM EDTA buffer (pH 7.4) containing either 0.1% bovine serum albumin (GR) or 1mM dithiothreitol (GPx). For GR analyses 10pL of homogenate or buffer blank was added to a 96-well plate and 200uL of 47 a reaction mixture containing 0.1 mM DTNB and 0.63mM NADPH in buffer was added to each well. Background absorbance was measured at 405nm for 5 min. The reaction was initiated by the addition of 10uL 0.33M G S S G to each well and the reduction of DTNB was monitored at 405nm for 5 min. For GPx analyses, 10uL of homogenate or buffer blank was added to a 96-well plate and 200pL of a reaction mixture containing 3.5mM G S H , 1mM sodium azide, 2U7mL G R and 0.12mM N A D P H was added to each well. The reaction was initiated by the addition of 10pL of 0.03% hydrogen peroxide and the oxidation of NADPH was monitored at 340nm for 5 min. GPx and G R were analyzed at 28°C (approximately room temperature), as well as at respective acclimation temperatures. Statistical Analyses The effects of temperature, time and B S O in the RTH-149 experiment were analyzed using 1-way or 2-way A N O V A , or regression as appropriate, and the effects of temperature in the killifish experiment were analyzed using 1-way A N O V A or regression followed by Tukey's post-hoc test using SigmaStat and SigmaPlot ( S P S S Inc.). Decay coefficients were determined by non-linear (exponential decay) regression analysis using SigmaPlot, and G S H turnover half-lives were calculated from the corresponding exponential decay equations. Comparisons of regression models were performed using analysis of covariance. Differences were considered significant if p<0.05. All data are presented as mean ± standard error of the mean. Results Experiment 3.1: Influence of incubation temperature on tGSH turnover in vitro In BSO-incubated cells, tGSH levels declined over time in all temperature groups (p<0.001, Figure 3.2A), and this decline was most pronounced in 26°C incubated cells. By 48 h cells incubated at 26°C had lower tGSH levels than those at the lower temperatures, and at 72 h tGSH levels differed between all temperature groups. tGSH levels in control cells fell within the range reported for primary cultures of rainbow trout hepatocytes (4.5 - 100 nmol/mg protein; Radice et al., 2001; Ferraris et al., 2002; Borgdanova et al., 2005; Figure 3.2B). However, control cell tGSH varied with time and temperature (p<0.001, Figure 3.2B). Cells incubated at 26°C had higher tGSH levels than those at lower temperatures at 0 and 24 h, but by 72 h there were no differences among temperature groups. tGSH levels in all temperature groups declined in control cells at 24 h, and remained depressed at 72 h in 10°C and 26°C incubated cells but not those incubated at 18°C. The relationship between time and tGSH levels (as nmol/mg protein) in BSO-incubated cells was modeled as an exponential decay of tGSH (Figure 3.3A). The overall rate of decline as described by exponential decay was different among all temperatures (F 1 6 8 =77.6) in the order of 26 oC>18°C>10°C, with corresponding tGSH apparent half-lives of 21.7, 48.5, and 73.7 h respectively. As tGSH levels varied in control cells, I also modeled the relationship between temperature and tGSH decline as percent of control cell levels (Figure 3.3B). Although decay coefficients were slightly lower when tGSH decline was calculated as percent of control cells, temperature increased the decay coefficients in a similar manner to when tGSH decline was calculated as absolute values. This indicates temperature-dependent 48 70 60 50 40 30 S 20 H 10 -4—* o l-l 0 0 s o 0 6 ~ 6 0 0 0 9 50 40 30 20 10 A b +BSO -o-- 10°C - o - 18°C • 26°C 20 40 time (h) 60 Figure 3.2: Total G S H levels in (A) G S H synthesis inhibitor (25mM BSO) incubated and (B) control RTH-149 cells held at different temperatures. Time is from media change and beginning of B S O incubation, 16 h after beginning of temperature incubation. Where letters differ (a,b,c), significant differences in temperatures within time and B S O exist. * indicates difference from time 0, within temperature and BSO. <)> indicates B S O different from controls, within time and temperature. n=6. 49 70 n 60 50 40 c '5 O 1-1 & W) s 1 30 H Si 2 0 10 o i-i c o o o X O 0 100 4 75 50 25 H Exponential decay: 10°C a: tGSH = 39.8 * e"0'009 * t i m e 18°C b: tGSH = 41.2 * e" 0 0 1 4 * t i m e 26°C C : tGSH = 52.6 * e" 0 0 3 2 * t i m e R =0.71 R=0.81 R =0.97 - o - - 10°C -® - 18°C • 26°C B Exponential decay: 10°C a: tGSH = 104.1 *e" 0 0 0 6 * t i m e 18°C b: tGSH = 101.0 * e" 0 0 1 3 * t , m e 26°C C : tGSH = 104.1 * e"0'026 * t i m e R=0.9C R =1.00 R2=0.97 0 20 40 60 time (h from begining of B S O incubation) Figure 3.3: The effect of incubation with a G S H synthesis inhibitor (25mM BSO) on total G S H levels over time in RTH-149 cells held at different temperatures. A) The relationship between tGSH levels (as nmol/mg protein) and time as expressed by exponential decay. B) The relationship between tGSH levels (as % of control levels at equal time points) and time as expressed by exponential decay. Where letters differ (a,b,c), significant differences in decay coefficients exist. n=6. 50 declines in tGSH levels were due to turnover differences, and not alterations in G S H levels due to temperature. The relationship between the exponential decay coefficients of tGSH decline and temperature could be described by the equation: decay coefficient = 0 . 0 0 3 3 x e 0 0 8 6 4 " t e m p e r a t u r e (Figure 3.4, R2=0.98). When the curve was extrapolated to 37°C and compared to those reported in mammalian liver, the estimated G S H decay coefficient (0.082) was only 26% of the average mammalian decay coefficient (0.14-0.70, average 0.31, Sekura and Meister, 1974; Meredith and Reed, 1982; Orrenius et al., 1982; Griffith and Meister, 1985; Potter and Tran, 1993). The calculated apparent half-life of RTH-149 cells extrapolated to 37°C was 8.5 h, which was 2.7 times longer than apparent half-lives reported in mammalian liver and hepatocytes (1-5 h, average 3.1 h, references as above). The relationship between G S H decay coefficients and temperature could also be described using a linear model, although the fit was not as good as that provided by the exponential model (decay coefficient = -0.007+0.0014xtemperature; R2=0.90). The linear model also resulted in a more conservative estimate of the decay coefficient at 37°C (0.045), with a corresponding apparent half-life of 15.4 h, five times that of estimated mammalian half-lives. As decay coefficients were slightly lower when tGSH decline was expressed as % of control, I also examined the relationship between temperature and decay coefficients calculated from % of control data. The models were similar to those of Figure 3.4, and fitted equally well to an exponential model as a linear model (R2=1.00 and 0.98 respectively). When these relationships were extrapolated to 37°C the estimated decay coefficients were similar, although slightly lower than those estimated using tGSH as nmol/mg protein (0.071, half-life 9.8 h for exponential, and 0.039, half-life 17.8 h for linear models). Total protein as an estimate of cell quantity per well is given in Table 3.1. There was an overall effect of temperature on total protein per well in both control and BSO-incubated cells (p=0.001). Cells incubated at 18°C had overall higher total protein than those at 10°C or 26°C. However, at individual time points, total protein in cells at 18°C was significantly greater that at 10°C at 24 h in BSO-incubated cells only. There was no effect of BSO on total protein quantity at any temperature (p=0.168). LDH released (% of total) as an estimate of cell viability is given in Table 3.2. In general, percent LDH released was well below 10% and there were minor inconsistent effects of temperature (p=0.001) and time (p=0.002) on LDH released. Cells incubated at 18°C appeared the healthiest overall as they had less LDH released than cells at 10°C and 26°C measured at several time points, and demonstrated little change in LDH released over time. BSO incubation had inconsistent effects on LDH release, depending on the temperature. There was no effect of BSO on LDH release at 18°C, B S O decreased LDH release at 26°C, and transiently increased LDH released at 10°C. Experiment 3.2: Influence of acclimation temperature on GSH levels and enzyme activity in vivo The average water temperatures and oxygen content, hematocrit, and amount eaten by killifish acclimated to different temperatures are presented in Table 3.3. The oxygen content was greatest in the 6°C tank, and was similar among the tanks at higher temperatures. Feeding rate increased with increasing temperature (p=0.014), although only differed between those fish held at 6°C and those held at 25, 30 and 33°C. Weight and length did not differ between fish held at different temperatures (8.7±4.0g, p=0.82, and 8.4±0.2cm, p=0.91 respectively). There was no clear relationship between hematocrit and 51 Figure 3.4: Decay coefficients of total G S H decrease during incubation with a G S H synthesis inhibitor (25mM BSO) in cultured RTH-149 cells as a function of incubation temperature. Decay coefficients of RTH-149 cells are calculated from this study, and the mammalian data averaged from previous reports of liver or hepatocytes (Sekura and Meister, 1974; Meredith and Reed, 1982; Orrenius et al., 1982; Griffith and Meister, 1985; Potter and Tran, 1993). The trend lines (linear - dashed, exponential - solid) are calculated from RTH-149 cell data points and are extrapolated to 37°C. Table 3.1: Total protein (pg/well) of RTH-149 cells over time, held at different temperatures with or without the G S H synthesis blocker B S O (25mM). Where letters differ (a,b), significant differences between temperatures, within time and BSO exist. * indicates significant difference from time 0, within temperature and B S O . n=6. Time (h) Control/BSO 10°C 18°C 26°C 0 Control/BSO 103.0 ±4.6 103.7 ±5 .4 104.3 ±2.1 4 B S O 82.6 ± 4.0* 99.7 ± 4.2 96.8 ± 3.0 8 BSO 82.5 ± 7.8 95.4 ± 5.0 106.2 ±8.1 24 Control 99.2 ± 4.2 118.3 ±3 .3 91.5 ±4 .4 BSO 91 .6±4 .6 a 114 .8±6 .6 b 100.4 ± 3 . 9 a b 48 B S O 105.2 ±4.4 109.4 ± 3.6 94.7 ±4 .1 72 Control 111.1 ±4.1 119.9 ±5 .3 101.0 ±7 .5 BSO 101.8 ± 2.6 115.2 ±5 .5 94.1 ± 3.5 52 Table 3.2: Lactate dehydrogenase released (% of total) from RTH-149 cells overt ime, held at different temperatures with or without the G S H synthesis blocker B S O (25mM). Where letters differ (a,b,c), significant differences in temperatures, within time and B S O exist. * indicates significant difference from time 0, within temperature and BSO. <|> indicates BSO significantly different from control, within time and temperature. n=6. Time (h) Control/BSO 10°C 18°C 26°C 0 Control/BSO 3.0 ± 0.2 3.0 ±0 .5 3.0 ± 0.3 4 B S O 4.6 ±0 .7 4.0 ±0 .2 3.2 ±0 .3 8 B S O 6.1 ±0 .5 a * 2.1 ± 0 . 4 b 1.7±0.3 b 24 Control 1.6±0.2 a * 3 . 3 ± 0 . 4 b 5.4 ± 0.2C* BSO 4 . 2 ± 0 . 7 a D 2.1 ± 0 . 4 b 3.2 ± 0 .2 a b D 48 BSO 3.7 ± 0.4 a 3.1 ± 0 . 3 a 5.7 ± 0.7 b* 72 Control 6.4 ± 0.2 a* 2 . 3 ± 0 . 7 b 4.1 ±0 .3 C BSO 5 .6±0 .7 a * 4 . 3 ± 0 . 9 a * 1 .5±0 .5 b D Table 3.3: Average acclimation temperature (°C, averaged over 10 d) and oxygen content of the water (mg/L (% saturation)), amount eaten (% body weight per d, measured over 3 d), and hematocrit (n=8) of killifish acclimated to different temperatures for three weeks. Where letters differ (a,b,c,d,e), significant differences between temperature groups exist. Tank Average Temperature Oxygen Amount Eaten Hematocrit 6 6.2±0.2 a 10.0 (107) 0.03±0.00 a 34.1±1.6a b 17 17.3±0.1 b 6.0 (85) 0.48±0.08 a b 39.8±1.6a 25 25.0±0.1 c 5.3 (78) 0.94±0.16 b 32.4±1.5a b 30 30.0±0.1 d 6.6 (88) 1.06±0.20b 37.9±3.0a b 33 32.7±0.2 e 5.7 (93) 0.94±0.32 b 31.0±1.2b 53 temperature, although fish held at 17°C had significantly greater hematocrit than those held at 33°C (p=0.017). Fish held at 6°C exhibited little observed physical activity, whereas those held at warmer temperatures were relatively active. Killifish liver tGSH levels were similar to those reported in a previous study (approximately 1pmol/g wet weight, Meyer et al., 2003). There was no effect of temperature on tGSH levels in the killifish gill (p=0.07), heart (p=0.37), or R B C (p=0.05, Figure 3.5). However, temperature had a significant effect on tGSH levels in the liver (p<0.001), brain (p<0.001), muscle (p<0.001), and plasma (p=0.02). In both the liver and brain, tGSH levels increased with increasing temperature (R2=0.79 and 0.72 respectively), although this correlation was significant for liver (p=0.04), but not brain (p=0.07). The observed increase in tGSH levels in these tissues could be equally explained by feeding rate (R2=0.80, p=0.04, and R 2=0.64, p=0.10 for liver and brain respectively). In the muscle, the pattern of increasing tGSH levels with increased temperature was less clear, as fish acclimated to 17°C had lower tGSH than those held at 6°C, and tGSH levels did not significantly differ between fish held at 25, 30 or 33°C. There was no clear relationship between of tGSH levels and temperature in the plasma and the only significant difference was between fish held at 6°C and 17°C. To compare tissue levels of tGSH of this study and those of previous mammalian studies, liver, brain and muscle tGSH were expressed as pmol/g wet weight, to be consistent with units used by the majority of mammalian in vivo studies. The linear relationship between temperature and tGSH levels (as pmol/g wet weight) of killifish liver, brain and muscle (Figure 3.6; R 2=0.70, 0.77 and 0.47 respectively) was similar to the relationship when tGSH was expressed as nmol/mg protein. Exponential modeling did not greatly improve the relationship between tissue tGSH and temperature (R 2=0.72, 0.81 and 0.52 for liver, brain and muscle respectively). When the models were extrapolated to 37°C and compared with those of published mammalian tissues (averaged from mice: Leeuwenburgh and J i , 1995; Shertzer et al., 1995; Leeuwenburgh and J i , 1998; rats: Kretzschmar and Muller, 1993; Malmezat et al., 2000; Zhang et al., 2002; Kolesnichenko et al., 2003; mice and hamsters: Brooks and Pong, 1981; mice and rats: Wallace, 1989; rats and guinea pigs: Ando et al., 1994), the extrapolated killifish levels were 14% of mammalian levels in the liver, 50% in the brain and 38% in the muscle. Temperature affected liver G S S G as percent of tGSH (p<0.001, Figure 3.7A), although overall G S S G levels were very low (less than 5% of tGSH). Fish acclimated to 6°C had higher % G S S G than those acclimated to 25, 30 or 33°C. Reduced liver G S H ( tGSH-2 *GSSG) increased with increasing temperature in a similar pattern as tGSH (Figure 3.7B). Temperature (R2=0.48) and feeding level (R2=0.50) could account for the increase in reduced G S H equally well. Killifish GPx activity was approximately three times higher and G R activity approximately 20% (Figure 3.8) of those previously reported for killifish (GPx=50-120, GR=8-12 nmol/min/mg protein, Meyer et al., 2003). However, G R and G P x levels fell within the range of G R and G P x activity reported for other fish, although these levels vary greatly (GPx: 2-1300, average 300 nmol/min/mg protein; GR: 0.3-30, average 11.5 nmol/min/mg protein; Wallace, 1989; Gallagher and Di Giulio, 1992; Rady, 1993; Otto and Moon, 1995; Lindstrom-Seppa et al., 1996; Almar et al., 1998; Lushchak et al., 2001; Stephensen et al., 2002; Meyer et al., 2003). There was no effect of temperature on liver G P x (p=0.40) or G R (p=0.43) activity as measured at 28°C (Figure 3.8A.B). However, when activity was measured at respective 54 Liver Brain Muscle Gill Heart RBC Plasma (uM) Figure 3.5: Tissue total G S H levels of killifish acclimated to 6, 17, 25, 30 or 33°C for three weeks. Where letters differ (a,b,c) significant differences between temperatures within tissues exist. There were no significant effects of temperature on gill, heart, or R B C tGSH levels. n=8. 55 3 •S 6.0 -5? S OO O > 3 3 1.5 0 0 O a 3 0 0 0 0 o CO 3 Killifish Mammals 1.5 A 1.0 0.5 4 0.0 2.0 1.0 2 0.5 0.0 0.6 0.4 H 0.2 0.0 B 10 15 20 25 30 acclimation temperature (°C) 35 40 Figure 3.6: Total G S H levels of killifish (A) liver, (B) brain, and (C) muscle expressed as a function of acclimation temperature (n=8). Trend lines (solid lines) and 95% confidence intervals (dashed lines) are calculated from killifish data and extrapolated to 37°C. Mammalian data are averaged from previous rodent studies (Brooks and Pong, 1981; Wallace, 1989; Kretzschmar and Muller, 1993; Andoe ta l . , 1994; Leeuwenburgh and J i , 1995; Shertzer et al., 1995; Leeuwenburgh and J i , 1998; Malmezat et al., 2000; Zhang et al., 2002; Kolesnichenko et al., 2003). 56 X oo O —^» o -»-» o O 0 0 0 0 O ab 17 25 30 33 14 o a a OO O U O -a 8 6 4 2 0 B b ab b T 17 25 30 33 acclimation temperature (°C) Figure 3.7: (A) Oxidized (GSSG) and (B) reduced G S H levels in killifish liver acclimated to different temperatures. Where letters differ (a,b,c) significant differences between temperatures exist. n=8. '33 -4—» o OH StO a a -4-> \> o a x P i O 500 400 300 200 100 0 400 300 200 100 0 T b 17 25 30 33 ^ 3 g '53 o 2 S 1 6 B 3 c o a Pi O B D ab b _JL_ ab 17 25 30 33 acclimation temperature (°C) Figure 3.8: Glutathione peroxidase (GPx) and glutathione reductase (GR) activity in the liver of killifish acclimated to different temperatures. (A) GPx activity measured at 28°C; (B) G R activity measured at 28°C; (C) G P x activity measured at respective acclimation temperatures; (D) G R activity measured at respective acclimation temperatures. Where letters differ (a,b,c) significant differences between temperatures exist. There were no significant effects of acclimation temperature on G P x or G R activity measured at 28°C. n=8. 57 acclimation temperatures, G P x activity increased with increasing temperature to 25°C, although the only significant differences were that fish acclimated to 6°C had lower G P x activity than those acclimated to other temperatures (p<0.001, Figure 3.8C). G R activity increased with increasing temperature to 30°C and decreased at 33°C (p<0.001, Figure 3.8D). GPx had a temperature coefficient ( Q 1 0 - ratio of increase in activity with a temperature increase of 10°C) of 1.4 from 6-25°C and 1.2 from 6-33°C. If the data from fish acclimated to 33°C were removed, GR had a Q 1 0 of 5.3, and if only the data from fish acclimated to 6-25°C were considered, G R had a Q 1 0 of 1.7. G P x activity from 6-25°C was linearly extrapolated to 37°C (R2=1.00) to give an estimated activity of 480 nmol/min/mg protein. The reported hepatic G P x activity of mammals varies greatly (110-1370, average 567.4±471.4 nmol/min/mg protein; mice and rats: Wallace, 1989; rats and guinea pigs: Ando et al., 1994; rats: Kosenko et al., 1997; Kolesnichenko et al., 2003), and the estimated activity calculated from the killifish data could account for approximately 85% of this activity on average. G R activity as a function of temperature could be expressed exponentially from 6-30°C (R2=0.93) or linearly from 6-30°C (R2=0.99). When G R activity was theoretically extrapolated to 37°C, the estimated activities were 8.6 or 1.3nmol/min/mg protein, respectively. As with G P x activity, the reported hepatic G R activity of mammals varies (7.7-109 nmol/min/mg protein, average 37.3 nmol/min/mg protein; mice and rats: Wallace, 1989; rats: Kosenko et al., 1997; Kolesnichenko et al., 2003). However, the effects of temperature could account for only 4-23% of the difference between G R activity between fish and mammals, on average. D i s c u s s i o n The goal of these studies was to determine the effects of temperature on G S H dynamics in two fish models, and to ascertain to what degree temperature could account for differences in G S H dynamics between fish and mammals. Increased incubation temperature increased turnover and transiently raised levels of tGSH in RTH-149 cells. In the killifish, higher acclimation temperatures increased GPx and G R activity in the liver (as measured at respective acclimation temperatures), increased tGSH levels in the liver, brain and to some extent the muscle, and decreased % G S S G in liver. However, there was little or no effect of temperature on tGSH levels in the gill, heart, R B C or plasma, and no effect of temperature on G P x or G R apparent quantity in the liver (as measured at a common temperature) of killifish. The changes observed in G S H dynamics with altered acclimation temperature were likely due to increased metabolism associated with increased temperature, as previously hypothesized (Wilhelm-Filho et al., 2000). Targett (1978) reported killifish undergo partial metabolic compensation for temperatures between 13-29°C. However, the compensation was not complete, and killifish did not compensate over 4-13°C, indicating temperature-induced increase in metabolically produced reactive oxygen species is likely. G S H dynamics may have been altered to compensate for potentially higher reactive oxygen species and lipid peroxidation production at higher temperatures, as observed in short-term hyperthermia in mammals (reviewed by Ohtsuka et al., 1994) and fish (Parihar et al., 1996; Kaur et al., 2005). This is consistent with previous studies that found seasonal temperature changes altered G S H levels in liver, brain and other tissues of ectotherms (Dziubek, 1987; Perez-Pinzon and Rice, 1995; Wilhelm-Filho et al., 2001; Gorbi et al., 2005). As well, short-term hyperthermia in mammals increased G S H levels after a 58 transient decrease (reviewed by Ohtsuka et al., 1994), and increased hepatic export of oxidized and reduced G S H (Skibba et al., 1991). Although feeding levels accounted for tGSH levels equally well as temperature, increased feeding is most likely due to increased metabolic demands at higher temperature. Although G S H synthesis activity was not measured, I demonstrated in Chapter 2 that G S H levels may be controlled by precursor availability rather than synthesis activity. Consequently, high amino acid supply from increased feeding levels would allow for increased G S H synthesis during acclimation to high temperatures. Unexpectedly, higher temperature only transiently increased tGSH levels in RTH-149 cells and did not affect tGSH levels in several killifish tissues (gill, heart, RBC) . In other study organisms, temperature or high metabolic rate altered G S H levels in these tissues (bullfrog blood - Rana temporaria, Dziubek, 1987; human R B C , Ohtsuka et al., 1994; catfish gill, Parihar et al., 1997; shrew heart - Blarina brevicauda; Stewart et al., 2005). The lack of change in tGSH levels in these tissues of this study may be due to the tissues metabolically compensating for increased temperature or by increasing other antioxidants in the tissues, such as catalase, superoxide dismutase, or various non-enzymatic antioxidants. Other temperature-associated physiological changes may have resulted in lack of temperature effects on G S H dynamics. For example, polyunsaturated fatty acid (PUFA) content of the membranes increases with decreasing temperature to maintain fluidity of cell membranes (see Hazel and Prosser, 1974). As P U F A are more susceptible to lipid peroxidation (see Sciuto, 1997), lower temperatures may increase the demand for antioxidants such as G S H , despite potential decreased metabolism-related demands. As well, fish are able to excrete hydrogen peroxide from the gill (Wilhelm-Filho et al., 1994), which may prevent the need for increased antioxidants in the gill, blood, and possibly the heart. It is clear that altered tGSH levels are not necessary to acclimate to different temperatures in every tissue. In the liver, G S S G was below 5% of tGSH levels and the pattern and quantity of reduced G S H was similar to that of total G S H . Killifish acclimated to 6°C had higher hepatic % G S S G than those fish acclimated to 25°C or greater, as well as higher tGSH muscle and plasma levels than those at 17°C. This may be due to increased oxidation of membrane P U F A at low temperatures (see above), or may be a response to the slight supersaturation of the water with oxygen at 6°C (107%, Table 3.3). Combined with lower enzyme activity of G R due to low temperature, this likely caused the relatively high hepatic % G S S G in fish acclimated to 6°C. This is consistent with an association found between low G R activity and high % G S S G in hibernating ground squirrels (Carey et al., 2003). When measured at a common temperature of 28°C, there was no effect of acclimation temperature on either GPx or G R activity. This implies that the total quantity of these enzymes was not altered by acclimation temperature. When activity was measured at the respective acclimation temperatures, activity increased with increasing temperature, although the effects differed between GPx and GR. G P x activity increased with temperature to a maximum at 25°C, after which temperature had no effect on G P x activity. The temperature coefficient (Qi 0) of G P x from 6-25°C was 1.4, lower than would be expected if the enzymes were fully thermolabile (Q 1 0=2-3, Hazel and Prosser, 1974). Although I cannot determine the cause of this relative thermal insensitivity, altered G P x quantity with acclimation is not likely to be a factor, based on the lack of significant differences in activity between acclimation groups 59 when the enzyme is assayed at a common temperature of 28°C. Alternatively, a shift in the isoforms of GPx expressed and/or post-translational modifications of the enzymes could account for the apparent thermal insensitivity of GPx from fish acclimated to temperatures of 25°C and higher. G R on the other hand was more thermolabile than G P x over 6-25°C (Q 1 0=1.7), and at 30°C activity greatly increased (Q 1 0 from 6-30°C=5.3). As with GPx , apparent quantity of G R did not change greatly with temperature. Thus, the relatively high Q 1 0 of this enzyme may be the result of differences in post-translational modification of G R across acclimation temperatures, or to expression of different isoforms of this enzyme. At 33°C, activity of G R greatly diminished to levels similar to those at 6-25°C, indicating 33°C was near the thermal maximum for enzyme activity. However, if G R activity was inadequate to meet cellular demands at this temperature, a concomitant increase in G S S G would be expected, as the ability of G R to reduce G S S G would be impaired. As this was not observed, G R activity was either sufficient to reduce existing levels of G S S G or G R was functioning at an adequate state in vivo but may have destabilized during processing and analysis. The Q 1 0 values reported here are consistent with reportedly high Q 1 0 values for GR and lower Q 1 0 values for antioxidant enzymes in temperate and Arctic fishes ( Q 1 0 over 1-6°C for GR: 2.2-4.0, catalase: 1.2-1.6, superoxide dismutase: 1.2-2.1, Speers-Roesch and Ballantyne, 2005). The difference in Q 1 0 between G R and G P x may indicate an increased need for the reduction of G S S G by G R relative to G P x at higher temperatures, or may be due to increased thermal sensitivity of G R compared to GPx. The decrease in tGSH levels over time in control RTH-149 cells grown at all temperatures was most likely due to the cells acclimating to the withdrawal of fetal calf serum from the media. The more rapid recovery of this decline, as well as small increases in total protein and decreases in LDH released in cells held at 18°C compared to those held at 10 or 26°C, indicates that cells at 18°C were slightly healthier than those at other temperatures. Although 10 and 26°C are within the proliferation zone of rainbow trout cell lines, 18°C is closer to the optimum temperature of 20°C (Bols et al., 1992) and is a smaller shift in temperature from the previous maintenance temperature of 20°C. This may have confounded the changes in tGSH levels with altered acclimation temperature. Temperature could account for only 26% of the difference in turnover between RTH-149 cells and mammalian liver. However, the difference between mammalian G S H turnover and that of the RTH-149 cells reported here may be less than the difference between fish and mammals in vivo. Scott et al. (1993) reported G S H turnover was slightly reduced in a cell line compared with isolated cells or tissues. In vivo turnover of G S H in fish may be greater than in RTH-149 cells, and consequently temperature may account for a larger portion of the difference in turnover between fish and mammals. In addition, mammalian studies used radiolabeled amino acid incorporation in the presence of B S O to determine G S H turnover. In the RTH-149 cells I cannot confirm that G C S activity was completely blocked. Consequently, apparent turnover of RTH-149 cells may be underestimated and the difference between fish and mammalian turnover overestimated. Even taking into account the above difficulties, the large difference between extrapolated turnover of RTH-149 cells at 37°C and those of mammals indicates that temperature cannot fully explain the difference in G S H turnover between these two organisms. In vivo, tGSH levels of killifish extrapolated to 37°C were only 50% of mammalian levels in the brain, 38% in the muscle, and 14% in the liver. In the liver, temperature could account for the majority of the differences between fish and mammals in G P x activity (85%), but poorly accounted for differences in 60 G R activity (4-23%). As well, there was no effect of temperature on tGSH levels in gill, heart or R B C of killifish. This observation demonstrates that temperature can account for only a portion of the differences between the G S H antioxidant system of fish and mammals in vivo. In addition to temperature, two other factors contribute to the difference in metabolic rate between ectotherms and endotherms. Both body size and the metabolic cost of endothermy are known to increase metabolism (Hulbert and Else, 1989) and consequently may increase G S H dynamics in mammals versus fish. Although I compared killifish data to those of small mammals (rodents), killifish are generally smaller in size than rodents. This size difference may confound the calculated differences in G S H dynamics between fish and mammals. However, Brooks and Pong (1981) reported body weight does not affect G S H levels in mice or hamsters, suggesting metabolic alterations due to body weight are not a significant factor in the differences in G S H dynamics between fish and mammals. However, the metabolic cost of endothermy may be an important factor in these differences. Metabolic rate of mammals is 4-5 times that of size and body temperature-matched ectotherms (Hulbert and Else, 1989). In specific organs, metabolism of ectotherms (as estimated by cytochrome oxidase activity) is 50-72% of mammalian rates in the brain, 43% in the muscle and 37% in the liver (Else and Hulbert, 1981; Hulbert and Else, 1989). These differences are similar to the calculated differences of G S H levels between fish and mammals in brain and muscle of this study (50% and 38% respectively), although less so for the liver (14%). Consequently, the metabolic costs of endothermy may play a role in the difference in G S H dynamics, and particularly in G S H levels, between fish and mammals. It is important to note that endothermic animals undergo physiological and metabolic acclimation to different temperatures to maintain sufficient energy for growth, reproduction, etc. (see Hazel and Prosser, 1974). The above models demonstrate there is a correlation between temperature and G S H levels, turnover, and enzyme activity in some tissues. However, whether this is due to metabolic rate and feeding level, to physiological and biochemical adjustments to different temperatures, or a combination of the above is not known. Summary • Increased incubation temperature increased tGSH turnover and transiently increased tGSH levels of RTH-149 cells in vitro. • Increased acclimation temperature increased hepatic GPx and G R activity and tGSH levels of liver, brain and muscle, but not of R B C , heart or gills of killifish in vivo. • Temperature could account for differences in G P x activity in vivo between fish models used and mammals, but only partially explained differences in G S H turnover in vitro and G R activity and tGSH levels in vivo between fish and mammals. The ability of fish to excrete hydrogen peroxide from their gills and the metabolic cost of endothermy in mammals are likely additional factors contributing to differences in G S H dynamics between mammals and fish. 61 References Almar, M., Otero, L , Santos, C , Gallego, J .G . , 1998. Liver glutathione content and glutathione-dependent enzymes of two species of freshwater fish as bioindicators of chemical pollution. Journal of Environmental Science and Health. Part B: Pesticides, Food Contaminants, and Agricultural Wastes 33: 769-783. Anderson, M.E., 1998. Glutathione: an overview of biosynthesis and modulation. Chemico-Biological Interactions 111-112: 1-14. Ando, M., Katagiri, K., Yamamoto, S., Asanuma, S., Usuda, M., Kawahara, I., Wakamatsu, K., 1994. Effect of hyperthermia on glutathione-peroxidase and lipid peroxidative damage in liver. Journal of Thermal Biology 19: 177-185. Bergmeyer, H.U. Methods of Enzymatic Analysis. Volume VI: Metabolites I: Carbohydrates. V C H Publishers, Deerfield Beach, FL. Bols, N.C., Mosser, D.D., Steels, G.B., 1992. Temperature studies and recent advances with fish cells in vitro. Comparative Biochemistry and Physiology. Part A, Physiology 103: 1-14. Borgdanova, A., Grenacher, B., Nikinmaa, M., Gassmann, M., 2005. Hypoxic responses of Na + /K + ATPase in trout hepatocytes. Journal of Experimental Biology 208: 1793-1801. Brooks, R.R., Pong, S.F. , 1981. Effects of fasting, body weight, methylcellulose, and carboxymethylcellulose on hepatic glutathione levels in mice and hamsters. Biochemical Pharmacology 30: 589-594. Carey, H.V., Rhoads, C.A., Aw, T.Y., 2003. Hibernation induces glutathione redox imbalance in ground squirrel intestine. Journal of Comparative Physiology. B, Biochemical, Systemic, and Environmental Physiology 173: 269-276. Davidson, J.F. , Schiestl, R.H., 2001. Mitochondrial respiratory electron carriers are involved in oxidative stress during heat stress in Saccharomyces cerevisiae. Molecular and Cellular Biology 21: 8483-8489. Dziubek, K., 1987. Glutathione metabolism in selected organs of Rana temporaria L. in annual cycle and under different stressors. Part II. Starvation and temperature effects. Acta Biologica Cracoviensia ser. Zoologica 29: 55-68. Else, P.L., Hulbert, A . J . , 1981. Comparison of the "mammalian machine" and the "reptile machine": energy production. American Journal of Physiology 240: R3-R9. Fangue, N.A., Hofmeister, M., Schulte, P.M., unpublished data. Intraspecific variation in thermal tolerance and heat shock protein gene expression in common killifish, Fundulus heteroclitus. Ferraris, M., Radice, S., Catalani, P., Francolini, M., Marabini, L., Chiesara, E., 2002. Early oxidative damage in primary cultured trout hepatocytes: a time course study. Aquatic Toxicology 59: 283-296. Fridovich, I., 2004. Mitochondria: are they the seat of senescence? Aging Cell 3: 13-16. Gallagher, E.P., Di Giulio, R.T., 1992. A comparison of glutathione-dependent enzymes in liver, gills and posterior kidney of channel catfish (Ictalurus punctatus). Comparative Biochemistry and Physiology. Part C, Pharmacology, Toxicology and Endocrinology 102: 543-547. Gallagher, E.P. , Hasspieler, B.M., Digiulio, R.T., 1992. Effects of buthionine sulfoximine and diethyl maleate on glutathione turnover in the channel catfish. Biochemical Pharmacology 43: 2209-2215. Gorbi, S. , Baldini, C , Regoli, F., 2005. Seasonal variability of metallothioneins, cytochrome P450, bile metabolites and oxyradical metabolism in the European eel Anguilla anguilla L. (Anguillidae) and striped mullet Mugil cephalus L. (Mugilidae). Archives of Environmental Contamination and Toxicology 49: 62-70. Griffith, O.W., 1980. Determination of glutathione and glutathione disulfide using glutathione reductase and 2-vinylpyridine. Analytical Biochemistry 106: 207-212. Griffith, O.W., Meister, A., 1979. Potent and specific inhibition of glutathione synthesis by buthionine sulfoximine (S-n-butyl homocysteine sulfoximine). The Journal of Biological Chemistry 254: 7558-7560. 62 Griffith, O.W., Meister, A., 1985. Origin and turnover of mitochondrial glutathione. Proceedings of the National Academy of Science of the United States of America 82: 4668-4672. Hayes, J.D., McLellan, L.I., 1999. Glutathione and glutathione-dependent enzymes represent a co-ordinately regulated defense against oxidative stress. Free Radical Research 31: 273-300. Hazel, J . , Prosser, C , 1974. Molecular mechanisms of temperature compensation in poikilotherms. Physiological Reviews 54: 620-677. Hulbert, A . J . , Else, P.L., 1989. Evolution of mammalian endothermic metabolism: mitochondrial activity and cell composition. American Journal of Physiology 256: R63-R69. Kaur, M., Atif, F., Ali, M., Rehman, H., Raisuddin, S., 2005. Heat stress-induced alterations of antioxidants in the freshwater fish Channa punctata Bloch. Journal of Fish Biology 67: 1653-1665. Kidd, P.M., 1997. Glutathione: systemic protectant against oxidative and free radical damage. Alternative Medicine Review 2: 155-176. Kolesnichenko, L., Kulinsky, V., Sotnikova, G. , Kovtun, V., 2003. Influence of changes in glutathione concentration on body temperature and tolerance to cerebral ischemia. Biochemistry (Moscow) 68: 534-540. Kosenko, E., Kaminsky, M., Kaminsky, A., Valencia, M., Lee, L., Hermenegildo, C , Felipo, V., 1997. Superoxide production and antioxidant enzymes in ammonia intoxication in rats. Free Radical Research 27: 637-644. Kretzschmar, M., Muller, D., 1993. Aging, training and exercise - a review of effects on plasma glutathione and lipid peroxides. Sports Medicine 15: 196-209. Lannan, C.N. , Winton, J.R., Fryer, J.L., 1984. Fish cell lines: establishment and characterization of nine cell lines from salmonids. In Vitro 20: 671-676. Lauterburg, B.H., Adams, J.D., Mitchell, J.R., 1984. Hepatic glutathione homeostasis in the rat: efflux accounts for glutathione turnover. Hepatology 4: 586-590. Leeuwenburgh, C., J i , L., 1995. Glutathione depletion in rested and exercised mice - biochemical consequence and adaptation. Archives of Biochemistry and Biophysics 316: 941-949. Leeuwenburgh, C , J i , L., 1998. Glutathione and glutathione ethyl ester supplementation of mice alter glutathione homeostasis during exercise. Journal of Nutrition 128: 2420-2426. Lind, C., Gerdes, R., Hamnell, Y., Schuppe-Koistinen, I., von Lowenhielm, H.B., Holmgren, A., Cotgreave, I.A., 2002. Identification of S-glutathionylated cellular proteins during oxidative stress and constitutive metabolism by affinity purification and proteomic analysis. Archives of Biochemistry and Biophysics 406: 229-240. Lindstrom-Seppa, P., Roy, S., Huuskonen, S., Tossavainen, K., Ritola, O., Marin, E., 1996. Biotransformation and glutathione homeostasis in rainbow trout exposed to chemical and physical stress. Marine Environmental Research 42: 323-327. Lunn, G. , Dale, G.L., Beutler, E., 1979. Transport accounts for glutathione turnover in human erythrocytes. Blood 54: 238-244. Lushchak, V.I., Lushchak, L.P., Mota, A.A., Hermes-Lima, M., 2001. Oxidative stress and antioxidant defenses in goldfish Carassius auratus during anoxia and reoxygenation. American Journal of Physiology 280: R100-R107. Malmezat, T., Breuille, D., Capitan, P., Mirand, P.P., Obled, C , 2000. Glutathione turnover is increased during the acute phase of sepsis in rats. Journal of Nutrition 130: 1239-1246. Meister, A., 1995. Strategies for increasing cellular glutathione. In: Packer, L. Cadenas, E. (Eds.), Biothiols in Health and Disease. M. Dekker, New York, pp. 165-188. Meredith, M.J., Reed, D.J., 1982. Status of the mitochondrial pool of glutathione in the isolated hepatocyte. The Journal of Biological Chemistry 257: 3747-3753. Meyer, J . , Smith, J . , Winston, G. , Di Giulio, R., 2003. Antioxidant defenses in killifish (Fundulus heteroclitus) exposed to contaminated sediments and model prooxidants: short-term and heritable responses. Aquatic Toxicology 65: 377-395. 63 Ohtsuka, Y., Yabunaka, N., Fujisawa, H., Watanabe, I., Agishi, Y., 1994. Effect of thermal-stress on glutathione metabolism in human erythrocytes. Journal of Applied Physiology 68: 97-91. Orrenius, S., Ormstad, K., Thor, H., Jewell, S.A., 1982. Turnover and functions of glutathione studied with isolated hepatic and renal cells. Federation Proceedings 42: 3177-3188. Otto, D.M.E., Moon, T.W., 1995. 3,3',4,4'-Tetrachlorobiphenyl effects on antioxidant enzymes and glutathione status in different tissues of rainbow trout. Pharmacology and Toxicology 77: 281-287. Parihar, M.S., Dubey, A.K., Javeri, T., Prakash, P., 1996. Changes in lipid peroxidation, superoxide dismutase activity, ascorbic acid and phospholipid content in liver of freshwater catfish Heteropneustes fossilis exposed to elevated temperature. Journal of Thermal Biology 21: 323-330. Parihar, M.S., Javeri, T., Hemnani, T., Dubey, A.K., Prakash, P., 1997. Responses of superoxide dismutase, glutathione peroxidase and reduced glutathione antioxidant defenses in gills of the freshwater catfish (Heteropneustes fossilis) to short-term elevated temperature. Journal of Thermal Biology 22: 151-156. Perez-Pinzon, M.A., Rice, M.E., 1995. Seasonal- and temperature-dependent variation in C N S ascorbate and glutathione levels in anoxia-tolerant turtles. Brain Research 705: 45-52. Potter, D.W., Tran, T., 1993. Apparent rates of glutathione turnover in rat tissues. Toxicology and Applied Pharmacology 120: 186-192. Radice, S., Ferraris, M., Marabini, L , Grande, S., Chiesara, E., 2001. Effect of iprodione, a dicarboximide fungicide, on primary cultured rainbow trout (Oncorhynchus mykiss) hepatocytes. Aquatic Toxicology 54: 51-58. Rady, A.A., 1993. Effect of change in environmental temperature on antioxidant enzyme activities and lipid peroxidation in red blood cells of carp. Comparative Biochemistry and Physiology. Part B, Biochemistry and Molecular Biology 104: 695-698. Sciuto, A .M. , 1997. Antioxidant properties of glutathione and its role in tissue protection. In: Baskin, S.I. Salem, H. (Eds.), Oxidants, Antioxidants, and Free Radicals. Taylor and Francis, Washington, pp. 171-192. Scott, R.D., Hughey, R.P., Curthoys, N.P., 1993. Role of apical and basolateral secretion in turnover of glutathione in Llc-Pk(1) cells. American Journal of Physiology 265: F723-F728. Sekura, R., Meister, A., 1974. Glutathione turnover in the kidney; considerations relating to the y-glutamyl cycle and the transport of amino acids. Proceedings of the National Academy of Sciences of the United States of America 71: 2969-2972. Shertzer, H., Vasiliou, V., Liu, R., Tabor, M., Nebert, D., 1995. Enzyme-induction by L-buthionine (S,R)-sulfoximine in cultured mouse hepatoma-cells. Chemical Research in Toxicology 8: 431-436. Skibba, J.L., Powers, R.H., Stadnicka, A., Cullinane, D.W., Almagro, U.A., Kalbfleisch, J .H. , 1991. Oxidative stress as a precursor to the irreversible hepatocellular injury caused by hyperthermia. International Journal of Hyperthermia 7: 749-761. Smith, P.K., Krohn, R.I., Hemanson, G.T., Mallia, A.K., Gartner, F.H., Provenzano, M.D., Fujimoto, E.K., Goeke, N.M., Olson, B.J. , Klenk, D.C., 1985. Measurement of protein using bicinchoninic acid. Analytical Biochemistry 150: 76-85. Speers-Roesch, B., Ballantyne, J .S . , 2005. Activities of antioxidant enzymes and cytochrome c oxidase in liver of Arctic and temperate teleosts. Comparative Biochemistry and Physiology. Part A, Physiology 140: 487-494. Stephensen, E., Sturve, J . , Fbrlin, L., 2002. Effects of redox cycling compounds on glutathione content and activity of glutathione-related enzymes in rainbow trout liver. Comparative Biochemistry and Physiology. Part C, Pharmacology, Toxicology and Endocrinology 133: 435-442. Stewart, J .M. , Woods, A.K., Blakely, J.A., 2005. Maximal enzyme activities, and myoglobin and glutathione concentrations in heart, liver and skeletal muscle of the Northern Short-tailed shrew (Blarina brevicauda; Insectivora: Soricidae). Comparative Biochemistry and Physiology. Part B, Biochemistry and Molecular Biology 141: 267-273. 64 Targett, T.E., 1978. Respiratory metabolism of temperature acclimated Fundulus heteroclitus (L): Zones of compensation and dependence. Journal of Experimental Marine Biology and Ecology 32: 197-206. Wallace, K.B., 1989. Glutathione-dependent metabolism in fish and rodents. Environmental Toxicology and Chemistry 8: 1049-1055. Wilhelm-Filho, D., Gonzalez-Flecha, B., Boveris, A., 1994. Gill diffusion as a physiological mechanism for hydrogen peroxide elimination by fish. Brazilian Journal of Medical and Biological Research 27: 2879-2882. Wilhelm-Filho, D., Torres, M.A., Marcon, J.L., Fraga, C .G. , Boveris, A., 2000. Comparative antioxidant defences in vertebrates - emphasis on fish and mammals. Trends in Comparative Biochemistry And Physiology 7: 33-45. Wilhelm-Filho, D., Torres, M.A., Tribess, T.B., Pedrosa, R . C , Soares, C.H.L., 2001. Influence of season and pollution on the antioxidant defenses of the cichlid fish acara (Geophagus brasiliensis). Brazilian Journal of Medical and Biological Research 34: 719-726. Zhang, Z., Juurlink, B.H.J., Lyon, A.W., Paterson, P.G. , 2002. Effects of dietary protein and L-2-oxothiazolidine-4-carboxylate on rat brain glutathione concentration. Nutrition Research 22: 1475-1486. 65 CHAPTER 4: Glutathione dynamics and metabolism in growth hormone transgenic salmon during feeding and starvation.1 Introduction Aerobic metabolism is the main source of energy production in animals, but it is not without cost. Oxygen consumption rates have been correlated to the rate of reactive oxygen species (ROS) leakage from the mitochondria (reviewed by Muradian et al., 2002; Fridovich, 2004). If left unchecked these highly reactive species can damage key components of a cell, such as lipid, DNA and protein molecules (reviewed by Kidd, 1997). Consequently, R O S production is balanced by a suite of antioxidants, which neutralize the oxidative power of R O S . As metabolism changes, it is expected that antioxidants would adjust to changing production of R O S . One such antioxidant system involves the tripeptide glutathione (GSH) and its associated enzymes. G S H , alone or as a cofactor for the G S H peroxidase enzymes (GPx), reduces R O S and products of oxidative stress such as lipid hydroperoxides. In the process, G S H is oxidized (oxidized glutathione - G S S G ) , and is returned to its reduced form by G S S G reductase (GR). There is mounting evidence that G S H levels may be proportional to metabolic rate in vertebrates. In Chapter 3, I demonstrated that some components of the G S H antioxidant system increased with increasing acclimation temperature in killifish, although whether this was associated with increased metabolism or other physiological alterations to temperature was not determined. In addition, some antioxidant enzymes have been correlated to metabolic rate in mice (Muradian et al., 2002), and activity of metabolic enzymes in shrews {Blarina brevicauda, Stewart et al., 2005) and deep sea fish (Janssens et al., 2000), although antioxidant enzyme activity was inversely correlated to activity of metabolic enzymes in Arctic and temperate fish (Speers-Roesch and Ballantyne, 2005). To examine whether components of the G S H system are proportional to metabolic rate in fish, I examined these systems in a fish model with accelerated growth. Insertion of a growth hormone transgene in salmon results in high metabolic rates and greatly accelerated growth rates compared to non-transgenic controls (Devlin et al., 1994; Cook et al., 2000a). Consequently, growth hormone transgenic salmon provide an interesting model with which to study the relationship between the G S H antioxidant system and metabolic rate. In such animals, it has not yet been determined if high metabolic rates are an intrinsic effect of the growth hormone transgene, or due to increased feeding and growth. As well, it is not known if the antioxidant system of the fish is adjusted to, or compromised by, the transgene and associated high growth rate. In mammals, growth hormone injection increased G S H , decreased G S S G and decreased levels of the antioxidant enzymes GPx , catalase and Mn-superoxide dismutase in various tissues of Ames dwarf mice (Brown-Borg and Rakoczy, 2003), and decreased G S S G in the hippocampus of aging rats (Donahue et al., 2006). In these studies, it was not determined whether these changes were from intrinsic effects of the growth hormone, or from other effects such as increased feeding and/or growth rates. ' Oxygen consumption data published as "Leggatt, R.A., Devlin, R.H., Farrell, A.P., and Randall, D.J. 2003. Oxygen uptake of growth hormone transgenic coho salmon during starvation and feeding. Journal of Fish Biology. 62: 1053-1066. 66 To explore the relationship between G S H dynamics and metabolism using this unique experimental model, I performed two separate experiments to examined oxygen consumption rates (as a determinant of aerobic metabolism) and G S H levels and activity of associated enzymes in growth hormone transgenic and control coho salmon (Oncorhynchus kisutch). Oxygen consumption rates and G S H dynamics were measured in three main groups of coho salmon: transgenic fish fed to satiation (maximum growth rate), size-matched control fish fed to satiation, and transgenic fish fed equal rations as control fish (control growth rate). This latter group was either fed over four days (oxygen consumption rate) or throughout their lifespan (GSH dynamics). In addition, oxygen consumption rates and G S H dynamics were measured in transgenic and control fish after 4 d or 1 month starvation respectively. Oxygen consumption rates and G S H dynamics (GSH levels, oxidation, and activity of GPx, GR, and G S H synthesis and breakdown enzymes) of transgenic and control fish during feeding and starvation were measured to address the following three objectives: Objective 4.1: To determine if increased metabolic rate is intrinsic to transgenic coho salmon or simply associated with increased feeding and growth. Objective 4.2: To determine if high growth rate and/or the growth hormone transgene alters G S H dynamics in coho salmon. Objective 4.3: To determine if G S H dynamics are correlated to metabolic rate in transgenic and control coho salmon. Materials and Methods Transgenic fish used in these experiments were produced and raised in a secure and contained transgenic facility at the West Vancouver Laboratory, B.C., Canada. The growth hormone transgenic gene construct used was OnMTGHI (containing type-1 growth hormone) fused to a metallothionein-B promoter, both from sockeye salmon (Devlin et al., 1994a). The gene was inserted into coho salmon derived from Chehalis River wild stock and all subsequent generations bred with normal wild Chehalis coho salmon. Fish used in Experiment 4.1 were F 3 generation and fish used in Experiment 4.2 were M77 strain, F 4 and F 5 generation. Control fish for both experiments were non-transgenic wild Chehalis coho salmon. Transgenesis was confirmed in fish by presence of rapid growth, or by P C R analysis. Fish were cultured in oxygen-saturated, 10°C well water, with water flow greater than 1L/min/kg fish and density less than 5kg/m 3 . Fish were fed stage-specific trout food from Skretting (Vancouver, BC, Canada) twice a day until start of the experiments. All experiments took place in the autumn (September-October). Experiment 4.1: Oxygen consumption during feeding and starvation To determine the effects of the growth hormone transgene on metabolism during starvation and feeding, oxygen consumption rates of transgenic and size-matched, one year older control coho salmon were measured during four days starvation and four days fed either a fixed ration or to satiation. The experiment was performed in an eight-cylinder respirometer originally described by Duval et al. (1981) and modified from Johansen and Geen (1990). Cylinders were placed in a water bath in a climate-controlled room to maintain the water temperature at 11.5°C. Each glass cylinder was approximately 8L 67 with the inflow regulated by a computer-controlled solenoid valve. Each cylinder contained an OxyGuard oxygen probe (Point Four Systems, North Vancouver, BC , Canada) and a propeller attached to a 12V motor to maintain water flow across the membrane. To determine oxygen consumption rates of each fish, water to each cylinder was flushed for 5 min, then oxygen decline in cylinders measured every 3 min over 37 min, followed by a 5 min water flush. In this way, oxygen consumption rates for each fish were determined every 42 min for the duration of the 8 d period. Fish were monitored in separate cylinders during three 8 d periods. In total, 12 transgenic and 11 control fish were used. The fish were fed, and then placed in the cylinders (one fish per cylinder) for eight days, consisting of four days of starvation followed by four days of feeding. During the feeding period, the fish were fed twice daily at 1030 and 1600 hours. One half of the fish received a fixed ration of 0.5% body weight (BW) of food per feeding (1.0% BW/day), the other half were fed to satiation. Food pellets were counted and dropped in one at a time and satiation was presumed when the fish let two to three pellets drop to the bottom of the cylinder uneaten. Any uneaten pellets were removed and counted. At the end of the eight days, the fish were sacrificed by an overdose of anesthetic (2g/L tricaine methanesulfonate, 4g/L sodium bicarbonate) and weighed. The background oxygen consumption rate of the empty cylinders was measured for 24 hours. The average day and night background oxygen consumption rate was calculated for each cylinder and these values subtracted from the oxygen consumption rate of all the fish batches. The oxygen consumption of the cylinders accounted for 5.2 ± 4.6% of the total oxygen consumption. Dafa Analysis Oxygen consumption rates in mg oxygen per kg offish per hour were calculated as: Oxygen consumption rate = (V x A 0 2 ) / (At x BW) where V is the volume (L) of the respirometer, A 0 2 is the change in oxygen content of the water (mg/L) during measurements minus the A 0 2 of the empty respirometer, At is the total time of measurements (h), and BW is the weight (kg) of the fish. Oxygen consumption rates of fish exhibiting spontaneous activity were considered routine oxygen consumption (RMR). These were calculated by the average oxygen consumption rate over one day/night or feeding period. In order to avoid discrepancies in the oxygen consumption rates of the two groups of fish due to differences in spontaneous activity, the resting oxygen consumption rates of the fish were calculated from frequency histograms of day and night periods or of feeding periods as described by Steffensen et al. (1994). Standard metabolic rate (SMR), defined as the oxygen consumption of starved, resting fish at a constant temperature, was calculated in this same way. Experiment 4.2: Glutathione dynamics during feeding and starvation The effect of a growth hormone transgene and feeding level on total G S H levels, oxidation level, and activity of associated enzymes was determined in three groups of size-matched fish: transgenic coho salmon (F 5 generation) fed to satiation for 9 months, control coho salmon fed to satiation for 21 months, and transgenic coho salmon (F 4 generation) fed a fixed ration equal to control salmon for 21 months. F 4 generation salmon were progeny from the same female as the control salmon, but with transgenic sires. The three groups of fish were held in three separate tanks. Eight fish of each group were sacrificed by an overdose of anaesthetic as above, weight and length recorded and blood taken by caudal severage. 68 Blood was centrifuged at 6800*g for 3 min and plasma removed and frozen in liquid nitrogen. Liver, muscle, and posterior kidney were then excised, rinsed in ice-cold phosphate-buffered saline, blotted dry and frozen in liquid nitrogen. To sample intestinal mucosa, approximately one inch of posterior intestine was removed, cut open, rinsed in ice-cold phosphate-buffered saline, gently blotted dry and the mucosa scraped off the musculature with a razor. All tissues were stored at -80°C until analysis. Tissues were sampled for total G S H (tGSH = GSH+2xGSSG) , liver sampled for G S S G and activity of GR, GPx, and the synthesis enzyme y-glutamylcysteine synthetase (GCS), and posterior kidney and intestinal mucosa sampled for G S H catabolic enzyme y-glutamyltranspeptidase (yGT). After sampling, control and transgenic fish fed to satiation were combined into one tank under water conditions outlined above. Fish were starved for one month and then eight fish from each group sampled for liver, intestinal mucosa, posterior kidney and plasma as above. Tissue Preparation and Analysis G R was purchased from Roche Diagnostics (Laval, Q C , Canada), and tricaine methanesulfonate was purchased from Syndel, Canada (Vancouver, BC , Canada). All other chemicals were purchased from Sigma-Aldrich (Oakville, ON, Canada). Tissues were homogenized (muscle tissue) or sonicated (all other tissue) in one of the following buffers: 100mM Tris-HCI (pH 8.0) with 1mM ethylenediaminetetraacetic acid (EDTA), 1pM pepstatin, 1pM leupeptin, 0.15pM aprotinin, and 0.5mM phenylmethylsulfonyl fluoride protease inhibitors (intestinal mucosa, kidney), or 125mM sodium phosphate buffer (pH 7.5) with protease inhibitors (liver G S H and GR, muscle) and supplemented with 1mM dithiothreitol (liver G C S and GPx). Tissues were homogenized at an approximate ratio of 0.1g/mL and supernatant obtained by centrifugation at 11 600*g at 4°C for 5 min. Portions of supernatant were sampled for total protein, liver supernatant was sampled for G C S , G R and GPx , and posterior kidney and intestinal mucosa was sampled for yGT. The remaining supernatant or plasma was added to an equal volume of 10% sulfosalicylic acid, centrifuged at 10 000*g at 4°C for 10 min and the supernatant was removed for tGSH and G S S G analysis. All analyses were performed in triplicate on 96-well microplates and measured using a SpectraMax spectrophotometer equipped with SoftmaxPro Software (Molecular Devices Corporation). Protein content was analyzed using the bicinchoninic acid method as per Smith et al. (1985), using bovine serum albumin as standards and analyzed at 516nm. tGSH and G S S G analyses were modified from Griffith (1980) as follows. For tGSH analyses, 6|uL of triethanolamine was added per 10Q\xL of tissue or cell supernatant and G S S G standards. 10jiL of samples or G S S G standard and 200rxL of reaction mixture were added to a 96 well plate. The reaction mixture contained 0.22mM reduced B-nicotinamide adenine dinucleotide phosphate (NADPH), 0.62mM 5,5'-dithiobis(2-nitrobenzoic acid) (DTNB), 0.56% triethanolamine and 0.5U/mL G R in 125mM sodium phosphate, 6mM EDTA buffer (pH 7.4). The change in absorbance was monitored at 412nm for 5 min. For G S S G analyses, 6pL of 2-vinylpyridine (to bind reduced GSH) and 20pL of 0.1% triethanolamine in phosphate-buffered saline was added per 100pL of sample or G S S G standard. Samples and standards were mixed vigorously for 1 min and then incubated for 50 min. G S S G analyses were then as per tGSH above except that the concentration of DTNB in the reaction mixture was 1pM. 69 G R and G P x activity of liver homogenates were analyzed by procedures modified from Stephensen et al. (2002), and G C S activity was analyzed by a procedure modified from Seelig and Meister (1985). Homogenates were diluted to 5mg protein/mL with 125mM sodium phosphate, 6mM EDTA buffer (pH 7.4) containing either 0.1% bovine serum albumin (GR) or 1mM dithiothreitol (GCS, GPx). For G C S analyses, 10uL of homogenate or buffer blank was added to a 96-well plate and 100uL of a reaction mixture was added to each well for final concentrations of 100mM KCI, 60mM MgCI 2, 20mM EDTA, 0.4mM N A D P H , 6.2U/ml_ pyruvate kinase, 8.8U/ml_ lactate dehydrogenase, 10mM phosphoenol pyruvate and 10mM L-glutamic acid, in 100mM Tris-HCI (pH 8.0). Background absorbance was monitored for 5 min at 340nm, 28°C, and then 100uL reaction starter was added to each well for final concentrations of 10mM ATP , 5mM dithiothreitol and 10mM cysteine, in 100mM Tris-HCI (pH 8.0). Activity was monitored at 340nm, 28°C for 5 min, and background absorbance subtracted for final G C S activity measurement. For GR analyses 10uL of homogenate or buffer blank was added to a 96-well plate and 200uL of a reaction mixture containing 0.1 mM DTNB and 0.63mM N A D P H in buffer was added to each well. Background absorbance was measured at 405nm for 5 min. The reaction was initiated by the addition of 10uL 0.33M G S S G to each well and the reduction of DTNB was monitored at 405nm, 28°C for 5 min. For G P x analyses, 10uL of homogenate or buffer blank was added to a 96-well plate and 200uL of a reaction mixture containing 3.5mM G S H , 1mM sodium azide, 2U/mL G R and 0.12mM NADPH was added to each well. The reaction was initiated by the addition of 10pL of 0.03% hydrogen peroxide and the oxidation of NADPH was monitored at 340nm for 5 min. yGT activity of tissue homogenates were analyzed by a procedure modified from Meister et al. (1981). 10uL of tissue homogenate and 200uL of a reaction mix was added to a 96-well plate. The reaction mix contained 0.1 M glycyl-glycine, 5mM L-y-glutamic acid p-nitroanilide, and 100mM Tris-HCI (pH 8.0) at a ratio of 1:1:2 respectively. The production of p-nitroaniline was measured by absorbance change at 410nm over 5 min. Statistical Analysis Frequency histograms of oxygen consumption were analyzed with TableCurve 2-D, and all other analyses were performed with SigmaStat (both from S P S S Inc.). Paired comparisons were performed with t-tests and proportion comparisons were performed with z-tests. Multiple comparisons were performed using 1-way or 2-way A N O V A or regression analysis as appropriate, followed by Tukey's posf-hoc test. Differences were considered significant if p<0.05. All data are presented as mean ± standard error of the mean. Results Experiment 4.1: Oxygen consumption during feeding and starvation Control (33.6±1.7g) and transgenic fish (35.6±2.1g) did not differ in their initial weight (p=0.46). While in the holding tanks, the transgenic and control fish displayed very different behaviours. The control fish tended to stay near the bottom corners of the tank, but became active when disturbed. The transgenic fish tended to stay near the surface of the water and were active most of the time. As well, there was more minor fin and eye damage in transgenic fish than control fish (p=0.05). 70 Representative oxygen consumption rates over the 8 d period for a single transgenic fish fed a fixed ration, a single transgenic fish fed to satiation, and a single control fish fed to satiation, measured every 42 min, are given in Figure 4.1. A representative histogram of oxygen consumption used to determine average resting oxygen consumption rate over a day/night or feeding time is given in Figure 4.2. Oxygen consumption rates were normally distributed over the lower end of the oxygen consumption range of each day/night or feeding period, and this lower range was used to estimate resting oxygen consumption rates. Rates higher than this normally distributed range were used to estimate active oxygen consumption rates. Starved Oxygen Consumption Rates Both the transgenic and control fish displayed decreasing resting oxygen consumption during the starvation period that could be described by a hyperbolic regression curve (Figure 4.3). The two lines were significantly different from each other (F 1 1 7 8 =9.41, p=0.002) with the control fish having a higher initial oxygen consumption rate than the transgenic fish (p=0.03, Figure 4.4), as well as a steeper slope. Standard metabolic rate: Both groups of fish reached an ostensible steady state in resting oxygen consumption by the fourth day of starvation (Figure 4.3). Control and transgenic fish did not differ in resting oxygen consumption rate over the final 24 h period of starvation (100.6±5.9 and 88.5+5.6 mg 0 2 /kg/h respectively, p=0.12, n=11 for control fish and n=12 for transgenic fish). When the lowest resting oxygen consumption rate of the day and night periods was considered to account for diurnal variation, control and transgenic oxygen consumption rates did not differ (p=0.33). These latter data were used as an estimate of the standard metabolic rate (SMR) of the fish (Figure 4.4). This S M R was similar to, although slightly higher than, those extrapolated from other similar-sized salmonids (60-70 mg 0 2 /kg/h, Beamish, 1964; Brett, 1964; McLean et al., 1993). Routine oxygen consumption rate: Routine starved metabolic rate (RMR) was estimated by an average of all oxygen consumption rates over the last 24 hours of starvation. Control fish had a significantly higher R M R when compared with transgenic fish (p=0.04, Figure 4.4). The scope for spontaneous activity, defined as the range of oxygen consumption of starved fish exhibiting spontaneous activity, was estimated over the last 24 hours of starvation. The scope for spontaneous activity was greater for control (176.1±23.4 mg 0 2 /kg/h) than for transgenic fish (114.4±9.3 mg 0 2 /kg/h, p=0.02). Transgenic and control fish did not differ in the minimum oxygen consumption rate of this range (p=0.60), but control fish had a higher maximum oxygen consumption rate than transgenic fish (p=0.02, Figure 4.4). Feeding Metabolism During the feeding period, the transgenic fish fed at a fixed level ate slightly, but not significantly, less than the ration level of 1% BW/day (0.75±0.07% BW/day, p=0.19, n=6). The transgenic fish fed to satiation ate significantly more than those on a fixed ration, averaging 1.81 ±0.21% BW/day (p=0.03, n=6). The control fish had low feeding levels during the experiment, with several fish eating only once or not at all and only four fish eating consistently. Those fish offered a fixed ration ate significantly less than 1.0% BW/day (0.27±0.08% BW/day, p=0.005, n=6). Those fed to satiation ate 0.10±0.05% BW/day (n=5), and the total feed consumption of the control fish was 0.19±0.08% BW/day (n=11). For both feeding regimens, the transgenic fish ate significantly more than the control fish (p<0.001). To calculate feeding 71 -starved- •fed-o G o G O o G <D W) o 400 300 200 100 0 400 300 200 100 0 400 300 200 100 0 lis i i i B I P ! Bwa BIBI! sftg? i l i l i i i * tu H i 111 time (day / night) Figure 4.1: Oxygen consumption rates of representative coho salmon over 4 days starvation and 4 days feeding, measured every 42 min. A) A growth hormone transgenic fish fed 1% body weight/day, B) a growth hormone transgenic fish fed to satiation, and C) a control fish fed 1% body weight/day. White bars indicate day, grey bars indicate night. 72 6 5 -SMR--RMR--active-o a* 3 2 H X: 100 120 140 160 180 oxygen consumption rate (mg 02/kg/h) 200 Figure 4.2: Frequency histogram of oxygen consumption rates of a representative coho salmon over one day/night or feeding period. Horizontal lines indicate data used to calculate resting/standard metabolic rate (SMR), routine metabolic rate (RMR), and active metabolic rate. 200 -J—» 0$ •2 x S ^ O CO c S <u M >. o O Control • Transgenic 150 100 20 40 60 80 time (h from start of starvation) 100 Figure 4.3: Resting oxygen consumption rates of growth hormone transgenic and control salmon during a four day acclimation and starvation period, calculated every day/night period. White/grey bars indicated day/night respectively. Inverse regression equations: transgenic fish: oxygen consumption rate = 84.7 + 480.9/time (R 2=0.12, n=12); control fish: oxygen consumption rate = 89.3 + 1117.7/time (R2=0.37, n=11). Significant differences among groups are indicated by differing letters (a,b,c). 73 300 i Figure 4.4: Oxygen consumption rates of growth hormone transgenic and control salmon. Initial metabolic rate (initial) was calculated over the first 12 h of transfer to respirometers (control: n=11, transgenic n=12). Standard metabolic rate (SMR), routine metabolic rate (RMR), minimum and maximum routine metabolic rates (min R M R and max R M R respectively) were calculated from the last 24 h of a four day starvation period (control: n=11, transgenic n=12). Metabolic rates when fed 0.5% body weight/feeding (0.5% BW) and when fed to satiation (satiation) were calculated from oxygen consumption rates over 6 h after each feeding (n=5-11). * indicates significant difference between transgenic and control fish. 74 metabolic rate, fish that ate at low feeding levels, i.e. two times or less over the four day period, were eliminated from the data. This consisted of three control fish fed a fixed ration, three control fish fed to satiation and one transgenic fish fed to satiation. Oxygen consumption rate at a fixed ration level: Only two of 12 control fish ate at 1.0% BW/day (or 0.5% BW/feeding). The resting oxygen consumption rate when fed at a fixed ration was 1.4 times greater for transgenic fish than control fish (p=0.03, Figure 4.5). Oxygen consumption rate when fed to satiation: As the control fish fed at a fixed level were not confined by their ration level (i.e., they consumed significantly less feed than the level provided to them), these data were also used to analyze metabolism when fed to satiation. The resting oxygen consumption rate of the transgenic and control fish fed to satiation as a function of feeding level is shown in Figure 4.5. Control fish increased oxygen consumption with increasing feeding levels to a maximum oxygen consumption rate of approximately 130 mg 0 2 /kg/h at 0.90% BW/feeding. Transgenic fish had a peak oxygen consumption rate of approximately 180 mg 0 2 /kg/h at 1.4% BW/feeding, over a feeding range of 0 to 2.5% BW/feeding. The average oxygen consumption rate when fish were fed to satiation was greater for transgenic fish than for the control fish by a factor of 1.7 (p<0.001, Figure 4.4). When compared over a common feeding range of 0 to 0.9% BW/feeding (Figure 4.5) transgenic fish had 1.4 times greater oxygen consumption rate than control fish when they had ingested the same %BW of food. In control fish, the average oxygen consumption rate in fish fed to satiation was significantly lower than the initial acclimation oxygen consumption rate when fish were first placed in the respirometer (p<0.001, Figure 4.4). In transgenic fish, there was no significant difference between initial oxygen consumption and oxygen consumption when fed to satiation (p=0.15, Figure 4.4). At low levels of feeding, some control fish had oxygen consumption rates less than that of their measured standard metabolic rate. When the oxygen consumption rate of control fish that had eaten two or less times over the 4 d period was examined (n=8), excluding periods when the fish had eaten, the decline in average oxygen consumption did not plateau until the start of day six. The resting, unfed oxygen consumption rate of these fish from day five to eight, taking into consideration diurnal variation, was 73.8±6.2 mg 0 2 /kg/h. This was not significantly different from the calculated S M R of the transgenic fish (p=0.41) or the calculated S M R value of the control fish (p=0.06). The routine oxygen consumption of the control fish calculated in the same manner (104.9±9.3 mg 0 2 /kg/h) did not differ from the calculated RMR of the transgenic fish (p=0.35), but was significantly smaller than the calculated RMR of the control fish (p=0.01). Experiment 4.2: Glutathione dynamics during feeding and starvation Weight and length of transgenic and control coho salmon under different feeding and starvation regimes are given in Table 4.1. There were no differences in weight between any group offish (p=0.86). However, transgenic fish fed to satiation were significantly smaller in length than transgenic fish fed a fixed ration or starved control fish (p=0.02). Tissue tGSH Levels and GCS Activity tGSH levels in the liver of transgenic fish fed to satiation were significantly higher than control fish (p<0.001), whereas tGSH levels of transgenic fish fed a fixed ration did not significantly differ from either 75 300 i 0 -J , , , , , 0.0 0.5 1.0 1.5 2.0 2.5 food consumption level (% BW/feeding) Figure 4.5: Resting oxygen consumption rates after feeding of growth hormone transgenic and control coho salmon at different feeding levels. Regression equations for transgenic fish: oxygen consumption rate = 121.8 + 90.1x(food consumption level)-35.3x(food consumption level) 2, R2=0.18. For control fish: oxygen consumption rate = 84.8 + 50.7x(food consumption level), R2=0.23. Table 4.1: Weight and length of fish from Experiment 4.2: growth hormone transgenic coho salmon fed to satiation, fed a fixed ration or starved for one month, and control coho salmon fed to satiation or starved for one month. Significant differences in length among groups are indicated by differing letters (a,b). There were no significant differences in weight among any fish groups. n=8. Group Weight (g) Length (cm) __ Transgenic satiation 72.9±5.0 17.1±0.4 a Control satiation 81.0±5.3 18 .6±0 .4 a b Transgenic fixed ration 74.3±6.5 19.0±0.5 b Starved Transgenic satiation 73.U6.0 18 .6±0 .5 a b Control satiation 76.1±6.4 19.1±0.4 b 76 transgenic or control fed to satiation (Figure 4.6A). After one month starvation, liver tGSH levels decreased to a similar extent in both transgenic and control fish (56 and 44% respectively), although levels were no longer significantly different from each other. In the posterior kidney, there was no effect of the transgene or ration level on tGSH levels during starvation or feeding (p=0.21, Figure 4.6B), and starvation significantly increased tGSH levels in both transgenic and control fish (5.4 and 4.5 times respectively, p<0.001). There was no significant effect of the transgene or ration level on intestinal mucosa tGSH level (p=0.05, Figure 4.6C). Starvation increased tGSH levels in the mucosa overall (p=0.002) although the only significant differences were between transgenic fish fed to satiation and starved control fish. Transgenic fish fed to satiation had significantly greater plasma tGSH levels than both fed control fish and transgenic fish fed a fixed ration (p<0.001, Figure 4.6D). After one month starvation, plasma tGSH levels of transgenic fish fell to similar levels as fed and starved control fish. There was no significant effect of starvation on plasma tGSH levels in control fish. Muscle tGSH levels were significantly higher in transgenic fish fed to satiation than control fish or transgenic fish fed a fixed ration (p<0.001, Figure 4.6E). Liver G C S activity was significantly lower in transgenic fish fed to satiation than control fish or transgenic fish fed a fixed ration (p=0.002, Figure 4.7), although transgenic and control groups did not differ after one month starvation. Liver GSSG, GSH and GR, GPx Activity The liver of transgenic fish fed to satiation had lower G S S G (as % of tGSH) than control fish during feeding and starvation, although this difference was only significant during starvation (p<0.001, Figure 4.8A). G S S G of transgenic fish fed a fixed ration did not differ from control fish or transgenic fish fed to satiation. Reduced G S H levels (tGSH - 2 * G S S G ) in the liver were similar to tGSH levels, with the exception that transgenic fish fed a fixed ration had significantly lower G S H levels than transgenic fish fed to satiation (Figure 4.8B). This indicates that the differences observed between fish groups are primarily due to increased G S H , and not G S S G levels. G R activity in the liver was significantly greater in transgenic fish fed to satiation and after one month starvation than all other groups during feeding and starvation (p<0.001, Figure 4.9A). There was no effect of starvation on G R activity in either group of fish (p=0.06). Liver G P x activity did not differ between fish groups during feeding or during starvation (p=0.58, Figure 4.9B), although starvation significantly decreased G P x activity in both transgenic and control fish (p<0.001). Intestinal Mucosa and Posterior Kidney yGT Activity yGT activity in the intestinal mucosa of transgenic salmon fed to satiation was significantly greater than that of control salmon or transgenic salmon fed a fixed ration (p<0.001, Figure 4.1 OA). Starvation decreased intestinal yGT in both transgenic and control salmon, and fish groups did not differ in yGT activity during starvation. Posterior kidney yGT activity did not differ between any fish groups during feeding (p=0.16, Figure 4.10B). Kidney yGT activity decreased in both groups off ish after one month starvation, although this was only significant for control fish. Relationship between oxygen consumption rate and GSH dynamics To determine if metabolic rates and G S H dynamics were correlated in control and transgenic coho salmon, resting oxygen consumption rates from Experiment 4-1 were compared to components of 77 30 -I 25 -2^0 i r, 15 ==io H A * be JL 10 -I 8 -6 -4 -Transgenic satiation ab J L Transgenic satiation be J_ Control satiation b I ab J _ Control satiation ac 12 -j 10 -X OO o m u 3 4 2 J Transgenic rationed ab Transgenic rationed B 35 3^0 X g20 « 15 B S 10 * 5^  o Transgenic satiation D a Transgenic satiation Control satiation Control satiation • Fed a Starved Transgenic rationed Transgenic rationed 2.5 2.0 X oo 0 1.5 -+-< 1 1.0 a 0.5 E ^ 0.0 'a .2 Ml .2 c ca l-l H 75 c 2 .2 c a O • r; O T 3 (D 5 i-i H Figure 4.6: Total G S H levels (nmol/mg protein) in A) liver, B) posterior kidney, C) intestinal mucosa, D) plasma, and E) white muscle in the following groups of control and growth hormone transgenic coho salmon: transgenic salmon fed to satiation, control salmon fed to satiation, transgenic salmon fed an equal ration as control salmon, transgenic salmon starved for one month, and control salmon starved for one month. Significant differences among groups, within tissues, are indicated by differing letters (a,b,c). n=8. 78 • § 2 5 n —^» o & 2 0 60 6 | B 0 0 O O 15 10 ab Transgenic satiation ab Control satiation Transgenic rationed Figure 4.7: y-Glutamylcysteine synthetase (GCS) activity in the liver of the following groups of control and growth hormone transgenic coho salmon: transgenic salmon fed to satiation, control salmon fed to satiation, transgenic salmon fed an equal ration as control salmon, transgenic salmon starved for one month, and control salmon starved for one month. Where letters differ (a,b) significant differences among fish groups exist. n=8. «S 8 1 ac J - ^ a Transgenic satiation be J L b 1 Control satiation abc _x ^ 3 0 c '53 o ^ 2 0 ^ 25 Transgenic rationed •3 15 B X oo 5 0 B a X. be Transgenic satiation be J L c Control satiation Transgenic rationed Figure 4.8: A) Oxidized (GSSG) and B) reduced (GSH) glutathione in liver of the following groups of control and growth hormone transgenic coho salmon: transgenic salmon fed to satiation, control salmon fed to satiation, transgenic salmon fed an equal ration as control salmon, transgenic salmon starved for one month, and control salmon starved for one month. Where letters differ (a,b,c) significant differences among fish groups exist. n=8. 79 1.6 1.2 .g 1.8 o a l . 4 OJO S .5 1-0 J | 0 .8 l 0 . 6 S0 .4 ^ 0 2 O 0.0 a J L Transgenic satiation b Control satiation • S 160 o I-I Transgenic rationed 140 M 1 2 0 I 1 0 0 S 80 1 60 E 40 20 0 x o B a T Transgenic satiation Control satiation Transgenic rationed Figure 4.9: A) Glutathione reductase (GR) and B) glutathione peroxidase (GPx) activities in the liver of the following groups of control and growth hormone transgenic coho salmon: transgenic salmon fed to satiation, control salmon fed to satiation, transgenic salmon fed an equal ration as control salmon, transgenic salmon starved for one month, and control salmon starved for one month. Where letters differ (a,b) significant differences among fish groups exist. n=8. 10 H 8 a u 6 _c %-» to 3 4 G Transgenic satiation Control satiation c 25 i 20 H o C S 10 Transgenic rationed B a ab Transgenic satiation Control satiation Transgenic rationed Figure 4.10: y-Glutamyltranspeptidase activity (yGT - nmol/min/mg protein) in A) intestinal mucosa and B) posterior kidney of the following groups of control and growth hormone transgenic coho salmon: transgenic salmon fed to satiation, control salmon fed to satiation, transgenic salmon fed an equal ration as control salmon, transgenic salmon starved for one month, and control salmon starved for one month. Where letters differ (a,b,c) significant differences among fish groups, within tissues, exist. n=8. 80 the G S H system from Experiment 4-2 (Figure 4.11). Oxygen consumption rates of fed and starved control and transgenic coho salmon were in the order of: starved control fish < starved transgenic fish < control fish fed to satiation < transgenic fish fed a fixed ration < transgenic fish fed to satiation. When components of the G S H system were ranked in a similar manner and this ranking compared to the oxygen consumption ranking, liver tGSH (Figure 4.11 A) and G S H levels, and intestinal yGT activity (Figure 4.11B) positively corresponded to oxygen consumption rates, and intestinal tGSH (p=0.02, Figure 4.11C) negatively corresponded to oxygen consumption rates. There were no apparent relationships between oxygen consumption rates and any other measured component of the G S H system in control and transgenic salmon. Discussion Oxygen consumption during feeding and starvation High oxygen consumption rates in growth hormone transgenic coho salmon were due to the effects of feeding, and not to an increase in basal metabolism. When transgenic coho salmon were fed to satiation to fuel the accelerated growth, they had metabolic rates 1.7 times greater than control fish fed to satiation. Similar differences in metabolic rates have been reported for growth hormone transgenic and control Atlantic salmon (Salmo salar) fed to satiation (Stevens et al., 1998; Cook et al., 2000a). However, when the oxygen consumption rates of fish fed equal levels were compared, transgenic fish had 1.4 times greater metabolic rate than control fish. This increase in oxygen consumption rate may be due to an increased ability of the transgenic fish to digest, absorb and process food. Growth hormone injections result in increased intestinal amino acid and lipid absorption (Menella, 1988; Sun, 1990) and possibly gluconeogenesis, protein synthesis and lipid turnover in fish (Oommen and Johnson, 1998). As well, growth hormone transgenic salmon have greater gut surface areas (Stevens et al., 1999; Stevens and Devlin, 2000, 2005), which could increase intestinal absorption. Raven et al. (2006) found growth hormone transgenic coho salmon had higher feed conversion efficiency and were more efficient at utilizing dietary protein and energy than control fish. Cook et al. (2000c) found growth hormone transgenic Atlantic salmon had greater ability to process food, although they do not differ from control fish in the extent to which they digest protein, dry matter or energy. An increased ability to process and absorb food would result in a larger proportion of the food eaten being digested by the fish and less being passed through the intestine undigested. Consequently the transgenic fish's post-prandial oxygen demand would increase at a fixed feeding level to process the larger proportion of food being digested, with a consequent increase in feeding metabolic rate above that of control fish. Transgenic fish did not differ from controls in SMR, indicating the growth hormone transgene alone does not affect metabolic rates in coho salmon. Transgenic fish also had similar RMR to controls after 6 d starvation, although they had lower RMR after 4 d starvation. In contrast, Cook et al. (2000a, 2000b) found growth hormone transgenic Atlantic salmon had greater routine oxygen consumption after 24 hours and through most of eight weeks of starvation. The discrepancies between studies may be attributed to differences in fish interactions and activity level between experiments, as well as differences in acclimation ability of control and transgenic fish. In Cook's experiments, oxygen consumption was 81 200 150 100 50 bO O bO 3, - d a t-l a o c o o c bo o 0 150 -100 -50 -0 150 100 50 0 J Oxygen consumption rate I I -• I GSH component T B C x o £ O GO I "3 5 bO oJ co +-» c. oo 1-1 H o U a o 0 3 &o -a ID c o c bO.y CO •<-> 0 3 p i l - i " H '3 .2 C « 03 CO S-l H 30 25 ^ 53 i n E o CO is 15 bO 3 10= g 4 o g E—i '53 0 ° -C bO '-£ 3 CQ j - 1 1 o •S B c 10 8 6 '53 ffi o CO 9 M cu 3 • 3 ^ u 3 .3 o Figure 4.11: Correlation between oxygen consumption rate (n=6-12) and A) liver total G S H , B) intestinal y-glutamyltranspeptidase (yGT), and C) intestinal total G S H (n=8) in growth hormone transgenic and control coho salmon during feeding and starvation. Note that oxygen consumption rates and G S H components were measured on separate groups of fish. 82 measured in groups of fish, whereas in my experiment oxygen consumption of single fish was measured. Increased fin and eye damage observed in our study indicates that the transgenic fish were more aggressive than control fish when reared in a group (Moutou et al., 1998). As well, the observed behaviour of the fish in the holding tanks (i.e. transgenic fish actively swimming at the surface and the control fish remaining at the bottom and only becoming active when disturbed) indicates that transgenic fish were actively searching for food most of the time, whereas control fish spent more time hiding. The growth hormone transgene appears to override predator avoidance behaviour of the fish by increasing feeding motivation. The increased routine oxygen consumption rate observed in transgenic Atlantic salmon in previous experiments may be due in part to increased activity and cohort interactions of these fish. As well, species differences and strain-to-strain variation in growth hormone gene incorporation may also be responsible for dissimilarities in RMR between studies (Devlin, 1997; Reichhardt, 2000). The higher RMR as well as scope for spontaneous activity of the control fish in this study was unexpected, as they were less active than transgenic fish when observed in the holding tanks. Although RMR of single fish of a comparable size have not been measured, it was expected that fish in this study would have lower RMR than those of grouped fish, as there would be no fish-to-fish interactions. Transgenic fish had a lower RMR, although the control fish had a similar RMR to those previously measured in grouped fish (approximately 140 mg 0 2 /kg/h, McLean et al., 1993; Cook et al., 2000a). High RMR in control fish, as well as high initial oxygen consumption upon placement in the respirometers of control fish, indicate the fish had additional metabolic costs of acclimation to a new environment. The control fish were from a wild stock, and consequently may have been less able to acclimate to artificial culture conditions, as seen in their holding tank behaviour. Although the transgenic salmon were originally from the same stock and controls, the increased feeding motivation of these fish may have more quickly overcome the stress of acclimating to the respirometer. The increased aerobic metabolic rate of growth hormone transgenic coho salmon appears to be a function of greater feeding levels and increased ability to process and absorb ingested food, and not to an additional basal metabolic cost of the transgene. However, transgenic fish had a decreased metabolic cost of acclimation to the experimental environment compared with control fish that may mask potential difference in the basal metabolic rate between the two groups off ish. Glutathione dynamics during feeding and starvation Both feeding level and the growth hormone transgene affected G S H dynamics in coho salmon. Transgenic fish fed to satiation had increased liver, muscle and plasma tGSH, decreased liver G S S G levels and G C S activity, increased liver GR activity and increased intestinal yGT activity than control fish fed to satiation, but did not differ from controls in kidney and intestinal t G S H , liver G P x activity or kidney yGT activity. With most G S H components, differences between transgenic and control fish were due to increased feeding and growth rate, and not necessarily to intrinsic effects of the transgene, as transgenic fish fed control rations had similar levels to fed control fish, and starvation eliminated significant differences between transgenic and control fish. These results are consistent with a mammalian study that found mice with increased growth rates had increased kidney and liver G S H , but 24 h starvation removed this increase (Kolatej et al., 1998). The high tGSH levels in transgenic fish in various tissues were likely due to increased supply of amino acid precursors due to high feeding levels. Transgenic fish 83 fed to satiation had higher activity of the G S H breakdown enzyme yGT per mg protein or per unit intestine (data not shown), as well as greater gut surface areas (Stevens and Devlin, 2000, 2005). High intestinal yGT activity would increase the ability of transgenic fish to metabolize dietary G S H to its transportable amino acid precursors and increase supply of precursors for G S H synthesis within tissues. In Chapter 2, it was demonstrated that G S H synthesis capacity of rainbow trout was limited by supply of precursors, and not necessarily to activity of the G S H synthesis enzyme G C S . Consequently, increased levels of G S H within various tissues of growth hormone transgenic fish was likely due to increased supply of precursors from high food intake and intestinal yGT activity. In the liver, elevated tGSH levels were associated with both the transgene and increased growth rate and feeding, as transgenic fish fed at the control level had tGSH levels between those of transgenic and control fish fed to satiation, and starvation did not greatly diminish the difference between transgenic and control liver tGSH levels. Surprisingly, the activity of G C S did not correspond to tGSH levels, as transgenic fish fed to satiation had lower activity than control and transgenic fish fed equal rations. This indicates that increased G S H levels were due solely to increased amino acid precursors and not synthesis rates. However, liver gene expression of G C S as measured by DNA microarray, although not significantly different between groups, was consistently in the order of: transgenic fish fed to satiation > transgenic fish fed the control ration > control fish fed to satiation (Rise et al., 2006). Increased liver tGSH in transgenic fish was likely due to both increased amino acid supply and increased synthesis capacity via G C S . G S H is the major redox regulator in cells (Arrigo, 1999), and thus the percent of total G S H in oxidized form can indicate tissue redox balance. In mammals, G S S G is less than 1% of tGSH in healthy cells, and levels of 10-50% can indicate oxidative stress (Forman and Liu, 1997; Arrigo, 1999; Pastore et al., 2003). Transgenic fish were expected to have higher % G S S G than control fish due to higher metabolic rates and potentially higher R O S production. However, transgenic fish had lower % G S S G than controls during feeding to satiation and starvation, and transgenic fish fed a fixed ration had % G S S G in between levels of control and transgenic fish fed to satiation. Decreased % G S S G in transgenic fish was likely due to observed higher G R activity, as G R is responsible for returning G S S G to its reduced and antioxidant active state. These data indicate that transgenic fish were able to compensate for potential metabolically-induced oxidative shifts in the G S H redox balance. Indeed, high G S H and low G S S G suggest transgenic fish fed to satiation were better able to maintain liver redox status than control fish, and maintained this advantage after one month starvation. In contrast, G P x activity was not affected by feeding level or the transgene, indicating this enzyme is either insensitive to growth hormone and accelerated growth, or basal levels are sufficient to cope with increased metabolic rate of transgenic fish. The effect of the growth hormone transgene on tGSH levels in this study are consistent with Brown-Borg and Rakoczy (2003), who found growth hormone injection in Ames dwarf mice increased G S H and decreased G S S G levels in liver, muscle and brain. However, unlike the current study, growth hormone administration decreased yGT activity of Ames mice in several tissues, but had inconsistent effects on G C S activity, indicating the cause of the high G S H levels was decreased G S H catabolism and not increased synthesis in this organism (Brown-Borg et al., 2005). As well, growth hormone injection decreased G P x and did not affect G R activity in Ames mice (Brown-Borg and Rakoczy, 2003), although 84 the effects of high feeding and growth were not separated from other effects of growth hormone in this study. Differences between the Ames mice studies and the current study indicate the effects of growth hormone on G S H dynamics depends on species type or whether growth hormone was injected or increased by transgenesis. Correlation between intestinal yGT and oxygen consumption rates suggest coho salmon could maintain high G S H levels during increased metabolic rate by increasing the ability to metabolize G S H precursors from the diet. Starvation decreased liver tGSH, G S H and GPx, and intestinal and kidney yGT in both transgenic and control fish, and decreased plasma tGSH in transgenic fish. Similar effects were reported in several studies in fish, reptiles and mammals that found starvation decreased G S H levels in various tissues (Dziubek, 1987; Ogasawara et al., 1989; Hum etal. , 1991; Gallagher et al., 1992; Di Simplicio et al., 1997; Szkudelski et al., 2004; Leggatt et al., 2006). In contrast to the current study, previous studies found starvation increased % G S S G or lipid peroxidation levels, increased G P x activity, and had inconsistent effects on G R activity (Ogasawara et al., 1989; Di Simplicio et al., 1997; Pascual et al., 2003; Morales et al., 2004; Altan et al., 2005). High G S S G and lipid peroxidation in previous studies suggest starvation results in oxidative stress in most organisms. Although lipid peroxidation or other end products of oxidative stress were not measured, starvation did not alter % G S S G in the liver of coho salmon, indicating fish maintained redox balance in this tissue. Lower metabolic rates observed during starvation in both control and transgenic fish could result in decreased reactive oxygen species production and consequently decrease demand for antioxidants in this species. Surprisingly, kidney tGSH levels increased with starvation in both transgenic and control fish. This may be due to decreased activity of kidney yGT, and consequently decreased catabolism of G S H in this tissue. It may also be due to unusually low levels of kidney tGSH during feeding and not to high levels during starvation. Posterior kidney tGSH of other salmonids are generally around 10nmol/mg protein (Chapter 2), whereas fed levels in this study were 4nmol/mg protein or lower. The cause of this is unclear, but may be due to unidentified environmental or physiological factors affecting the fish. The effects of metabolic rate on G S H dynamics of coho salmon are difficult to distinguish from those of the growth hormone transgene, feeding, and growth rate. Consequently, determining which of these factors directly influences G S H dynamics is challenging. As well, oxygen consumption rates fluctuate greatly over short periods of time, depending on activity levels (see Figure 4.1). G S H levels are not expected to correlate with these fluctuations, but may be related to standard metabolic rate under different conditions. Transgenic fish fed to satiation had higher feeding levels, higher oxygen consumption rates and higher levels of G S H components than control fish, and starvation decreased oxygen consumption rates and levels of G S H components in both control and transgenic fish, indicating there is a relationship between feeding level, metabolic rate and G S H dynamics. However, only increased liver G S H and intestinal yGT and decreased intestinal tGSH corresponded to increased oxygen consumption rates in transgenic and control coho salmon. In the case of intestinal t G S H , this was likely due to increased G S H catabolism by high yGT levels, and not directly due to high metabolism. Transgenic fish fed a fixed ration had higher metabolic rates, without differing from growth and feeding rate of control fish and are therefore useful in determining direct effects of metabolic rate on G S H dynamics. The higher metabolic rate of rationed transgenic fish did not correspond to higher G S H 85 components than control fish in any measured parameter, although rationed transgenic fish had similar liver tGSH levels than transgenic fish fed to satiation. This suggests that metabolic rate alone did not affect most components of the G S H system. Liver tGSH levels appear to correspond most consistently to metabolic rates in transgenic and control fish. This is consistent with a correlation between acclimation temperature and liver tGSH levels in killifish (Chapter 3). The liver is involved in GSH-dependent detoxification, and exports G S H to the plasma for transport to other tissues in mammals (Kaplowitz et al., 1995) and likely fish (Chapter 2). As such, the liver is expected to be more sensitive to changes in G S H demand by factors that alter metabolic rate (e.g. temperature and growth rates). It is important to note that fish measured for G S H dynamics differed in several ways from those measured for oxygen consumption rate. Fish measured for G S H components were twice as large and were fed a fixed ration or starved for longer periods of time than those measured for oxygen consumption rates, and G S H components were measured on fish held in a group whereas oxygen consumption was measured on fish held singly. All of these factors may have affected both metabolism and G S H dynamics. Consequently, the relationships between G S H and metabolism, or lack thereof, should be interpreted with caution. However, comparisons between these two groups are useful for examining general trends in metabolic rates and G S H dynamics. The effect of metabolic rate on liver G S H levels is particularly intriguing and requires further examination. Individually, these experiments both demonstrate that changes in metabolism and most components of the G S H system in growth hormone transgenic fish are not due to direct effects of increased growth hormone, but are associated with increased feeding and accelerated growth. A similar trend was observed in Chapter 3, where increased tissue tGSH levels were equally correlated in acclimation temperature and feeding rates in killifish. High G S H levels and G R enzyme activity likely relieves oxidative stress brought on by high metabolic and growth rates in transgenic fish. Whether the relationship between the metabolic rate and G S H levels is cause and effect or both due to increased feeding and growth is difficult to discern. Indeed, the effects of a growth hormone transgene on growth rate, feeding level, metabolism and G S H dynamics are likely inextricably linked. Summary • Increased metabolic rates of growth hormone transgenic coho salmon were due to increased feeding and ability to process food, and not to an intrinsic effect of the transgene and high growth hormone levels. • High levels of G S H and associated enzyme activity in transgenic salmon were due solely to increased feeding and growth in most tissues, whereas in the liver there appeared to be additional intrinsic effects of the transgene. • Metabolic rate alone corresponded to liver tGSH levels and intestinal yGT activity, but did not correspond to other components of the G S H antioxidant system in transgenic and control salmon. However, high feeding and growth rates of transgenic salmon were associated with both increased metabolism and components of the G S H antioxidant system. 86 References Altan, O., Oguz, I., Erbayraktar, Z., Yimaz, 0 . , Bayraktarl, H., Sis, B., 2005. The effect of prolonged fasting and refeeding on G S H - P x activity, plasma glucose and cholesterol levels and welfare of older hens from different genotypes. Archiv Fur Geflugelkunde 69: 185-191. Arrigo, A . -P . , 1999. Gene expression and the thiol redox state. Free Radical Biology & Medicine 27: 936-944. Beamish, F.W.H., 1964. Respiration of fishes with special emphasis on standard oxygen consumption. II. Influence of weight and temperature on respiration of several species. Canadian Journal of Zoology 42: 177-188. Brett, J.R., 1964. The respiratory metabolism and swimming performance of young sockeye salmon. Journal of the Fisheries Research Board of Canada 21: 1183-1226. Brown-Borg, H.M., Rakoczy, S .G. , 2003. Growth hormone administration to long-living dwarf mice alters multiple components of the antioxidative defense system. Mechanisms of Ageing and Development 124: 1013-1024. Brown-Borg, H.M., Rakoczy, S .G. , Uthus, E.O., 2005. Growth hormone alters methionine and glutathione metabolism in Ames dwarf mice. Mechanisms of Ageing and Development 126: 389-398. Cook, J.T., McNiven, M.A., Sutterlin, A .M. , 2000a. Metabolic rate of pre-smolt growth-enhanced transgenic Atlantic salmon {Salmo salar). Aquaculture 188: 33-45. Cook, J.T., Sutterlin, A .M. , McNiven, M.A., 2000b. Effect of food deprivation on oxygen consumption and body composition of growth-enhanced transgenic Atlantic salmon (Salmo salar). Aquaculture 188: 47-63. Cook, J.T., McNiven, M.A., Richardson, G.F., Sutterlin, A .M. , 2000c. Growth rate, body composition and feed digestibility/conversion of growth-enhanced transgenic Atlantic salmon (Salmo salar). Aquaculture 188: 15-32. Devlin, R., Yesakl, T., Biagl, C , Donaldson, E., 1994. Extraordinary salmon growth. Nature 371: 209-210. Devlin, R.H., 1997. Transgenic salmon: generation and use. In: Houdebine, L.M. (Eds.), Transgenic Animals: Generation and Use. Hardwood Academic Publishers, France, pp. 105-117. Di Simplicio, P., Rossi, R., Falcinelli, S., Ceserani, R., Formento, M.L., 1997. Antioxidant status in various tissues of the mouse after fasting and swimming stress. European Journal of Applied Physiology 76: 302-307. Donahue, A .N . , Aschner, M., Lash, L.H., Syversen, T., Sonntag, W.E. , 2006. Growth hormone administration to aged animals reduces disulfide glutathione levels in hippocampus. Mechanisms of Ageing and Development 127: 57-63. Duval, W.S. , Fink, R.P., Geen, G.H. , 1981. An aquatic respirometer for use in sublethal studies. Water Research 15: 817-822. Dziubek, K., 1987. Glutathione metabolism in selected organs of Rana temporaria L. in annual cycle and under different stressors. Part II. Starvation and temperature effects. Acta Biologica Cracoviensia ser. Zoologica 29: 55-68. Forman, H.F., Liu, R.-M., 1997. Glutathione cycling in oxidative stress. In: Clerch, B. Massaro, D.J. (Eds.), Oxygen, Gene Expression And Cellular Function. M. Dekker, New York, pp. 99-121. Fridovich, I., 2004. Mitochondria: are they the seat of senescence? Aging Cell 3: 13-16. Gallagher, E.P., Hasspieler, B.M., Digiulio, R.T., 1992. Effects of buthionine sulfoximine and diethyl maleate on glutathione turnover in the channel catfish. Biochemical Pharmacology 43: 2209-2215. Griffith, O.W., 1980. Determination of glutathione and glutathione disulfide using glutathione reductase and 2-vinylpyridine. Analytical Biochemistry 106: 207-212. Hum, S., Robitaille, L., Hoffer, L.J., 1991. Plasma glutathione turnover in the rat: effect of fasting and buthionine sulfoximine. Canadian Journal of Physiology and Pharmacology 69: 581-587. Janssens, B.J. , Childress, J . J . , Baguet, F., Rees, J . -F. , 2000. Reduced enzymatic antioxidative defense in deep-sea fish. The Journal of Experimental Biology 203: 3717-3725. 87 Johansen, J.A., Geen, G.H. , 1990. Sublethal and acute toxicity of the ethylene glycol butly ether ester formulation of triclopyr to juvenile coho salmon Onhcorhyncus-kisutch. Archives of Environmental Contamination & Toxicology 19: 610-616. Kaplowitz, N., Ookhtens, M., Lu, S . C . - L , 1995. Sinusoidal efflux of hepatic G S H : significance, physiology, and regulation. In: Packer, L. Cadenas, E. (Eds.), Antioxidants in Health and Disease, Biothiols in Health and Disease. M. Dekker, New York, pp. 213-229. Kidd, P.M., 1997. Glutathione: systemic protectant against oxidative and free radical damage. Alternative Medicine Review 2: 155-176. Kolajaj, A., Ryihska, J . , Flak, P., 1998. The influence of selection on stress reactivity in mice. Part VI. Effect of starvation and T 3 injection on the level of glutathione in liver and kidney. Journal of Animal Breeding and Genetics 115: 407-411. Leggatt, R.A., Scheer, K.W., Afonso, L.O.B., Iwama, G.K., 2006. Triploid and diploid rainbow trout do not differ in their stress response to transportation. North American Journal of Aquaculture 68: 1-8. McLean, W.E. , Jensen, J.O.T., Alderdice, D.F., 1993. Oxygen consumption rates and water flow requirements of Pacific salmon (Oncorhynchus spp.) in the fish culture environment. Aquaculture 109: 281-313. Meister, A., Tate, S .S. , Griffith, O.W., 1981. g-Glutamyl transpeptidase. Methods in Enzymology 77: 237-253. Menella, J.Y., 1988. Effets de I'administration d'hormone de croissance ovine sur I'utilisation metabolique des nutriments chez la truite arc-en-ciel (Salmo gairdneri R.). In: (Eds.), Lab. nutrition des poissons: Institut national dela recherche agronomique. Saintp-pee-sur-nivelle, France, pp. 67. Morales, A .E . , Perez-Jimenez, A., Hidalgo, M . C , Abellan, E., Cardenete, G. , 2004. Oxidative stress and antioxidant defenses after prolonged starvation in Dentex dentex liver. Comparative Biochemistry and Pysiology. Part C, Pharmacology, Toxicology and Endocrinology 139: 153-161. Moutou, K.A., McCarthy, I.D., Houlihan, D.F., 1998. The effect of ration level and social rank on the development of fin damage in juvenile rainbow trout. Journal of Fish Biology 52: 756-770. Muradian, K.K., Utko, N.A., Mozzhukhina, T.G. , Litoshenko, A.Y., Pishel, I.N., Bezrukov, V.V., Fraifield, V .E . , 2002. Pair-wise linear and 3D nonlinear relationships between the liver antioxidant enzyme activities and the rate of body oxygen consumption in mice. Free Radical Biology and Medicine 33: 1736-1739. Ogasawara, T., Ohnhaus, E.E., Hoensch, H.P., 1989. Glutathione and its related enzymes in the small intestinal-mucosa of rats - effects of starvation and diet. Research in Experimental Medicine 189: 195-204. Oommen, O.V., Johnson, B., 1998. Metabolic effects of ovine growth hormone in a teleost, Anabas testudineus. In: Unknown (Eds.), Trends in Comparative Endocrinology and Neurobiology: From Molecular to Interactive Biology, pp. 380-381. Pascual, P., Pedrajas, J.R., Toribio, F., Lopez-Barea, J . , Peinado, J . , 2003. Effect of food deprivation on oxidative stress biomarkers in fish (Sparus aurata). Chemico-Biological Interactions 145: 191-199. Pastore, A., Federici, G. , Bertini, E., Piemonte, F., 2003. Analysis of glutathione: implication in redox and detoxification. Clinica Chimica Acta 333: 19-39. Raven, P.A., Devlin, R.H., Higgs, D.A., 2006. Influence of dietary digestible energy content on growth, protein and energy utilization and body composition of growth hormone transgenic and non-transgenic coho salmon (Oncorhynchus kisutch). Aquaculture 254: 730-747. Reichhardt, T., 2000. Will souped up salmon sink or swim? Nature 406: 12. Rise, M.L., Douglas, S .E . , Sakhrani, D., Williams, J . , Ewart, V.K., Rise, M., Davidson, W.S. , Koop, B.F., Devlin, R.H. 2006 (supplemental material). Multiple microarray platforms utilized for hepatic gene expression profiliing of growth hormone transgenic coho salmon with and without ration restrition. Journal of Molecular Endocriniology. In press. Seelig, G.F. , Meister, A., 1985. Glutathione biosynthesis - gamma-glutamylcysteine synthetase from rat-kidney. Methods in Enzymology 113: 379-390. 88 Smith, P.K., Krohn, R.I., Hemanson, G.T., Mallia, A.K., Gartner, F.H., Provenzano, M.D., Fujimoto, E.K., Goeke, N.M., Olson, B.J. , Klenk, D.C., 1985. Measurement of protein using bicinchoninic acid. Analytical Biochemistry 150: 76-85. Speers-Roesch, B., Ballantyne, J .S. , 2005. Activities of antioxidant enzymes and cytochrome c oxidase in liver of Arctic and temperate teleosts. Comparative Biochemistry and Physiology. Part A, Physiology 140: 487-494. Steffensen, J . , Bushnell, P., Schurmann, H., 1994. Oxygen consumption in four species of teleosts from Greeland: no evidence of metabolic cold adaptaion. Polar Biology 14: 49-54. Stephensen, E., Sturve, J . , Forlin, L , 2002. Effects of redox cycling compounds on glutathione content and activity of glutathione-related enzymes in rainbow trout liver. Comparative Biochemistry and Pysiology. Part C, Pharmacology, Toxicology and Endocrinology 133: 435-442. Stevens, E.D., Devlin, R.H., 2000. Intestinal morphology in growth hormone transgenic coho salmon. Journal of Fish Biology 56: 191-195. Stevens, E.D., Devlin, R.H., 2005. Gut size in GH-transgenic coho salmon is enhanced by both the GHtransgene and increased food intake. Journal of Fish Biology 66: 1633-1648. Stevens, E.D., Sutterlin, A., Cook, T., 1998. Respiratory metabolism and swimming performance in growth hormone transgenic Atlantic salmon. Canadian Journal of Fisheries & Aquatic Sciences 55: 2028-2065. Stevens, E.D., Wagner, G.N. , Sutterlin, A., 1999. Gut morphology in growth hormone transgeic Atlantic salmon. Journal of Fish Biology 55: 517-526. Stewart, J .M. , Woods, A.K., Blakely, J.A., 2005. Maximal enzyme activities, and myoglobin and glutathione concentrations in heart, liver and skeletal muscle of the Northern Short-tailed shrew (Blarina brevicauda; Insectivora: Soricidae). Comparative Biochemistry and Physiology. Part B, Biochemistry and Molecular Biology 141: 267-273. Sun, L , 1990. Effect of bovine growth hormone on fish growth and intestinal amino acid absorption. Dissertation Abstracts International Part B, The Sciences and Engineering 51: 172. Szkudelski, T., Okulicz, M., Bialik, I., Szkudelska, K., 2004. The influence of fasting on liver sulfhydryl groups, glutathione peroxidase and glutathione-S-transferase activities in the rat. Journal of Physiology and Biochemistry 60: 1-6. 89 CHAPTER 5 Glutathione levels, oxidation and activity of associated enzymes during an acute heat stress in rainbow trout in vivo and in vitro Introduction Fish live in a variable environment where factors such as temperature can fluctuate on a daily and seasonal basis. There are many physiological systems in place to maintain homeostasis in such environments. The antioxidant glutathione (GSH) and its associated enzymes represent one such system that maintains the redox balance of cells. In Chapter 3, I demonstrated that G S H levels, turnover and activity of GSH-associated enzymes in some tissues were altered proportionally to acclimation temperature changes in fish. Increasing temperature may result in an increase in metabolically produced reactive oxygen species, and hence increase the demand for the reducing power of G S H . Whether short-term changes may also challenge the redox status of cells and hence challenge the G S H system has not been well examined in fish. Acute temperature changes beyond the normal range a fish is exposed to results in the fish entering a stressed state. This may affect G S H dynamics in several ways. Stress can increase oxygen consumption (see Wendelaar Bonga, 1997) which can increase production of reactive oxygen species by the mitochondria (Muradian et al., 2002). This would increase demand for antioxidants such as G S H . G S H may also be affected by heat stress through interactions with components of the generalized stress response, such as the stress hormones. Fish respond to a variety of stressors, such as acute temperature change, by eliciting a generalized stress response that includes increasing plasma hormones such as the adrenocorticotropic hormone, catecholamines and corticosteroids (see Barton and Iwama, 1991). These in turn result in metabolic, hematological and other physiological changes (see Barton and Iwama, 1991) which act to bring about a new homeostasis within the organism. Glucocorticoids and catecholamines have been reported to increase G S H synthesis and activity of antioxidant enzymes such as glutathione peroxidase (GPx), superoxide dismutase, catalase and glucose-6-phosphate dehydrogenase in the liver and lungs of various organisms, including fish (Ignatius and Oommen, 1990; Lu et al., 1992; Speck et al., 1993; Walther et al., 1996). Glucocorticoids and epinephrine have been reported to decrease G S H levels in the liver (Simmons et al., 1991) and in cultured brain cells (Patel et al., 2002) of mammals. Stress hormones may regulate G S H dynamics by direct effects on expression of G S H enzymes (Lu et al., 1992; Voss et al., 1996), or G S H dynamics may be altered to reduce oxidative products of stress hormone metabolism (Baez et al., 1997). In fish, G S H levels are altered in response to stressors that directly threaten the redox balance of a cell, such as oxidative or toxicant stressors (Bell and Cowey, 1990; Otto and Moon, 1995; Almar et al., 1998; Pena-Llopis et al., 2001; Stephensen et al., 2002; Hughes and Gallagher, 2004). However, few studies have examined the effect of acute heat stress on G S H dynamics in fish. Acute heat stress in catfish (Heteropneustes fossilis) decreased G S H levels and G P x activity in gills (Parihar et al., 1997), and increased lipid peroxidation in liver (Parihar et al., 1996) during large temperature increases, but not 90 during more moderate changes. In the snakehead (Channa punctata), heat stress decreased G S H levels and G R activity in the kidney and gills, decreased G S H levels and increased G R activity in the liver, and increased lipid peroxidation in all three tissues (Kaur et al., 2005). The effect of an increase in temperature on G S H dynamics is better examined in endotherms, and indicates acute heat stress alters G S H dynamics, although the nature of this alteration is inconsistent between studies (Harris et al., 1991; Skibba et al., 1991; Kondo et al., 1993; Ohtsuka et al., 1994; Arechiga et al., 1995; Mirochnitchenko et al., 1995; Mahmoud and Edens, 2003). In addition, heat stress increases lipid peroxidation levels and hydrogen peroxide production from the mitochondria (Ohtsuka et al., 1994; Arechiga et al., 1995; Davidson and Schiestl, 2001), indicating heat stress-induced changes in G S H dynamics are the result of oxidative stress. Alterations in G S H dynamics in response to non-oxidative or toxicant stressors such as heat may indicate that G S H has a role in the generalized response to stress in fish. To test the hypothesis that the G S H antioxidant system is altered by heat stress in fish, either by interactions with components of the generalized stress response or by indirect oxidative stress, I examined G S H dynamics in various rainbow trout models (Oncorhynchus mykiss) after exposure to acute heat stress. Rainbow trout were exposed to abrupt temperature changes within the temperature tolerance zone of rainbow trout in vivo and within the lethal zone in vitro (as defined by Bols et al., 1992), and the effect on G S H levels, oxidation, and activity of associated enzymes (GPx and glutathione reductase - GR) was measured. In addition, I compared these effects to those of an acute oxidative stress in vitro to address the following objectives: Objective 5.1: To determine if an acute heat stress alters G S H dynamics in rainbow trout either in vivo or in vitro. Objective 5.2: To determine if changes in G S H dynamics in response to a heat stress match changes in response to an oxidative stress in a rainbow trout hepatoma cell line in vitro. Materials and Methods A rainbow trout hepatoma cell line (RTH-149) was purchased from the American Type Culture Collection (ATCC#: CRL-1710, Manassas, VA, USA). Rainbow trout were purchased from Sun Valley Trout Farms (Mission, BC , Canada). Cell culture medium for RTH-149 cells was purchased from Invitrogen Corporation (Burlington, O N , Canada). Glutathione reductase (GR) was purchased from Roche Diagnostics (Laval, Q C , Canada), antibodies were purchased from Stressgen Corporation (Victoria, BC, Canada), tricaine methanesulfonate was purchased from Syndei, Canada (Vancouver, BC, Canada), and all other chemicals were purchased from Sigma-Aldrich (Oakville, O N , Canada). Experiment 5.1 took place in the summer (in vivo, August) and spring (in vitro, April). Experimental Series 5.1: GSH dynamics in response to a heat stress in vivo and in vitro In Vivo The effects of an acute heat stress on G S H levels and activity of associated enzymes in vivo was determined with the following experiment. 25g rainbow trout were held in eight 70L tanks, supplied with 13°C aerated, dechlorinated City of Vancouver municipal fresh water. Four tanks of fish were then transferred to 70L tanks containing 23°C aerated static water (heat group), and four tanks of fish were 91 transferred to identical tanks containing 13°C aerated static water (control group) for two hours and then returned to their original holding tanks (tanks per treatment: n=4). Four fish per tank were sampled pre-stress (-2 h), and 0, 6, 12 and 24 h post-stress (fish per tank: m=4). Sampled fish were sacrificed by overdose of anesthetic (2g/L tricaine methanesulfonate, 4g/L sodium bicarbonate). Blood was taken by caudal puncture, centrifuged at 2900xg for 3 minutes, plasma removed and red blood cells (RBC) frozen in liquid nitrogen. The liver, posterior kidney, spleen and gill were then excised and frozen in liquid nitrogen. All samples were stored at -80°C and analyzed for total G S H (tGSH = G S H + 2x 0xidized G S H - G S S G ) within one month. In addition, liver of fish from one tank was analyzed for G R and GPx (n=4). In Vitro - Primary Culture of Hepatocytes To examine the effects of an acute heat stress on G S H levels and oxidation in vitro, hepatocytes were cultured from a 500g rainbow trout as per Mommsen et al. (1994). Cells were plated on 6-well plates at 1.5x106cells/well in 2mL sodium bicarbonate-buffered L-15 media supplemented with 1% penicillin, streptomycin, and amphotericin B. Cells were incubated at 18°C in air, and media changed every 48 h. 96 h after hepatocyte isolation, the medium was changed on all cells, and one half of the plates moved to a 30°C incubator (heat group) and the other half moved the same distance and then returned to the 18°C incubator (control group). After one hour all plates were moved and returned to the 18°C incubator. Cells were sampled immediately before heat stress (-1 h) and 0, 4, 8, and 24 h post-heat stress. Cells were scraped into a 2mL tube and gently mixed to evenly distribute cells within media. 10pL of cell solution was then added to 500pL trypan blue for viability analysis. The remaining cell solution was centrifuged at 10 OOOxg for 20 sec, washed once in ice-cold phosphate-buffered saline and remaining cell pellet frozen on dry ice. Cells were analyzed for tGSH, G S S G , total protein (n=6) and viability (n=4). Experiment 5.2: GSH dynamics in response to heat and oxidative stress in vitro To compare the effects of an acute heat stress with that of an acute oxidative stress on G S H dynamics in rainbow trout in vitro, RTH-149 cells were used. RTH-149 cells were cultured in 18.3mM H E P E S and 0.75% sodium bicarbonate buffered Dulbecco's Modified Eagle's Medium (DMEM) with 10% fetal calf serum (FCS), 0.1mM non-essential amino acids and 1% antibiotic/antimycotic (penicillin, streptomycin, amphotericin B), pH 7.4. During growth, cells were held at 20°C in air. For the following experiments cells were plated in 6-well plates at a density of 2.5x105 cells/2ml_/well and grown for 3-4 days until approximately 80% confluent. Serum (e.g. FCS) is sometimes removed from the media during exposure to stressors in cultured cells (e.g. Barhoumi et al., 1996; Baliharova et al., 2003; Shi and Zheng, 2005), in part to prevent unknown elements in the serum from interacting with chemical stressors or influencing the cellular stress response. However, serum withdrawal reportedly induces apoptosis, alters antioxidant levels, and increases expression of stress proteins such as heat shock protein 70 (Hsp70) in mammalian cell lines (Mampuru et al., 1996; Kim et al., 2000; Lewis and Hughes-Fulford, 2000). To determine whether temporary serum deprivation influences the stress response of RTH-149, I compared tGSH and Hsp70 levels after heat and oxidative stressors in cells incubated with or without F C S . Four days after plating RTH-149 cells in 6-well plates, the cell medium was changed. Half the cells received normal media (with FCS) and the other half received normal media lacking F C S (no FCS) . Cells were then further divided 92 into control, heat and oxidative groups as follows: 1) control group cells remained in a 20°C incubator and received 20uL D M E M per well, 2) heat group cells were moved to a 33°C incubator, and 3) oxidative group cells received 20uL D M E M supplemented with hydrogen peroxide at final concentration of 10mM. One hour later, all groups were sampled (3 wells per F C S and stressor group). All cell samples were stored at -80°C until analysis for tGSH and Hsp70 (n=3). Based on the results of this experiment, the following experiments were performed with serum during exposure to stressors. To determine if the changes in G S H dynamics after a heat stress in vitro were similar to changes brought about by oxidative stress, RTH-149 cells were exposed to acute heat and oxidative stressors. Three days after plating RTH-149 cells in 6-well plates, the medium was changed and 4 h later cells received control, heat, or oxidative treatments as described above. After the treatment, the medium was changed in the oxidative group to remove residual hydrogen peroxide, and all cells were returned to the 20°C incubator. To allow time for sampling and medium change, treatments were staggered. The control group was treated first, followed by the heat group 15 min later and the oxidative group 2.5 h later. Cells were sampled immediately after the start of the stressors (0 h), 30 min into the stressors (0.5 h) and 0, 1, 5, and 18 h after the end of the stressors (1, 2, 6, and 19 h respectively). 6 wells per treatment and time were sampled. To sample wells, 100uL of the medium was removed and frozen on dry ice for lactate dehydrogenase (LDH) analysis. Cells were then scraped into 2mL tubes, centrifuged at 11 600*g for 20 sec, 100pL medium retained for LDH analysis and the cell pellet and medium frozen on dry ice. All samples were stored at -80°C until analysis for the following: cell protein, LDH, tGSH, G S S G , and GR, and medium LDH (n=6). Sample Preparation and Analysis In vivo tissues for tGSH analysis were sonicated in 5% sulfosalicylic acid at a ratio of 0.1g/mL and supernatant obtained by centrifugation at 10 000*g at 4°C for 10 min. All other tissues and cells were sonicated in 125mM sodium phosphate buffer (pH 7.5) with 1mM ethylenediaminetetraacetic acid (EDTA), 1pM pepstatin, 1uM leupeptin, 0.15uM aprotinin, and 0.5mM phenylmethylsulfonyl fluoride protease inhibitors. Homogenate buffer for GPx analyses was supplemented with 1mM dithiothreitol. Supernatant was obtained by centrifugation at 10 000*g at 4°C for 5 min. Portions of supernatant were removed for protein, LDH, GR, GPx , and Hsp70 analyses, and the remaining supernatant added to an equal volume of 10% sulfosalicylic acid and supernatant for G S H analyses obtained by centrifugation at 10 000xg at 4°C for 10 min. Supernatant for Hsp70 analysis was added to an equal volume of 2 times Laemmli's buffer (20% glycerol, 4% sodium dodecyl sulfate, 0.25% bromophenol blue, 10% B-mercaptoethanol in 0.5M Tris-HCI), boiled for 3 min and stored at -20°C. All analyses except Hsp70 were performed in triplicate on 96-well microplates and measured using a SpectraMax spectrophotometer equipped with SoftmaxPro Software (Molecular Devices Corporation). Protein content was analyzed using the bicinchoninic acid method using Smith et al. (1985), with bovine serum albumin standards and analyzed at 516nm. LDH activities in RTH-149 cell homogenate and media were analyzed as previously described (Bergmeyer, 1985). In brief, cell homogenate was diluted 11 times in 100mM H E P E S , pH 7.4. 10uL of diluted cell homogenate or 20uL of media were added to a 96-well plate, with appropriate H E P E S or media blanks. 0.4mM reduced B-nicotinamide adenine dinucleotide (NADH) in H E P E S buffer was added 93 to each well for a final volume of 190uL. The plates were covered and incubated at 28°C for 15 min for cell homogenate and 30 min for media. Background absorbance decline was then read for 5 min at 340nm at 28°C. The reaction was initiated by the addition of 10pL of 28mM pyruvate and oxidation of NADH to N A D + was measured by absorbance decline at 340nm for 5 min at 28°C. tGSH and G S S G analyses were modified from Griffith (1980) as follows. For tGSH analyses, 6jaL of triethanolamine was added per 100|uL of tissue or cell supernatant and G S S G standards. 10u.L of samples or G S S G standard and 200(iL of reaction mixture were added to a 96 well plate. The reaction mixture contained 0.22mM reduced 3-nicotinamide adenine dinucleotide phosphate (NADPH), 0.62mM 5,5'-dithiobis(2-nitrobenzoic acid) (DTNB), 0.56% triethanolamine and 0.5U7ml_ G R in 125mM sodium phosphate, 6mM EDTA buffer (pH 7.4). The change in absorbance was monitored at 412nm for 5 min. For G S S G analyses, 6pL of 2-vinylpyridine (to bind reduced GSH) and 20pL of 0.1% triethanolamine in P B S was added per 100pL of sample or G S S G standard. Samples and standards were mixed vigorously for one minute and then incubated for 50 min. G S S G analyses were conducted as described above except that the concentration of DTNB in the reaction mixture was 1 pM. G R and G P x activities were analyzed by procedures modified from Stephensen et al. (2002). Homogenates were diluted to 5mg protein/mL (in vivo) or 1.6mg/mL (in vitro) with 125mM sodium phosphate, 6mM EDTA buffer (pH 7.4) containing either 0.1% bovine serum albumin (for G R analysis) or 1mM dithiothreitol (for G P x analysis). For GR analyses 10pL of homogenate or buffer blank was added to a 96-well plate and 200pL of a reaction mixture containing 0.1 mM DTNB and 0.63mM NADPH in buffer was added to each well. Background absorbance was measured at 405nm for 5 min. The reaction was initiated by the addition of 10pL 0.33M G S S G to each well and the reduction of DTNB was monitored at 405nm, 28°C for 5 min. For G P x analyses, 10pL of homogenate or buffer blank was added to a 96-well plate and 200pL of a reaction mixture containing 3.5mM G S H , 1mM sodium azide, 2U/mL GR and 0.12mM N A D P H was added to each well. The reaction was initiated by the addition of 10pL of 0.03% H Y D R O G E N P E R O X I D E and the oxidation of NADPH was monitored at 340nm for 5 min. Hsp70 was analyzed by S D S - P A G E and Western blotting as per Forsyth et al. (1997), with the following modifications. 20pL of cell homogenate diluted in Laemmli's buffer to 0.5mg protein/mL was added to gel wells in duplicate. Cell homogenate from RTH-149 cells 18 h after exposure to a 33°C heat stress was added to each gel as a reference sample. After transfer of separated proteins onto nitrocellulose membrane, the membrane was incubated with rabbit polyclonal anti-chinook salmon Hsp70 primary antibody, followed by alkaline phosphatase-conjugated goat-anti rabbit secondary antibody. Final band intensity was determined using SigmaGel ( S P S S Inc.) and Hsp70 data reported relative to the reference value. Data were analyzed by 1 or 2-way A N O V A as appropriate, followed by Tukey's post-hoc test, using SigmaStat ( S P S S Inc.). If normality or equal variance failed, data were square root or inversely transformed before analyses. If normality or equal variance was still not satisfied, Kruskal-Wallis non-parametric A N O V A on ranks was used, followed by Dunn's post-hoc comparison test. In Experiment 5.1 in vivo the data from all fish were averaged (m=4) and these averages used in data analysis (n=4). Differences were considered significant if p<0.05. All data are presented as mean ± standard error. 94 Results Experimental Series 5.1: GSH dynamics in response to a heat stress in vivo and in vitro In Vivo There were few effects of a 2 h, 23°C (+10°C, ambient temperature 13°C) heat stress on tGSH levels and enzyme activity of rainbow trout in vivo. Liver tGSH levels were lower overall in heat-stressed fish (p=0.02) although they did not differ from control fish at any one time point (p=0.40, Figure 5.1). Kidney tGSH levels were not affected by heat stress (p=0.65), although they increased with time overall (p=0.01, Table 5.1). Neither heat stress, nor time altered tGSH levels in R B C (p=0.79, 0.15), gill (p=0.45, 0.77) or spleen (p=0.91, 0.13, respectively, Table 5.1). In a subsample of liver tissues, heat stress significantly increased G P x activity overall (p=0.046), although there was no significant difference between control and heat-stressed groups at any one time point (Figure 5.2A). G P x activity decreased in both control and heat-stressed fish over time (p<0.001). There was no effect of heat stress or time on liver G R activity in rainbow trout (p=0.07, 0.10, respectively, Figure 5.2B). In Vitro There was a significant interaction between time and a 1 h, 30°C (+15°C, ambient temperature 15°C) heat stress on tGSH levels of cultured rainbow trout hepatocytes in vitro (p=0.024, Figure 5.3A). Heat stress transiently increased tGSH levels in hepatocytes 4 h after the end of the stressor (p<0.001), although levels in both control and heat stress cells were higher at 24 h post-stress compared to pre-stress levels. There was no effect of heat stress on G S S G as percent of tGSH (p=0.56, Figure 5.3B). There was a significant effect of time on % G S S G (p=0.03), although there were no individual differences between time points (p=0.10). The pattern of reduced G S H (tGSH - 2 * G S S G ) did not differ greatly from tGSH (Table 5.2) and heat stress transiently increased G S H levels at 4 h post-heat stress (p<0.001). Percent viability of hepatocytes as measured by trypan blue exclusion was unaffected by heat stress (p=0.67) and did not vary over time (p=0.25, Table 5.2). Experiment 5.2: GSH dynamics in response to heat and oxidative stress in vitro 1 h serum withdrawal during heat and oxidative stressors altered baseline levels of tGSH and Hsp70, and the response to stress in RTH-149 cells. There was a significant interaction between stressor treatment and F C S withdrawal on tGSH levels (p=0.01, Figure 5.4A). F C S withdrawal increased tGSH levels in the control and oxidative groups (10mM hydrogen peroxide) 1.8-fold, but not in the heat-stressed (33°C, +13°C, ambient temperature 20°C) group. Cells exposed to oxidative stress had lower tGSH levels than control cells, regardless of F C S . However, cells exposed to heat stress had lower tGSH than control cells only during F C S withdrawal. There was a significant interaction between stress treatment and F C S withdrawal on Hsp70 levels (p=0.04, Figure 5.4B). F C S withdrawal increased Hsp70 12-fold in the control group, 7-fold in the oxidative group, and 4-fold in the heat group. There were no differences in Hsp70 among stressor groups within F C S treatment. In RTH-149 cells exposed to a 1 h control, heat, or oxidative treatment (as above), tGSH levels fluctuated over time with no clear pattern (Figure 5.5A). There was a significant interaction between time and treatment (p=0.006). There were no differences in tGSH levels among treatment groups at any one time point, although tGSH levels in the oxidative group generally increased post-stress. There was a 95 Figure 5.1: Total G S H levels in the liver of rainbow trout after a 2 h, 23°C heat stress (+10°C, control 13°C). Grey bar indicates time of heat stress. Although liver tGSH levels were lower in heat-stressed than control fish (p=0.02), there were no individual differences at any one time point. n=4, m=4. Table 5.1: Tissue total G S H levels (umol/g tissue) in rainbow trout exposed to a 2 h, 23°C heat stress (+10°C, control 13°C). Time is h from end of heat stress. Although there was a significant effect of time on kidney tGSH levels (p=0.01), there were no individual differences between groups in this or any other tissue. n=4, m=4. Kidney R B C Gill Spleen time Control Heat Control Heat Control Heat Control Heat Pre 1.84±0.07 1.80±0.05 1.52±0.10 1.73±0.06 0.99±0.05 1.08±0.04 1.44±0.07 1.44±0.05 0 1.89±0.06 1.77±0.10 1.75±0.09 1.77±0.15 1.13±0.05 1.10±0.04 1.67±0.09 1.45±0.11 6 1.66±0.24 1.93±0.06 1.79±0.11 1.53±0.08 1.06±0.02 1.09±0.03 1.70±0.07 1.59±0.01 12 2.04±0.16 1.93±0.04 1.73±0.11 1.73±0.22 1.08±0.05 1.14±0.02 1.60±0.08 1.70±0.02 24 2.12±0.07 2.21±0.06 1.98±0.10 1.90±0.15 1.06±0.02 1.06±0.05 1.51±0.10 1.53±0.05 96 Figure 5.2: A) Glutathione peroxidase (GPx) and B) glutathione reductase (GR) activity in the liver of rainbow trout after a 2 h, 23°C heat stress (+10°C, control 13°C). Grey bar indicates time of heat stress. Significant differences in G P x activity are indicated by differing letters (a,b,c,d,e). There were no significant differences in G R activity. n=4. 97 o s 0 J 1 1 1 1 1 1 0 5 10 15 20 25 time (h after end of heat stress) Figure 5.3: A) Total G S H and B) oxidized G S H (GSSG) in a primary culture of rainbow trout hepatocytes after a 1 h, 30°C heat stress (+15°C, control 15°C). Grey bar indicates time of heat stress. Significant differences are indicated by differing letters (a,b,c). There were no significant differences in G S S G . n=6. 98 Table 5.2: Reduced G S H (nmol/mg protein, n=6) and viability (% of cells excluding trypan blue, n=4) in a primary culture of rainbow trout hepatocytes after a 1 h, 30°C heat stress (+15°C, control 15°C). Significant differences in G S H levels are indicated by differing letters (a,b,c). There were no significant differences in cell viability. G S H Viability Time (h post-stress) Control Heat Control Heat Pre-stress 19.0±1.0 a 73.0±2.6 0 24.8±3.4 a b c 23.8±1.6 a b c 66.5±2.6 66.7+1.1 4 20.6±1.2 a c 31.9±0.7 b 65.5±5.5 65.4±6.4 8 25.6±1.1 a b c 26.3±2.6 a b c 73.6±1.4 70.5±3.9 24 28.0±2.3 b o 30.1±1.8 b 63.0±4.0 59.7±9.9 FCS noFCS FCS noFCS Figure 5.4: A) Total G S H and B) Hsp70 levels in RTH-149 cells exposed to a 1 h heat (33°C, +13°C) or oxidative (10mM hydrogen peroxide) stress (control 20°C), with or without fetal calf serum (FCS) in the medium. Cells were sampled at the end of the 1 h stressor period. Significant differences among treatment and F C S groups are indicated by differing letters (a,b,c,d). n=3. 99 120 c '55 o •— ft so £ O 100 8 0 ^ 60 £ s 40 H 2 0 ab I a i i acl i f o XS ca ab & JL ac I. a o a 0 0 a 3 0 20 H B 1 0 a x a aT a nti i 1 M I I Control • • • Heat i i Oxidative -r a a T a T a JUT a ' © H—< o S -ft SO -a 100 i 80 4 60 40 20 1 ab I u JO "i f ab 0 30 60 duration of stressor (min) y I ab ab I 6 18 recovery time (h from end of stressor) o ca T ° Figure 5.5: A) Total G S H , B) oxidized G S H (GSSG) , and C) reduced G S H in RTH-149 cells exposed to 1 h heat (33°C, +13°C), oxidative (10mM hydrogen peroxide), or control (20°C) treatment. Significant differences are indicated by differing letters (a,b,c). n=6. 100 significant interaction between treatment and time on G S S G as percent of tGSH (p<0.001, Figure 5.5B). Cells exposed to oxidative stress had % G S S G 10 times greater than control and 20 times greater than heat-stressed fish at the end of the stressor (p<0.001), although there were no other differences among time or treatment. Heat stress did not significantly affect % G S S G in RTH-149 cells. Reduced G S H followed similar patterns as tGSH (Figure 5.5C). There was a significant interaction between time and treatment on G R activity in RTH-149 cells (p<0.001, Figure 5.6). G R increased with time in both control and heat-stressed cells, although heat-stressed cells increased G R activity at an earlier time than control cells (0 and 1 h post-stress respectively). RTH-149 cells in the oxidative group had higher G R activity than control or heat groups initially. However, cells in the oxidative group were staggered up to 2.5 h behind the control group and if levels were compared in real time (e.g. 0 h in oxidative group versus 2 h in control and heat groups) there were no differences among treatment groups initially. Unlike control and heat groups, cells in the oxidative group did not increase G R activity with time and had lower GR activity than control and heat groups at 19 h. There was a significant interaction between treatment and time on percent LDH released (p<0.001, Figure 5.7). LDH release increased greatly in cells during and after exposure to an oxidative stress, and transiently increased after a heat stress. Data transformations failed to bring normality or equal variance to the data and consequently a non-parametric test was used to analyze the data (Dunn's). Using this conservative test, the only significant differences were that the oxidative group had higher LDH released at 18 h post-stress than any time group in the control group and at 18 h in the heat group. Discussion Heat stress resulted in few alterations in G S H dynamics in rainbow trout in vivo or in vitro. In rainbow trout, acute heat stress of the magnitude used here (23°C, +10°C above ambient temperature of 13°C) resulted in a 6-fold increase in plasma Cortisol, a 2-fold increase in plasma glucose, and a 6-fold increase in liver Hsp70 (Afonso, unpublished data). However, acute heat stress did not significantly alter tissue tGSH levels or activity of hepatic G R or GPx of rainbow trout in vivo at any time point measured. Consequently, G S H is likely not affected by components of the generalized stress response in rainbow trout. In vitro, heat stress resulted in a transient increase in G S H levels in a primary culture of rainbow trout hepatocytes (30°C, +15°C above ambient temperature of 15°C), and transiently increased GR activity in a rainbow trout hepatoma cell line (RTH-149, 33°C, +13°C above ambient temperature of 20°C), but did not greatly alter G S H levels or oxidation of RTH-149 cells. In addition, G S H dynamics in RTH-149 cells exposed to an acute heat stress were not altered in parallel to those of RTH-149 cells exposed to an acute oxidative stress (10mM hydrogen peroxide). G S H levels in hepatocytes and GR activity in RTH-149 cells increased with time in both control and heat-stressed groups, and the transient increase in G S H or G R activity post-heat stress was not significantly different from control levels measured 4 h and 1 h later, respectively. In both cases, heat stress did not necessarily increase the magnitude of G S H levels or G R activity, but accelerated a natural increasing trend within the cells. Heat stress may have accelerated metabolic changes with time in primary and cell line culture that resulted in altered G S H dynamics. 101 g —-O — & 12 i 10 H & 4 8 2 Pi O Control Heat Oxidative be T 3 b e f a a 3 a a 0 30 60 duration of stressor (min) e I l ed | be. 1 6 18 recovery time (h from end of stressor) Figure 5.6: Glutathione reductase (GR) activity in RTH-149 cells exposed to a 1 h heat (33°C, +13°C), oxidative (10mM hydrogen peroxide), or control (20°C) treatment. Significant differences are indicated by differing letters (a,b,c,d,e). n=6. s 60 CD O a o o C3 —I 40 20 i Control Heat Oxidative ab abn ab •LL a l [ I a ab I ab a B 0 30 60 duration of stressor (min) ab ab a b 4LJ 6 18 recovery time (h from end of stressor) Figure 5.7: Lactate dehydrogenase (LDH) released from RTH-149 cells exposed to a 1 h heat (33°C, +13°C), oxidative (10mM hydrogen peroxide), or control (20°C) treatment. Significant differences are indicated by differing letters (a,b). n=6. 102 Unlike changes in acclimation temperature in killifish (Chapter 3), acute increases in temperature did not greatly affect the G S H antioxidant system in rainbow trout, through indirect oxidative stress or other means. This could indicate that G S H levels and enzyme activity in rainbow trout may be high enough to combat acute heat stress, or that heat stress did not greatly challenge the G S H antioxidant system in rainbow trout. These results differ from previous studies that reported heat stress increased (Harris eta l . , 1991; Kondo et al., 1993; Mahmoud and Edens, 2003) or decreased (Ohtsuka etal . , 1994; Arechiga et al., 1995; Parihar et al., 1997; Kaur et al., 2005) G S H levels in mammals, chickens and fish, although are consistent with a lack of G S H change with heat stress reported in mice (Mirochnitchenko et al., 1995). The lack of consistency in G S H levels in response to a heat stress within the literature and this chapter may be due to the severity of heat stress used. Heat stress significantly affected G S H levels in a primary culture of rainbow trout hepatocytes only. These cells were also exposed to the largest change in temperature with heat stress (+15°C, compared with +13°C in RTH-149 cells and +10°C in vivo), although RTH-149 cells had the highest magnitude of heat stress (33°C, compared with 30°C in hepatocytes and 23°C in vivo). The change in temperature, rather than the final temperature level of heat stress may affect G S H dynamics in vertebrates. However, at the acclimation temperatures used, a +15°C heat stress would have resulted in mortality in vivo and in RTH-149 cells (Brett, 1971, Chapter 6 respectively). A combination of low acclimation temperature followed by a large increase in temperature may result in altered G S H dynamics without mortality, but smaller changes in temperature did not affect G S H in rainbow trout. Species differences, as well as in vivo and in vitro differences, may also result in the inconsistent effects of heat stress on G S H dynamics in both the literature and this chapter. An increase in temperature of +7°C for 1 h was sufficient to decrease G S H levels in the gills of catfish (Parihar et al., 1997), whereas an increase of +10°C for 2 h did not affect tGSH levels in rainbow trout. Although oxidative stress associated with acute heat stress may influence G S H levels in other organisms, it does not appear to be a factor in rainbow trout. Oxidative stress altered G S H dynamics in RTH-149 cells, by increasing G S S G (as % of tGSH) and preventing G R activity increase over time. Hydrogen peroxide used to induce the oxidative stress likely directly oxidized G S H , resulting in higher % G S S G . In addition, the poor viability of cells due to the oxidative stress may have prevented the increase in G R activity over time as observed in other groups. Serum deprivation increased tGSH and Hsp70 levels in RTH-149 cells. Although the effects of short-term serum deprivation have not been examined, long-term serum deprivation increased Hsp70 mRNA (Lewis and Hughes-Fulford, 2000) and altered G S H levels in mammalian cell lines (Kim et al., 2000; Zhuge and Cederbaum, 2006). Serum contains growth factors as well as transport proteins, spreading factors and stabilization and detoxification factors (van der Valk et al., 2004) that maintain cell growth and health. The sudden removal of serum may shift the cells from proliferation stage to cell cycle arrest. The consequent cellular changes involved in this, or effects of other components of F C S , may result in the noted increase in tGSH and Hsp70 levels. Heat stress prevented the increase in tGSH levels with F C S withdrawal, resulting in an ostensible decrease in tGSH levels with heat stress. These results suggest removing F C S during exposure to stressors in cell culture may alter the response of cells to those stressors. The minor changes in tGSH levels and activity of associated enzymes observed with heat stress 103 in vivo and in vitro were of equal or lower magnitude than changes observed in control groups over time. Fluctuations in components of the G S H system over time have been observed in other studies. Increased posterior kidney tGSH levels and decreased liver GPx activity in vivo are consistent with that observed in coho salmon during starvation (Chapter 4). As fish were not fed after heat stress, changes observed in G S H dynamics in vivo were likely due to changes associated with starvation. Ferraris et al., (2002) reported culture of rainbow trout hepatocytes resulted in increased oxidative stress and decreased G S H levels over the first 24 h of culture. While cells recovered from this stress after the first day of culture, increased G S H levels observed in the present experiment over time may be due to cells responding to this earlier oxidative stress, or acclimating to other aspects of the culture environment. In RTH-149 cells, G S H levels fluctuated over time and G R activity increased over time. The cause of this is unknown, but may be due to different stages of the cell cycle over time as cells became more confluent (Ohara and Terasima, 1969; Lewis et al., 1988). As heat stress-induced alterations in G S H dynamics were less than natural fluctuations over 1-2 days, this lends further evidence that heat stress does not greatly influence G S H dynamics in rainbow trout. Heat stress did not significantly alter G S H dynamics in rainbow trout in vivo or in vitro. Consequently, neither increased stress hormones such as Cortisol, nor increased metabolic rate due to stress, likely affects G S H dynamics during heat stress in this organism. Although G S H is important during oxidative stress and long-term temperature changes in fish, the lack of effect of heat stress on G S H dynamics suggests that altered G S H is not necessarily a component of the generalized stress response in fish. However, whether G S H dynamics influence other components of the stress response in fish is not yet determined. Summary • Heat stress did not consistently alter G S H dynamics in various in vivo and in vitro rainbow trout models. • Heat stress did not result in similar alterations in G S H dynamics as observed in oxidative stress in vitro. • Acute heat stress does not result in sufficient oxidative stress to challenge the G S H antioxidant system in rainbow trout, nor are G S H dynamics affected by heat induced components of the generalized stress response in rainbow trout. 104 References Afonso, L.O.B., unpublished data. Gender and stress in rainbow trout: does heat shock affect the response differently in males and females? Almar, M., Otero, L , Santos, C , Gallego, J .G . , 1998. Liver glutathione content and glutathione-dependent enzymes of two species of freshwater fish as bioindicators of chemical pollution. Journal of Environmental Science and Health. Part B: Pesticides, Food Contaminants, and Agricultural Wastes 33: 769-783. Arechiga, C.F. , Ealy, A.D., Hansen, P.J. , 1995. Evidence that glutathione is involved in thermotolerance of preimplantation murine embryos. Biology of Reproduction 52: 1296-1301. Baez, S., Segura-Aguilar, J . , Widersten, M., Johansson, A . -S . , Mannervik, B., 1997. Glutathione transferases catalyse the detoxication of oxidized metabolites (o-quinones) of catecholamines and may serve as an antioxidant system preventing degenerative cellular processes. Biochemical Journal 324: 25-28. Baliharova, V., Skalova, L , Maas, R.F.M., De Vrieze, G. , Bull, S. , Fink-Gremmels, J . , 2003. The effects of benzimidazole anthelmintics on P4501A in rat hepatocytes and HepG2 cells. Research in Veterinary Science 75: 61-69. Barhoumi, R., Mouneimne, Y., Phillips, T.D., Safe, S.H. , Burghardt, R.C., 1996. Alteration of oxytocin-induced calcium oscillations in clone 9 cells by toxin exposure. Fundamental and Applied Toxicology 33: 220-228. Barton, B.A., Iwama, G.K., 1991. Physiological changes in fish from stress in aquaculture with emphasis on the response and effects of corticosteroids. Annual Reviews of Fish Diseases 1: 3-26. Bell, G. , Cowey, C.B. , 1990. Oxidation of glutathione during hydroperoxide metabolism in isolated hepatocytes of rainbow trout (Salmo gairdneri). Fish Physiology and Biochemistry 8: 11-17. Bergmeyer, H.U. Methods of Enzymatic Analysis. Volume VI: Metabolites I: Carbohydrates. V C H Publishers, Deerfield Beach, FL. Bols, N.C., Mosser, D.D., Steels, G.B., 1992. Temperature studies and recent advances with fish cells in vitro. Comparative Biochemistry and Physiology. Part A, Physiology 103: 1-14. Brett, J.R., 1971. Energetic responses of salmon to temperature. A study of some thermal relations in the physiology and freshwater ecology of sockeye salmon (Oncorhynchus nerka). American Zoologist 11: 99-113. Davidson, J .F. , Schiestl, R.H., 2001. Mitochondrial respiratory electron carriers are involved in oxidative stress during heat stress in Saccharomyces cerevisiae. Molecular and Cellular Biology 21: 8483-8489. Ferraris, M., Radice, S., Catalani, P., Francolini, M., Marabini, L., Chiesara, E., 2002. Early oxidative damage in primary cultured trout hepatocytes: a time course study. Aquatic Toxicology 59: 283-296. Forsyth, R.B., Candido, E.P.M. , Babich, S.L., Iwama, G.K., 1997. Stress protein expression in coho salmon with bacterial kidney disease. Journal of Aquatic Animal Health 9: 18-25. Griffith, O.W., 1980. Determination of glutathione and glutathione disulfide using glutathione reductase and 2-vinylpyridine. Analytical Biochemistry 106: 207-212. Harris, C , Juchau, M.R., Mirkes, P.E., 1991. Role of glutathione and hsp-70 in the acquisition of thermotolerance in postimplantation rat embryos. Teratology 43: 229-239. Hughes, E., Gallagher, E., 2004. Effects of b-naphthoflavone on hepatic biotransformation and glutathione biosynthesis in largemouth bass (Micropterus salmoides). Marine Environmental Research 58: 675-679. Ignatius, J . , Oommen, O.V., 1990. Effects of corticosteroids and protein synthesis inhibitors on activities of oxidative enzymes in a bony fish, Anabas testudineus (Bloch). General & Comparative Endocrinology 78: 303-310. Kaur, M., Atif, F., AN, M., Rehman, H., Raisuddin, S., 2005. Heat stress-induced alterations of antioxidants in the freshwater fish Channa punctata Bloch. Journal of Fish Biology 67: 1653-1665. 105 Kim, Y.H. , Takahashi, M., Noguchi, N., Suzuki, E., Suzuki, K., Taniguchi, N., Niki, E., 2000. Inhibition of c-jun expression induces antioxidant enzymes under serum deprivation. Archives of Biochemistry and Biophysics 374: 339-346. Kondo, T., Yoshida, K., Urata, Y., Goto, S., Gasa, S., Taniguchi, N., 1993. Gamma-glutamylcysteine synthetase and active transport of glutathione s-conjugate are responsive to heat shock in K562 erythroid cells. Journal of Biological Chemistry 268: 20366-20372. Lewis, A.D., Hayes, J.D., Wolf, C.R., 1988. Glutathione and glutathione-dependent enzymes in ovarian adenocarcinoma cell lines derived from a patient before and after the onset of drug resistance: intrinsic differences and cell cycle effects. Carcinogenesis 9: 1283-1287. Lewis, M.L., Hughes-Fulford, M., 2000. Regulation of heat shock protein message in jurkat cells cultured under serum-starved and gravity-altered conditions. Journal of Cellular Biochemistry 77: 127-134. Lu, S.C. , J . , G., Kuhlenkamp, J . , Kaplowitz, N., 1992. Insulin and glucocorticoid dependence of hepatic y-glutamylcysteine synthetase and glutathione synthesis in the rat. Journal of Clinical Investigation 90: 524-532. Mahmoud, K.Z., Edens, F.W., 2003. Influence of selenium sources on age-related and mild heat stress-related changes of blood and liver glutathione redox cycle in broiler chickens (Gallus domesticus). Comparative Biochemistry and Physiology. Part B, Biochemistry and Molecular Biology 136: 921-934. Mampuru, L.J., Chen, S .J . , Kalenik, J.L., Bradley, M.E., Lee, T.C., 1996. Analysis of events associated with serum deprivation-induced apoptosis in C3H/Sol8 muscle satellite cells. Experimental Cell Research 226: 372-380. Mirochnitchenko, O., Palnitkar, U., Philbert, M., Inouye, M., 1995. Thermosensitive phenotype of transgenic mice overproducing human glutathione peroxidases. Proceedings of the National Academy of Sciences of the United States of America 92: 8120-8124. Mommsen, T.P., Moon, T.W., Walsh, P.J. , 1994. Hepatocytes: isolation, maintenance and utilization. In: Hochachka, P.W. Mommsen, T.P. (Eds.), Biochemistry and Molecular Biology of Fishes, Vol 3. Elsevier Science B.V., pp. 355-373. Muradian, K.K., Utko, N.A., Mozzhukhina, T.G. , Litoshenko, A.Y., Pishel, I.N., Bezrukov, V.V., Fraifield, V .E . , 2002. Pair-wise linear and 3D nonlinear relationships between the liver antioxidant enzyme activities and the rate of body oxygen consumption in mice. Free Radical Biology and Medicine 33: 1736-1739. Ohara, H., Terasima, T., 1969. Variations of cellular sulfhydryl content during cell cycle of HeLa cells and its correlation to cyclic change of x-ray sensitivity. Experimental Cell Research 58: 182-185. Ohtsuka, Y., Yabunaka, N., Fujisawa, H., Watanabe, I., Agishi, Y., 1994. Effect of thermal-stress on glutathione metabolism in human erythrocytes. Journal of Applied Physiology 68: 97-91. Otto, D.M.E., Moon, T.W., 1995. 3,3',4,4'-Tetrachlorobiphenyl effects on antioxidant enzymes and glutathione status in different tissues of rainbow trout. Pharmacology and Toxicology 77: 281-287. Parihar, M.S., Dubey, A.K., Javeri, T., Prakash, P., 1996. Changes in lipid peroxidation, superoxide dismutase activity, ascorbic acid and phospholipid content in liver of freshwater catfish Heteropneustes fossilis exposed to elevated temperature. Journal of Thermal Biology 21: 323-330. Parihar, M.S., Javeri, T., Hemnani, T., Dubey, A.K., Prakash, P., 1997. Responses of superoxide dismutase, glutathione peroxidase and reduced glutathione antioxidant defenses in gills of the freshwater catfish (Heteropneustes fossilis) to short-term elevated temperature. Journal of Thermal Biology 22: 151-156. Patel, R., Mcintosh, L.J., McLaughlin, M., Brooke, S., Nimon, V., Sapolsky, R.M., 2002. Disruptive effects of glucocorticoids on glutathione peroxidase biochemistry in hippocampal cultures. Journal of Neurochemistry 82: 118-125. Pena-Llopis, S., Pena, J .B. , Sancho, E., Fernandez-Vega, C , 2001. Glutathione-dependent resistance of the European eel Anguilla anguilla to the herbicide molinate. Chemosphere 45: 671-681. Shi, L.Z., Zheng, W., 2005. Establishment of an in vitro brain barrier epithelial transport system for pharmacological and toxicological study. Brain Research 1057: 37-48. 106 Simmons, H.F., James, R.C., Harbison, R.D., Patel, D.G., Roberts, S .M. , 1991. Examination of the role of catecholamines in hepatic glutathione suppression by cold-restraint in mice. Toxicology 67: 29-40. Skibba, J.L., Powers, R.H., Stadnicka, A., Cullinane, D.W., Almagro, U.A., Kalbfleisch, J .H. , 1991. Oxidative stress as a precursor to the irreversible hepatocellular injury caused by hyperthermia. International Journal of Hyperthermia 7: 749-761. Smith, P.K., Krohn, R.I., Hemanson, G.T., Mallia, A.K., Gartner, F.H., Provenzano, M.D., Fujimoto, E.K., Goeke, N.M., Olson, B.J., Klenk, D.C., 1985. Measurement of protein using bicinchoninic acid. Analytical Biochemistry 150: 76-85. Speck, R.F., Schranz, C., Lauterburg, B.H., 1993. Prednisolone stimulates hepatica glutathione synthesis in mice: protection by prednisolone against acetaminophen hepatotoxicity in vivo. Journal of Hepatology 18: 67. Stephensen, E., Sturve, J . , Forlin, L , 2002. Effects of redox cycling compounds on glutathione content and activity of glutathione-related enzymes in rainbow trout liver. Comparative Biochemistry and Physiology. Part C, Pharmacology, Toxicology and Endocrinology 133: 435-442. van der Valk, J . , Mellor, D., Brands, R., Fischer, R., Gruber, F., Gstraunthaler, G. , Hellebrekers, L, Hyllner, J . , Jonker, F.H., Prieto, P., Thalen, M., Baumans, V., 2004. The humane collection of fetal bovine serum and possibilities for serum-free cell and tissue culture. Toxicology in Vitro 18: 1-12. Voss, S.H. , Park, Y., Kwon, S.-O., Whalen, R., Boyer, T.D., 1996. Role of interleukin 6 and corticosteroids in the regulation of expression of glutathione S-transferases in primary cultures of rat hepatocytes. Biochemical Journal 317: 627-632. Walther, F.J. , David-Cu, R., Mehta, E.I., Polk, D.H., Jobe, A.H. , Ikegami, M., 1996. Higher lung antioxidant enzyme activity persists after single dose of corticosteroids in preterm lambs. American Journal of Physiology 15: L187-L191. Wendelaar Bonga, S .E . , 1997. The stress response in fish. Physiological Reviews 77: 591-625. Zhuge, J . , Cederbaum, A.I., 2006. Serum deprivation-induced HepG2 cell death is potentiated by C Y P 2 E 1 . Free Radical Biology and Medicine 40: 63-74. 107 CHAPTER 6 The effects of glutathione levels on components of the generalized and cellular stress responses in rainbow trout in vivo and in vitro Introduction Fish live in fluctuating environments where they are exposed to daily and seasonal changes in water temperature, oxygen and chemical composition, as well as to physical stressors such as water turbulence and predation. To cope with environmental fluctuations beyond their normal range, fish mount a coordinated, integrated stress response that acts at both the whole animal and cellular level (reviewed by Iwama et al., 1999). Fish respond to a wide variety of stressors through a generalized stress response that includes the release of stress hormones such as adrenocorticotropic hormone, catecholamines and corticosteroids (see Barton and Iwama, 1991). These in turn bring about metabolic, hematological and other secondary changes that act to maintain homeostasis within the fish (see Barton and Iwama, 1991). Stressors that compromise protein integrity, such as temperature and chemical stressors, also initiate a cellular stress response. A suite of molecular chaperones, termed heat shock proteins (Hsp's), are upregulated to maintain protein conformation and repair proteins damaged by stress (see Iwama et al., 1998). The maintenance of the redox status during stress has not been well examined in fish. A major component of the cellular redox balance is the antioxidant glutathione (GSH) and its associated enzymes. In Chapter 5, I determined that although G S H dynamics are altered in fish in response to oxidative and toxicant stressors (Bell and Cowey, 1990; Otto and Moon, 1995; Almar et al., 1998; Pena-Llopis et al., 2001; Stephensen et al., 2002; Hughes and Gallagher, 2004), they are not affected by a stressor, such as temperature, that doesn't directly compromise the redox status of rainbow trout (Oncorhynchus mykiss). However, G S H has been associated with several components of the generalized and cellular stress responses in vertebrates, such as the glucocorticoid receptor, Hsc70, and the small Hsp's (Esposito et al., 1995; Zavialov et al., 1998; Hoppe et al., 2004). Although G S H dynamics may not be altered during all types of stress, adequate levels of G S H may be necessary to maintain generalized and cellular responses to a wide variety of stressors. G S H may be involved in the stress response in several ways: by maintaining the redox balance when compromised by stress by removing toxic products and metabolites of stress, and by maintaining protein integrity and function of key components of the stress response by interacting with their protein thiols. The effect of altered G S H status on the generalized stress response has not been well examined in vertebrates. Increased or decreased G S H levels resulted in proportional changes in Cortisol released after exposure to pesticides in rainbow trout in vitro (Dorval and Hontela, 2003). In addition, antioxidant supplements have been shown to suppress heat-induced Cortisol release in chickens (Mahmoud et al., 2004), and alterations in G S H levels were proportional to gluconeogenesis activity in rat kidney cells (Winiarska et a.l., 2003). Decreased levels of G S H are also reported to decrease the DNA binding of the glucocorticoid receptor and hence the physiological effects of Cortisol in mammalian cells (Esposito et al., 108 1995). The effects of altered G S H levels on the cellular Hsp response have not been examined in fish, although there has been extensive research on this subject in mammals in vitro. The majority of these studies demonstrate that G S H status can affect the Hsp response to heat, chemical and other types of stressors. However, the influence of G S H status on the Hsp response differs greatly between studies, depending on the type, severity and duration of the stressor, as well as the cell type considered. Up-regulation of Hsp's is initiated by activation of heat shock factors (e.g. HSF1) which then bind to the heat shock elements of Hsp gene promoters (see Iwama et al., 1998). G S H levels may affect the Hsp response by direct action on HSF activation and DNA-binding (Rokutan et al., 1996; Ito et al., 1998; Bijur et al., 1999; Will et al., 1999), or indirectly by altering protein thiol status and ultimately protein stability during stress (Liu et al., 1996; Freeman et al., 1997; McDuffee et al., 1997; Zou et al., 1998). In general, altered G S H levels have an inversely proportional effect on Hsp levels after exposure to chemical or oxidative stressors that may require direct action of G S H for detoxification and protein protection (Abe et al., 1994; Rokutan et al., 1996; Ito et al., 1998; Bijur et al., 1999; Gaubin et al., 2000). However, the relationship between G S H and Hsp during heat stress is less clear, as decreased G S H levels increased, decreased or had no effect on Hsp levels, and increased G S H levels increased Hsp expression post-heat stress, depending on the cell type and the severity of the heat shock (Russo et al., 1984; Freeman et al., 1988; Harris eta l . , 1991; Freeman etal . , 1993; Rokutan etal . , 1996; Ito eta l . , 1998). The inconsistent effects of altered G S H on the Hsp response of mammals in vitro suggest that G S H does not act directly on induction pathways of Hsp's. This inconsistency may also be due to the severity of heat shock used between studies, where adequate G S H levels may be more important during exposure to heat stressors of greater intensity. In fish, tissue levels of G S H are easier to manipulate than in mammals (Chapter 2). Consequently, fish may be an interesting model for examining the relationship between G S H and the stress response in vertebrates. The generalized and cellular stress responses are integrated in vivo (Ackerman et al., 2000; Basu et al., 2001; Boone et al., 2002; Boone and Vijayan, 2002; Basu et al., 2003), and consequently the effects of G S H on either response may influence this integration. For example, the glucocorticoid receptor is associated with a number of Hsp's that facilitate its folding and nuclear trafficking, and this association is increased by stress (Basu et al., 2003). The binding domains of the glucocorticoid receptor, including Hsp binding, are thiol-regulated (Esposito et al., 1995; Makino et al., 1999) and may therefore be affected by G S H . Consequently, effects of G S H on the stress response in fish may differ in vivo compared with in vitro. In Chapter 2,1 determined that, unlike mammals, G S H levels of several rainbow trout tissues did not match potential synthesis rate, but were limited by precursor supply. Factors that affect metabolic rate (i.e. increased acclimation temperature and growth rates) also increased feeding rates, increased the ability to take up G S H precursors from the diet and increased G S H levels in several tissues (Chapters 4 and 5). Whether artificially increasing G S H levels to match synthesis capacity would impart an advantage to the fish is not clear. The effects of altered G S H levels on the ability of fish to respond to stress have not been determined, but could have potential interest in aquaculture and environmental research. I hypothesize that increased G S H levels will ameliorate, and decreased G S H levels will augment the generalized and cellular responses to stress in rainbow trout. To address this, I altered G S H levels of 109 rainbow trout in vivo and a rainbow trout hepatoma cell line (RTH-149) in vitro. G S H levels were increased by injection or incubation with reduced G S H , and G S H levels were decreased by injection or incubation with the G S H synthesis blocker buthionine sulfoximine (BSO). Rainbow trout were then exposed to heat and chasing stresses in vivo and varying intensities of heat stress in vitro to address the following objectives: Objective 6.1: To determine if altered G S H levels affect components of the stress response (as indicated by plasma Cortisol and glucose levels, and tissue Hsp levels) to heat and chasing stress in rainbow trout in vivo. Objective 6.2: To determine if altered G S H levels affect components of the cellular stress response (Hsp70 and 30) to heat stress in rainbow trout in vitro. Materials and Methods Rainbow trout (32g) were obtained from Spring Valley Trout Hatchery (Mission, BC , Canada), and RTH-149 cells were purchased from the American Type Culture Collection (ATCC#: CRL-1710, Manassas, VA, USA). Primary and secondary antibodies were obtained from Stressgen Biotechnologies (Victoria, BC , Canada), glutathione reductase was purchased from Roche Diagnostics (Laval, Q C , Canada) and tricaine methanesulfonate (MS-222) was purchased from Syndel, Canada (Vancouver, BC , Canada). All cell culture media were from Invitrogen Corporation (Burlington, O N , Canada). All other chemicals were purchased from Sigma-Aldrich Corporation (Oakville, O N , Canada). In vivo experiments took place in the spring (April, mild heat stress experiment) or summer (August, severe heat stress and handling stress experiments). Experimental Series 6.1: The effects of altered GSH levels on the generalized and cellular stress responses in vivo Rainbow trout were held in aerated 70L tanks supplied with dechlorinated city of Vancouver municipal fresh water at ambient temperature. Fish were fed to satiation with a 1.5mm commercial trout feed (Skretting, Vancouver, BC , Canada) daily until the start of the experiment. G S H levels were altered in fish (one tank per injection group and stressor treatment) by injection once a day for three days with 0.8mmol GSH/kg fish to increase G S H levels, 1.8mmol BSO/kg fish to decrease G S H levels, or saline as control, as per Otto et al. (1997). All injection mixtures were made in physiological saline (100mM NaCl, 2.5mM KCI, 1.5mM CaCI 2 , 1.0mM MgCI 2 , 0.5mM N a H 2 P 0 4 , 5mM N a H C 0 3 , pH 6.8). Fish were removed from the tanks and lightly anaesthetized with 50mg/L MS-222 and 100mg/L sodium bicarbonate. Fish were then weighed and injected intraperitoneally with 10rxL of injection mixture per g of fish. Fish were allowed to recover in a 100L aerated holding tank and then returned to their respective tanks. Six to eight hours after the final injection, one tank from each injection group were exposed to one of the following treatments: 1) 2 h, 20.5°C heat stress (+12°C, ambient temperature 8.5°C), 2) 2 h, 25.5°C heat stress (+10.5°C, ambient temperature 15.0°C), or 3) 3 min chasing stress until exhaustion. Heat-stressed fish (treatments 1 and 2) were net transferred into aerated, 50L treatment tanks with elevated temperature and static water for two hours before returning them back to their original tanks. Temperature was closely monitored throughout. For treatment 3, fish were transferred to a 10L bucket of ambient, aerated water 110 and chased with a net until exhaustion (approximately 3min). For treatment 1, a control group was included that were transferred to similar tanks as the heat-stressed group, but at ambient temperature. For treatments 2 and 3, stressed groups were compared with pre-stress levels. Fish were sampled before first injection, immediately before exposure to the stressors and 1 and 10-12 hours post-stress. These latter two times were chosen as time frames post-stress known to display elevated plasma Cortisol and tissue Hsp levels, respectively (Ackerman et al., 2000; Basu et al., 2001). Sampled fish were removed quickly from the tank four at a time and killed by overdose of anesthetic (2g/L MS-222, 4g/L sodium bicarbonate). Their blood was removed by caudal puncture, centrifuged at 2900xg for 3 minutes and the plasma frozen in liquid nitrogen. The liver and posterior kidney were removed, rinsed with ice-cold phosphate-buffered saline (pH 7.4), blotted dry and frozen in liquid nitrogen. All samples were stored at -80°C until the following analyses: plasma Cortisol and glucose, and tissue total G S H ( tGSH=GSH+2GSSG), Hsp70 and 30 (n=8) and oxidized glutathione ( G S S G , n=4). Experimental Series 6.2: The effects of altered GSH levels on the cellular stress response in vitro To further examine the effects of altered G S H levels on the cellular response to heat stress in rainbow trout, a rainbow trout hepatoma cell line (RTH-149) was used. RTH-149 cells were cultured in 18.3mM H E P E S and 0.75% sodium bicarbonate buffered Dulbecco's Modified Eagle's Medium (DMEM) with 10% fetal calf serum, 0.1mM non-essential amino acids and 1% antibiotic/antimycotic (penicillin, streptomycin, amphotericin B), pH 7.4. During growth, cells were held in air, at room temperature (approximately 20°C). For the following experiments, cells were plated in 6-well plates at a density of 2.5x10s cells/2ml_/well and grown for 3-4 days at 20°C until approximately 80% confluent. To determine the time after heat stress at which Hsp70 levels are maximally elevated, preliminary experiments were conducted where RTH-149 cells were exposed to either a 28°C (+8°C), or 33°C (+13°C, ambient temperature 20°C) 1 h heat stress and Hsp70 protein levels were measured pre-stress and 3-24 h or 16-20 h post-stress. In both experiments, Hsp70 levels peaked at approximately 18 h post-heat stress (Figure 6.1, n=3). Consequently, for the following set of experiments, cells were sampled for Hsp's at 18 h post-heat stress. To alter G S H levels in RTH-149 cells, 48 h before the heat stress, the medium on all cells was replaced with normal medium, except the medium for the depleted G S H group was supplemented with 25mM BSO. 6 h before the heat stress, 60uL of 0.2mM G S H in D M E M was added to the enhanced group for a final concentration of 6mM G S H . 60uL of unsupplemented D M E M was also added to control cells and 60uL of D M E M supplemented with 25mM BSO was added to the B S O group. These concentrations and times (25mM B S O for 48 h, 6mM G S H for 6 h) have been observed to significantly alter G S H levels (Leggatt, unpublished data). Preliminary experiments were performed to determine the minimum temperature needed to increase Hsp70 levels in RTH-149 cells. Cells were divided into control, BSO and G S H groups as above, and then exposed to a 1 h ambient (20°C) or heat stress (24.5/+4.5°C, or 27°C/+7°C) temperatures for 1 h before being returned to 20°C, and Hsp70 levels measured 18 h after heat stress (n=6). Based on these experiments, a temperature of 24°C was used as approximately threshold induction temperature for Hsp70. The effect of altered G S H levels on the magnitude of the Hsp response to heat stress in RTH-149 111 0 5 10 15 20 25 0 5 15 20 time (h from end of heat stress) Figure 6.1: Hsp70 levels in RTH-149 cells following exposure to a A) 28°C (+8°C) or B) 33°C (+13°C) 1 h heat stress (ambient temperature 20°C). Grey bar indicates time of heat stress. Significant differences within experiments are indicated by differing letters (a,b,c). n=3. 112 cells was determined in the following experiments. Cells were exposed to 1 h ambient temperature (20°C), or heat stresses of varying degrees (24, 33, or 35°C, +4, +13, or +15°C respectively) before being returned to 20°C temperature. The latter two temperatures were chosen as temperatures observed to induce Hsp30, and result in decreased viability in RTH-149 cells respectively (Leggatt, unpublished data). Both ambient and heat-stressed cells were sampled 18 h post-stress. To sample cells, 100pL of the medium was removed and frozen on dry ice. Cells were then scraped into 2ml_ tubes, centrifuged at 11 600*g for 20 sec, washed once in phosphate-buffered saline and the cell pellet and 100pL of supernatant frozen on dry ice. All samples were stored at -80°C and analyzed for the following: cell protein, lactate dehydrogenase (LDH), tGSH, and Hsp70 and 30, and medium LDH (n=6). The effect of altered G S H levels on the timing of the Hsp response to heat stress in RTH-149 cells was determined in the following experiment. G S H levels were altered in RTH-149 cells as above, with the addition of a second B S O group that only received B S O for 6 h. This group was used to compare direct effects of low G S H levels versus direct effects of B S O , as G S H levels would not have decreased significantly after 6 h incubation with B S O . Cells were then exposed to 1 h ambient (20°C) or heat stress (33°C, +13°C) temperature before returning to ambient temperature. Cells were sampled and analyzed pre-stress and 6, 12, 18 and 24 h post-stress as above (n=6). Sample Preparation and Analysis Tissues and cells were sonicated in 100mM Tris (in vivo Hsp analyses) or 125mM sodium phosphate buffers (all other analyses) with 1mM ethylenediaminetetraacetic acid, 1pM pepstatin, 1pM leupeptin, 0.15pM aprotinin, and 0.5mM phenylmethylsulfonyl fluoride protease inhibitors (pH 7.5). Supernatant of tissue homogenate was obtained by centrifugation at 10 OOOxg at 4°C for 5 min. Portions of tissue supernatant or cell homogenate were removed for protein, LDH, and Hsp70 analyses, and the remaining supernatant added to an equal volume of 10% sulfosalicylic acid and supernatant for G S H analyses obtained by centrifugation at 10 OOOxg at 4°C for 10 min. Homogenate for Hsp analysis was added to an equal volume of Laemmli's buffer (20% glycerol, 4% sodium dodecyl sulfate, 0.25% bromophenol blue, 10% 6-mercaptoethanol in 0.5M Tris-HCI), boiled for 3 min and stored at -20°C. All analyses except Hsp70 were performed in duplicate or triplicate on 96-well microplates and measured using a SpectraMax spectrophotometer equipped with SoftmaxPro Software (Molecular Devices Corporation). Protein content was analyzed using the bicinchoninic acid method as per Smith et al. (1985), using bovine serum albumin as standards and analyzed at 516nm. Plasma Cortisol levels were measured using a commercially available enzyme-linked immunosorbant assay (ELISA, Neogen Corporation, Lasing, Ml, USA) and analyzed at 650nm. Plasma glucose levels were measured using a commercially available kit (Sigma Diagnostics, Oakville, O N , Canada) based on Trinder (1969). tGSH and G S S G analyses were modified from Griffith (1980) as follows. For tGSH analyses, 6rxL of triethanolamine was added per 100nL of tissue or cell supernatant and G S S G standards. 10fxL of samples or G S S G standard and 200(iL of reaction mixture were added to a 96 well plate. The reaction mixture contained 0.22mM reduced 3-nicotinamide adenine dinucleotide phosphate, 0.62mM 5,5'-dithiobis(2-nitrobenzoic acid), 0.56% triethanolamine and 0.5U/mL glutathione reductase in 125mM sodium phosphate, 6mM ethylenediaminetetraacetic acid buffer (pH 7.4). The change in absorbance was 113 monitored at 412nm for 5 min. For G S S G analyses, 6uL of 2-vinylpyridine (to bind reduced GSH) and 20ul_ of 0.1% triethanolamine in phosphate-buffered saline was added per 100uL of sample or G S S G standard. Samples and standards were mixed vigorously for 1 min and then incubated for 50 min. G S S G analyses were then as per tGSH above except that the concentration of dithiobis(2-nitrobenzoic acid) in the reaction mixture was 1pM. LDH activities in RTH-149 cell homogenate and media were analyzed as previously described (Bergmeyer, 1985). In brief, cell homogenate was diluted 11 times in 100mM H E P E S , pH 7.4. 10uL of cell homogenate or 20uL of media were added to a 96-well plate, with appropriate H E P E S or media blanks. 0.4mM reduced B-nicotinamide adenine dinucleotide (NADH) in H E P E S buffer was added to each well for a final volume of 190uL. The plates were covered and incubated at 28°C for 15 min for cell homogenate and 30 min to 2 h for media. Background absorbance decline was then read for 5 min at 340nm at 28°C. The reaction was initiated by the addition of 10pL of 28mM pyruvate and oxidation of NADH to N A D + was measured by absorbance decline at 340nm for 5 min at 28°C. Hsp70 was analyzed by S D S - P A G E and Western blotting as per Forsyth et al. (1997), with the following modifications. 20uL of cell homogenate diluted in Laemmli's buffer to 1mg protein/mL (in vivo) or 0.5mg protein/mL (in vitro) was added to gel wells in duplicate. Liver homogenate from rainbow trout exposed to a 22°C (in vivo Hsp70) or 25.5°C (in vivo Hsp30) heat stress, or homogenate from RTH-149 cells exposed to a 33°C heat stress (in vitro) were added to each gel as a reference sample. After transfer of separated proteins onto a nitrocellulose membrane, the membrane was incubated with rabbit polyclonal anti-chinook salmon Hsp70 primary antibody or mouse monoclonal anti-rainbow trout Hsp30 primary antibody, followed by appropriate alkaline phosphatase-conjugated secondary antibodies. Final band intensity was determined using SigmaGel ( S P S S Inc.) and Hsp data reported relative to the reference value. Data were analyzed by 1 or 2-way A N O V A as appropriate, followed by Tukey's post-hoc test, using SigmaStat ( S P S S Inc.). If normality or equal variance failed, data were square root or natural log transformed before analyses to satisfy normality and equal variance. If transformation failed to normalize data or bring about equal variance, data were analyzed by Kruskal-Wallis non-parametric A N O V A on ranks, followed by Dunn's post-hoc comparison test. Differences were considered significant if p<0.05. All data are presented as mean ± standard error of the mean. Results tGSH levels in rainbow trout models after injection or incubation with GSH or BSO in Experiments 6.1 and 6.2 respectively tGSH levels for the following experiments are given in Figures 6.2 to 6.5. In rainbow trout exposed to a 20.5°C heat stress (+12°C, ambient temperature 8.5°C), G S H injection increased tGSH levels and B S O injection decreased G S H levels in both liver and posterior kidney (p<0.001, Figure 6.2). In rainbow trout exposed to a 25.5°C heat stress (+10.5°C, ambient temperature 15°C), G S H injection increased tGSH levels in both liver and kidney, and B S O injection decreased tGSH levels in the kidney (p<0.001), but not until after the heat stress in the liver (Figure 6.3). G S S G , as percent of tGSH for 114 '5 O 1-o , 0 0 e 30 20 10 30 20 10 bde bd ad AS I ae I I B H Control a GSH R ^ ^ l BSO I pre-injection pre-stress 8.5°C 20.5°C lOh post-heat stress Figure 6.2: Total G S H in A) liver and B) posterior kidney of rainbow trout injected once a day for three days with saline (control), G S H or BSO, and then exposed to a 2 h, 20.5°C heat stress (+12°C, ambient temperature 8.5°C). tGSH levels were measured pre-injection, immediately pre-stress (6 h after last injection), and 10 h after the end of the heat stress. Significant differences between groups, within tissues are indicated by differing letters (a,b,c,d,e). n=8. 115 30 20 S 10 g '5 -*-» o S-l OH GO O B 25 H 20 15 10 D Control • GSH BSO a * b a X a X be I ad X ab bd X ad pre-injection pre-stress 12h post-stress 15°C 25.5°C Figure 6.3: Total G S H in A) liver and B) posterior kidney of rainbow trout injected once a day for three days with saline (control), G S H or BSO, and then exposed to a 2 h, 25.5°C heat stress (+10.5°C, ambient temperature 15°C). tGSH levels were measured pre-injection, immediately pre-stress (8 h after last injection), and 12 h after the end of the heat stress. Significant differences between groups are indicated by differing letters (a,b,c,d). n=8. 116 rainbow trout in vivo are given in Table 6.1. G S H and BSO injection did not affect liver % G S S G in either experiment (p=0.18 and 0.54 respectively) or posterior kidney % G S S G in the 20°C heat stress experiment (p=0.07). In the 25.5°C heat stress experiment, BSO injection increased % G S S G in the kidney above that of saline-injected fish (p<0.001). However, % G S S G was below 5% in all groups, indicating the injections did not result in oxidative stress in the fish and the majority of tGSH alterations were due to altered reduced G S H . In RTH-149 cells exposed to 1 h heat stresses of varying intensity, cells incubated with G S H had higher tGSH levels, and BSO-incubated cells had lower tGSH levels than control cells. These levels remained as such 18 h after heat stress in all BSO and G S H groups, although the increase was insignificant in GSH-incubated cells in 20°C (ambient) and 24°C (+4°C) groups (Figure 6.4). In addition, BSO-incubated cells exposed to 24°C (+4°C) heat stress had lower G S H levels than B S O cells exposed to other temperatures, and GSH-incubated cells exposed to a 33°C (+13°C, ambient temperature 20°C) heat stress had higher G S H levels than G S H cells exposed to other temperatures. In RTH-149 cells exposed to a 1 h, 33°C (+13°C) heat stress only, immediately prior to heat stress GSH-incubated cells had higher tGSH levels than control cells, and cells incubated with B S O for 48 h, but not 6 h, had lower tGSH levels than control cells (p<0.001, Figure 6.5). Experimental Series 6.1: The effects of altered GSH levels on the generalized and cellular stress responses in vivo The Generalized Stress Response Plasma Cortisol levels pre-injection and pre-stress (Figures 6.6 and 6.8) were within the range reported for unstressed states in salmonids less than 30-40ng/mL, reviewed by Barton and Iwama, 1991), indicating injection or sampling did not cause significant stress in the fish. Control fish exposed to a 20.5°C heat stress (+12°C, ambient temperature 8.5°C) had increased levels of plasma C o r t i s o l 1 h post-heat stress (p<0.001, Figure 6.6). However, plasma Cortisol levels in fish previously injected with G S H or B S O did not significantly increase after heat stress, although the overall levels were not significantly different from control levels (p=0.37). The 20.5°C (+12°C) heat stress did not significantly increase plasma glucose levels in any group (p=0.80), although fish injected with G S H had significantly higher plasma glucose than control fish (p=0.001, Figure 6.7). Severe chasing (3 min chasing) or heat stress (25.5°C, +10.5°C, ambient temperature 15°C) greatly increased plasma Cortisol levels 1 h post-stress in all injection groups (p<0.001, Figure 6.8). There were no significant differences between injection groups before or after exposure to these stressors (p=0.54). The Cellular Stress Response G S H or B S O injection alone did not affect liver or posterior kidney Hsp70 levels in either the 20.5°C or 25.5°C heat stress experiments (Figures 6.9 and 6.10). In the 20.5°C heat stress experiment (+12°C, ambient temperature 8.5°C), there was a significant interaction between injection type and heat stress in the liver (p=0.04), but not the kidney (p=0.05, Figure 6.9). Heat stress significantly increased Hsp70 levels in all groups and tissues (p<0.001) with the exception of posterior kidney of control fish. However, the magnitude of increase was greater in G S H - (liver: 3.5-fold, kidney: 3.8-fold) and B S O - (liver: 4.2-fold, kidney: 2.8-fold) injected fish than control fish (liver: 2.2-fold, kidney: 1.9-fold), although these levels were significantly different in the liver only. There was no difference in Hsp70 levels post-20.5°C 117 Table 6.1: Oxidized glutathione ( G S S G , % of total GSH) in rainbow trout injected once a day for three days with saline (control), G S H or B S O . % G S S G levels for control, G S H , and B S O groups were measured immediately before heat stress. Significant differences within kidney of 25.5°C heat stress experiment are indicated by differing letters (a,b,c). There were no significant differences in any other tissue or experiment. n=4. 20.5°C heat stress (ambient 8.5°C) 25.5°C heat stress (ambient 15°C) Liver Kidney Liver Kidney Pre-injection 2.53±0.57 1.26±0.15 2.43±0.54 1.00±0.11 a D Control 1.24±0.12 0.88±0.15 3.73±0.63 1.71±0.42a G S H 0.88+0.13 0.7U0.17 3.76+0.84 1.60±0.57b c BSO 1.40±0.12 1.12±0.11 2.8510.37 2.42±0.34 c 80 I 60 o S-l PH I 40 H S | 20 I Control GSH W77~\ BSO ab i d ab ab 111 Pllll pre-stress 20 24 33 heat stress temperature (°C) 0 d 35 Figure 6.4: Total G S H levels in RTH-149 cells incubated with normal media (control), media containing G S H for 6 h, or media containing B S O for 48 h, and then exposed to a 1 h heat stress of varying temperatures (ambient temperature 20°C). tGSH was measured pre-stress (n=2) and 18 h after the end of the heat stress (n=6). Significant differences between groups post-heat stress only are indicated by differing letters (a,b,c,d,e,f). 118 60 Figure 6.5: Total G S H levels in RTH-149 cells incubated with normal media (control), media containing G S H for 6 h, media containing B S O for 48 h, or media containing BSO for 6 h. tGSH levels were measured immediately prior to a 1 h, 33°C heat stress (+13°C, ambient temperature 20°C). Significant differences between groups are indicated by differing letters (a,b,c). n=6. c t; o o ca J2 100 i 80 •{ 60 -\ 40 20 ac pre-injection be T ac T a I i Control GSH abc ac I BSO injection group Figure 6.6: Plasma Cortisol levels in rainbow trout injected once a day for three days with saline (control), G S H or B S O , and then exposed to a 2 h, 20.5°C (+12°C) heat stress or maintained at ambient temperature (8.5°C). Grey bar is level before injection, and white bars and scatter plots are 1 h after the end of the heat stress. Significant differences between groups are indicated by differing letters (a,b,c). n=8. 119 CD CO O O as S 14 12 H 10 8 6 pre-injection Control GSH BSO injection group Figure 6.7: Plasma glucose levels in rainbow trout injected once a day for three days with saline (control), G S H or B S O , and then exposed to a 2 h, 20.5°C (+12°C) heat stress or maintained at ambient temperature (8.5°C). Grey bar is level before injection, and white bars and scatter plots are 1 h after end of heat stress. Significant differences between groups are indicated by differing letters (a,b,c). n=8. 140 n pre-injection pre-stress chasing heat Figure 6.8: Plasma Cortisol levels in rainbow trout injected once a day for three days with saline (control), G S H or B S O , and then exposed to a 3 min chasing stress, or 2 h, 25.5°C heat stress (+10.5°C, ambient temperature 15°C). Plasma Cortisol of treatment groups was measured 1 h after the end of the heat or chasing stress. Significant differences between groups are indicated by differing letters (a,b). n=8. 120 Figure 6.9: Liver and posterior kidney Hsp70 levels in rainbow trout injected once a day for three days with saline (control), G S H or B S O , and then exposed to a 2 h, 20.5°C (+12°C) heat stress, or maintained at ambient temperature (8.5°C). Hsp70 was measured 10 h after the end of the heat stress. Significant differences between groups within tissue are indicated by differing letters (a,b,c). n=8. 121 > CD O C cu S-i o o >-. CD !-O CO 20 15 10 5 1 0 1.2 1.0 0.8 0.6 0.4 0.2 0.0 B X b I Control 1 1 1 1 GSH Y//A BSO • Pre-stress 1 1 25.5°C a a x , JL b JU Liver Kidney Figure 6.10: Liver and posterior kidney A) Hsp70 and B) Hsp30 levels in rainbow trout injected once a day for three days with saline (control), G S H or B S O , and then exposed to a 2 h, 25.5°C heat stress (+10.5°C, ambient temperature 15°C). Hsp70 was measured pre- or 10 h post-heat stress. Significant differences between groups within tissue are indicated by differing letters (a,b,c). n=8. 122 (+12°C) heat stress between G S H - and BSO-injected fish in either liver or kidney. In the 25.5°C heat stress experiment (+10.5°C, ambient temperature 15°C) there was a significant interaction between injection group and heat stress on Hsp70 levels in the kidney (p<0.001), but not in the liver (p=0.20), although there was a significant injection effect in the liver (p=0.001, Figure 6.10). Heat stress greatly increased Hsp70 levels and induced Hsp30 in all groups in both liver and posterior kidney (p<0.001). However, the magnitude of increase was reduced by G S H and BSO to differing degrees depending on the tissue and Hsp molecule. The 25.5°C (+10.5°C) heat stress increased Hsp70 to a similar extent in control and G S H groups in both the liver (approximately 60-fold) and kidney (approximately 80-fold). However, the increase was lower in the BSO group in both the liver (50-fold) and kidney (35-fold). The increase in Hsp30 after a 25.5°C (+10.5°C) heat stress was lower in the G S H group than in the control group in both liver and kidney (approximately 50% of control groups). The B S O group also had a lower increase in Hsp30 compared with the control group in the liver (approximately 60% of control) but not in the posterior kidney (approximately 95% of control). Experimental Series 6.2: The effects of altered GSH levels on the cellular stress response in vitro In two preliminary experiments, RTH-149 cells with altered G S H levels were exposed to either 24.5°C (+4.5°C) or 27°C (+7°C, ambient temperature 20°C) 1 h heat stresses. 18 h after both heat stresses, Hsp70 levels significantly increased above those of cells held at ambient temperature in all media groups (p<0.001, Figure 6.11). There was no effect of altered G S H levels on Hsp70 levels in ambient or heat-stressed cells in the 24.5°C (+4.5°C) experiment (p=0.89). However, in the 27°C (+7°C) experiment there was a significant interaction between altered G S H and heat stress (p<0.001). G S H -incubated cells had higher Hsp70 than control cells at ambient temperature, and lower Hsp70 than both control and BSO-incubated cells after a 27°C (+7°C) heat stress. The effects of altered G S H levels on Hsp levels were determined after a mild heat stress (24°C, +4°C), a heat stress known to induce Hsp30 (33°C, +13°C), and a lethal heat stress (35°C, +15°C, ambient temperature 20°C, Figure 6.12). There was a significant interaction between altered G S H levels and heat stress temperature on Hsp70 levels (p<0.001). Hsp70 levels in control and GSH-incubated RTH-149 cells increased significantly 18 h after 33°C (+13°C) and 35°C (+15°C) heat stresses, but not after a 24°C (+4°C) heat stress (ambient temperature 20°C, Figure 6.12A). Surprisingly, Hsp70 levels did not significantly increase in BSO-incubated cells after any heat stress. Hsp30 levels were only increased after the 33°C (+13°C) heat stress in control and GSH-incubated cells (p=0.004, Figure 6.12B), but not the BSO-incubated cells. Percent LDH released from cells as an indication of reduced viability is given in Figure 6.13. There was a significant interaction between altered G S H levels and heat stress temperature on LDH released (p=0.001). LDH released was significantly greater by 6-fold in BSO-incubated cells than control cells after 24°C (+4°C) heat stress and 2.5-fold after a 33°C (+13°C) heat stress, but not at ambient temperature (20°C) or after a 35°C (+15°C) heat stress. The G S H group had 1.3-fold higher LDH released and Hsp30 levels approximately 60% of control cells after a 33°C (+13°C) heat stress, although neither of these differences were significant. LDH release significantly increased in the B S O -incubated cells after all measured heat stresses, but only after the lethal heat stress in the G S H and control cells. When the above experiment was repeated with only the 33°C (+13°C) heat stress and measuring 123 stress heat stress temperature (°C) Figure 6.11: Hsp70 levels in RTH-149 cells incubated with normal media (control), media containing G S H for 6 h, or media containing B S O for 48 h, and then exposed to a 1h, 24.5 (+4.5°C) or a 27°C (+7°C) heat stress, or maintained at ambient temperature (20°C). Hsp levels were measured 18 h after the end of the heat stress. Significant differences between groups, within experiment are indicated by differing letters (a,b,c,d). n=6. O c CD •~ CD o CD _> 13 o 4^ a 1 to 0 a a ab 0 abcabc a ^ d 20 (A 1 J Control 3 GSH 7777] BSO abc d c X be 24 33 35 heat stress temperature (°C) CD 13 > CD O a CD CD CD ^ > 13 1-4 O ab X 33 Figure 6.12: A) Hsp70 and B) Hsp30 levels in RTH-149 cells incubated with normal media (control), media containing G S H for 6 h, or media containing B S O for 48 h, and then exposed to a 1 h heat stress of varying temperatures, or maintained at ambient temperature (20°C). Hsp levels were measured 18 h after the end of the heat stress. Significant differences between groups, within Hsp type, are indicated by differing letters (a,b,c,d,e). n=6. 124 heat stress temperature (°C) Figure 6.13: Percent lactate dehydrogenase (LDH) released from RTH-149 cells incubated with normal media (control), media containing G S H for 6 h, or media containing B S O for 48 h, and then exposed to a 1 h heat stress of varying temperatures (ambient temperature 20°C). LDH release was measured 18 h after the end of the heat stress. Significant differences between groups are indicated by differing letters (a,b,c). n=6. 125 Hsp30 and 70 levels at 6, 12, 18 and 24 h post-heat stress, the magnitude of increase in Hsp70 and 30 in control and G S H groups was similar to the above (Figure 6.14, 6.15). Unlike the above experiment, cells incubated with B S O for either 6 or 48 h before heat stress had similar Hsp70 or 30 levels as control and G S H groups at all measured time points. In addition, B S O incubation did not affect percent LDH released at any measured time point post4ieat stress (Figure 6.16). G S H incubation did not alter Hsp70 or 30 levels post-heat stress, but had approximately 3-fold higher LDH released than control cells at all time points in the 33°C (+13°C) exposed cells, and at 6, 12, and 18 h in 20°C incubated cells. Discussion This is the first study to examine the effects of altered G S H levels on the cellular and generalized stress responses of any vertebrate in vivo, and the first study to investigate the effects of altered G S H on the cellular stress response of fish cells in vitro. In almost all experiments and time points, G S H injection or incubation increased G S H levels and B S O injection or incubation decreased G S H levels in rainbow trout both in vivo and in vitro. This supports the idea that rainbow trout may be a good model for examining the role of G S H within vertebrates. As in Chapter 5, heat stress did not greatly affect tGSH levels in rainbow trout either in vivo or in vitro, although there were some minor, inconsistent effects of heat stress in B S O - and GSH-incubated cells in vitro. Both increased and decreased G S H levels influenced the generalized and cellular stress responses in rainbow trout, but this influence differed depending on the severity of the stressor, the component of the stress response examined, and whether studies were conducted in vivo or in vitro. This indicates that G S H affects the stress response in rainbow trout, but the relationship is not direct and is influenced by additional factors. Altered G S H levels in either direction were associated with decreased plasma Cortisol release after a 20.5°C (+12°C, ambient temperature 8.5°C) heat stress. This is consistent with a previous study that found antioxidant supplements decreased plasma Cortisol increase after heat stress in chickens (Mahmoud et al., 2004), and decreased G S H decreased Cortisol secretion in rainbow trout cells exposed to pesticides (Dorval and Hontela, 2003), although the latter also reported increased G S H levels increased Cortisol secretion to pesticides. Alterations in GSH-influenced redox status in either direction may have inhibited a step in Cortisol synthesis in rainbow trout exposed to a 20.5°C (+12°C) heat stress. Components of steroid synthesis pathways, including those for Cortisol, are reportedly altered by redox status, as ascorbic acid reportedly suppressed adrenocortical steroidogenesis in poultry (reviewed by Mahmoud et al., 2004), and decreased G S H levels suppressed plasma estradiol and progesterone in rats (Kang and Uthus, 1996). However, a balanced G S H status was not essential for stress-induced Cortisol release, as altered G S H levels did not influence plasma Cortisol levels after 25.5°C (+10.5, ambient temperature 15°C) heat stress or the physical stressor of chasing. The increased plasma glucose observed in rainbow trout injected with G S H , regardless of heat stress, is consistent with previous reports that increased G S H levels increased gluconeogenesis in rabbits (Winiarska et al., 2003). The above data indicate G S H modulates the generalized stress response to milder stressors. Although the mechanisms of this are unknown, G S H does not likely have a direct effect on the generalized stress response as altered G S H levels did not influence the response to more severe stressors. 126 2.0 1.5 3 1.0 0.5 0.0 T 12h 18h time (h from end of heat shock) r. • Control 1 GSH 6h T77n BS0 48h BS0 6h 24h Figure 6.14: Hsp70 levels in RTH-149 cells incubated with normal media (control), media containing G S H for 6 h, media containing B S O for 48 h, or media containing B S O for 6 h, and then exposed to a 1 h, 33°C heat stress (+13°C, ambient temperature 20°C). Significant differences between groups within time are indicated by differing letters (a,b). n=6. -> u S3 a o > -2 o m 7 n 5 1 4 1 3 i 2 A l H I I Control I I GSH 6h 17771 BS0 48h RVs^l BS0 6h I X I 6h 12h 18h time (h from end of heat shock) 24h Figure 6.15: Hsp30 levels in RTH-149 cells incubated with normal media (control), media containing G S H for 6 h, media containing BSO for 48 h, or media containing B S O for 6 h, and then exposed to a 1 h, 33°C heat stress (+13°C, ambient temperature 20°C). There were no significant differences among groups. n=6. 127 — OJ •J~. -I-Q — J 12 10 bcde cde tlcdef ac cde Tde ad ft I I I Control I I GSH 6h {7771 BS0 48h RV^I BS0 6h 6h 12h 18h 24h time (h from end of heat stress) Figure 6.16: Percent lactate dehydrogenase (LDH) released from RTH-149 cells incubated with normal media (control), media containing G S H for 6 h, media containing B S O for 48 h, or media containing BSO for 6 h, and then exposed to a 1 h, 33°C heat stress (+13°C, ambient temperature 20°C). Significant differences between groups are indicated by differing letters (a,b,c,d,e,f). Groups not significantly different from any other group are have no letters. n=6. 128 This is the first reported study of the effects of altered G S H levels on the heat shock response in a vertebrate in vivo, although this relationship has been examined extensively in mammals in vitro. Rainbow trout with altered G S H levels had higher Hsp70 levels after a relatively mild heat stress in vivo (20.5°C, +12°C, ambient temperature 8.5°C), indicating altered G S H status may decrease the temperature threshold of Hsp induction. This is consistent with mammalian studies in vitro that found increased (Abe et al., 1999), or decreased (Freeman et al., 1993) G S H levels decreased the temperature at which Hsp70 synthesis increased. However, in vitro, there was no difference in magnitude of Hsp response between control, G S H or B S O groups after a mild heat stress that significantly increased Hsp70 levels (24.5°C, +4.5°C), or after one that did not (24°C, +4°C, ambient temperature 20°C). This indicates that altered G S H levels do not influence the Hsp induction temperature in rainbow trout in vitro. The different effects of altered G S H on the Hsp response to milder heat stresses in vivo versus in vitro may be due to interactions with components of the generalized stress response in vivo which are not present in vitro. In vivo, lower Hsp70 levels in saline-injected fish correspond to higher plasma Cortisol levels compared to G S H - and BSO-injected fish. In fish, previous studies report high plasma C o r t i s o l levels decreased the Hsp response after a heat stress (Ackerman et al., 2000; Basu et al., 2001; Boone and Vijayan, 2002; Basu et al., 2003). Higher levels of Hsp70 in G S H - and BSO-injected fish in vivo may not have been due to direct effects of G S H status on Hsp induction, but to the removal of Cortisol induced-suppression of the Hsp response post-heat stress. Both increased and decreased G S H levels were associated with a diminished increase in Hsp levels after a 25.5°C heat stress in vivo (+10.5°C, ambient temperature 15°C), although the effects were not consistent between tissues or different Hsp types. Increased G S H decreased Hsp30 levels after a 25.5°C (+10.5°C, ambient temperature 15°C) heat stress in both liver and kidney, but did not affect Hsp70 levels. Hsp30 is only induced during severe heat stress, and its function as a molecular chaperone is not well examined in fish. In mammals, altered G S H levels affect the conformation of other small molecular weight Hsp's (e.g. Hsp25, Zavialov et al., 1998), and consequently alter the chaperoning ability of these molecules. Such a relationship has not been examined between the fish Hsp30 and G S H . However, if fish Hsp30 shares similar properties with mammalian small Hsp's, increased G S H levels may increase chaperoning ability and consequently decrease the total levels of Hsp30 required post-heat stress. This appears consistent with the hypothesis that increased G S H levels will ameliorate the cellular stress response. However, in vitro studies suggest otherwise. Hsp30 levels in RTH-149 cells after a 33°C (+13°C, ambient temperature 20°C) heat shock show a similar, although insignificant, decrease with increased G S H levels as in vivo. GSH-incubated cells had similar LDH released as control cells after this heat stress, indicating increased G S H does not impart a survival advantage to trout. Increased G S H also diminished the increase in Hsp70 levels after a 27°C (+7°C, ambient temperature 20°C) heat shock in vitro. However, in this experiment G S H also increased resting levels of Hsp70, which was not observed in any other experiment in vivo or in vitro. Additional factors may have influenced the relationship between G S H and Hsp70 in this group, which resulted in alterations in Hsp70 levels that were not observed in other experiments. Decreased G S H levels reduced all Hsp levels after a 25.5°C (+10.5°C, ambient temperature 15°C) heat shock in vivo with the exception of kidney Hsp30 levels. This may have occurred through a 129 number of mechanisms. Lower Hsp levels could indicate decreased G S H has a stabilizing effect on cellular proteins and consequently decreases the requirement for the protein chaperoning ability of Hsp's. However, previous mammalian studies in vitro have shown decreased G S H inhibited or suppressed the Hsp response to heat stress by inhibiting overall protein synthesis (Russo et al., 1984; Freeman et al., 1988) or by inhibiting HSF1 activation (Rokutan et al., 1996). In vitro, decreased G S H levels were inconsistently associated with a reduction in Hsp70 and 30 and increased LDH release relative to controls after a severe heat stress (33°C, +13°C, ambient temperature 20°C). This indicates decreased G S H levels may have reduced the ability of cells to withstand more severe heat stresses, which may impair or suppress the Hsp response to severe heat stress. The effect of decreased G S H on Hsp levels after a severe heat stress in vitro was inconsistent between experiments, as it is within the mammalian literature. The results of the final RTH-149 experiment, and those of one mammalian study (Harris et al., 1991) showed no effect of decreased G S H on Hsp levels after a severe heat stress, whereas other mammalian studies reported decreased G S H increased the Hsp response to severe heat stress (Freeman et al., 1993; Ito et al., 1998). These results suggest that decreased G S H does not alter the Hsp response to severe heat stress directly, but possibly through several indirect pathways. Although both increased and decreased G S H levels were associated with altered general and cellular stress responses in rainbow trout, neither alteration in G S H status appeared to impart an advantage to the fish. G S H appears to modulate the stress responses differently depending on the severity of the stressor and on interactions between the cellular and generalized stress responses. Other unidentified factors may modulate the effect of G S H on the cellular stress response in rainbow trout. For example, the differing effects of G S H status may be influenced by acclimation temperature. Ambient temperature differed between in vivo (8.5°C for the 20.5°C heat stress experiment, and 15°C for the 25.5°C heat stress experiment) and in vitro (20°C) experiments. A +10.5°C (25.5°C) heat stress resulted in significantly larger increases in plasma Cortisol and tissue Hsp levels compared to a +12°C (20.5°) heat stress, indicating acclimation temperature has a significant effect on the change in temperature required to cause stress in fish. This is consistent with Howell et al. (1991), who reported myofibrils from warm-acclimated fish required a smaller change in temperature before protein denaturation that myofibrils from cold-acclimated fish. Acclimation to different temperatures results in a suite of changes including altered protein and lipid conformation. For example, fish acclimated to different temperatures have altered composition of membrane phospholipids to maintain membrane fluidity (Hazel and Prosser, 1974; Bowden et al., 1996). This may in turn alter heat-induced damage to cellular membranes which can increase the Hsp response to heat stress (Samples et al., 1999). G S H , along with the glutathione peroxidase enzymes, can stabilize lipid membranes (see Hayes and McLellan, 1999). G S H status may consequently ameliorate a potential lipid-induced portion of the Hsp response differently at different temperatures. The effects of altered G S H differed greatly between similar experiments in vitro. Possible factors that may have influenced the effect of G S H status on the heat shock response in vitro could be thermal history, stage of cell cycle and composition of fetal calf serum within the culture media. These in turn may affect the redox status or different enzyme expression of the cells, which may have influenced the importance of G S H during heat stress. While there were some similarities in the effects of altered G S H 130 levels on Hsp levels in vivo versus in vitro, in general these effects were more significant in vivo. This suggests G S H may be involved in interactions between cellular and other physiological processes after heat stress that are not present in a reduced preparation such as a cell line. In Chapter 5 I demonstrated that G S H dynamics are not affected by stressors such as heat that do not directly alter cellular redox status. Although altered G S H levels influenced both the generalized and cellular stress responses, this influence varied between experiments, suggesting G S H is involved in, but not consistently a key component of, the generalized stress response in rainbow trout. As G S H levels can be altered by factors such as acclimation temperature, growth rate, and feeding level (Chapters 3 and 4), this may impact how the fish responds to stress. Summary • Both decreased and increased G S H levels prevented the increase in plasma Cortisol levels after a 20.5°C (+12°C, ambient temperature 8.5°C) heat stress, but did not affect the C o r t i s o l response to more severe heat (25.5°C, +10°C, ambient temperature 15°C) or chasing stresses in rainbow trout in vivo. At the cellular level, both decreased and increased G S H levels lowered the induction temperature of Hsp70 in rainbow trout. As well, increased G S H levels decreased Hsp30, and decreased G S H levels decreased Hsp70 and 30 after a severe heat stress in vivo. • In vitro, increased G S H levels did not greatly affect the Hsp response, but decreased G S H levels inconsistently lowered the upper temperature at which Hsp levels increase. • G S H does not consistently affect the ability of rainbow trout to respond to stress, but appears to indirectly influence components of both the cellular and generalized stress response, indicating it is a minor component of the integrated stress response in these fish. 131 References Abe, T., Gotoh, S., Higashi, K., 1999. Higher induction of heat shock protein 72 by heat stress in cisplatin-resistant than in cisplatin-sensitive cancer cells. Biochimica et Biophysica Acta 1445: 123-133. Abe, T., Konishi, T., Katoh, T., Hirano, H., Matsukuma, K., Kashimura, M., Higashi, K., 1994. Induction of heat shock 70 mRNA by cadmium is mediated by glutathione suppressive and non-suppressive triggers. Biochimica et Biophysica Acta 1201: 29-36. Ackerman, P.A., Forsyth, R.B., Mazur, C.F. , Iwama, G.K., 2000. Stress hormones and the cellular stress response in salmonids. Fish Physiology and Biochemistry 23: 327-336. Almar, M., Otero, L , Santos, C , Gallego, J .G . , 1998. Liver glutathione content and glutathione-dependent enzymes of two species of freshwater fish as bioindicators of chemical pollution. Journal of Environmental Science and Health. Part B: Pesticides, Food Contaminants, and Agricultural Wastes 33: 769-783. Barton, B.A., Iwama, G.K., 1991. Physiological changes in fish from stress in aquaculture with emphasis on the response and effects of corticosteroids. Annual Reviews of Fish Diseases 1: 3-26. Basu, N., Kennedy, C .J . , Iwama, G.K., 2003. The effects of stress on the association between hsp70 and the glucocorticoid receptor in rainbow trout. Comparative Biochemistry and Physiology. Part A, Physiology 134: 655-663. Basu, N., Nakano, T., Grau, E.G., Iwama, G.K., 2001. The effects of Cortisol on heat shock protein 70 levels in two fish species. General & Comparative Endocrinology 124: 97-105. Bell, G. , Cowey, C.B. , 1990. Oxidation of glutathione during hydroperoxide metabolism in isolated hepatocytes of rainbow trout (Salmo gairdneri). Fish Physiology and Biochemistry 8: 11-17. Bergmeyer, H.U. Methods of Enzymatic Analysis. Volume VI: Metabolites I: Carbohydrates. V C H Publishers, Deerfield Beach, FL. Bijur, G.N. , Davis, R.E., Jope, R.S., 1999. Rapid activation of heat shock factor-1 DNA binding by H 2 0 2 and modulation by glutathione in human neuroblastoma and Alzheimer's disease cybrid cells. Molecular Brain Research 71: 69-77. Boone, A .N . , Vijayan, M.M., 2002. Glucocorticoid-mediated attenuation of the hsp70 response in trout hepatocytes involves the proteasome. American Journal of Physiology - Regulated and Integrated Comparative Physiology 283: R680-R687. Boone, A .N . , Ducouret, B., Vijayan, M.M., 2002. Glucocorticoid-induced glucose release is abolished in trout hepatocytes with elevated hsp70 content. Journal of Endocrinology 172: R1-R5. Bowden, L.A., Restall, C .J . , Rowley, A.F. , 1996. The influence of environmental temperature on membrane fluidity, fatty acid composition and lipoxygenase product generation in head kidney leucocytes of the rainbow trout, Oncorhynchus mykiss. Comparative Biochemistry and Physiology. Part B, Biochemistry and Molecular Biology 115B: 375-382. Dorval, J . , Hontela, A., 2003. Role of glutathione redox cycle and catalase in defense against oxidative stress induced by endosulfan in adrenocortical cells of rainbow trout (Oncorhynchus mykiss). Toxicology and Applied Pharmacology 192: 191-200. Esposito, F., Cuccovillo, F., Morra, F., Russo, T., Cimino, R., 1995. DNA binding activity of the glucocorticoid receptor is sensitive to redox changes in intact cells. Biochimica et Biophysica Acta 1260: 308-314. Forsyth, R.B., Candido, E.P.M. , Babich, S.L., Iwama, G.K., 1997. Stress protein expression in coho salmon with bacterial kidney disease. Journal of Aquatic Animal Health 9: 18-25. Freeman, M.L., Meredith, M.J., Laszlo, A., 1988. Depletion of glutathione, heat-shock protein-synthesis, and the development of thermotolerance in chinese-hamster ovary cells. Cancer Research 48: 7033-7037. Freeman, M.L., Sierra-Rivera, E., Voorhees, G .J . , Eisert, D.R., Meredith, M.J., 1993. Synthesis of hsp-70 is enhanced in glutathione-depleted hep G2 cells. Radiation Research 135: 387-393. Freeman, M.L., Huntley, S.A., Meredith, M.J., Senisterra, G.A., Lepock, J.R., 1997. Destabilization and denaturation of cellular protein by glutathione depletion. Cell Stress and Chaperones 2: 191-198. 132 Gaubin, Y., Vaissade, F., Croute, F., Beau, B., Soleihavoup, J . -P . , Murat, J . -C . , 2000. Implication of free radicals and glutathione in the mechanism of cadmium-induced expression of stress proteins in the A549 human lung cell-line. Biochimica et Biophysica Acta 1495: 4-13. Griffith, O.W., 1980. Determination of glutathione and glutathione disulfide using glutathione reductase and 2-vinylpyridine. Analytical Biochemistry 106: 207-212. Harris, C., Juchau, M.R., Mirkes, P.E., 1991. Role of glutathione and hsp-70 in the acquisition of thermotolerance in postimplantation rat embryos. Teratology 43: 229-239. Hayes, J.D., McLellan, L.I., 1999. Glutathione and glutathione-dependent enzymes represent a co-ordinately regulated defense against oxidative stress. Free Radical Research 31: 273-300. Hazel, J . , Prosser, C., 1974. Molecular mechanisms of temperature compensation in poikilotherms. Physiological Reviews 54: 620-677. Hoppe, G. , Chai, Y .C . , Crabb, J.W., Sears, J . , 2004. Protein s-glutathionylation in retinal pigment epithelium converts heat shock protein 70 to an active chaperone. Experimental Eye Research 78: 1085-1092. Howell, B.K., Matthews, A.D., Donnelly, A .P. , 1991. Thermal stability of fish myofibrils: a differential scanning calorimetric study. International Journal of Food Science and Technology 26: 283-295. Hughes, E., Gallagher, E., 2004. Effects of b-naphthoflavone on hepatic biotransformation and glutathione biosynthesis in largemouth bass (Micropterus salmoides). Marine Environmental Research 58: 675-679. Ito, H., Okamoto, K., Kato, K., 1998. Enhancement of expression of stress proteins by agents that lower the levels of glutathione in cells. Biochimica et Biophysica Acta 1397: 223-230. Iwama, G.K., Thomas, P.T., Forsyth, R.B., Vijayan, M.M., 1998. Heat shock protein expression in fish. Reviews in Fish Biology and Fisheries 8: 35-56. Iwama, G.K., Vijayan, M.M., Forsyth, R.B., Ackerman, P., 1999. Heat shock proteins and physiological stress in fish. American Zoology 39: 901-909. Kang, Y . J . , Uthus, E.O., 1996. Suppression of plasma estradiol and progesterone concentrations by buthionine sulfoximine in female rats. Biochemical Pharmacology 51: 567-570. Liu, H., Lightfoot, R., Stevens, J.L., 1996. Activation of heat shock factor by alkylating agents is triggered by gluathione depletion and oxidation of protein thiols. The Journal of Biological Chemistry 271: 4805-4812. Mahmoud, K.Z., Edens, F.W., Eisen, E.J. , Havenstein, G.B., 2004. Ascorbic acid decreases heat shock protein 70 and plasma corticosterone response in broilers (Gallus gallus domesticus) subjected to cyclic heat stress. Comparative Biochemistry and Physiology. Part B, Biochemistry and Molecular Biology 137: 35-42. Makino, Y., Yoshikawa, N., Okamoto, K., Hirota, K., Yodoi, J . , Makino, I., Tanaka, H., 1999. Direct association with thioredoxin allows redox regulation of glucocorticoid receptor function. The Journal of Biological Chemistry 274: 3182-3188. McDuffee, A T . , Senisterra, G. , Huntley, S., Lepock, J.R., Sekhar, K.R., Meredith, M.J., Borrelli, M.J., Morrow, J.D., Freeman, M.L., 1997. Proteins containing non-native disulfide bonds generated by oxidative stress can act as signals for the induction of the heat shock response. Journal of Cellular Physiology 171: 143-151. Otto, D.M.E., Moon, T.W., 1995. 3,3',4,4'-Tetrachlorobiphenyl effects on antioxidant enzymes and glutathione status in different tissues of rainbow trout. Pharmacology and Toxicology 77: 281-287. Otto, D.M.E., Sen, C.K., Hidiroglou, N., Madere, R., Moon, T.W., 1997. Role of exogenous glutathione in teleost fish and its effects on antioxidant defense responses in rainbow trout exposed to 3,3',4,4'-tetrachlorobiphenyl. Fish Physiology and Biochemistry 16: 449-457. Pena-Llopis, S., Pena, J .B. , Sancho, E., Fernandez-Vega, C , 2001. Glutathione-dependent resistance of the European eel Anguilla anguilla to the herbicide molinate. Chemosphere 45: 671-681. Rokutan, K., Hirakawa, T., Teshima, S., Honda, S., Kishi, K., 1996. Glutathione depletion impairs transcriptional activation of heat shock genes in primary cultures of guinea pig gastric mucosal cells. Journal of Clinical Investigation 97: 2242-2250. 133 Russo, A., Mitchell, J .B. , McPherson, S., 1984. The effects of glutathione depletion on thermotolerance and heat stress protein synthesis. British Journal of Cancer 49: 753-758. Samples, B.L., Pool, G.L., Lumb, R.H., 1999. Polyunsaturated fatty acids enhance the heat induced stress response in rainbow trout (Oncorhynchus mykiss) leukocytes. Comparative Biochemistry and Physiology. Part B, Biochemistry and Molecular Biology 123: 389-397. Smith, P.K., Krohn, R.I., Hemanson, G.T., Mallia, A.K., Gartner, F.H., Provenzano, M.D., Fujimoto, E.K., Goeke, N.M., Olson, B.J. , Klenk, D.C., 1985. Measurement of protein using bicinchoninic acid. Analytical Biochemistry 150: 76-85. Stephensen, E., Sturve, J . , Fbrlin, L , 2002. Effects of redox cycling compounds on glutathione content and activity of glutathione-related enzymes in rainbow trout liver. Comparative Biochemistry and Physiology. Part C, Pharmacology, Toxicology and Endocrinology 133: 435-442. Trinder, P., 1969. Determination of glucose in blood using glucose oxidase with an alternative oxygen acceptor. Annals of Clinical Biochemistry 6: 24-27. Will, 0 . , Mahler, H . - C , Arrigo, A. -P. , Epe, B., 1999. Influence of glutathione levels and heat-shock on the steady-state levels of oxidative DNA base modifications in mammalian cells. Carcinogenesis 20: 333-337. Winiarska, K., Drozak, J . , Wegrzynowicz, M., Jagielski, A.K., Bryla, J . , 2003. Relationship between gluconeogenesis and glutathione redox state in rabbit kidney-cortex tubules. Metabolism: Clinical and Experimental 52: 739-746. Zavialov, A.V., Gaestel, M., Korpela, T., Zav'yalov, V .P . , 1998. Thiol/disulfide exchange between small heat shock protein 25 and glutathione. Biochimica et Biophysica Acta 1388: 123-132. Zou, J . , Salminen, W.F. , Roberts, S.M. , Voellmy, R., 1998. Correlation between glutathione oxidation and trimerization of heat shock factor 1, an early step in stress induction of the H S P response. Cell Stress and Chaperones 3: 130-141. 134 CHAPTER 7 General Discussion and Future Directions Although some studies have examined individual factors of the glutathione (GSH) system in fish, this thesis represents the first large body of work examining G S H dynamics in fish beyond its importance during exposure to toxic and oxidative stressors. By addressing the following three objectives, I examined the mechanisms by which G S H is transported in fish tissues, the influence of metabolism on G S H dynamics, and the role of G S H in the stress response in fish. Objective I: To determine the mechanisms by which GSH is transported in fish. Rainbow trout were able to take up exogenous G S H by various mechanisms, depending on the tissue. The liver was unique in its ability to directly uptake G S H by apparent gradient-dependent transport. The posterior kidney, liver and gill took up G S H by extracellular breakdown and intracellular synthesis, indicating endogenous G S H levels were not sufficient to inhibit G S H synthesis. This is the first report of any vertebrate taking up significant quantities of G S H by such a pathway. There was little uptake of exogenous G S H evident in the brain, muscle, or intestine of rainbow trout. Objective II: To examine the importance of GSH during conditions that alter metabolism. Both altered acclimation temperature and accelerated growth rates influenced G S H dynamics in various fish models to varying degrees. Temperature influenced G S H dynamics in fish, as G S H turnover was proportional to incubation temperature in RTH-149 cells in vitro, and hepatic G S H enzyme activity and liver, brain and muscle tGSH levels were proportional to acclimation temperature in killifish in vivo, although acclimation temperature did not affect tGSH levels in heart, gill or plasma. These were the first experiments to examine the effects of temperature on G S H turnover in any model, and the first to examine the specific effects of acclimation temperature on G S H dynamics in an ectotherm. Chapter 4 represented the first study examining the effects of a growth hormone transgene on G S H dynamics in an ectotherm. Growth hormone transgenic coho salmon had higher G S H levels and activity of associated enzymes than control fish in several tissues. In the liver, this was related to metabolic rate, while in other tissues altered G S H dynamics were better associated with increased feeding and growth rate in these fish. Objective III: To examine the role of GSH as a component of the generalized and cellular stress responses in fish. Heat stress had little effect on G S H dynamics in rainbow trout in vivo or in vitro. The response of rainbow trout differs from previous studies in fish that found heat stress decreased G S H levels in catfish (Heteropneustes fossilis, Parihar et al., 1997), and snakeheads (Channa punctata, Kaur et al., 2005). However, altered G S H levels indirectly influenced the generalized stress response of rainbow trout in vivo, and the cellular response to heat stress in vivo and in vitro. These were the first experiments to demonstrate G S H influences the primary and cellular stress responses in a vertebrate in vivo, and the first to demonstrate G S H influences the cellular stress response in fish in vitro. While the G S H antioxidant system of rainbow trout may not be altered in response to stress in general, adequate 135 G S H levels may be necessary to mount both generalized and cellular stress responses in some circumstances. These experiments demonstrate several unique aspects of G S H metabolism and function in various fish models. Specifically, G S H levels in fish were lower than the potential synthesis rate in some tissues. Conditions that influence metabolism (e.g. increasing temperature and growth rate) resulted in increased G S H utilization (e.g. turnover), increased G S H levels and activity of several associated enzymes, and an increased ability to metabolize G S H from the diet. High potential for G S H synthesis in some fish tissues may allow for rapid increases in G S H levels during times of high demand and increased precursor supply. Unlike mammalian systems (see Ballatori et al., 2005), direct import of exogenous G S H was observed in the liver of rainbow trout (Chapter 2). However, this is likely not the result of a unique G S H transporter in fish, but possibly by gradient-driven transporters which normally function as G S H exporters, similar to those of mammals. In mammals, the liver, in conjunction with these transporters, acts to maintain plasma G S H levels at approximately 20pM (Anderson et al., 1980; Lauterburg et al., 1984). The liver of fish likely has a similar role, and despite relatively low tissue levels, maintains plasma levels similar to those of mammals (10-30pM, Chapter 2-4). The role of the G S H and GSH-S-conjugate transporter identified in skate liver (Rebbeor et al., 2000) should be examined in this respect. Rainbow trout may have a unique method of G S H catabolism, as blocking the G S H catabolic enzyme y-glutamyltranspeptidase (yGT) with acivicin did not prevent G S H uptake by breakdown and resynthesis in the kidney (Chapter 2). However, this possibility should be viewed with caution as such an alternate system has not been identified in any other organism, and acivicin-induced inhibition of yGT influenced G S H uptake in other tissues (liver and to a lesser extent the gill). As well, a number of problems are associated with acivicin, such as its ability to bind non-specifically to the y-glutamyl binding sites of other proteins, suggesting non-specific effects of this inhibitor could exist and thus results should be interpreted with caution. G S H levels and activity of associated enzymes in the liver, brain and muscle of various fish were altered by factors that influence metabolic rate, specifically acclimation temperature in killifish (Chapter 3), and growth rate in coho salmon (Chapter 4). Hepatic G S H levels were correlated to metabolic rates in growth hormone transgenic and non-transgenic coho salmon, indicating G S H levels may be upregulated to reduce metabolically produced reactive oxygen species (ROS) in this tissue. Although regulation of G S H levels in fish may be metabolically-driven in the liver, other tissues did not show the same changes in G S H dynamics with altered temperature or growth rate. As well, a number of other factors associated with increasing acclimation temperature and growth rate may affect G S H dynamics. Fish compensate for increased acclimation temperature by altering lipid and protein conformation, and often metabolically compensate for temperature changes to maintain adequate activity and growth over a large range of temperatures (see Hazel and Prosser, 1974). Alterations in lipid conformation may influence demands on G S H in changing environmental temperatures. Partial metabolic compensation for temperature in killifish over some temperatures (Targett, 1978) suggests that metabolic rate alone does not explain effects of temperature on G S H dynamics in this fish. 136 The gill and posterior kidney are postulated to have similar GSH-dependent functions as the liver in fish (Gallagher and Di Giulio, 1992). Combined with the ability to take up exogenous G S H by breakdown and resynthesis, this suggests G S H function and regulation would be altered by similar conditions as the liver. Yet G S H levels in these tissues were unaffected by acclimation temperature (killifish gill, Chapter 3) or growth rate (coho salmon kidney, Chapter 4). Potential compensation in metabolic rates or other unidentified factors in these tissues may remove the increased demand for G S H in these tissues, or other antioxidants may be upregulated instead of G S H . Another intriguing possibility is that the ability to excrete hydrogen peroxide in fish may prevent the need for increased G S H levels in these tissues. When examining hydrogen peroxide excretion in fish, Wilhelm-Filho et al. (1994) did not determine if hydrogen peroxide excretion was specifically or exclusively via the gills. Fish in freshwater produce copious amounts of dilute urine to maintain ionic balance. This may enable them to excrete hydrogen peroxide produced in the kidney out through the urine, as well as excretion through the gill, to combat metabolically-induced or other increases in hydrogen peroxide production in these tissues. The experiments in Chapters 5 and 6 demonstrated that although G S H dynamics in rainbow trout are not altered by a stressor that does not directly cause oxidative stress (e.g. temperature), altered G S H levels influence components of the generalized and cellular stress responses to heat stress. However, the influence of G S H on these systems was inconsistent between studies, indicating there is no direct relationship between G S H and components of the stress response in rainbow trout. The influence of G S H on the response to a heat stress may depend on the acclimation temperature, the magnitude of increase in temperature, as well as in vivo and in vitro differences. Although lowered G S H levels were detrimental in some studies, whether increased G S H levels were beneficial or detrimental to the fish was inconclusive. Consequently, artificially increasing endogenous G S H levels to match tissue synthesis rates may not lend an advantage to the fish, at least in terms of the ability to respond to heat or handling stress. Although G S H may be important to recover from heat and other stressors, its role in the generalized and cellular stress responses remains inconclusive. I examined G S H dynamics in fish as an indicator of antioxidant potential under changing conditions. However, it is important to note that, although G S H is a key antioxidant due to its high intracellular levels, it is part of an integrated system composed of several antioxidants. Different antioxidants neutralize different R O S , and as such are difficult to compare. Environmental fluctuations may impose antioxidant-specific alterations in fish, and other antioxidants may influence the response of G S H to these fluctuations. For example, superoxide dismutase (SOD) is responsible for reducing the superoxide radical to hydrogen peroxide, which is then reduced to water by G S H , G P x and catalase (CAT). SOD levels are relatively high in fish blood, whereas CAT and G P x are low (see Wilhelm-Filho et al., 2000). The ability to excrete hydrogen peroxide through the gills may decrease the necessity for antioxidants responsible for its reduction, while the ability to reduce other R O S remains in high demand. Although G S H levels in the gill, R B C or kidney were not altered by factors that influenced metabolism, S O D activity may have been increased to compensate for potential higher production of the superoxide radical from mitochondrial respiration under those conditions. CAT and G S H / G P x appear to neutralize hydrogen peroxide from different sources in fish. Janssens et al. (2000) suggested CAT is more important in house-keeping removal of hydrogen peroxide, possibly due to its localization in the 137 peroxisomes, while G P x is involved in metabolically produced hydrogen peroxide. As such, G S H / G P x may be more important in maintaining redox balance during altered environmental and physiological conditions. Finally, G S H interacts with several other antioxidants and consequently the impact of G S H may be influenced by their concentration. For example, Jurima-Romet et al. (1996) found Vitamins C and E could replace the antioxidant roles of G S H during G S H depletion, and Wefers and Sies (1988) found protection against lipid peroxidation by G S H and Vitamin C was dependent on Vitamin E. Different levels of endogenous Vitamin C and E may have resulted in the inconsistent influence of G S H on components of the stress response in Chapter 6. The objectives of my thesis focused on examining the role of G S H as an antioxidant during changing physiological conditions in fish. G S S G and tGSH levels and G R activity were altered by temperature and growth rate in fish, indicating GSH 's function as an antioxidant is significant under these conditions. However, G S H also has an important role in glutathione-S-transferase- (GST) dependent detoxification. The focus of G S T research in fish has examined its role during exposure to exogenous contaminants such as pesticides and general pollution. However, G S T and G S H are also important in detoxifying endogenously-produced metabolites, and this may be influenced by fluctuating environmental conditions. For example, stress can increase levels of plasma catecholamines which are then broken down in the gills, kidney and liver (see Randall and Perry, 1992). Catecholamine catabolism can result in production of electrophilic-catecholamine intermediates that require G S H and G S T for detoxification and excretion (Baez et al., 1997). Activity of G S T enzymes may be increased to cope with fluctuating environmental conditions in tissues involved in GSH-dependent detoxification, such as the liver, kidney and gills, and consequently requires examination. Due to equipment restrictions, in several of my experiments (Experiments 2.1, 2.2, 3.2, 4.2, 6.1), each treatment was represented by fish in a single tank. As such, tank effects were not taken into consideration during statistical analyses. Where replicate tanks were used (Experiment 5.1), there was no statistical tank effect on G S H components measured. These same tanks were used for several other experiments (Experiments 2.1, 2.2, 6.1), and consequently, tank effect likely did not play a significant part in differences observed between treatments. However, it is important to note the statistical flaw in these experiments. All future experiments should be run in replicate tanks wherever possible. Meyer et al. (2003) found offspring of killifish living in a contaminated stream had higher levels of G S H components than offspring of animals living upstream of the contaminated site. This indicates that G S H dynamics in fish may be important to adapt to different environmental conditions on an evolutionary scale. In this thesis I found that increasing growth rate in coho salmon, and acclimation temperature in killifish and a rainbow trout cell line, altered G S H components over several weeks to months. This indicates that G S H dynamics are adjusted to changing environmental and physiological conditions on an acclimation scale. However, acute heat stress did not affect G S H dynamics in rainbow trout over the short-term (hours). Components of the G S H system in fish can adapt and acclimate to deal with changing environments, although their importance during short-term environmental perturbations may be restricted to oxidative and toxicant stressors. 138 Future Directions I have examined the pathways by which G S H is transported in fish, and established that G S H dynamics are adjusted to cope with altered physiological conditions in fish. However, several specific questions remain unanswered. In Chapter 2, acivicin, a non-specific inhibitor of yGT, did not prevent G S H uptake by breakdown and resynthesis in the kidney. This indicates that fish may have an alternate and unique means of G S H breakdown. However, apparent yGT-independent uptake of G S H in the kidney may be due to difficulties with the acivicin blocker, and not to a novel system of G S H catabolism. Several other inhibitors of yGT are available, but most have similar problems as associated with acivicin (see Burg and Mulder, 2002). London and Gabel (2001) developed a borate-inhibitor (ABBA) of yGT that is more specific than acivicin, and could be used to address whether yGT-independent uptake of G S H takes place in the kidney. However, preliminary studies suggest A B B A is quickly broken down in vivo (London, pers. comm.). Consequently future studies would be restricted to in vitro models. I have postulated that the ability to excrete hydrogen peroxide from the gills, and possible the posterior kidney, is the major factor influencing the lack of-GSH alterations by acclimation temperature or growth rates in these tissues. Future experiments would first address whether freshwater fish are able to excrete hydrogen peroxide in their urine, as well as through the gills. In addition, determining whether hydrogen peroxide excretion is influenced by acclimation temperature would address the above hypothesis. Examining S O D activity, as well as G S H , G P x and CAT activity in these tissues would determine if altered environmental conditions result in upregulation of antioxidants required for removal of unexcretable R O S (e.g. superoxide), and result in a lack of upregulation of antioxidants required for removal of excretable R O S (e.g. hydrogen peroxide). The results of Chapters 6 demonstrated that G S H may play a minor role in the generalized and cellular responses to stress in fish. However, I was unable to determine whether increased G S H levels impart an advantage to fish or if decreased G S H levels put fish at a disadvantage during stress. By determining the effect of altered G S H levels on the production of oxidatively damaged biological molecules during stress (e.g. lipid peroxidation, protein oxidation), I would be better able to define the role of G S H in the redox balance of fish during short-term perturbations such as temperature stress. In many respects, G S H acts as a non-specific redox buffer of the cellular environment. As such, G S H may influence many systems in the cell that are controlled by redox status, and G S H may be adjusted to maintain the redox balance in fluctuating environmental and physiological conditions. In some respects, this makes G S H metabolism and cellular roles difficult to discern with certainty, as G S H likely influences many cellular processes and this influence may be enhanced or diminished by the activity of other antioxidants. Future research examining the role of G S H in maintaining homeostasis of fish in fluctuating environments needs to address G S H not as an individual system, but as a component of the integrated antioxidant system off ish. 139 References Anderson, M.E., Bridges, R.J . , Meister, A., 1980. Direct evidence for inter-organ transport of glutathione and that the non-filtration renal mechanism for glutathione utilization involves g-glutamyl transpeptidase. Biochemical and Biophysical Research Communications 96: 848-853. Baez, S., Segura-Aguilar, J . , Widersten, M., Johansson, A . -S . , Mannervik, B., 1997. Glutathione transferases catalyse the detoxication of oxidized metabolites (o-quinones) of catecholamines and may serve as an antioxidant system preventing degenerative cellular processes. Biochemical Journal 324: 25-28. Ballatori, N., Hammond, C.L., Cunningham, J.B. , Krance, S .M. , Marchan, R., 2005. Molecular mechanisms of reduced glutathione transport: role of the M R P / C F T R / A B C C and OATP/SLC21A families of membrane proteins. Toxicology and Applied Pharmacology 204: 238-255. Burg, D., Mulder, G .J . , 2002. Glutathione conjugates and their synthesis derivatives as inhibitors of glutathione-dependent enzymes involved in cancer and drug resistance. Drug Metabolism Reviews 34: 821-863. Gallagher, E.P., Di Giulio, R.T., 1992. A comparison of glutathione-dependent enzymes in liver, gills and posterior kidney of channel catfish (Ictalurus punctatus). Comparative Biochemistry and Physiology. P a r t e , Pharmacology, Toxicology and Endocrinology 102: 543-547. Hazel, J . , Prosser, C , 1974. Molecular mechanisms of temperature compensation in poikilotherms. Physiological Reviews 54: 620-677. Janssens, B.J. , Childress, J . J . , Baguet, F., Rees, J. -F. , 2000. Reduced enzymatic antioxidative defense in deep-sea fish. The Journal of Experimental Biology 203: 3717-3725. Jurima-Romet, M., Abbott, F.S., Tang, W., Huang, H.S., Whitehouse, L.W., 1996. Cytotoxicity of unsaturated metabolites of valproic acid and protection by vitamins C and E in glutathione-depleted rat hepatocytes. Toxicology 112: 69-85. Kaur, M., Atif, F., Ali, M., Rehman, H., Raisuddin, S., 2005. Heat stress-induced alterations of antioxidants in the freshwater fish Channa punctata Bloch. Journal of Fish Biology 67: 1653-1665. Lauterburg, B.H., Adams, J.D., Mitchell, J.R., 1984. Hepatic glutathione homeostasis in the rat: efflux accounts for glutathione turnover. Hepatology 4: 586-590. London, R., Gabel, S. , 2001. Development and evaluation of a boronate inhibitor of g-glutamyl transpeptidase. Archives of Biochemistry and Biophysics 385: 250-258. Meyer, J . , Smith, J . , Winston, G. , Di Giulio, R., 2003. Antioxidant defenses in killifish (Fundulus heteroclitus) exposed to contaminated sediments and model prooxidants: short-term and heritable responses. Aquatic Toxicology 65: 377-395. Parihar, M.S., Javeri, T., Hemnani, T., Dubey, A.K., Prakash, P., 1997. Responses of superoxide dismutase, glutathione peroxidase and reduced glutathione antioxidant defenses in gills of the freshwater catfish (Heteropneustes fossilis) to short-term elevated temperature. Journal of Thermal Biology 22: 151-156. Randall, D.J., Perry, S.F. , 1992. Catecholamines. In: Hoar, W.S. , Randall, D.J. Farrell, A . P . (Eds.), Fish Physiology Vol 12B: The Cardiovascular System. Academic Press, San Diego, pp. 255-300. Rebbeor, J .F. , Connolly, G .C. , Henson, J .H. , Boyer, J.L. , Ballatori, N., 2000. ATP-dependent G S H and glutathione S-conjugate transport in skate liver: role of an Mrp functional homologue. American Journal of Physiology-Gastrointestinal and Liver Physiology 279: G417-G425. Targett, T.E. , 1978. Respiratory metabolism of temperature acclimated Fundulus heteroclitus (L): Zones of compensation and dependence. Journal of Experimental Marine Biology and Ecology 32: 197-206. Wefers, H., Sies, H., 1988. The protection by ascorbate and glutathione against microsomal lipid peroxidation is dependent on vitamin E. European Journal of Biochemistry 174: 353-357. Wilhelm-Filho, D., Gonzalez-Flecha, B., Boveris, A., 1994. Gill diffusion as a physiological mechanism for hydrogen peroxide elimination by fish. Brazilian Journal of Medical and Biological Research 27: 2879-2882. 140 Wilhelm-Filho, D., Torres, M.A., Marcon, J . L , Fraga, C .G . , Boveris, A., 2000. Comparative antioxidant defences in vertebrates - emphasis on fish and mammals. Trends in Comparative Biochemistry And Physiology 7: 33-45. 141 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0092848/manifest

Comment

Related Items