UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Impact of iron-catalyzed dietary lipid peroxidation on growth performance, general health and flesh proximate… Sutton, Jill 2004

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

Item Metadata

Download

Media
831-ubc_2004-0647.pdf [ 4.46MB ]
Metadata
JSON: 831-1.0091636.json
JSON-LD: 831-1.0091636-ld.json
RDF/XML (Pretty): 831-1.0091636-rdf.xml
RDF/JSON: 831-1.0091636-rdf.json
Turtle: 831-1.0091636-turtle.txt
N-Triples: 831-1.0091636-rdf-ntriples.txt
Original Record: 831-1.0091636-source.json
Full Text
831-1.0091636-fulltext.txt
Citation
831-1.0091636.ris

Full Text

IMPACT OF IRON-CATALYZED DIETARY LIPID PEROXIDATION ON GROWTH PERFORMANCE, GENERAL H E A L T H AND FLESH PROXIMATE AND FATTY ACID COMPOSITION OF ATLANTIC SALMON (Salmo salar) IN SEAWATER.  by JILL SUTTON  B.Tech. (Environmental Engineering), British Columbia Institute of Technology, 2002 B.Sc. (Fisheries Resource Management), University of Northern British Columbia, 2000  A THESIS SUBMITTED IN PARTIAL FULFILMENT OF T H E REQUIREMENTS FOR THE DEGREE OF  MASTER OF SCIENCE in THE F A C U L T Y OF G R A D U A T E STUDIES (Food Science)  We accept this thesis as conforming to the required standard  THE UNIVERSITY OF BRITISH C O L U M B I A September 2004 © Jill Sutton, 2004  FACULTY OF GRADUATE STUDIES  THE UNIVERSITY OF BRITISH COLUMBIA  Library Authorization  In p r e s e n t i n g this t h e s i s in partial f u l f i l 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 a n a d v a n c e d d e g r e e a t t h e U n i v e r s i t y o f British C o l u m b i a , I a g r e e t h a t t h e L i b r a r y shall m a k e it f r e e l y a v a i l a b l e f o r r e f e r e n c e a n d s t u d y . I f u r t h e r a g r e e t h a t p e r m i s s i o n f o r e x t e n s i v e c o p y i n g o f this t h e s i s f o r s c h o l a r l y p u r p o s e s m a y b e g r a n t e d b y t h e h e a d o f m y d e p a r t m e n t o r b y h i s o r h e r r e p r e s e n t a t i v e s . It is u n d e r s t o o d t h a t c o p y i n g o r p u b l i c a t i o n o f this t h e s i s f o r f i n a n c i a l g a i n s h a l l n o t b e a l l o w e d w i t h o u t m y w r i t t e n p e r m i s s i o n .  Jill S u t t o n  06/10/2004  N a m e o f A u t h o r (please print)  Date (dd/mm/yyyy)  Title o f T h e s i s :  Impact of iron-catalyzed dietary lipid peroxidation on growth performance, general  health and flesh proximate and fatty acid composition of Atlantic salmon (Salmo salar) in seawater  Degree:  M.Sc.  Department of  Food Science  2004  T h e U n i v e r s i t y o f British C o l u m b i a Vancouver, B C  Canada  grad.ubc.ca/forms/?formlD=THS  page 1 of 1  last updated: 5-Oct-04  ABSTRACT Post-juvenile Atlantic salmon (Salmo salar) were fed to satiation, twice daily for 126 days,with diets supplemented with copper (10 or 35 mg/kg), iron (333 or 1000 mg/kg) 'and unoxidized or /  oxidized dietary lipid measured as peroxide value (PV, <5 or <5+ or 35+ meq/kg). The dietary effects measured include growth performance, general health, flesh proximate and fatty acid composition, and in situ oxidative stress. Lipid peroxidation in the fishmeal and fish oil was controlled by adding ethoxyquin (150 mg/kg) to diets containing unoxidized dietary lipid (<5 meq/kg). A l l diets contained the required level of vitamin E (30 IU/kg). Diets supplemented with the highest level of iron, without anti-oxidant protection (no ethoxyquin), showed the greatest loss of eicosapentaenoic (EPA) and docosahexaenoic acids (DHA) and vitamin E (p<0.05). Fish fed these diets displayed significantly (p<0.05) decreased growth, feed intake, feed efficiency, and utilization of gross energy and protein that were attributable to the losses of the preceding essential nutrients and possibly dietary protein modification. Dietary vitamin E concentrations decreased in all diets, except those with low levels of iron and peroxide values, during the 126-day study period. Diets without anti-oxidant supplementation had significantly (p<0.05) higher lipid peroxidation levels as compared with anti-oxidant-containing diets. However, fish hepatic and muscle tissue lipid peroxidation were unaffected. Autopsy-based assessments of general health indicate that fish fed diets influenced by iron-catalyzed lipid peroxidation exhibited clinical signs of poor health indicative of dietary vitamin E and omega-3 highly unsaturated fatty acid deficiency. These results suggest that fish diets undergoing ironcatalyzed lipid peroxidation with attendant deficiencies in omega-3 highly unsaturated fatty acids and vitamin E adversely influence the growth performance, general health, and fillet proximate  and fatty acid composition of post-juvenile Atlantic salmon. The consumption of oxidized dietary lipids by the experimental fish did not influence in situ oxidative stress parameters as measured in this study.  T A B L E OF CONTENTS Page Abstract  ii  Table of Contents  iv  List of Tables  vii  List of Figures  ix  List of Appendices  xi  Acknowledgements  xii  1.0  Introduction  1  2.0  Literature review  4  2.1  Aquaculture  2.2 Lipid peroxidation 2.2.1 Mechanisms of lipid peroxidation 2.2.2 Stages 2.2.3 Polyunsaturated fatty acid (PUFA) peroxidation 2.2.4 Techniques 2.2.4.1 Loss of substrates 2.2.4.2 Total peroxide measurement 2.2.4.3 Enzymatic assays 2.2.4.4 Product separation 2.2.4.5 End products 2.2.4.6 Physical measurement 2.2.5 Factors affecting lipid peroxidation 2.2.5.1 Pro-oxidants 2.2.5.1.1 Iron 2.2.5.1.2 Copper 2.2.5.2 Anti-oxidants 2.2.5.2.1 Vitamin E 2.2.5.2.2 Vitamin C 2.2.5.2.3 Manganese 2.2.5.2.4 Zinc 2.2.5.2.5 Ethoxyquin 2.2.5.2.6 Selenium-dependent glutathione peroxidase 2.2.5.2.7 Catalase 2.2.5.2.8 Superoxide dismutase 2.2.5.3 Biological interactions of anti-oxidants  4 5 7 7 8 9 9 10 10 11 11 12 12 12 13 14 15 16 17 18 18 18 19 20 20 21 iv  2.2.6 Biological impacts and significance 2.2.7 Fate of oxidized dietary lipids 2.2.8 Health implications of oxidized dietary lipids 2.2.8.1 Atherosclerosis 2.2.8.2 Cancer 2.2.8.3 Liver damage 2.3 Fish nutrition 2.3.1 Energy 2.3.2 Protein 2.3.3 Lipids  21 23 23 23 24 25 25 26 27 28  3.0 Impact of iron-catalyzed dietary lipid peroxidation on growth performance, general health and flesh proximate and fatty acid composition of Atlantic salmon (Salmo salar) in seawater 37 Abstract  37  3.1  38  Introduction  3.2 Materials and methods 3.2.1 Experimental diets 3.2.2 Diet storage trial 3.2.3 Experimental fish and dietary trial conditions 3.2.4 Sample collection and biometric analyses 3.2.5 Proximate composition of experimentalfishand diets 3.2.6 Fatty acid methyl esters 3.2.7 Fatty acid analyses 3.2.8 Measurement of malondialdehyde in dietary treatments 3.2.9 Determination of vitamin E content 3.2.10 Determination of selenium-dependent glutathione peroxidase 3.2.11 Determination of conjugated dienes 3.2.12 Determination of ethoxyquin 3.2.13 Histopathology 3.2.14 Metal and trace element determination 3.2.15 General health 3.2.16 Experimental design and statistical analysis  40 40 42 42 43 44 44 45 46 47 47 48 48 48 49 49 49  3.3 Results 3.3.1 Malondialdehyde content of dietary treatments 3.3.2 Body weight gains and feed efficiency 3.3.3 Feed intake, PER, PPD, GEU and proximate composition 3.3.4 Fatty acid composition of dietary lipid 3.3.5 Fatty acid composition of muscle tissue 3.3.6 Dietary vitamin E 3.3.7 Tissue vitamin E 3.3.8 Measures of oxidative stress 3.3.9 General health observations  50 51 51 52 52 53 54 54 55 55  3.4 Discussion 3.4.1 Antagonistic behaviour ofpro-oxidants 3.4.2 Dietary iron effects 3.4.3 Protein modification  55 56 57 58 v  3.4.4 3.4.5 3.4.6 3.4.7 3.5  Fatty acid deficiency Nutritional quality offillets Vitamin E deficiency Feed intake Conclusion  58 59 60 60 61  4.0  General Conclusion  78  5.0  References  81  vi  LIST OF TABLES Chapter 2  Page  Table 2.1 Nutrient requirements for Atlantic salmon as percentages of diet, milligrams per kilogram of diet, or international units (IU) per kilogram of diet (as-fed basis). Adapted from N R C (1993). R = nutrient is required but the level is not determined. N T = not tested.  36  Chapter 3 Table 3.1 Ingredient composition of the dietary treatments fed to Atlantic salmon over a 126-day period.  67  Table 3.2 68 Initial dietary concentrations of proximate constituents, ethoxyquin, copper and iron on a dry weight basis and initial peroxide values. Copper was in the form of CuS04-5H20 and iron was in the form of F e S 0 7 H 0 . 4  2  Table 3.3 69 Mineral compositions of the dietary treatments measured as mg per kg diet (dry weight basis). Table 3.4 70 Growth-related performance data of fish fed diets 1 through 12. SGR represents specific growth rate and F E represents feed efficiency. Values are means,± standard deviation, of 3 replicate tanks for each diet. Means for a given parameter that share a common superscript letter were not significantly different (ns). Table 3.5 71 Feed intake (g fish" ), protein efficiency ratios (PER, [wet weight gain (g) • protein consumption (g) '])> percent protein deposition (%PD, [protein gained in fish (g) • total protein consumed (g)" • 100]), and gross energy utilization (GEU, [gross energy gained by the fish (MJ) • total gross energy consumed (MJ)" • 100]), and final proximate composition of fish (% body weight basis). Initial muscle and whole body proximate compositions for all experimental fish were 19.0% and 16.5% for protein, 1.8% and 6.8% for lipid, 69.2% and 65.5% for moisture, and 7.4% and 8.6% for ash, respectively. Values are means, ± standard deviation, of 3 replicate tanks for each diet. Means for a given parameter that share a common superscript letter were not significantly different. 1  1  vn  Table 3.6a Relative fatty acid composition (% of total fatty acids) for diets 1 through 12 on day 0 of storage trial. * Represents all fatty acids in this category; some are not shown. Means for a given parameter that share a common superscript letter were not significantly different (p<0.05). Values are means, ± standard deviation, of 3 replicate tanks for each diet. Not significant is represented by ns. Table 3.6b Relative fatty acid composition (% of total fatty acids) for diets 1 through 12 on day 42 of storage trial. *Represents all fatty acids in this category; some are not shown. Means for a given parameter that share a common superscript letter were not significantly different (p<0.05). Values are means, ± standard deviation, of 3 replicate tanks for each diet. Not significant is represented by ns. Table 3.6c Relative fatty acid composition (% of total fatty acids) for diets 1 through 12 on day 84 of storage trial. *Represents all fatty acids in this category; some are not shown. Means for a given parameter that share a common superscript letter were not significantly different (p<0.05). Values are means, ± standard deviation, of 3 replicate tanks for each diet. Not significant is represented by ns. Table 3.6d Relative fatty acid composition (% of total fatty acids) for diets 1 through 12 on day 126 (end) of storage trial. *Represents all fatty acids in this category; some are not shown. Means for a given parameter that share a common superscript letter were not significantly different (p<0.05). Values are means, ± standard deviation, of 3 replicate tanks for each diet. Not significant is represented by ns. Table 3.7 Relative fatty acid composition (% of total fatty acids) of muscle tissue for fish fed diets 1 through 12. The data only represent day 126 of the growth trial. *Represents all fatty acids in this category; some are not shown. Means for a given parameter that share a common superscript letter were not significantly different (p<0.05). Not significant is represented by ns. Table 3.8 Vitamin E (Vit. E) concentrations in the diets at the beginning (day 0) and day 126 of the storage trial. Vitamin E concentrations of the muscle and liver tissues of the fish fed diets 1 through 12 at the end of the 126-day growth trial are also shown. Values are means, ± standard deviation, of 3 replicate tanks for each diet. Means, within each column, with common superscript letters were not significantly different (p<0.05)  LIST OF FIGURES  Chapter 2  Page  Figure 2.1 World aquaculture production from 1950-2001 (FAO, 2002).  30  Figure 2.2 Auto-oxidation of an olefinic compound. R H represents an olefinic compound where the H is attached to an allylic carbon atom; R* represents the substrate radical; R O O H represents a hydroperoxide; ROO* represents a peroxyl radical; RO* represents an alkoxy radical; O H * represents a hydroxyl radical. Initiation of auto-oxidation can be catalyzed by light, heat or pro-oxidants in the presence of oxygen. Figure modified from Gunstone (1996) and Kanner and Rosenthal (1992).  31  Figure 2.3 Chemical changes during auto-oxidation of olefinic compounds and some examples of methods commonly used to characterize the changes.  32  Figure 2.4 The important mechanisms of transition metal ion (i.e. iron) acceleration of lipid peroxidation. Fe represents iron; R O O H represents a hydroperoxide; ROO* represents a peroxyl radical; RO* represents an alkoxy radical; R* represents the alkyl radical; O H * represents a hydroxyl radical.  33  Figure 2.5 The basic reaction between vitamin E and radicals. A H represents vitamin E; A * represents oxidized vitamin E; R O O H represents a hydroperoxide, R*, RO*, and ROO* represent radicals; R H , R O H , R A , R O A , R O O A represent stabilized lipid substrates.  34  Figure 2.6 The oxidation reduction pathways of glutathione. GSSG represents the oxidized form of glutathione (glutathione disulfide); GSH represents the reduced form of glutathione. One molecule of hydrogen peroxide is reduced to 2 molecules of water while 2 molecules of glutathione (GSH) are oxidized in a reaction catalyzed by the selenoenzyme, glutathione peroxidase. Oxidized glutathione (GSSG) may be reduced by the flavin adenine dinucleotide (FAD)-dependent enzyme, glutathione reductase. Adapted from Meister (1976).  35  IX  Chapter 3 Figure 3.1 Malondialdehyde (MDA) content of the dietary treatments at the start (day-0, black bars) and end (day-126, grey bars) of the growth trial. Values are means, ± standard deviation, of 3 replicate tanks for each diet. Means with a common superscript letter were not significantly different (p<0.05) on day 0 (a-b) and day 126 (A-E).  63  Figure 3.2 Relationship between final feed efficiency and final malondialdehyde (MDA) content of the dietary treatments. Values are means, ± standard deviation, of 3 replicate tanks for each diet. Spearman rank correlation of r= -0.71, p<0.05.  64  Figure 3.3 Relationship between final dietary n-3 polyunsaturated fatty acids (n-3 H U F A ) and final malondialdehyde (MDA) content of dietary treatments. Values are means, ± standard deviation, of 3 replicate tanks for each diet. Spearman rank correlation of r= -0.85, p<0.05.  65  Figure 3.4 Relationship between muscle vitamin E concentration and final malondialdehyde (MDA) content of dietary treatments. Values are means, ± standard deviation, of 3 replicate tanks for each diet. Spearman rank correlation of r= -0.68, p<0.05.  66  x  LIST OF APPENDICES Appendix 1 90 Data represent total mortalities for the fish fed diets 1-12 at each sampling interval and for the entire 126-day growth trial. *Mortalities due to technical problem with air line to tank.  Appendix 2 91 In situ measures of oxidative stress. Values are means, ± standard deviation, of 3 replicate tanks for each diet. Appendix 3 92 General health observations recorded throughout the 126-day growth trial. Symbols represent; An-anemia, Fh-fin hemorrhage, Gp-pale gills, Lh-liver hemorrhage, Ly-yellow liver, Se-enlarged spleen, Sb-black spleen, Od- uncharacteristic odour and (#)-# fish >1.  XI  ACKNOWLEDGEMENTS Many people helped in the completion of this thesis project including Shannon Balfry Shannon Chan, Erin Friesen, Navneet Gill, Chen-Huei Huang, David Kitts, Rosalind Leggatt, John Lumsden, Janice Oakes, Mahmoud Rowshandeli, Nancy Richardson, Yssac Vargas, and Sherman Yee. I appreciate the enthusiasm and effort that all of you brought to this project. I have enormous gratitude to my supervisors, Dave and Brent. I cannot overstate the level of support, encouragement, sound advice, good teaching, and good company that both of you provided. The past few years have been made easier by the encouragement, friendship and support of many people including Deepa, Elizabeth, Erin, Jabus, Jake, Jessica, Lisa, Mareike, Rosalind, Shannon, Shawna, and Wendy. Thank you for helping me through the difficult times, and for all the emotional support, humour and entertainment. I would not have attempted, let alone completed, this thesis without the endless support of the people I consider family. M y father, who taught me to enjoy nature and my mother, who loves the ocean and a good jigs dinner, are to blame for my compassion for nature and encouraging me to study fish in the first place. David, like a good brother, has constantly challenged me with his passion for knowledge and its communication. Reanna, Kaleb, Nathyn, and Sharon though newly acquired, have shown me that life can be complicated and simple at the same time, and to enjoy those subtleties. Anne and Del always made me feel like I had a second family by their welcoming, supportive, and humorous nature. Friends like Trent and Shelley, who are always ready to offer "constructive" criticism and insight into the matters of science, art, war and love, are invaluable and I cannot overstate how influential their presence has been. And finally, I owe tremendous gratitude to Steve, for more than I can express in words and for helping me understand that I should not be afraid of what I do not know.  1.0  INTRODUCTION  The concept of sustainable aquaculture, including spatial and temporal dimensions of environmental, economic, and social elements, has been driven primarily by critics and opponents of aquaculture (Pauly et al., 2003). The aquaculture industry is growing rapidly worldwide but continues to contend with issues of disease, product quality, feed contamination, environmental impacts and public outrage (Stokstad, 2004). Although opposition to aquaculture has hindered its growth, the industry has benefited by being pushed to become productive under constrained conditions. These constraints are beneficial because they influence the safety and product quality of aquaculture products; conditions that can lever their market value.  Fish nutrition and its impact on animal welfare is an important aquaculture issue. The original development of purified and practical, formulated diets for intensive aquaculture resulted in high mortality rates related to a poor understanding of fish nutrition (Halver, 2002). For several hundred years, nutrient-deficient and imbalanced diets have been recognized as a causative factor in the development of disease (Halver, 2002). However, the nutrient requirements and effect of processing on the digestibility and nutritive value of protein and lipid sources for aquatic species are not clearly understood (Halver, 2002). Further, the growth of aquaculture nutrition and the aquafeed sector must contend with rapidly approaching biological limits on key feed ingredients, such as fishmeal and oil from wild fish stocks, and the need to identify and develop alternatives (Hardy et a l , 2001).  Diets deficient in essential fatty acids (EFA) cause most of the nutritional disease outbreaks in farmed fish (Tacon, 1996). E F A , viz. eicosapentaenoic acid (EPA) and docosahexanoic acid (DHA), are essential for the optimal growth and development of marine fish (Sargent et al., 1999; Sargent et al., 2002). However, n-3 highly unsaturated fatty acids, such as E P A and D H A , are susceptible to lipid peroxidation and i f this occurs to an appreciable extent, feed intake and utilization can be adversely affected and non-infectious and infectious disease can result (Mourente et al., 2002; Tacon, 1996).  Lipid peroxidation causes beneficial and detrimental changes in the biochemical, sensory and nutritional properties of food. Several studies have linked the consumption of lipid peroxidation products to disease development; however, the current literature fails to address in depth how the bioavailability, uptake, distribution and mechanisms associated with the different types and concentrations of lipid peroxidation products influence fish health. Accordingly, the nutritional value, safety and sensory quality of the edible flesh from fish fed diets influenced by lipid peroxidation has not been examined.  The primary objective of this study was to examine the impact of pro-oxidant catalyzed lipid peroxidation on the (1) growth performance, (2) fatty acid, proximate composition and vitamin E content, (3) in situ lipid peroxidation, and (4) general health of post-juvenile Atlantic salmon (Salmo salar). Previous research has examined the impact of pro-oxidant catalyzed lipid peroxidation on rainbow trout (Desjardins et al., 1987). In addition, the effect of previously oxidized lipids has been addressed in coho and Atlantic salmon (Forster et al., 1988; Koshio et al., 1994) and other fish (Mourente et al., 2002). However, those experiments did not examine how pro-oxidant catalyzed lipid peroxidation influenced the nutritive value of the feed for salmon. 2  In this regard, the secondary objective of this study was to examine how diet formulation and duration of diet storage influenced (1) the vitamin E content and (2) the proximate and fatty acid compositions of the dietary treatments. Vitamin E and unsaturated fatty acids are susceptible to lipid peroxidation. Therefore, temporal monitoring was necessary for to gain a complete understanding of how changes in these dietary components could potentially influence the nutritional status of the experimental fish and subsequently, their health and flesh quality.  This research was designed to assist in understanding how diets undergoing varying rates of prooxidant-catalyzed lipid peroxidation, due to the presence or absence of ethoxyquin and two dietary levels of the pro-oxidants iron and copper, can influence the welfare and production of post-juvenile Atlantic salmon (i.e. 60-300 g).  3  2.0  2.1  LITERATURE REVIEW  Aquaculture  Aquaculture continues to develop in response to the uncertainty and instability of the worlds' fish production in natural systems (Lovell, 1998). Historically, the worldwide demand for fish products was easily matched by supply but was hindered by technological limitations (Higgs and Dong, 2000; Lovell, 1998). Presently, demand for fish products is no longer hindered by technology, but by the finite and fully utilized supply of wild fish stocks (Lovell, 1998). As a result, the aquaculture industry has become a significant source of fish products. However, the flesh quality and safety of cultured fish continues to endure significant scrutiny (Koshio et al., 1994; Stokstad, 2004).  Global aquaculture production has increased markedly over the past two decades (Naylor et al., 2000) with China contributing to 75% of the yield (Figure 2.1). Concurrently, interest in aquaculture nutrition has developed as links between nutritional status, disease and production became apparent (Waagbo, 1997).  The most common nutritional diseases in farmed fish are due to diets deficient in essential fatty acids (EFA) or improper formulation of dietary lipids (Tacon, 1996). Carnivorous fish must obtain high amounts of n-3 highly unsaturated fatty acids (n-3 H U F A ) from their diet due their inability or varying ability to desaturate and elongate linolenic acid (C18:3n-3) to eicosapentaenoic acid (EPA, C20:5n-3) and docosahexanoic acid (DHA, C22:6n-3) (Hamre et al., 2001; Mourente et al., 2002; Sargent et al., 1999). The n-3 H U F A are essential for the 4  optimal growth and development of marine fish (Sargent et al., 1999; Sargent et al., 2002). However, feeds rich in these fatty acids are susceptible to lipid peroxidation and the consumption of the peroxidation products may lead to disease (Mourente et al., 2002; Tacon, 1996).  Recent human studies focusing on the health implications of ingesting oxidized dietary lipids have tried to determine their biological fate by examining the biological activity and metabolic pathways of lipid peroxidation products and how they could promote disease (Dobarganes and Marquez-Ruiz, 2003). Strong correlations have been made between consumption of oxidized lipids and disease promotion, but the bioavailability of lipid peroxidation products in homeotherms and poikilotherms is still poorly understood (Billek, 2000; Dobarganes and Marquez-Ruiz, 2003; Hamre et al., 2001; Haumann, 1993; Riedemann and Ward, 2002). For example, oxidized dietary lipids have been linked to being a source of plasma hydroperoxides (Ursini et al., 1998; Williams et al., 1999), but whether these compounds were produced in the diet, or produced in vivo, still remains to be determined (Dobarganes and Marquez-Ruiz, 2003; Hamre et al., 2001).  2.2  Lipid peroxidation  Lipid peroxidation, involving the reaction of molecular oxygen with lipids at double bond sites, has gained scientific interest due its impact on food quality and role in biological processes (Kanner and Rosenthal, 1992; Nakamura et al., 1998; Porter et al., 1995; Yang, 1992). Biologically, lipid peroxidation has been linked with functional and structural alteration of membranes, hormones, vitamins (Kanner and Rosenthal, 1992; Nakamura, et al., 1998), D N A and protein (Gershon, 1988; Halliwell and Chirico, 1993; Marnett, 1987), tumour initiation (Marnett, 1987), arteriosclerosis (Hessler et al., 1983) and aging (Cheeseman and Slater, 1993)  and alternatively, decreasing tissue damage during sepsis (Riedemann and Ward, 2002). Lipid peroxidation has also been directly correlated with flavour and aroma deterioration, declining food wholesomeness, and food safety (Kanner and Rosenthal, 1992). Accordingly, the influence of dietary lipid peroxidation products has gained interest due to the possible link between nutrition and biological oxidative stress (Addis, 1986; Kubow, 1992).  Despite the complexity of lipid peroxidation reactions, both in vivo and in dietary lipids, the key mechanisms and variables involved in photo-oxidation, auto-oxidation, and enzymatic oxidation are recognized and accepted (Frankel, 1998). Current research has focused on either the identity and quantitation of the oxidation compounds involved in the various oxidation reactions, given the large amount and variety of compounds that can be potentially formed from each oxidizable substrate (Dobarganes and Marquez-Ruiz, 2003), or it has aimed to establish a relationship between dietary lipid peroxidation and biological oxidative stress (Billek, 2000).  At the present time, there is little research on the effects of consuming oxidized lipids, which is essential to developing an understanding whether moderate consumption of these lipids is sufficient to cause any physiological effects (Dobarganes and Marquez-Ruiz, 2003). Current research indicates that lipid peroxidation does not represent a significant human health risk since chronic or acute exposure levels required for toxicological development would alter the sensory quality and the food item would likely be rejected (Dobarganes and Marquez-Ruiz, 2003). Ironically, baking and frying processes involving thermo-oxidation impart aromas and flavours that are highly desired by consumers (Dobarganes and Marquez-Ruiz, 2003; Haumann, 1993).  6  2.2.1 Mechanisms of lipid peroxidation In a peroxide-free system, the initiation of lipid peroxidation occurs when a hydrogen atom is lost from a methylene group (> CH2 group) of an unsaturated fatty acid (Benzie, 1996; Halliwell and Gutteridge, 1989). Although hydrogen atoms can be abstracted from fatty acids by any chemical species that has enough reactivity, such as reduced iron, the peroxyl and hydroxyl radicals are the most significant contributors to continual hydrogen abstraction during lipid peroxidation (Benzie, 1996; Halliwell and Chirico, 1993). The loss of hydrogen atoms is easier in polyunsaturated fatty acids (PUFA), because of the relative ease with which a hydrogen atom can be removed from a double bond structure, making the double bonds highly susceptible to radical attack (Halliwell and Chirico, 1993).  2.2.2 Stages Lipid peroxidation is a complex process that occurs in multiple stages (Halliwell and Gutteridge, 1989). It is comprised of three main stages that are known as initiation, propagation and termination (Figure 2.2).  The initiation stage (auto-oxidation) begins with an induction period that is generally very slow and is succeeded by a period of more rapid reaction; the propagation stage (Gunstone, 1996). Auto-oxidation of lipids is a radical chain process (Gunstone, 1996), where one radical begets the production of another (Halliwell and Chirico, 1993). Hydroperoxides, formed as a consequence of the radical chain process, can damage hormones, vitamins and pigments, readily decompose into secondary oxidation compounds that are responsible for rancid flavours, and have been linked to the production of toxic compounds (Halliwell and Chirico, 1993). However, the initiation stage is still not fully understood due to the complicated nature of three possible reactions including (1) metal-catalyzed radical production, (2) photo-oxygenation and the 7  production of hydroperoxides, and (3) thermal stimulation of the initiation stage (Gunstone, 1996).  The propagation stage succeeds the initiation stage when molecular oxygen is added to the alkyl radical (R*) (Gunstone, 1996; Porter et al., 1995). The duration of this stage is dependent upon many factors including the ratio of lipids to proteins, the fatty acid composition, and the presence and concentrations of one or more anti-oxidants (Halliwell and Chirico, 1993). Propagation is also more complicated than the scheme in Figure 2.2 suggests and can include "radical coupling with oxygen, atom or group transfer, fragmentation, rearrangement and cyclization" (Porter et a l , 1995). As a result, the reaction products of auto-oxidized lipids are very difficult to characterize (Porter et al., 1995).  The termination reaction is only activated when there are sufficient numbers of radicals present; an event that does not commonly occur biologically (Bors et al., 1990). However, inhibition of the auto-oxidation of lipids is generally attempted during the induction step or by promoting the termination reaction before the propagation stage can proceed through more than a few cycles (Gunstone, 1996). Anti-oxidants are used to inhibit the induction stage and promote the termination reaction (Gunstone, 1996).  2.2.3  Polyunsaturated fatty acid (PUFA) peroxidation  Lipid peroxidation requires the constant generation of the primary oxidation products known as hydroperoxides (Dobarganes and Velasco, 2002). Hydroperoxides decompose to create an assortment of non-volatile and volatile compounds such as polymeric triacylglycerides and aldehydes, respectively (Figure 2.3) (Dobarganes and Velasco, 2002). The polyunsaturated fatty 8  acid moieties or olefinic acids, including linolenate and linoleate, are the most susceptible to oxidation (Dobarganes and Velasco, 2002).  Both isolated P U F A and those integrated into lipids and lipid matrices are readily attacked by radicals causing impairment of cell membrane function, changes in membrane fluidity, inactivation of receptors and enzymes, and increased cell membrane permeability to ions (Halliwell and Chirico, 1993). This complicates lipid peroxidation in biological systems due to the assortment of lipids in matrices that also contain proteins, carbohydrates, enzymes, nucleic acids and metals that may stimulate or diminish its effects (Cheeseman, 1990). Further, there are several varieties of radicals, including the hydroxyl and superoxide radicals, that have unique chemical reactions (Halliwell and Chirico, 1993). For example, the hydroxyl radical reacts with most molecules at very high rates (10 -10 M ' V ) , whereas the superoxide radical has a much 9  10  1  lower reactivity (10 NT's" ) (Eberhardt, 2001). As a result, no single assay method can be used 3  1  to fully describe the impact of lipid peroxidation in a biological system (Cheeseman, 1990).  2.2.4 Techniques Several techniques are commonly used to measure the rate of oxidation, each one measuring something different (Halliwell and Gutteridge, 1989). The following section generally describes the most commonly applied techniques for analysis of lipid peroxidation.  2.2.4.1 Loss of substrates Lipid peroxidation causes loss of unsaturated fatty-acid side-chains (Halliwell and Gutteridge, 1989). Hence, a simple way to measure the overall rate of peroxidation is to measure the loss of each fatty acid (Halliwell and Gutteridge, 1989). The lipids must be hydrolyzed to release the 9  fatty acids which can be measured by high performance liquid chromatography (Halliwell and Gutteridge, 1989). Analysis by gas-liquid chromatography requires the hydrolyzed fatty acids to be further converted to volatile products (methyl esters) prior to injection (Halliwell and Gutteridge, 1989). A l l sample preparation should be done under nitrogen gas to prevent false results that could arise from oxidation reactions (Halliwell and Gutteridge, 1989).  2.2.4.2 Total peroxide measurement Hydroperoxides, the primary products formed by the reaction of oxygen with unsaturated fatty acid moieties, can be measured based on their ability to oxidize iodine from potassium iodide or ferrous to ferric ions (Dobarganes and Velasco, 2002; Lezerovich, 1985). The hydroperoxide concentration is usually expressed as peroxide value (PV or POV) and the concentration is expressed in terms of milli-equivalents of peroxide per 1 kg of sample (Hara and Totani, 1988). The most widely used method to determine P V involves volumetric measurements of iodine produced from potassium iodide (Lezerovich, 1985; Yildiz et al., 2003).  2.2.4.3 Enzymatic assays Enzymatic reactions can be used as sensitive and specific methods for the analysis of hydroperoxides (Dobarganes and Velasco, 2002). These assays require very small amounts of sample but they are more complex than those based on chemical methods since they provide information on what classes of hydroperoxides are present in the sample (Dobarganes and Velasco, 2002). However, these assays are only applicable to biological samples since they reflect the activity of enzymes present in the in vivo anti-oxidant defence system (Dobarganes and Velasco, 2002).  10  2.2.4.4 Product separation The objective of these methods is to provide specific information about the compounds present in a sample (Halliwell and Chirico, 1993). Dobargnes and Velasco (2002) pointed out that the methodological advancement of chromatographic techniques by combining gas chromatography and mass spectrometry (GC-MS) allowed for the understanding of oxidation pathways. These methodologies have been used to elucidate the various mechanisms and compositions of lipids undergoing photo-oxidation, thermal oxidation, auto-oxidation and enzymatic oxidation. These techniques have also allowed for the detailed characterization of oxidation products that may be related to clinical diagnosis of disease.  2.2.4.5 End products End products are the compounds that are produced by the decomposition of primary peroxidation products (Shahidi and Wanasundara, 2002). The measurement of these products is more appropriate for food systems since they generally emit an odour (Shahidi and Wanasundara, 2002). Most tests to determine end products are theoretically simple but in practice, it is very difficult to ensure accurate values (Benzie, 1996). Problems often arise due to contamination and post-sampling changes that can confound and invalidate the results (Benzie, 1996).  The most commonly used end product assay is the measurement of thiobarbituric reactive substances (TBARS), such as malondialdehyde. In general terms, the assay involves reacting thiobarbituric acid (TBA) with malondialdehyde to form a colored chromophore (Gunstone, 1996). This assay, which can be performed by either distillation or extraction, is very useful for monitoring changes in the same products with different treatments (Shahidi and Wanasundara, 2002).  2.2.4.6 Physical measurement The transformation of P U F A to hydroperoxides and the subsequent production of low molecular weight breakdown products allow for direct measurement of the physical changes due to lipid peroxidation (Dobarganes and Velasco, 2002; Gray and Monahan, 1992). However, detection of lipid peroxidation products using physical methods is not widely used (Dobarganes and Velasco, 2002; Robards, et al., 1988). Although measurement of conjugated dienes is relatively inexpensive, it is limited to the earlier stages of peroxidation and the presence of compounds absorbing in the same region may interfere with determination (Halliwell and Chirico, 1993; Shahidi and Wanasundara, 2002). Other methods including Fourier Transform Infrared Spectroscopy and Exclusion Chromatography could be very useful but the equipment is currently too expensive for regular or quality assurance use (Dobarganes and Velasco, 2002).  2.2.5  Factors affecting lipid peroxidation  Lipid peroxidation has been recognized as an important reaction in nutrition and food safety as well as in physiological and toxicological processes (Schaich, 1992). Several factors including presence of oxygen and the level of H U F A , anti-oxidants and pro-oxidants can influence lipid peroxidation. Interrelationships between these dietary components can have synergistic or antagonistic effects in the diet, during absorption and within the gastrointestinal tract, or in vivo (Lygren et al., 1999).  2.2.5.1 Pro-oxidants Generally, pro-oxidants are elements or compounds that catalyze lipid peroxidation. Oxygen, a bi-radical with two unpaired electrons, reacts poorly with non-radicals due to spin restriction (Gutteridge and Halliwell, 1988). Therefore, transition metals become important because they 12  are able to change oxidation states by single-electron transfers (Gutteridge and Halliwell, 1988; Schaich, 1992). This association between oxygen utilization and transition metal catalysis has evolved throughout aerobic life with numerous examples of iron and copper acting as redox catalysts at "the active sites of oxygenases, oxidases, anti-oxidants, oxygen transport, and electron transport proteins" (Gutteridge and Halliwell, 1988)  Metal catalysis of lipid peroxidation in foods and tissues is chemically and physically complex and generally does not follow classic model-predicted patterns (Schaich, 1992). This has been attributed to the complexity of food and tissue environments and the multiple reaction pathways that are difficult to control and reproduce (Gutteridge and Halliwell, 1988; Schaich, 1992). For example, lipid peroxidation can become inhibited due to high metal concentration. This is thought to result from the oxidation and reduction of radicals by the transition metal ions, iron and copper (Schaich, 1992). These relationships, however, are poorly understood (Schaich, 1992).  A l l transition metal ions, except titanium, have the ability to exist in two or more oxidation states (Eberhardt, 2001). Therefore, the oxidation of these metal ions can proceed through a one- or two-electron transfer (Eberhardt, 2001). Reactions involving transition metal ions in lipid peroxidation are often simplified to one-electron transfer processes (Figure 2.4) although the chemistry is far more complex (Halliwell and Gutteridge, 1990).  2.2.5.1.1 Iron Information regarding the absorption and metabolism of iron in aquatic organisms is limited (Lall, 2002). The concentration of iron required is dependent on the species, life history stage, 13  nutritional status, health, iron status, and digestive physiology (Lall, 2002). To complicate matters further, the form and bioavailability of iron from different feedstuffs is highly variable and can affect the dietary requirements (Lall, 2002). As a general guideline, 30-60 mg of iron per kg is supplemented to Atlantic salmon diets (Table 2.1).  Iron is required in cellular respiratory processes (Andersen et al., 1998) and can be found complexed to heme and non-heme compounds and heme enzymes (Lall, 2002). Although not common in cultured fish, iron deficiency can "suppress hematocrit, hemoglobin, and plasma iron levels" (Lall, 2002).  Iron is also involved in the formation of hydro- and radical peroxides (Figure 2.4) in the presence of oxygen (Andersen et al., 1998; Ozben, 1998). Desjardins et al. (1987), suggest that the concurrent effects of diet rancidity and iron-catalyzed dietary lipid peroxidation influenced the growth performance and physiology of rainbow trout. They also suggested that the dietary concentration of iron, in the presence of lipid peroxidation reactions, can elicit toxic effects at a lower level (86 mg/kg) than i f diet rancidity were controlled (1380 mg/kg). Iron toxicity can cause pathology of liver cells and adversely affect trout growth performance, feed efficiency and mortality rate (Desjardins et al., 1987; Lall, 2002).  2.2.5.1.2 Copper Copper is essential for the activities of several enzymes including cytochrome c oxidase, superoxide dismutase, tyrosinase, lysyl oxidase, ceruloplasmin, and dopamine P-hydroxylase (Lall, 2002). Copper is also found in ceruloplasmin, a copper-protein complex that is involved in  14  iron utilization and transportation (Lall, 2002). Atlantic salmon require dietary supplementation of 5 mg of copper per kg to sustain life processes (Table 2.1).  Excessive dietary copper can lead to negative health effects and toxicity at relatively low (34 mg/kg) concentrations (Berntssen et al., 1999). Copper is considered to be a strong pro-oxidant with the ability to produce hydroperoxides (Piriou et al., 1987) causing organ and neurological degeneration, cytotoxicity, and death (Venugopal and Luckey, 1978).  2.2.5.2 Anti-oxidants Anti-oxidants are compounds that inhibit or interrupt lipid peroxidation (Cheeseman and Slater, 1993; Dobarganes and Marquez-Ruiz, 2003; Hardy and Roley, 2000). There are several different types of anti-oxidants but they can generally be categorized as preventative, sacrificial and peroxide destroyers (Hardy and Roley, 2000). Preventative anti-oxidants sequester or chelate metals (i.e. copper and iron) that act as catalysts to initiate lipid peroxidation (Cheeseman and Slater, 1993; Hardy and Roley, 2000). However, preventative anti-oxidants cannot influence the propagation stage and consequently do not have an impact on the rate or extent of autooxidation (Hardy and Roley, 2000). Sacrificial anti-oxidants donate a hydrogen to radicals, converting them to unreactive forms, and thereby inhibit the continuation of the propagation stage (Cheeseman and Slater, 1993; Hardy and Roley, 2000). Peroxide destroyers, as the name implies, remove peroxides to limit the generation of new radicals and reduce the rate of autooxidation (Cheeseman and Slater, 1993; Hardy and Roley, 2000).  Temperature (Dobarganes and Marquez-Ruiz, 2003; Verleyen et al., 2001) and the degree of lipid unsaturation (Dobarganes and Marquez-Ruiz, 2003; Hardy and Roley, 2000) can influence 15  the level of natural and supplemented anti-oxidants present in a food. High temperatures cause heat labile anti-oxidants to be rapidly lost, independent of the degree of lipid unsaturation (Dobarganes and Marquez-Ruiz, 2003; Verleyen et al., 2001).  Conversely, the degree of lipid unsaturation can influence the rate of anti-oxidant consumption and this is generally independent of temperature. The detection of sensory rancidity in food can be negatively correlated with the concentration of residual anti-oxidants present (Martin-Polvillo et al., 1996), independent of temperature and degree of unsaturation (Dobarganes and MarquezRuiz, 2003). However, anti-oxidant presence does not entirely preclude the formation of lipid peroxidation products (Dobarganes and Marquez-Ruiz, 2003).  2.2.5.2.1 Vitamin E Vitamin E (tocopherols and tocotrienols) are the most important lipophilic quenchers of radicals (Packer and Fuchs, 1993). Their importance as radical quenchers is imparted by their ability to donate phenolic hydrogen atoms and consequently, stabilize and terminate the lipid peroxidation chain reaction (Sargent et al., 2002). Of the vitamers E, formally known as tocopherols, atocopherol is the most biologically active form (Halver, 2002). The biological activity of 1 mg of (synthetic) DL-a-tocopheryl acetate is defined as 1 international unit (nj), whereas 1 mg of naturally occurring D-a-tocopherol is equivalent to 1.49 IU (Makoto et al., 1993). The structure of tocopherols includes a 6-chromanol ring structure with varying amounts of methyl groups at the 5, 7, and 8 positions (Makoto et al., 1993). The fused chromanol ring is able to donate a hydrogen atom to radicals (Figure 2.5); reducing them to less reactive compounds (Mukai, 1993).  16  The tocopherols act as intra- and intercellular anti-oxidants (Halver, 2002). Generally, the vitamin E requirement for salmonids is 30 IU/kg dry diet but this requirement can vary depending upon the amount of polyunsaturated fatty acids (PUFA) in the lipid component of the diet (Halver, 2002). The reported physiological requirements of vitamin E were determined under controlled conditions (Halver, 2002) and alteration of the P U F A content or level of lipid peroxidation in a diet may influence diet formulation.  Deficiency symptoms of vitamin E in salmonids include anemia, ascites, exophthalmia, poor growth, poor feed conversion, epicarditis, ceroid deposits in spleen and liver, muscular dystrophy and increased mortalities (Cowey et al., 1983; Halver, 2002; Helland et al., 1991).  2.2.5.2.2 Vitamin C Vitamin C, or ascorbic acid, is a water-soluble nutrient that is necessary for the synthesis of collagen and normal cartilage, as well as normal bone and tooth development and maintenance, and wound repair (Halver, 2002). Vitamin C also plays an important anti-oxidant role in lipid peroxidation by scavenging radicals formed in the water phase and maintaining the activity of glutathione peroxidase and superoxide dismutase in synergy with vitamin E and selenium (Halver, 2002). Vitamin C is similar to vitamin E in that it is able to hinder peroxidation by donating a hydrogen to scavenge radicals (Hamre et al., 2004). Vitamin E, however, is regenerated by vitamin C, while vitamin C is regenerated by glutathione which in turn is regenerated by N A D P H (Hamre et al., 2004). If vitamins C and E are not regenerated, remaining in their ascorbyl and tocopheryl radical forms, and allowed to accumulate, they can also act as a pro-oxidants and catalyze peroxidation reactions (Bowry et al., 1992).  17  2.2.5.2.3 Manganese Manganese can function as a cofactor for some enzymes or as a vital part of metabolism (Lall, 2002). Manganese is a transition metal that can appear in a number of transition states and has the ability to transfer electrons to other compounds giving rise to radical formation (Hamre et al., 2004). However, it is also a functional part of redox centers for anti-oxidant enzymes (Hamre et al., 2004). It has an active role in lipid peroxidation by decreasing the activities of copper and superoxide dismutase (Lall, 2002).  2.2.5.2.4 Zinc Zinc is an integral constituent of many enzymes and regulates several metabolic processes (Lall, 2002). Moreover, it has a potential structural role in nucleoproteins (Lall, 2002). Although zinc is a transition metal, it does not form radicals (Bury et al., 2003). The ubiquity of zinc allows it to interact with a wide range of cellular entities (Bury et al., 2003). Furthermore, zinc is considered redox inert, allowing it to form relatively stable associations with potential antioxidant roles (Bury et al., 2003)  2.2.5.2.5 Ethoxyquin Ethoxyquin (l,2-dihydro-2,2,4-trimethylquinoline-6-yl ethyl ether) is commonly used as an additive in the preparation offish meal. This is due to its capacity to prevent peroxidation of highly unsaturated fatty acids (HUFA) at low levels and its ability to be efficiently distributed throughout a product (de Koning, 2002). Dispersal of ethoxyquin in a product, however, requires the use of a lipid carrier due to its hydrophobic nature (de Koning, 2002). Accordingly, animals fed diets supplemented with ethoxyquin tend to concentrate the bioavailable antioxidant, or its metabolites, in fat stores (Wilson et al., 1981). 18  During storage, ethoxyquin-supplemented fishmeal can also contain oxidized ethoxyquin compounds including a quinolone compound and a dimer (He and Ackman, 2000). The effects of these compounds are not fully understood, however, residue levels have been found to be significantly below regulatory levels (<0.5 ppm)(He and Ackman, 2000). Ethoxyquin (Dunkley et al., 1968; Skaare and Nafstad, 1979; Wilson et al., 1981) and its metabolites (ter Meulen et al., 1980) are rapidly and almost completely excreted in the urine and feces. For example, He (1998) found that Atlantic salmon fed a diet containing 150 mg of ethoxyquin per kg contained less than 0.02 mg/kg in their tissues, while de Koning (2002) found that broiler chickens fed a diet containing 125 mg of ethyoxyquin per kg contained less than 0.005 mg/kg in their tissues.  2.2.5.2.6 Selenium-dependent glutathione peroxidase Glutathione is an essential co-factor for the anti-oxidant enzyme selenium-dependent glutathione peroxidase (Kidd, 1997). Glutathione peroxidase (GPox) detoxifies peroxides, such as hydrogen peroxide (H2O2), by reacting them with glutathione and reducing them to water (H2O) (Figure 2.6). This reaction has an important role in protecting cells from damage by radicals which are formed by peroxide decomposition. Figure 2.6 shows the oxidation reduction pathway of glutathione and the important role that selenium-dependent GPox has in the prevention of radical formation.  Selenium-dependent glutathione peroxidase consists of a selenocysteine moiety in its active site (Wendel, 1980). The selenocysteine moiety donates an electron to the peroxide substrate, becoming oxidized in the process. Reduced glutathione is then used by the enzyme to regenerate the selenocysteine moiety (Wendel, 1980). 19  2.2.5.2.7 Catalase Catalase is present in the peroxisomes of nearly all aerobic cells (Eberhardt, 2001). It is an important heme-based enzyme that acts as a catalyst to reduce H^Ch, to H2O and/or alcohol (Eberhardt, 2001). Catalase molecules are tetramers of four polypeptide chains, with each chain composed of more than 500 amino acids (Voet et al., 2002). Located within this tetramer are four porphyrin heme groups which are responsible for catalase's enzymatic activity (Voet et al., 2002). Catalase is recognized for having the most efficient catalytic activity and is reputed to catalyze a reaction upon encounter with a substrate molecule (Eberhardt, 2001; Voet et al., 2002). The effectiveness of catalase in limiting lipid peroxidation is enhanced by the presence of superoxide dismutase and vice versa (Eberhardt, 2001).  2.2.5.2.8 Superoxide dismutase Superoxide dismutases (SOD) are represented by a group of metalloenzymes that contains either zinc, copper or manganese at the active site of the enzyme (Eberhardt, 2001). The primary function of SOD is to dismutate the superoxide radical to form H2O2 that can be susbsequently reduced by catalase or selenium-dependent glutathione peroxidase (Eberhardt, 2001). However, production of high amounts of H2O2 inhibits SOD, creating a negative feedback system (Eberhardt, 2001). Further, increased concentration of H2O2, in the presence of the superoxide radical, increases its likelihood of acting as an oxidizing agent through the Haber-Weiss reaction (Eberhardt, 2001).  The enzymes catalase, selenium-dependent glutathione peroxidase and SOD are tightly regulated by a feedback system (Eberhardt, 2001; Voet et al., 2002). Excessive superoxide inhibits glutathione peroxidase and catalase to modulate the conversion of H2O2 to H2O (Eberhardt, 20  2001). Increased H2O2 slowly inactivates SOD while catalases and glutathione peroxidase, by reducing H 0 , conserves SOD (Eberhardt, 2001). SOD, by reducing superoxide, conserves 2  2  catalase and glutathione peroxidase (Eberhardt, 2001). Through this feedback system, constant low levels of SOD, glutathione peroxidase, and catalase, as well as superoxide and H2O2 are conserved, keeping the entire system fully functioning (Eberhardt, 2001; Fridovich, 1993). However, any disruption of this system would lead to promotion of oxidation (Eberhardt, 2001).  2.2.5.3 Biological interactions of anti-oxidants The complexity of the in vivo anti-oxidant defence system elucidates the importance of proper dietary nutrient balance in reducing oxidative stress of an organism (Hamre et al., 2004). Endogenous anti-oxidants such as glutathione, catalase, superoxide dismutase, and dietary antioxidants, vitamins E and C, protect cells against oxidative damage (Lygren et al., 1999), whereas, essential nutrients such iron and copper can exhibit pro-oxidant activity at elevated concentrations. Although numerous interactions between individual pro- and anti-oxidants have been studied, establishing relationships beyond one or two nutrients is problematic due to the complexity of nutrient inter-relationships (Hamre et al., 2004).  2.2.6  Biological impacts and significance  Early studies on the toxicity of oxidized lipids focused mainly on thermo-oxidized oils and fats. Roffo (1939) fed rats fats and oils that had been heated to between 250-350 °C for an unspecified time. He claimed that stomach carcinomas developed, but these results were never confirmed (Billek, 2000). Most research conducted after Roffo (1939) did not find that the heated oils were toxic, but rather indigestible, with the polymer fraction causing severe diarrhea in mammals (Billek, 2000). Firestone et al. (1961), found that thermally abused lipids could not be absorbed 21  and postulated that these oils may interfere with the absorption of other nutrients leading to dehydration, nutrient deficiencies and death in small rodents.  Oxidation compounds can be formed during any processing unit operation from raw product selection to final preparation before consumption (Dobarganes and Marquez-Ruiz, 2003). At each step, the compounds produced can vary dramatically depending upon the history and combination of temperature, light and oxygen exposure in effect (Addis, 1986; Billek, 2000; Clark and Serbia, 1991; Dobarganes and Marquez-Ruiz, 2003). For example, hydroperoxides are almost absent above a temperature of 150°C, due to decreased oxygen solubility at higher temperatures (Dobarganes, 1998), whereas triacylglycerol monomers, dimers and polymers can be formed from the very early stages of heating (Billek, 2000; Velasco and Dobarganes, 2002). The presence of the triacylglycerol compounds in oxidized lipids, however, often yields undesirable sensory attributes and are not consumed (Dobarganes and Marquez-Ruiz, 2003).  In a well known, long-term rat study, Lang et al. (1978), used commercial frying techniques to heat soybean and groundnut oil to a temperature of 175 °C prior to feeding. Over the ten years of their multi-generation studies, they did not find any adverse effects in the rats that ingested the thermally abused oils. As a result, researchers have begun to examine the methods employed in earlier studies and they have postulated that the observed toxicity of the thermally oxidized oils may have been due to alteration of other variables (Dobarganes and Marquez-Ruiz, 2003). For example, most studies that have reported mortalities used diets that were nutritionally incomplete. For instance, some studies have employed diets that lacked the appropriate levels of essential fatty acids and vitamin E (Meyer, 1978). Moreover, fats and oils have been used as the sole nutrient (Boyd, 1973), or oxidized oils have been placed directly into the stomach using a tube, thus, bypassing sensory discrimination (Shue et al., 1968). 22  2.2.7  Fate of oxidized dietary lipids  Recent studies focusing on the health implications of oxidized dietary lipids have tried to determine their biological fate by examining the biological activity and metabolic pathways of lipid peroxidation and how oxidized lipids could promote cardiovascular disease and cancer (Dobarganes and Marquez-Ruiz, 2003). Strong correlations have been made between consumption of oxidized lipids and disease promotion, but the bioavailability of lipid peroxidation products is poorly understood (Billek, 2000; Dobarganes and Marquez-Ruiz, 2003; Haumann, 1993; Riedemann and Ward, 2002). For example, oxidized dietary lipids have been linked to being a source of plasma hydroperoxides (Ursini et al., 1998; Williams et al., 1999), but the distinction between whether these compounds were produced in the diet and or produced in vivo, still remains to be determined (Dobarganes and Marquez-Ruiz, 2003).  Kanazawa and Ashida (1998a; b) suggested that hydroperoxides are hydrolysed by pancreatic lipase and then converted to the hydroxy and/or aldehydic compounds before they are absorbed. Kanner and Lapidot (2001) suggested that human gastric fluid may enhance lipid peroxidation in the presence of catalysts commonly found in food. Although both of these studies are suggestive that lipid peroxidation products can be absorbed, the fate and the extent of influence of these absorbed compounds needs further clarification.  2.2.8  Health implications of oxidized dietary lipids  2.2.8.1 Atherosclerosis Oxidized dietary lipids have been shown to promote the onset and development of atherosclerosis in animals (Staprans et al., 2003; Staprans et al., 1996; Staprans et al., 2000), but the mechanisms in vivo are complex and require greater clarity (Brown and Jessup, 1999; 23  Chisolm, Steinberg, 2000; Dobarganes and Marquez-Ruiz, 2003). Some experimental evidence shows that specific unsaturated fatty acids can influence the development of a pro-inflammatory environment (Toborek et al., 2002), but these data are in conflict with the accepted benefit of substituting polyunsaturated fat for saturated fat, despite the susceptibility of the former to peroxidation (Schaefer and Brousseau, 1998).  Studies have also shown a link between consumption of oxidized lipids and the development of atherosclerosis in diabetic patients (Staprans et al., 1999; Turek et al., 2003). The development of accelerated atherosclerosis in diabetic patients may be due to nonenzymatic glycation of serum lipoproteins, which would increase their susceptibility to peroxidation (Staprans et al., 1999).  2.2.8.2 Cancer Kanazawa et al. (2002), found effects of peroxyl radicals on D N A were depurination, depyrimidation, and base modification. Subsequently, the rate of D N A mutation was anticipated to contribute to carcinogenesis. Accordingly, the highly reactive malondialdehyde, 2-alkenal and 4-hydroxy-2-alkenal, have also shown significant cytotoxic and mutagenic effects in vitro (Esterbauer et al., 1991).  Recent research has examined the issue of potential carcinogenic effects of oxidized lipids through the examination of specific oncogenes and tumour suppressor gene expression. Perjesi et al. (2002) suggested that oxidized dietary lipids could be involved in tumour promotion and initiation. However, greater clarity in understanding the genes responsible for cancer  24  development as well as the mechanisms and pathways for regulating cell growth and development is required (Milner, 2002).  2.2.8.3 Liver damage Hepatic metabolism of lipid peroxidation products has been linked to increased levels of oxidative stress within the liver (Liu et al., 2000). However, many other variables could also be influencing oxidative stress in the liver including vitamin E or C deficiency, high cholesterol intake, smoking, over-exercising, etc. (Chisolm and Steinberg, 2000; Dobarganes and MarquezRuiz, 2003; Staprans et al., 2003).  2.3  Fish nutrition  Dietary protein and lipid quality are comprimised during pellet manufacturing of the feed since it requires heat and pressure, especially during extrusion processing (Hardy and Barrows, 2002). In addition, increased use of the finite global supplies of fishmeal and oil to meet the increasing demands for these commodities for aqua-feeds necessitate identification and/or development of suitable alternative protein and lipid sources of plant and/or animal origin (Rust, 2002). In either case, fish nutrition, feed manufacture, or fish health may be compromised unless substitutes and their processing methodology are based on good scientific knowledge (Rust, 2002).  Fish growth and health can be compromised by dietary factors such as energy content, lack of essential nutrients, and improper balances of the major energy-yielding nutrients (Higgs et al., 1995). Fortunately, nutrient and energy requirements are well established for some key salmonid, warmwater, and marine species which facilitates the design of diets for new species until more specific information is known (NRC, 1993). Careful analysis of the natural diets of  new fish species may also aid in diet design (Higgs et al., 1995). Proper development of diets for any species, considering any possible physical, chemical or biological losses that may influence a fish's nutritional status, should meet the health and production goals of the cultured fish (Hardy and Barrows, 2002). However, comprehensive knowledge of the species requirements for digestible nutrients and energy and of the acceptable dietary levels of an array of plant and animal protein and lipid sources would be required to ensure the cost-effectiveness of diets.  2.3.1 Energy A l l biological organisms need a continuous supply of energy to sustain themselves (Bureau et al., 2002). Fish and other poikilotherms require less energy than homeotherms because they do not need to maintain a constant internal temperature (Bureau et al., 2002). The metabolic rate offish is dependent on environmental factors, such as water temperature (Bureau et al., 2002) as well as physiological changes that occur at different stages in their life history (Higgs et a l , 1995). Accordingly, adequate dietary levels of digestible energy need to be included by considering all factors that may influence metabolism.  The energy and nutrient requirements for Atlantic salmon have not been thoroughly investigated (Table 2.1). Diets have been historically formulated based on the biology, life history and relatedness of Atlantic salmon to rainbow trout and Pacific salmon (Hardy and Barrows, 2002). Atlantic salmon are carnivorous and have a low requirement for carbohydrates (Watanabe, 1982). Therefore, dietary protein and lipids are commonly used as a source of biochemical energy for Atlantic salmon (Dabrowski and Guderley, 2002). The ratio of digestible protein to lipid is shifted downward as the fish increase in size (age) to ensure that most of the dietary  26  protein is used for new tissue formation rather than as a major source of energy (Dabrowski and Guderley, 2002).  2.3.2 Protein Protein is composed of 25 amino acids, and these can be a major dietary energy source for salmonids (Cho and Kaushik, 1985; Luquet and Watanabe, 1986). Amino acids are also used for tissue repair, replacement and generation especially when there are suboptimal ratios of digestible protein to non-protein energy (Gill, 2002; Sargent et al., 2002). Although salmonids do not have a fixed dietary protein requirement, they do require a well balanced mixture of indispensable (essential) and dispensable amino acids (Table 2.1) (Sargent et al., 2002) and an optimal ratio of digestible protein to other energy-yielding nutrients, such as lipids, at specific life history stages for cost effective fish production (Table 2.1) (Higgs et al., 1995).  It is imperative to ensure that optimal ratios between digestible protein and energy are used in the diet formulation. It is also important to ensure the use of high quality and digestible protein sources to enhance dietary protein utilization and consequently minimize nitrogen excretion (Anderson et al., 1997). Protein efficiency ratio (PER) is often used in studies that are examining the function of protein consumption on growth as a measure of dietary protein utilization, although the nature of the gain is not defined since it is calculated as weight gain divided by protein intake (Anderson et al., 1997). Atlantic salmon require approximately 3547% digestible protein (dry weight basis) in their diet for optimal growth (NRC, 1993) and generally have PER values that range from 2.2-3.1 (Anderson et al., 1997; Chaiyapechara, et al., 2003; Opstvedt et al., 2003; Vangend and Hemre, 2003).  27  2.3.3 Lipids Dietary lipids are mostly composed of fatty acids, that are hydrocarbon molecules comprised of 12-24 carbon units in tandem, and are commonly a part of triacylglycerides, wax esters, sterols, sphingolipids and phospholipids (Sargent et al., 2002). The nutritional value of a fatty acid is determined in part by the number of carbon units, quantity and position of unsaturation sites (carbon-carbon double bonds), and the number of carbon atoms between the terminal methyl group and site of unsaturation (Forster, 1987; Sargent et al., 2002). For example, eicosapentaenoic acid or C22:5n-3 is a fatty acid that contains a carbon chain of 22 units with 5 double bonds with the closest double bond to the terminal methyl group being a distance of 3 carbons away.  Fatty acids serve several metabolic functions in all vertebrates. First, they are a good source of energy (Sargent et al., 2002). Second, they provide essential constituents to cell membranes (Sargent et al., 2002). Finally, they facilitate the uptake of fat-soluble vitamins and carotenoid pigments and provide the components for synthesis of new cellular lipid which is required for growth, reproduction, and health (Forster, 1987; Sargent et al., 2002). Every vertebrate species has a specific dietary requirement for certain essential fatty acids, which is not restricted to but often includes E P A and D H A (Sargent et al., 2002). Some vertebrate species are able to convert the CI8 polyunsaturated fatty acids (PUFA) to the higher C20 and C22 highly unsaturated fatty acids (HUFA) through desaturation and elongation reactions. Most carnivores have lost considerable capacity to utilize this pathway possibly due to the high amount of H U F A present in their natural diets (Rivers et al., 1975; Sargent et al., 2002). However, research suggests that Atlantic salmon are still capable of C20 and C22 H U F A synthesis from their n-3, n-6, and n-9 precursors (Sargent et al., 2002). Nevertheless, these reaction pathways are energetically demanding and are only activated when there is an E F A deficiency (Sargent et al., 2002). 28  Consequently, H U F A (viz. E P A and DHA) should be supplied in the diet of Atlantic salmon. Most salmonids require 1-2% of their diet as n-3 fatty acids (Table 2.1) (Sargent et al., 2002; Takeuchi et al., 1979) and generally n-3 H U F A should represent 10% of the dietary lipid level (Higgs and Dong, 2000). Requirements for crude lipid content in the diet vary depending on the type of lipid and other energy sources used (Table 2.1) (Sargent et al., 2002). Signs of essential fatty acid deficiency of various fishes include fin rot, shock, myocarditis, scoliosis, decreased growth rate, decreased feed efficiency, decreased reproductive performance and increased mortality (NRC, 1993; Tacon, 1996).  2.3.4  Vitamin and mineral requirements  Essential minerals and vitamins are required for the maintenance and formation of skeletal structure, preservation and functioning of enzymes, hormones and osmoregulation, control of acid-base equilibriums and bio-metabolic pathway function (Lall, 2002). Fish have the ability to absorb these essential nutrients from their diet and, unlike their terrestrial counterparts, from the surrounding environment (Lall, 2002). Table 2.1 describes the mineral and vitamin requirements for Atlantic salmon. A l l of the listed minerals and vitamins must be provided in the postjuvenile Atlantic salmon diet except for calcium which is readily absorbed from seawater (Lall, 2002).  29  30  Initiation  RH + ? R'-CH=CH-R'  ^R* ^ROOH  induction period  Propagation  R* + 0 ROO* + R H ROOH  -*ROO* - ^ R O O H + R* -*RO* + O H *  fast reaction rate-determining step  Termination  R* + R* ROO* + ROO*  ->Non-radical products -> R1-CO-R2 + R , - C H O H - R +0  2  2  2  Figure 2.2. Auto-oxidation of an olefinic compound. R H represents an olefinic compound where the H is attached to an allylic carbon atom; R* represents the substrate radical; R O O H represents a hydroperoxide; ROO* represents a peroxyl radical; R O * represents an alkoxy radical; O H * represents a hydroxyl radical. Initiation of auto-oxidation can be catalyzed by light, heat or pro-oxidants in the presence of oxygen. Figure modified from Gunstone (1996) and Kanner and Rosenthal (1992).  31  Chemical Changes  Test Method (example)  Olefinic acids/esters  Accelerated tests measured at accelerated temperatures  0  induction period (long)  2  Allylic hydroperoxides (highly reactive species)  Peroxide Value  Short chain compounds (aldehydes) -> volatile, lower molecular wt. -> same chain length, rearranged ->higher molecular wt., polymer  Anisidine Value Thiobarbituric acid reactive substances  •  Tertiary - short chain acids  Rancimat TM  Figure 2.3. Chemical changes during auto-oxidation of olefinic compounds and some exampli of methods commonly used to characterize the changes.  Acceleration of hydroperoxide decomposition to form peroxy radical and alkoxy radical Fe + ROOH -» F e + ROO* + H Fe + ROOH Fe + R O * + OH" 3+  2+  2+  3+  Formations of alkyl radical by direct reaction Fe + RH ^ Fe + R * + H 3+  2+  +  +  Formation of hydroxyl radical (Fenton reaction) Fe + H 0 -» F e + O H " + O H * 2+  3+  2  2  Figure 2.4. The important mechanisms of transition metal ion (i.e. iron) acceleration of lipid peroxidation. Fe represents iron; R O O H represents a hydroperoxide; R O O * represents a peroxyl radical; R O * represents an alkoxy radical; R * represents the alkyl radical; O H * represents a hydroxyl radical; O H " represents a hydroxide ion.  33  R* + RO* + ROO* +  AH AH AH  -> ->  R* + RO* + ROO* +  A* A* A*  -> ->  RH + A* ROH + A* ROOH+ A * RA ROA ROOA  Figure 2.5. The basic reaction between vitamin E and radicals. A H represents vitamin E; A * represents oxidized vitamin E; R O O H represents a hydroperoxide, R*, RO*, and ROO* represent radicals; R H , R O H , R A , R O A , R O O A represent stabilized lipid substrates.  34  Oxidized Glutathione (GSSG) 2  NADPH+H+  H2O Glutahione peroxidase  Glutathione reductase Riboflavin  Selenium  H2O2  V  (FAD)  NADP*  (2 GSH) ~* Reduced Glutathione  Figure 2.6. The oxidation reduction pathways of glutathione. GSSG represents the oxidized form of glutathione (glutathione disulfide); G S H represents the reduced form of glutathione. One molecule of hydrogen peroxide is reduced to 2 molecules of water while 2 molecules of glutathione (GSH) are oxidized in a reaction catalyzed by the selenoenzyme, glutathione peroxidase. Oxidized glutathione (GSSG) may be reduced by the flavin adenine dinucleotide (FAD)-dependent enzyme, glutathione reductase. Adapted from Meister (1976).  35  Table 2.1. Nutrient requirements for Atlantic salmon as percentages of diet, milligrams per kilogram of diet, or international units (IU) per kilogram of diet (as-fed basis). Adapted from N R C (1993). R = nutrient is required but the level is not determined. N T = not tested. Atlantic salmon Energy Base (kcal De/kg diet)  Source  -3,600  N R C , 1993  55  Wison, 2002  Arginine  4.6  Wilson, 2002  Histidine  1.8  Wilson, 2002  Isoleucine  3.2  Wilson, 2002  Leucine  5.2  Wilson, 2002  Lysine  6.1  Wilson, 2002  Methionine + cystine  3.1  Wilson, 2002  Phenylalanine + tyrosine  5.8  Wilson, 2002  Threonine  3.2  Wilson, 2002  R  N R C , 1993; Wilson 2002  Protein, crude (digestible), percent Amino acids (percent of protein)  Tryptophan Valine Lipid, crude (digestible), percent  3.9  Wilson, 2002  10-3 5  Sargent, Tocher and Bell, 2002  Fatty acids n-3 fatty acids (percent of diet)  1 -2  n-3 H U F A (percent of lipid)  >10  N R C 1993; Sargent, Tocher and Bell, 2002 Sargent, Tocher and Bell, 2002  Calcium  NT  N R C 1993, Lall 2002  Chlorine  NT  N R C 1993  Magnesium  0.04  N R C 1993, Lall 2002  Phosphorus  0.6  N R C 1993, Lall 2002  R  N R C 1993, Lall 2002  NT  N R C 1993 N R C 1993, Lall 2002  Macrominerals (percent of diet)  Potassium Sodium Microminerals (mg/kg) Copper  5  Iodine  0.6-1.1  N R C 1993  Iron  30-60  N R C 1993, Lall 2002  Manganese  10  N R C 1993, Lall 2002  37-67  N R C 1993, Lall 2002  R>0.15  N R C 1993  A , IU/kg  2,500  N R C 1993, Halver 2002  D, IU/kg  2,400  N R C 1993, Halver 2002  E , IU/kg  30  N R C 1993, Halver 2002  K, mg/kg  10  N R C 1993, Halver 2002  Zinc Selenium Fat-soluble vitamins  Water-soluble vitamins (mg/kg) Riboflavin  7  N R C 1993  Pantothenic acid  20  N R C 1993  Niacin  30  N R C 1993  Vitamin B i  9  N R C 1993  Choline  3,000  Halver 2002  Biotin  2  1-1.5  N R C 1993, Halver 2002  Folate  6-10  Halver, 2002  Thiamin  10-15  Halver 2002  100-150  Halver 2002  Vitamin C  36  3.0  IMPACT OF IRON-CATALYZED DIETARY LIPID PEROXIDATION ON  GROWTH PERFORMANCE, GENERAL H E A L T H AND FLESH PROXIMATE AND FATTY ACID COMPOSITION OF ATLANTIC SALMON (SALMO SALAR) IN SEAWATER.  Abstract Post-juvenile Atlantic salmon (Salmo salar) were fed to satiation, twice daily for 126 days, with diets supplemented with copper (10 or 35 mg/kg), iron (333 or 1000 mg/kg) and unoxidized or oxidized dietary lipid measured as peroxide value (PV, <5 or <5+ or 35+ meq/kg). The dietary effects measured include growth performance, general health, flesh proximate and fatty acid composition, and in situ oxidative stress. Lipid peroxidation in the fishmeal and fish oil was controlled by adding ethoxyquin (150 mg/kg) to diets containing unoxidized dietary lipid (<5 meq/kg). A l l diets contained the required level of vitamin E (30 IU/kg). Diets supplemented with the highest level of iron, without anti-oxidant protection (no ethoxyquin), showed the greatest loss of eicosapentaenoic (EPA) and docosahexaenoic acids (DHA) and vitamin E (p<0.05). Fish fed these diets displayed significantly (p<0.05) decreased growth, feed intake, feed efficiency, and utilization of gross energy and protein that were attributable to the losses of the preceding essential nutrients and possibly dietary protein modification. Dietary vitamin E concentrations decreased in all diets, except those with low levels of iron and peroxide values, during the 126-day study period. Diets without anti-oxidant supplementation had significantly (p<0.05) higher lipid peroxidation levels as compared with anti-oxidant-containing diets. However, fish hepatic and muscle tissue lipid peroxidation were unaffected. Autopsy-based assessment of general health indicate that fish fed diets influenced by iron-catalyzed lipid 37  peroxdiation exhibited clinical signs of poor health indicative of dietary vitamin E and omega-3 highly unsaturated fatty acid deficiency. These results suggest that fish diets undergoing active lipid peroxidation with attendant deficiencies in omega-3 highly unsaturated fatty acids and vitamin E adversely influence the growth performance, general health, and fillet proximate and fatty acid composition of post-juvenile Atlantic salmon. The consumption of oxidized dietary lipids by the experimental fish did not influence in situ oxidative stress parameters as measured in this study.  3.1  Introduction  The most common nutritional diseases in farmed fish can be attributed to diets deficient in essential fatty acids (EFA) or improper formulation of dietary lipids (Tacon, 1996). Carnivorous fish must obtain high amounts of n-3 highly unsaturated fatty acids (n-3 H U F A ) from their diet due to inherent deficiencies in fatty acid biosynthesis (Hamre et al., 2001; Mourente et al., 2002; Sargent et al., 1999). The n-3 H U F A viz., eicosapentaenoic acid (EPA) and docosahexanoic acid (DHA), are essential for the optimal growth and development of marine fish (Sargent et al., 1999; Sargent et al., 2002). However, feed rich in n-3 H U F A , and other polyunsaturated fatty acids (PUFA), are susceptible to lipid peroxidation (Mourente et al., 2002; Tacon, 1996).  Lipid peroxidation, involving the reaction of molecular oxygen with fatty acids at double bond sites, has attracted scientific interest due its impact on food quality and role in biological processes (Kanner and Rosenthal, 1992; Nakamura et al., 1998; Porter et al., 1995; Yang, 1992). Lipid peroxidation can influence functional and structural alteration of membranes, hormones, vitamins (Kanner and Rosenthal, 1992; Nakamura et a l , 1998), D N A and protein modification (Gershon, 1988; Halliwell and Chirico, 1993; Marnett, 1987), tumour initiation (Marnett, 1987),  arteriosclerosis (Hessler et al., 1983), and aging (Cheeseman and Slater, 1993). Alternatively, lipid peroxidation has exhibited a positive biological role by decreasing tissue damage during sepsis (Riedemann and Ward, 2002); suggesting that its biological role is complex. Lipid peroxidation has also been implicated in flavour and aroma deterioration, decreasing food wholesomeness, and food safety (Kanner and Rosenthal, 1992). Accordingly, the influence of dietary lipid peroxidation products has generated scientific interest due to the possible link that exists between nutrition and biological oxidative stress (Addis, 1986; Kubow, 1992).  Recent studies focusing on the health implications of oxidized dietary lipids have attempted to understand their biological fate by examining the biological activity and metabolic pathways of lipid peroxidation and their relationship with cardiovascular disease and cancer (Dobarganes and Marquez-Ruiz, 2003). Strong positive correlations have been made between consumption of oxidized lipids and disease promotion, but the bioavailability of lipid peroxidation products is still poorly understood (Billek, 2000; Dobarganes and Marquez-Ruiz, 2003; Hamre, et al., 2001; Haumann, 1993; Riedemann and Ward, 2002). For example, it has been suggested that oxidized dietary lipids are a source of plasma hydroperoxides; compounds that can increase the susceptibility of atherogenic lipoproteins to peroxidation and lead to the formation of atherosclerotic lesions (Ursini et al., 1998; Williams et al., 1999). However, the distinction between whether these compounds were produced in the diet or produced in vivo, still remains to be determined (Dobarganes and Marquez-Ruiz, 2003; Hamre et al., 2001).  The objectives of this study were to examine the impacts of varying the dietary pro-oxidants, iron and copper, and the anti-oxidant ethoxyquin, on post-juvenile Atlantic salmon (Salmo salar). Experimental diets were designed to investigate the effect of pro-oxidant-catalyzed dietary lipid peroxidation on fish growth performance, proximate composition, fatty acid methyl 39  ester (FAME) composition, vitamin E content, selenium-dependent glutathione peroxidase activity, in situ oxidative stress, and general fish health. Diets were also analyzed for changes in vitamin E content, proximate composition and F A M E over the course of the study.  3.2  Materials and methods  3.2.1  Experim ental diets  Standard commercial crude herring oil (Clupea harengus) and mixed rockfish meal were supplied by JS McMillan Fisheries Ltd. from Prince Rupert, British Columbia to the Department of Fisheries and Oceans, Canada, West Vancouver Laboratory (WVL). Anti-oxidants were not added to the oil or the fishmeal during processing. Nitrogen flushing was applied to displace any oxygen from the oil upon arrival at the W V L . At that time 200 ppm of ethoxyquin (91% purity, Monsanto), a standard commercial anti-oxidant,' were added to one quarter of the fishmeal for later use in the preparation of control diets. Subsequently, all fishmeal was vacuum-packed into 3 mil (0.003 inch thickness), low oxygen/nitrogen (O2 transmission; 5.2 cc/sq.m/24 hours @ 23°C, H 0 transmission: 8 g/sq.m/24 hours @ 37.8°C @ 90% relative humidity) 2.5 kg plastic bags 2  (Westcoast Foodpak Systems Ltd., Vancouver, BC). Packaged fish meal was stored at -20 °C to minimize lipid peroxidation activity. Peroxide values (PV) of the fish oil and meal were determined at the time of packaging and recorded as <0.4 meq/kg using the A O A C (2000, 965.33) protocol.  The fish oil was oxidized by placing 60 L of the oil into three 23 L glass carboys and incubating at 40°C while passing oxygen (Medical Grade 99% purity, Praxair, Delta) through the oil via ceramic diffusers at a rate of 2.5 L/min for 1 month (Koshio et al., 1994) until a P V of 35 meq/kg was attained, as confirmed by the A O A C (2000, 965.33) protocol. Subsequently, the oxidized 40  oil was transferred into 4 L, darkened glass bottles, flushed with nitrogen, sealed and stored for 1 week at 15°C prior to incorporation into their respective diets.  Twelve experimental diets were designed by modifying a practical semi-purified basal diet formula and all of the test diets were steam pelleted to produce a single pellet size (3 mm). The experimental diets were supplemented with copper (10 or 35 mg/kg) and iron (333 or 1000 mg/kg) and either unoxidized or oxidized dietary lipid (peroxide values; <5 or <5+ or 35+ meq/kg)(Table 3.1). The experimental variables have been denoted throughout the text with the following symbols; I and i represent high and low iron; C and c represent high and low copper; E and e represent ethoxyquin presence and absence; and + indicates that the dietary oil had undergone forced oxidation prior to diet incorporation. For example, diet 9, which contained low levels of copper and iron, no ethoxyquin and previously oxidized oil, was denoted as "ice+".  The basal diet contained groundfish meal, wheat gluten meal, casein, and gelatin as the sources of protein. Equal levels of vitamin and mineral supplements were included in all of the test diets to ensure that the vitamin and mineral requirements of the post-juvenile Atlantic salmon were met (NRC, 1993). Supplemental levels were determined by taking the difference between the required and target (iron and copper) nutrient levels and the concentrations present in the analyzed dietary constituents. Target nutrient levels for iron and copper were arrived at by ensuring that the minimum values for each nutrient were above their respective requirements (NRC, 1993) whereas, the maximum values were selected near their respective toxic threshold (Berntssen et al., 1999; Desjardins et al., 1987). The diets were steam-pelleted (Higgs et al., 1979) under reduced moisture content at about 80-85 °C for 3-5 seconds with a 3mm die (California model C L 2 laboratory pellet mill) 5 days prior to the beginning of the experimental  41  trial. Part of the dietary lipid (77%) was reserved from the diet mixture before pelleting, and this was subsequently sprayed onto the pellets.  3.2.2  Diet storage trial  The pelleted diets were stored at 15 °C under low humidity in plastic-lined paper bags (Westcoast Foodpak Systems Ltd.) and continually mixed throughout the entire trial. Diets were sampled at 21-day intervals and these samples were transferred into 3 mil, high barrier, low oxygen/nitrogen (O2 transmission; 2.3 cc/sq.m/24 hours @ 23°C, H2O transmission; 7.8 g/sq.m/24 hours @ 37.8°C @ 90% relative humidity) 2.5 kg aluminum-lined bags. Thereafter, these bags were vacuum-sealed and stored at -40°C until diet analysis.  3.2.3  Experimentalfishand dietary trial conditions  In April, 2002, 4500 Atlantic salmon (Salmo salar) fry, vaccinated against Vibrio anguillarum and Aeromonas salmonicida, were obtained from N O R A M (Little Bear Creek, Vancouver Island, BC) and transferred to W V L (49° 15'N, 123° 10'W), where the growth trial was conducted. The fish were transferred to saltwater approximately one month later. The fish were fed a commercial diet (EWOS Canada Ltd.) prior to the growth trial. At the start of the study, the age of the fish was < 1 year and they had a mean weight of 64.7 ± 9.4 g. Between July 30 and August 1, 2002, the fish were randomly transferred to 36 1100 L tanks that were each provided with flow-through filtered (11 L/min), oxygenated (8 ppm), ambient seawater (30 ppt) that ranged in temperature between 11.6 and 13°C. The 12 diet treatments were assigned to triplicate tanks, that each contained 44 fish, using a randomized complete block design. The lighting followed natural daylight hours. Supplemental aeration was provided to each tank by compressed air passed through a diffuser. The fish were fed to satiation twice daily (once in 42  morning, once in afternoon) for 126 days, and their daily feed consumption was recorded as feed intake [feed dispensed (g) - (pellet number • mean pellet weight (g))]. Mortality and general health were recorded throughout the trial.  3.2.4  Sample collection and biometric analyses  Live body weights and lengths of individual fish from each experimental group were determined at 42-day intervals. The fish were starved 24 hours prior to sampling to evacuate gut contents so that this would not be a confounding factor in weight estimates. A l l fish from each tank were anesthetized using a dual anesthetic procedure that involved (1) the addition of clove oil (Hill Tech Canada. Inc.) into the experimental tanks (0.5 ppm), (2) removal of the fish from the tanks and (3) their placement into an aerated 30 L basin containing (80 ppm) tricainemethanesulfonate (MS-222, Syndel Laboratories Ltd., Vancouver).  During the initial sampling, 10 fish were randomly sampled for determination of whole-body proximate composition and 10 additional fish were randomly sampled and filleted for determination of muscle proximate composition. Every 42 days, two fish from each tank were randomly sampled and filleted for muscle proximate composition determinations. During the final sampling (day-126), fish were randomly sampled from each replicate group for whole-body proximate composition (n=5), fish muscle proximate composition (n=5), musclemalondialdehyde (n=3), muscle-vitamin E (n=5), muscle-selenium-dependent glutathione peroxidase (Se-GPox)(n=3), muscle-conjugated diene (n=3), muscle-histopathology (n=3), liverhistopathology (n=3), liver-malondialdehyde (n=3), liver-Se-Gpox (n=3) and liver-vitamin E (n=3). Fish that were sampled for proximate composition were immediately stored on ice postslaughter and immediately transferred into 3 mil, high barrier, low oxygen/nitrogen transmission 43  (O2 transmission; 2.3 cc/sq.m/24 hours @ 23°C, H2O transmission: 7.8 g/sq.m/24 hours @ 37.8°C @ 90% relative humidity) aluminum-lined bags. Thereafter, the bags were vacuumsealed and stored at -40°C until sample analysis.  3.2.5  Proximate composition of experimental fish and diets  Proximate analyses of diets, whole fish and fish fillets were measured according to the procedures of Higgs et al. (1979). In this regard, the protein concentrations of the diets were measured using a Technicon Autoanalyzer II, S.C colorimeter (Pulse Instrumentation Ltd., Saskatoon). A factor of 6.25 was used to convert percent nitrogen to percentage protein for each sample. Total lipid was extracted with chloroform/methanol (2:1 v/v) post-homogenization according to the Bligh and Dyer (1959) procedure.  3.2.6  Fatty acid methyl esters  The chloroform layer that contained lipid from the above described lipid extraction (section 3.2.5) was collected and stored in sealed glass vials (Varian) at -40 °C prior to methyl esterification. Working at room temperature, a calculated volume of the chloroform layer containing 0.3 ml of lipid was pipetted into a 20 ml (15 cm height, 1.5 cm inner diameter) glass test tube and placed into a 30 °C water bath. The sample was concentrated to < 1 mL by gently blowing medical grade nitrogen gas (Praxair, 99%) over the surface of the chloroform solution. Fatty acid methylation required addition of 1 mL of benzene (Anachemia) and 0.2 mL of sodium methoxide (Aldrich Chemicals, 0.5 % sodium metal in methanol) to the solution and heating of the test tube in a 50 °C water bath for 10 minutes. Post-methylation, the non-polar fatty acid methyl esters (FAME) were separated from the other lipid constituents by the addition of 5 mL of hexane (Anachemia, > 99.9 %), 5.0 mL of distilled water, and 0.10 mL of acetic acid 44  (Anachemia, > 95.0 %). After separation, the hexane layer was pipetted into a new test tube along with 5 ml of hexane. A small amount (tip of spatula) of anhydrous sodium sulfate (Sigma) was added to remove any residual water. This solution was then pipetted into a new test tube, placed in a 30°C water bath, and concentrated to < 2ml with nitrogen gas. Once concentrated, the sample was transferred to a GC vial (Varian) and stored at -18°C prior to GC analysis.  3.2.7  Fatty acid analyses  The relative fatty acid compositions of the fish muscle and feed samples were determined by methylating the fatty acids in the samples as described above in section 3.2.6, and then the F A M E were separated and quantified with a Varian model 3400 gas chromatograph equipped with a Varian CP-SIL capillary column (50 m X 0.25mm id, Varian, Missisauga, Ontario) and flame ionization detector. The column temperature, initially set at 160°C was raised, 4°C/minute until it reached 220°C where it was held for 18 minutes. The injector and detector temperatures were both set at 250°C. Helium was used as the carrier gas. Individual F A M E were identified by comparison with known external standards ( F A M E mixes 1, 2, and 37; Supelco, Inc.) and quantified by means of a direct linked computer with Varian Star Chromatograph Workstation (version 5.51) software. F A M E concentrations were reported on the basis of area percent and each fatty acid was subsequently reported as a percentage of the total identifiable fatty acids (>92%). Thereafter, the relative percentages of fatty acids were used to evaluate the lipid in fishmeal and the oxidized and unoxidized oil before diet production, the freshly made diets, and the diets and the muscle fillets at 42-day intervals during the 126-day trial.  45  3.2.8  Measurement of malondialdehyde in dietary treatments  The level of lipid peroxidation was measured colorimetrically to indicate the presence of malondialdehyde (MDA), using a modification of the procedure of Tarladgis et al. (1960). Five grams of tissue were blended in a Waring blender with 25 ml of deionized and distilled (dd) H 0 2  for 2 minutes and the mixture was quantitatively transferred into a 500 ml roundbottom flask by rinsing with 24 ml of ddH 0. The pH of the solution was lowered to 1.5 by adding 1.25 ml of 2  H C L solution consisting of H C L / d d H 0 (1:2 v/v). A few anti-bumping granules were added 2  before heating. The roundbottom flask was placed into a simple distillation system containing an electric heating element and the solution was boiled for about 10 minutes. Five ml of the distillate were transferred into a 20 ml pyrex screw capped test tube along with 5 ml of thiobarbituric acid (TBA) reagent that contained 0.02 M 2-TBA (Sigma) in 90% glacial acetic acid. The tubes were then placed in boiling water for 35 minutes. The absorbance of the resulting solution was compared to a M D A standard curve at 532 nm with a spectrophotometer (Spectronic 1001+, Milton Roy). The standard curve was constructed with 1,1-3,3tetraethoxypropane (TEP)(Sigma).  The M D A content was also measured using a modification of the procedure described by Lemon (1975). In this regard, five grams of tissue were blended in a homogenizer (Omni Mixer) with 10 ml of an extraction solution that contained 7.5% (w/v) trichloroacetic acid, 0.1% (w/v) propyl gallate (Sigma) and 0.1 % (w/v) ethylenediaminetetraacetic acid (Sigma) in ddH 0. The 2  resulting homogenate was gravity-filtered using Whatman #1 filter paper for 15 minutes. Subsequently, 5ml of the clear filtrate were mixed with 5 ml of 0.02 M 2-TBA in a 20 ml pyrex screw capped test tube and placed in boiling water for 40 minutes. The absorbance of the sample was determined at 532 nm with a spectrophotometer (Spectronic 1001+, Milton Roy) and  46  compared with that of a M D A standard curve. The standard curve was constructed with 1,1-3,3tetraethoxypropane (TEP)(Sigma).  3.2.9  Determination of vitamin E content  Total vitamin E (alpha-tocopherol) concentrations in the fishmeal, oil, feed and liver and muscle tissues were measured as IU/kg by high performance liquid chromatography (HPLC) following the protocol outlined by Tangney et al. (1981). Vitamin E concentrations in the fresh oil and fishmeal were determined to ensure that the vitamin E requirement for Atlantic salmon of 30 IU/kg diet was met in all cases. Vitamin E concentrations in muscle tissue were analyzed from 6 fish at the beginning (day 0) of the trial and three composite samples of 5 fish/tank at the end of the 126-day trial. The diets were sampled at the beginning (day-0), middle (day 63) and at the end of the growth trial (day 126).  3.2.10 Determination of selenium-dependent glutathione peroxidase Selenium-dependent glutathione peroxidase activity was measured as nmol/min/ml following the protocol outlined in the Glutathione Peroxidase Cellular Activity Assay kit (Sigma, 2003). A l l chemicals were purchased through Sigma. Selenium-dependent glutathione peroxidase kinetic activity was measured using a microplate reader at an absorption of 340 nm (SpectraMax 340PC, Molecular Devices) and was recorded and analyzed using SoftMaxPro (version 4.6, Molecular Devices) software.  47  3.2.11 Determination of conjugated dienes Lipid containing chloroform from the above described lipid extraction (section 3.2.5) was concentrated by gently blowing medical grade nitrogen gas (Praxair, 99%) over the surface of the chloroform solution. The chloroform-free lipids were then re-dissolved in cyclohexane at a concentration of 200 ug/ml as described by Corongiu and Banni (1994). The conventional U V absorption spectra were taken at 233 and 242 nm using an U V spectrophotometer (Spectronic 1001+, Milton Roy) and the heights of the two signals were added together. The results were expressed as ug/mg lipid.  3.2.12 Determination of ethoxyquin Ethoxyquin content was determined in the oil, fishmeal, and diets at the start (day-0) and end (day-126) of the dietary storage trial following the A O A C (2000) fluorometric method 963.07 (Novus International, St. Louis, Missouri).  3.2.13 Histopathology Liver and muscle samples were collected from 3 fish/tank at the end of the growth trial for histopathological analysis. Muscle tissue was collected as small cubes (3 cm ) from the right fillet under the dorsal fin. A l l tissue samples were immediately transferred into tubes containing buffered formalin (10%, Sigma) for preservation. After 30 days, preserved samples were transferred into tubes containing ethanol (95%, Sigma) for shipping. Samples were analyzed courtesy of Dr. Lumsden (Ontario Veterinary College, Guelph, ON) for granulomas, ceridosis indicative of liver lipoid disease, and oxidative stress.  48  3.2.14 Metal and trace element determination Metals and trace elements for the fresh fishmeal, wheat gluten meal, casein, gelatin, pelleted diets, and liver tissue were analyzed at Norwest Labs using ultrasonic nebulization inductively coupled plasma-atomic emission spectrometry (USEPA, 1994).  3.2.15 General health General health in the experimental fish during the growth trial was recorded following a modification of an autopsy-based assessment developed by Geode and Barton (1990). At each sampling interval, the general external appearances of all fish were examined for fin, eye, and gill damage, changes in skin pigmentation, and hemorrhaging. External and internal assessments were conducted on every fish that did not survive the growth trial. Internal assessments include examination of the spleen, liver, kidney, hind gut and gall bladder for inflammation and hemorrhaging.  3.2.16 Experimental design and statistical analysis The dietary treatments were arranged in a 2x2x3 factorial design with the factors being supplemental copper (10 or 35 mg/kg) and iron (333 or 1000 mg/kg) and either unoxidized or oxidized herring oil as distinguished by peroxide values (<5 or <5+ or 35+ meq/kg). Differences between means for the various parameters were analyzed by A N O V A or the Kruskall-Wallace test for non-parametric data followed by the Tukey-Bonferroni or non-parametric multiple comparison test, respectively.  49  Growth performance variables examined were specific growth rate (SGR=[ln final weight (g)-ln initial weight (g) • number of experimental day" • 100]), dry feed intake (DFI, total dry feed 1  intake (g) • fish" ), feed efficiency ratios (FE=[wet weight gained (g) • dry feed consumption 1  (g)" ]), protein efficiency ratios (PER, [wet weight gain (g) • protein consumption (g)" ]), percent 1  1  protein deposition (PPD, [protein gained in fish (g) • total protein consumed (g)" • 100]), and 1  gross energy utilization (GEU, [gross energy gained by the fish (MJ) • total gross energy consumed (MJ)" • 100]). 1  Differences among proximate analysis data were determined using A N C O V A with weight as the covariate. Vitamin E, M D A , fatty acid, and growth performance data were analyzed using a two-way A N O V A to test for differences between replicates. Correlations between dietary M D A and dietary n-3 fatty acids, in situ vitamin E, and values for FE of the experimental fish were established using the Spearman rank correlation.  Data were analyzed using SPSS 11.0 (SPSS Inc., Chicago, Illinois). Values were expressed as means and differences between means, and considered to be significant when p<0.05 (Zar, 1996).  3.3  Results  At the start of the study, it was determined that all diets contained similar concentrations of proximate constituents (Table 3.2) and nutritionally important minerals (Table 3.3), except for the varying levels of iron and copper.  50  3.3.1  Malondialdehyde content of dietary treatments  Dietary M D A data are shown in Figure 3.1. The dietary M D A contents on day 126 were significantly (p<0.05) higher in diets 11 (Ice+) and 12 (ICe+) than all other diets. Diets 11 (Ice+) and 12 (ICe+) contained high iron levels and previously oxidized oil. Diets 7 (Ice) and 8 (ICe), which also contained high iron without ethoxyquin protection, had the next level of M D A values followed by diet 6 (iCe), and then diets 5 (ice), 9 (ice+) and 10 (iCe+). Diets 1 (icE), 2 (iCE), 3 (IcE) and 4 (ICE) showed minimal levels of M D A on day 126. Dietary M D A increased significantly in all diets over the course of the 126-day trial.  3.3.2  Body weight gains and feed efficiency  The growth trial data including fork lengths, live weights, mortality, specific growth rates (SGR), feed efficiencies (FE), at the 42-day sampling intervals, fresh liver weights at day 126, as well as the values for FE and SGR for the entire 126-day trial are shown in Table 3.4. The first significant decreases in fish growth due to a dietary treatment took place within the first 42 days of the trial (Interval 1). Fish fed diet 11 (Ice+) exhibited significantly poorer growth than all other groups during the first 42 day period that can be attributed to decreased feed efficiency. Growth performance data for days 42 to 84 (Interval 2) also indicated that fish fed diet 11 (Ice+) generally continued to have poor weight gain along with groups fed diets 7 (Ice), 8 (ICe) and 12 (ICe+). Fish fed diets 7 (Ice), 8 (ICe), 11 (Ice+) and 12 (ICe+) had values for SGR that were significantly below the SGR values observed for fish fed all other diets. Feed efficiency values remained significantly depressed in fish fed diets 11 (Ice+) and 12 (ICe+) during the interval and there was a trend for reduced feed efficiency in groups fed diets 7 (Ice) and 8 (ICe). Between day 84 and 126 (Interval 3), groups fed diets 11 (Ice+), and 12 (ICe+) continued to exhibit the lowest weight and mean length growth rates, with groups fed diets 7 (Ice) and 8 (ICe) following 51  closely behind. On day-126, fish weights followed the sequence: diets 11 (Ice+) and 12 (ICe+) < diets 7 (Ice) and 8 (ICe) < diets 1-6, 9 and 10. Similarly, SGR and F E values of the fish over the entire experimental trial followed the sequence: diets 11 (Ice+) and 12 (ICe+) < diets 7 (Ice) and 8 (ICe) < diets 1-6, 9 and 10. Mortality rates were not influenced by dietary treatment (Appendix 1).  Figure 3.2 illustrates a significant negative correlation (r= -0.71) that was found between feed efficiency and level of dietary lipid peroxidation as measured by M D A concentration. Postjuvenile Atlantic salmon fed diet 11 (Ice+), with the highest level of iron and a previously oxidized oil source, had the lowest feed efficiency (pO.OOl).  3.3.3  Feed intake, PER, PPD, GEU and proximate composition  Feed intake and PER values of fish fed diets 11 (Ice+), and 12 (ICe+) were lower from day 42 to day 126 of the growth trial (Table 3.5) as compared to fish fed the other diets. Considering the entire trial period, fish fed diets 7 (Ice), 8 (ICe), 11 (Ice+), and 12 (ICe+) exhibited significantly poorer G E U and feed intake, whereas fish fed diets 11 (Ice+) exhibited significantly poorer PPD (Table 3.5) as compared to fish fed the other diets.  The proximate composition of both the  muscle and whole body samples were not significantly different when weight was used as a covariate (Table 3.5).  3.3.4  Fatty acid composition of dietary lipid  The fatty acid compositions of the 12 diets were similar at the beginning (day 0) of the trial (Table 3.6a). On day 42, there were significant differences in the fatty acid composition of the diets (Table 3.6b). In particular, the percentages for n-3 H U F A were significantly lower in diet 52  11 (Ice+) relative to the respective values for all other diets. Comparisons of dietary fatty acid compositions on day 84 (Table 3.6c) revealed that the percentages of n-3 fatty acids were significantly lower especially in diet 12 (ICe+), followed by diet 11 (Ice+) and diet 7 (Ice). Percentages for these fatty acids in diet 8 (ICe) showed a trend towards reduction but significant differences were not found relative to respective values measured in diets 1-6, 9 and 10. Clear differentiation of diets 7 (Ice), 8 (ICe), 11 (Ice+) and 12 (ICe+), with respect to n-3 H U F A percentages relative to those noted for other diets, were clearly evident on day 126 (Table 3.6d).  Figure 3.3 shows the negative correlation between percentages for n-3 H U F A and the level of dietary lipid peroxidation (r= -0.85, p<0.05) as measured by M D A concentration. Lipid peroxidation, due to pro-oxidant action and/or oxygen pre-treatment of herring oil, decreased the levels of dietary n-3 H U F A . Diets supplemented with the highest level of iron and lacking antioxidant protection (diets 7 [Ice], 8 [ICe], 11 [Ice+] and 12[ICe+]) demonstrated the greatest loss of E P A and D H A (p<0.02).  3.3.5  Fatty acid composition of muscle tissue  Similar to the findings reported above (section 3.3.4) for dietary fatty acid composition in relation to the treatments, the fatty acid data for fish fed diet 11 (Ice+) at day-126 had significantly lower n-3 H U F A then all other groups except for fish fed diet 12 (ICe+). The percentages of n-3 H U F A in the muscle lipids of fish fed diet 12 (ICe+) were significantly lower relative to values for fish fed diets 4 (ICE), 5 (ice), 8 (ICe), and 9 (ice+). Interestingly, percentages of n-3 H U F A and total P U F A were not significantly different from those fish ingesting the other dietary treatments except where noted above (Table 3.7). A n interesting  53  increase in the percentage of oleic acid (C18:ln9c) was observed in the muscle tissue for fish fed diets 11 (Ice+) and 12 (ICe+) by day 126 (Table 3.7).  3.3.6  Dietary vitam in E  The vitamin E content of the diets were not significantly different (-30 IU/kg) at the beginning of the storage trial (Table 3.8). At the end of the 126-day storage trial, the vitamin E concentrations of diets 1 (icE) and 2 (iCE) remained at -30 IU/kg while values measured in diets 3-12 were significantly decreased (Table 3.8). Diets 3, 4 and 5 had the lowest vitamin E concentrations at the end of the storage trial with values of 9.94, 8.81, and 12.07 IU/kg, respectively (Table 3.8). The remaining diets (6-12) had vitamin E concentrations in the range of 12.90 to 16.08 IU/kg (Table 3.8)  3.3.7  Tissue vitamin E  The vitamin E content of the muscle tissue sampled from the experimental fish at day-126 was significantly affected by the dietary treatments, whereas liver vitamin E content was unaffected. Fish fed diets supplemented with the highest levels of iron and copper and lacking anti-oxidant protection (diets 8 [ICe] and 12 [ICe+]) demonstrated the lowest concentrations of vitamin E in their muscle tissue (p<0.05) (Table 3.8). However, the vitamin E content of the muscle tissues did not reflect the corresponding content of dietary vitamin E (Table 3.8). Instead, the vitamin E content of the muscle tissue appeared to be inversely related to the level of dietary lipid peroxidation (Figure 3.4). Figure 3.4 shows a negative correlation (r= -0.68) between total vitamin E (IU/kg) in the experimental fish muscle tissue and the level of dietary lipid peroxidation as measured by M D A concentration in the feed.  54  3.3.8  Measures of oxidative stress  Histopathological parameters and in situ levels of conjugated dienes and selenium-dependent glutathione peroxidase activity measured in the liver and muscle tissues of the fish fed the test diets 1-12 were not significantly different (Appendix 2).  3.3.9  General health observations  Fish fed diets 7 [Ice], 8 [ICe], 11 [Ice+] and 12 [ICe+] exhibited clinical signs of poor health including light pigmentation, anemia, anorexia, yellowish livers, petechial hemorrhaging at the base of the fins and on the liver, and enlarged spleens (Appendix 3).  3.4  Discussion  The dietary treatments in this study were designed to assess the potential effects of pro-oxidant catalyzed lipid peroxidation, as promoted by the presence of the pro-oxidants, iron and copper, and lack of ethoxyquin supplementation, on the growth performance, health and in vivo oxidative stress of post-juvenile Atlantic salmon (Salmo salar). Detailed effects of the treatments on the cellular and humoral immune responses of the fish were considered to be beyond the scope of this thesis and will be published elsewhere (Balfry et al., unpublished data).  Diets; 7 (Ice), 8 (ICe), 11 (Ice+) and 12 (ICe+), all lacking ethoxyquin protection and having high levels of iron, negatively impacted the specific growth rates and feed efficiencies of the experimental fish (Table 3.4). The inclusion of previously oxidized lipid in diets 11 (Ice+) and 12 (ICe+) influenced the aforementioned performance parameters in the experimental fish earlier and to a greater degree than the lipids in diets 7 (Ice) and 8 (ICe), indicating that the extent of 55  dietary lipid peroxidation was an influential factor on the growth performance of Atlantic salmon. It should be noted that diets 7 (Ice) and 11 (Ice+) influenced the experimental fish earlier than their oxidation-level counterparts, diets 8 (ICe) and 12 (ICe+), respectively.  It therefore appears that the level of dietary lipid peroxidation adversely influenced the nutritive value of the dietary treatments and consequently depressed the growth performance of the experimental fish. Figures 3.2, 3.3, and 3.4 clearly indicated negative relationships between dietary M D A concentration and feed efficiency, dietary n-3 H U F A , and muscle vitamin E concentrations, respectively. The extent of the dietary lipid peroxidation in each diet was influenced by the presence of pro- and anti-oxidants and the initial level of lipid peroxidation (Figure 3.1). Iron was a strong catalyst of lipid peroxidation and the pro-oxidant effect was not exacerbated by a high dietary concentration of copper. Indeed, the high copper concentration lessened the pro-oxidant effect of iron with respect to feed efficiency values for the salmon (Table 3.5; effects of diets 11 and 12).  3.4.1 Antagonistic behaviour ofpro-oxidants The antagonistic relationship between copper and iron has been widely reported in biological systems (White et al., 2004; Sergent et al. 1999; Ouzounidou et al., 1998; Schaich, 1992; Halliwell and Gutteridge, 1984). Schaich (1992) discussed the possible antagonistic role of metals such as copper on pro-oxidant activity. She suggested that the oxidation-reduction of radicals produced by iron and/or copper, in high concentrations, can result in an overall reduction of the oxidation rate due to interruption of the radical chain. Therefore, high copper concentrations lessening the pro-oxidant effect of iron in the present study is not unique, but has  56  important implications in understanding how trace biometal interactions influence the development of disease.  3.4.2  Dietary iron effects  Desjardins et al. (1987), suggested that the combination of iron overload and diet rancidity could induce signs of toxicity in rainbow trout at an iron concentration as low as 86 mg/kg whereas iron toxicity alone required a concentration of 1380 mg/kg. The data from the current study suggest that iron-catalyzed dietary lipid peroxidation contributed to eliciting a toxicological effect in post-juvenile Atlantic salmon (Table 3.4). However, the toxicity was not related to iron overload. Dietary treatments protected with the anti-oxidant ethoxyquin (Table 3.2), including those with high iron levels (1450 mg/kg) did not result in any sign of toxicity in the experimental fish. Dietary treatments that did elicit a toxicological response in the experimental fish were those that were undergoing iron-catalyzed lipid peroxidation and these were associated with deficiencies in dietary vitamin E (Table 3.8) and n-3 H U F A (Table 3.6d).  Diets, lacking ethoxyquin protection and having high levels of iron (i.e. diets 7 (Ice), 8 (ICe), 11 (Ice+) and 12 (ICe+), negatively impacted the feed intake, PER, G E U , and PPD of the experimental fish (Table 3.5). PER and PPD are measures of the dietary protein quality whereas G E U reflects a ratio between energy gained and energy consumed. Therefore, treatment differences for these measures would suggest that the dietary protein and its bioavailability had been affected (Dabrowski and Guderley, 2002).  As described in section 3.3.3, differences in G E U values for salmon fed diets 7 (Ice), 8 (ICe), 11 (Ice+) and 12 (ICe+) were not apparent but were significantly different from all other diets. PPD 57  data followed the sequence; 11 (Ice+) < 12 (ICe+) and 7 (Ice) < 8 (ICe) < all other diets while PER data followed the sequence; 11 (Ice+) and 12 (ICe+) < 7 (Ice) < all other diets. Although there were minor discrepancies as to how strong the association was between diets 7 (Ice), 8 (ICe), 11 (Ice+) and 12 (ICe+) are for the preceeding parameters, it is obvious that diets with high iron were influential. Consequently, these data suggest that iron-catalyzed dietary lipid peroxidation likely adversely influenced dietary protein quality and consequently, fish growth.  3.4.3  Protein modification  Lipid peroxidation products formed in the presence of protein can influence protein quality. Lipid peroxidation is known to catalyze cross-linking reactions between the formed aldehydes and amino groups (Fenaille, et al., 2003; Leaver, et al., 1999; L i , King, 1999). Lysine is the amino acid group most often influenced by lipid peroxidation reactions (Fenaille, et al., 2003; Leaver, et al., 1999) with aldehyde-lysine adducts immunohistochemically found in atherosclerotic plaques (Itakura, et al., 2003). Both protein cross-linking and lipid peroxidationlysine reactions lower the protein quality of food/feed (Fenaille, et al., 2003; Leaver, et al., 1999) and consequently, influences protein (amino acid) metabolism (Dabrowski and Guderley, 2002). Therefore, the significantly lower values for PER, G E U , and PPD in fish fed diets 7 (Ice), 8 (ICe), 11 (Ice+) and 12 (ICe+) could be due to decreased dietary protein quality.  3.4.4  Fatty acid deficiency  The data presented in Table 3.6a-d and 3.7 strongly suggest that dietary fatty acid content greatly influenced the fatty acid composition of the muscle in the fish fed diets 11 (Ice+) and 12 (ICe+). The experimental fish that consumed diets 11 (Ice+) and 12 (ICe+), which were deficient in n-3 58  H U F A due to peroxidation of the dietary lipids, also had reduced percentages of n-3 H U F A in their muscle. The experimental fish seemed to compensate for the loss of n-3 H U F A in the diets by increasing the absorption of oleic acid (C18:ln-9c). Elevation of oleic acid in the muscle tissue, or alteration in the metabolism of n-9 fatty acids, has been previously reported to accompany E F A deficiency (Henderson and Tocher, 1987; Watanabe, 1982).  Forster et al. (1988), found that growth rate and feed efficiency of coho salmon (Oncorhynchus kisutch) were decreased due to sub-optimal concentrations of dietary n-3 H U F A but did not find evidence for any other toxic factors due to oxidized dietary lipid. Similarly, the growth performance data from this growth trial indicated that the experimental fish were influenced by the level of n-3 H U F A in their diets and possibly by differences of other nutrients such as vitamin E.  3.4.5  Nutritional quality of fillets  The beneficial effects of E P A and D H A for fish (Sargent et al., 2002; Takeuchi et al., 1979; Higgs and Dong, 2000; N R C , 1993; Tacon, 1996) and humans (Bjerregaard et al., 2004; Din et al., 2004; Tocher et al., 2003) are well documented. A n important factor influence the nutritional quality of salmon fillets is the level of E P A and D H A present in the muscle tissue (Rasmussen, 2001). Decreases in the levels of the n-3 H U F A in salmon flesh can influence the potential health benefits associated with consuming salmon (Bell et al., 2003).  As mentioned in section 3.4.4, the experimental fish fed diets deficient in n-3 H U F A , viz. diets 11 (Ice+) and 12 (ICe+), had decreased levels of n-3 H U F A in their muscle tissue. This result, linking dietary fatty acid composition to muscle fatty acid composition, is not uncommon (Bell 59  et al., 2003; Chaiyapechara et al., 2003; Hamre et al., 2004; Izquierdo et al. 2003). However, diets 11 (Ice+) and 12 (ICe+) became deficient in n-3 H U F A post- formulation of the diets. This implies that factors influencing dietary lipid peroxidation, including imbalanced diets and improper storage conditions, could decrease the nutritional quality of the salmon, and ultimately their market value.  3.4.6  Vitamin E deficiency  As mentioned above, vitamin E deficiency may have adversely influenced the growth performance of the Atlantic salmon in this study. Although histopathological signs of deficiency were not observed, clinical signs of vitamin E deficiency (Halver, 2002) were observed. These included lean bodies, lighter skin pigmentation, anemia, anorexia, and yellowish livers.  Oxidized feed is known to induce pathologies in fish especially in the presence of low vitamin E concentrations (Hamre, et al., 2001; Tacon, 1996). Pathological development can be hindered by incorporating higher dietary concentrations of vitamin E (Baker and Davies, 1997; Hamre, et al., 2001; Hung, et al., 1981; Tacon, 1996; Tocher, et al., 2003). Oxidized oil is often linked to decreased vitamin E levels in fish muscle tissues (Baker and Davies, 1997; Hung, et al., 1981). However, it has remained unclear as to whether the oxidized oil decreased the vitamin E level prior to its absorption or whether the vitamin E concentration was decreased in vivo (Hamre, et al., 2001).  3.4.7  Feed intake  As mentioned in section 3.3.3, fish fed diets 7 (Ice), 8 (ICe), 11 (Ice+), and 12 (ICe+) exhibited significantly lower feed intake than the fish fed the other experimental diets. However, the fish 60  fed these diets also had compromised growth rates and feed efficiencies possibly due to essential fatty acid (section 3.4.4) and vitamin E deficiencies (section 3.4.6); compromising fish size. Feed intake is generally influenced by palatability and fish size (Hardy and Barrows, 2002), whereas, fish size, growth and health are compromised by the lack of essential nutrients and improper balances of the major energy-yielding nutrients (Higgs et al., 1995). Although the palatability of the feed, used in this study, may have been adversely affected by the pro-oxidantcatalyzed lipid peroxidation, the presence of Finnstim™, an appetite enhancer, should have minimized the effects of sensory changes in the diet. Correspondingly, Hamre et al. (2001) found that the feed intake was not influenced in Atlantic salmon fed diets coated with oxidized herring oil. Hence, feed intake differences observed in this experimental trial were likely related to fish size, while fish size likely reflected diet quality in terms of decreased levels of dietary n-3 H U F A , vitamin E and protein as a consequence of iron-catalyzed lipid peroxidation.  3.5  Conclusion  This study has demonstrated that iron-catalyzed dietary lipid peroxidation adversely affected the growth performance of post-juvenile Atlantic salmon. High dietary iron concentrations were the major influential factor in promoting high dietary levels of lipid peroxidation. Subsequently, the iron-catalyzed dietary lipid peroxidation elicited toxicological impacts (i.e.poor growth and health) on the experimental fish; however, the level of iron itself was not found to be toxic under the conditions of the study. Lipid peroxidation can influence many components of a feed that can negatively impact its digestibility, safety, and nutritional value with the latter involving alterations in vitamins, proteins and fatty acid composition. Previous studies have suggested that the consumption of oxidized dietary lipids can influence in situ oxidative stress, however, the liver and muscle tissue in this study were not affected as indicated by histological analysis. In  this study the data indicate that the observed growth depression and poor health of the salmon probably resulted from dietary protein modification, deficiencies of n-3 H U F A and/or vitamin E, but not lipid peroxidation product toxicity. Further, the data implies that these factors, which were influenced by dietary lipid peroxidation, can decrease the nutritional quality and market value of the salmon. This suggests that the impact of oxidized dietary lipids and its effects on the nutritional and market value of salmon can be minimized by adequate dietary supplementation of anti-oxidants such as vitamin E and ethoxyquin.  62  50  1 (icE)  2 iCE  3  4  5  (IcE) (ICE) (ice)  6  7  (iCe) (Ice)  8  9  10  11  12  (ICe) (ice+) (iCe+) (lce+)(ICe+)  Diet Figure 3.1. Malondialdehyde (MDA) content of the dietary treatments at the start (day-0, black bars) and end (day-126, grey bars) of the growth trial. Values are means, ± standard deviation, of 3 replicate tanks for each diet ± SD. Means with a common superscript letter were not significantly different (p<0.05) on day 0 (a-b) and day 126 (A-E).  63  35  40  50  MDA (|jmol/kg) in feed  Figure 3.2. Relationship between final feed efficiency and final malondialdehyde (MDA) content of the dietary treatments. Values are means, ± standard deviation, of 3 replicate tanks for each diet ± SD. Spearman rank correlation of r= -0.71, p<0.05.  64  14 12 519 10'  IO  u- 6  11* 12  10  — i —  — i —  — i —  — i —  — i —  15  25  30  35  40  20  45  50  MDA ([jmol/kg) in feed  Figure 3.3. Relationship between final dietary n-3 highly unsaturated fatty acids (n-3 HUFA) and final malondialdehyde (MDA) content of dietary treatments. Values are means, ± standard deviation, of 3 replicate tanks for each diet ± SD. Spearman rank correlation of r= -0.85, p<0.05.  65  MDA (jjmol/kg) in feed Figure 3.4. Relationship between muscle vitamin E concentration and final malondialdehyde (MDA) content of dietary treatments. Values are means, ± standard deviation, of 3 replicate tanks for each diet ± SD. Spearman rank correlation of r= -0.68, p<0.05.  66  Table 3.1. Ingredient composition of the test diets fed to Atlantic salmon over a 126 day period.  1 (icE) 448.6 na 190.0 33.0 4.5 42.4 42.4 15.0 30.0 5.0 7.0 6.0 0.4 145.7 na 13.0 15.0 0.058 2.0  Ingredients Mixed groundfish meal, ethoxyquin (EQ) Mixed groundfish meal without EQ Wheat gluten meal Vitamin-free casein Gelatin Pre-gelatinized wheat starch Raw wheat starch Vitamin supplement Mineral supplement Vitamin E supplement Iron (Fe)/copper (Cu) supplement Choline chloride (50%) Vitamin C monophosphate (35%) Fresh herring oil (no EQ) Moderately oxidized herring oil (no EQ) Permapell Finnstim™ EQ DL-methionine 1  3  4  2 (iCE) 448.6 na 190.0 33.0 4.5 42.4 42.4 15.0 30.0 5.0 7.0 6.0 0.4 145.7 na 13.0 15.0 0.058 2.0  3 (IcE) 448.6 na 190.0 33.0 4.5 42.4 42.4 15.0 30.0 5.0 7.0 6.0 0.4 145.7 na 13.0 15.0 0.058 2.0  4 (ICE) 448.6 na 190.0 33.0 4.5 42.4 42.4 15.0 30.0 5.0 7.0 6.0 0.4 145.7 na 13.0 15.0 0.058 2.0  Diet (g/kg dry weight basis) 7 5 6 8 (ice) (iCe) (Ice) (ICe) na na na na 448.6 448.6 448.6 448.6 190.0 190.0 190.0 190.0 33.0 33.0 33.0 33.0 4.5 4.5 4.5 4.5 42.4 42.4 42.4 42.4 42.4 42.4 42.4 42.4 15.0 15.0 15.0 15.0 30.0 30.0 30.0 30.0 5.0 5.0 5.0 5.0 7.0 7.0 7.0 7.0 6.0 6.0 6.0 6.0 0.4 0.4 0.4 0.4 145.7 145.7 145.7 145.7 na na na na 13.0 13.0 13.0 13.0 15.0 15.0 15.0 15.0 0 0 0 0 2.0 2.0 2.0 2.0 5  9 (ice+) na 448.6 190.0 33.0 4.5 42.4 42.4 15.0 30.0 5.0 7.0 6.0 0.4 na 145.7 13.0 15.0 0 2.0  10 (iCe+) na 448.6 190.0 33.0 4.5 42.4 42.4 15.0 30.0 5.0 7.0 6.0 0.4 na 145.7 13.0 15.0 0 2.0  11 12 (Ice+) (ICe+) na na 448.6 448.6 190.0 190.0 33.0 33.0 4.5 4.5 42.4 42.4 42.4 42.4 15.0 15.0 30.0 30.0 5.0 5.0 7.0 7.0 6.0 6.0 0.4 0.4 na na 145.7 145.7 13.0 13.0 15.0 15.0 0 0 2.0 2.0  The vitamin supplement provided the following amounts of kg" of diet on a dry weight basis: vitamin A acetate, 5000 IU; cholecalciferol (D ), 2400 IU; menadione (as MSBC) 18.0 mg; D-calcium pantothenate, 168.0 mg; pyridoxine HC1, 49.3 mg; riboflavin, 60.0 mg; niacin 300.0; folic acid 15.0 mg; thiamine mononitrate, 56.0 mg; biotin, 1.5 mg; cyanocobalamin (B ), 0.09 mg; inositol, 400.0 mg. Wheat starch was used as the carrier. The mineral supplement provided the following (mgkg'diet on dry weight basis): P (as KH P0 ), 2500; magnesium (MgS0 7H 0), 600; manganese (as MnSCyH 0), 70.0; zinc (as ZnS0 -7H 0), 82.0; cobalt (as CoCl -6H 0), 3.0; iodine (as K I 0 and KI, 1:1), 10.0; fluorine (asNaF), 5.0; selenium (as Na Se0 ), 0.15; potassium (as K S 0 and K C 0 , 1:1), 2801 and as K H P 0 , 3156. Supplemental vitamin E (as DL-a-tocopheryl acetate) in diets 1-8 was 3.6 IU/kg diet and in diets 9-12 was 15.2 IU/kg. Alpha-cellulose was the carrier. The supplemental amount of iron as FeS0 7H 0 was 333 ppm in diets 1, 2, 5, 6, 9, and 10 and was 1000 ppm in all other diets. The supplemental amount of copper as CuS0 -5H 0 was 10 ppm in diets 1, 3, 5, 7, 9, and 11 and was 35 ppm in all remaining diets. Alpha-cellulose was used as the carrier. Refer to Table 3.2 for determined concentrations of iron and copper in all of the dietary treatments. na = not added 1  1  3  12  2  2  2  2  4  3  2  2  2  4  2  3  4  4  4  5  2  2  2  4  3  2  4  4  2  Table 3.2. Initial dietary concentrations of proximate constituents, ethoxyquin, copper and iron on a dry weight basis and initial peroxide values. Copper was in the form of CuSCv5H20 and iron was in the form of FeS04-7H 0. 2  Diet  Protein %  Ash %  Lipid %  Ethoxyquin mg/kg  Copper mg/kg  Iron mg/kg  Peroxide Value meq/kg  1 (icE) 2 (iCE) 3 (IcE) 4 (ICE) 5 (ice) 6 (iCe) 7 (Ice) 8 (ICe) 9 (ice+) 10 (iCe+) 11 (Ice+) 12 (ICe+)  51.4 51.7 50.6 51.6 50.3 50.4 51.6 50.5 49.4 50.4 48.9 50.4  14.2 14.5 14.4 14.4 14.4 14.4 14.8 14.8 14.0 14.1 14.9 14.7  19.4 18.8 19.0 19.1 19.4 19.0 18.1 19.1 19.4 19.5 19.0 18.6  150 150 150 150 na na na na na na na na  11 38 8 36 11 34 7 38 9 24 14 40  724 667 1477 1405 694 732 1450 1543 709 760 1487 1557  <5 <5 <5 <5 <5 <5 <5 <5 35 35 35 35  68  Table 3.3. Mineral compositions of the test diets measured as mg per kg diet (dry weight basis). Diet  Calcium  Cobalt  Copper  Iron  Magnesium  Manganese  Phosphorus  Potassium  Sodium  Zinc  1 (icE) 2 (iCE) 3 (IcE) 4 (ICE) 5 (ice) 6 (iCe) 7 (Ice) 8 (ICe) 9 (ice+) 10 (iCe+) 11 (Ice+) 12 (ICe+)  40150 36550 39200 40000 38367 37450 40950 39567 38200 40233 40867 40433  2 2 3 2 2 1 2 2 1 2 2 2  11 38 8 36 11 34 7 38 9 24 14 40  724 667 1477 1405 694 732 1450 1543 709 760 1487 1557  1790 1740 1737 1615 1780 1640 1780 1870 1650 1627 1750 1863  72 81 71 63 64 61 72 51 61 65 82 81  24050 23050 23700 24450 23033 22800 24900 24133 22650 23900 23300 23833  8215 8075 8100 8010 7853 7335 8470 8780 7230 7340 8063 8673  3465 3405 3340 3160 3707 3400 5035 3860 3245 3257 3840 3850  117 114 110 100 125 116 119 128 140 101 118 123  Table 3.4. Growth-related performance data of fish fed diets 1 through 12. SGR represents specific growth rate and FE represents feed efficiency. Values are means, ± standard deviation, of 3 replicate tanks for each diet. Means for a given parameter that share a common superscript letter were not significantly different (ns). 1 (icE)  2 (iCE)  3 (IcE)  4 (ICE)  5 (ice)  6 (iCe)  7 (Ice)  8 (ICe)  9 (ice+)  10 (iCe+)  11 (Ice+)  Length (cm)  18.63  18.60  18.74  18.74  18.67  18.76  18.49  18.70  18.78  18.63  18.73  18.63  ns  Live weight (g)  64.00  63.81  65.81  64.29  65.06  65.97  62.98  65.16  64.78  64.45  65.70  64.45  ns  21.81  22.28  19.68  22.04  21.3 l  22.24  22.22  21.30  21.3 l  pO.OOl  Diet  12 (ICe+)  Signif.  Day 0  Interval  Length (cm)  22.25  Live weight (g)  120.81  bc  d  21.80  22.48  b  c  b  111.52 °  122.10  109.65  b  b  b  d  b  bc  c  bc  119.66  123.00  116.80  b  b  d  d  bcd  c  118.65  cd  bc  118.61  cd  bc  115.63  bcd  9933a  p<0.001  1.03 ab  p<0.001  1.33  1.47  1.26  1.45  1.48  1.46  1.44  1.44  1.38  0.93  FE  1.24  1.13  1.16  1.08  1.18  1.26  1.14  1.21  1.15  1.09  0.86  b  b  26.69  Live weight (g)  209.58  SGR  1.31  FE  1.21  26.15  d  d  b  bc  b  b  b  b  b  b  b  a  b  b  ab  a  p<0.001  Q 9 2  b  26.73  26.04  d  bc  26.74  26.86  d  25.72  d  25.75  b  26.45  b  bc  26.4 l  b c  23.98  24.42  a  p<0.001  a  182.47  183.55  201.97  150.37  157.49"  p<0.001  1.39  1.28  1.35  b  1.34  1.31  1.07  1.03  a  1.27  1.33  1.02  1.10  p<0.001  1.21  1.28  b  1.22  1.22  1.23  1.10  1.07  ab  1.17  1.20  0.99  1.03  pO.OOl  30.20°  30.02  29.89°  26.38  303.78°  302.88°  210.34  3.56  3.91  b  2.98  a  b  0.79  a  199.66  cd  209.12  d  b  b  b  b  193.40  bc  209.71  d  b  213.17  d  b  b  b  a  b  ab  b  cd  201.93  cd  b  b  b  b  3  a  a  a  a  3 - Day 84 to 126  Length (cm)  30.35  Live weight (g)  313.57  Liver weight (g)  3.62  SGR  0.98  FE  1.10  Entire  b  b  2 - Day 42 to 84  Length (cm)  Interval  b  a  97.71  1.51  b  b  a  3  SGR  Interval  o  1 - Day 0 to 42  Trial  29.91  c  c  ab  c  304.1 l  c  309.32  b  3.94  b  0.90  3.56  b  1.04  c  1.13  c  b  c  300.75 3.83  30.45  c  b  c  314.11 3.86  30.55  c  b  c  318.99 3.95  b  0.96  c  c  3.32  28.47  b  253.44  b  ab  1.04"  0.98  1.08°  1.10  c  1.06  1.10  0.96  b  b  28.31  c  b  0.81  c  a  b  258.48 3.46  30.08  b  ab  b  c  ab  a  0.82  1.01  0.96  0.93  b  1.09°  1.04°  0.80  1.24°  1.22°  0.91  1.13  l.ll  0.88  b  26.38  a  a  225.37 3.05 0.87  p<0.001  a  a  p<0.001  a  p<0.001  ab  p<0.001  a  0.93  b  pO.OOl  a  1.00  ab  pO.OOl  bc  pO.OOl  - Day 0 to 126  SGR  1.27°  1.25  c  1.22°  1.22°  1.25  1.25  1.10  FE  1.17°  1.16  bc  1.17°  1.14 °  1.14  1.18°  1.04  b  c  bc  c  b  abc  1.10 j abc b  Q4  bc  b c  a  0.96  Table 3.5. Feed intake (g fish" ), protein efficiency ratios (PER, [wet weight gain (g) • protein consumption (g)' ]), percent protein deposition (%PD, [protein gained in fish (g) • total protein consumed (g)" • 100]), and gross energy utilization (GEU, [gross energy gained by the fish (MJ) • total gross energy consumed (MJ)" • 100]), and final proximate composition of fish (% body weight basis). Initial muscle and whole body proximate compositions for all experimental fish were 19.0% and 16.5% for protein, 1.8% and 6.8% for lipid, 69.2% and 65.5% for moisture, and 7.4% and 8.6% for ash, respectively. Values are means, ± standard deviation, of 3 replicate tanks for each diet. Means for a given parameter that share a common superscript 1  1  1  1  Interval  2 (iCE)  3 (IcE)  4 (ICE)  5 (ice)  6 (iCe)  7 (Ice)  8 (ICe)  9 (ice+)  10 (iCe+)  11 (Ice+)  12 (ICe+)  45.8  42.2  48.7  42.1  46.1  45.4  47.2  44.5  46.9  46.6  37.3  38.1  2.4  2.2  2.3  2.1  2.4  2.5  2.2  2.4  2.3  2.2  1.7  1.8  1  P E R (%)  73.5  P E R (%)  2.4  1  1  2.3  Trial  96.1 2.1  P E R (%)  68.3  c  2.5  b  bc  68.4 2.4  b  bc  b  74.0  73.2  59.7  2.4  2.4  2.1  C  C  b  b  60.4  ab  2.1  a  ab  a  71.0  bc  71.9  bc  2.4"  2.4  98.2  96.5  53.4 2.0  b  56.6  a  2.0  a  p<0.05 p<0.05  a  a  97.5  b  2.2  b  89.5  b  2.1  b  96.8  b  2.1  b  100.6  96.3  2.1  2.2  b  b  b  b  77.0  b  81.5  a  1.9'  b  a  2.2  1.8  a  b  2.1  b  72.8  b  75.5  a  a  1.8*  1.7  b  a  p<0.05 p<0.05  - Day 0 to 126  Feed Intake (g fish" ) 1  215.2  P E R (%)  2.3  P P D (%)  39.3  G E U (%)  47.7  Final  73.l  C  b  3 - Day 84 to 126  Feed Intake (g fish" )  Entire  ns ns  2 - Day 42 to 84  Feed Intake (g fish" )  Interval  Signif.  1 - Day 0 to 42  Feed Intake (g fish" )  Interval  1 (icE)  proximate  b  212.5 2.2  b  b  206.5 2.3  b  C  38.9  C  44.7  C  b  b  b  207.1 2.2  41.7°  39.7  45.6  45.4  bc  b  220.5 2.3  b  C  bc  b  40.0 46.0  214.6 2.3  b  C  bc  b  37.1 46.1  183.6  3  2.0  b  b  bc  ab  36.5  ab  38.4  a  186.2 jab  215.8  38.3  39.4  39.6  44.9  44.4  a  2  bc  37.6  a  2.3  b  214.8 2.2  b  C  b  b  163.3  3  1.8"  b  C  b  31.9  a  29.2  a  169.9  a  1.9  a  34.0  ab  33.8  a  p<0.05 p<0.05 p<0.05 p<0.05  composition  4.2  3.9  4.4  4.9  4.1  3.2  3.4  Muscle - Lipid (%)  4.7  Muscle - Protein (%)  18.0  20.5  20.0  19.3  18.9  18.6  19.0  18.9  Muscle - Moisture (%)  4.2  2.7  2.2  18.9  18.1  15.5  16.5  4.4  65.7  63.6  65.0  65.8  65.4  65.3  66.4  66.9  65.3  66.3  69.8  69.2  Muscle - Ash (%)  9.1  8.9  8.5  8.9  9.0  8.8  8.8  8.7  9.0  8.9  9.0  8.7  Whole Body- Lipid (%) Whole Body - Protein (%)  9.7  8.9  8.6  9.1  9.2  9.6  7.4  6.9  9.2  9.1  6.2  6.7  17.1  17.2  17.6  17.7  17.4  16.0  17.5  18.0  17.0  17.7  17.3  17.5  Whole Body - Moisture (%)  60.8  61.7  61.7  60.3  62.3  62.9  62.9  63.6  62.0  61.3  64.8  61.2  9.6  10.1  9.6  10.5  8.3  8.8  9.8  8.9  8.9  9.3  9.3  11.7  Whole Body - Ash (%)  ns ns ns ns ns ns ns ns  Table 3.6a. Relative fatty acid composition (% of total fatty acids) for diets 1 through 12 on day 0 of storage trial. *Represents all fatty acids in this category; some are not shown. Means for a given parameter that share a common superscript letter were not significantly different (p<0.05). Values are means, ± standard deviation, of 3 replicate tanks for each diet. Not significant is represented by ns. Fatty  Acid  C14:0  1 (icE) 5.70  2 (iCE) 5.12  3 (IcE)  4 (ICE)  5 (ice)  6 (iCe)  7 (Ice)  8 (iCe)  9 (ice+)  10 (iCe+)  11 (Ice+)  12 (ICe+)  Sig.  5.22  4.84  4.62  5.03  4.77  5.09  4.55  5.33  5.31  4.96  ns ns  C15:0  0.46  0.42  0.41  0.37  0.36  0.41  0.37  0.44  0.42  0.41  0.46  0.44  C16:0  18.06  17.15  17.71  16.49  15.65  16.60  15.97  17.04  16.32  17.38  17.45  16.40  ns  2.77  2.77  2.90  2.54  2.44  2.70  2.57  2.75  2.65  2.76  2.80  3.23  ns  28.80  27.32  28.44  32.50  36.84  32.81  34.73  29.69  30.18  30.41  28.80  31.27  ns  C16:l  6.55  5.95  5.88  5.34  5.79  6.51  5.91  7.39  6.65  6.92  6.99  7.71  ns  C18:ln9t  0.65  0.63  0.63  0.40  0.50  0.43  0.60  0.38  0.36  0.43  0.62  0.59  ns  C18:ln9c  23.93  24.49  23.39  21.52  19.75  22.22  20.86  22.74  20.40  23.60  24.32  22.29  ns  C18:ln7  3.98  3.59  4.05  3.90  3.81  3.91  4.18  3.90  3.60  4.03  3.85  3.96  ns  C20:l  8.52  8.83  8.49  7.86  7.58  8.69  7.68  8.48  7.43  8.67  8.97  8.04  ns  C22:ln9  7.46  8.60  8.32  7.50  6.59  7.38  7.67  8.10  8.59  7.66  8.04  7.97  ns  52.93  53.99  52.74  47.99  45.21  50.20  48.02  52.23  48.16  52.61  54.19  51.75  ns  Total dienes*  1.28  1.50  1.36  0.74  1.24  0.28  0.62  0.40  1.15  0.31  0.76  0.39  ns  C18:2n6t  0.40  0.03  0.04  0.22  0.36  0.29  0.37  0.36  0.29  0.39  0.48  1.01  ns  C18:2n6c  4.19  4.16  4.97  5.16  4.24  4.28  4.60  4.71  5.96  4.26  4.34  4.61  ns  0.09  0.11  0.26  ns  C18:0 Total saturated*  Total monoenes*  C20:4n6  0.06  0.11  0.11  0.88  0.66  0.25  0.16  0.31  0.62  Total n-6 P U F A *  5.00  4.68  5.52  6.57  5.44  5.05  5.29  5.58  7.06  5.14  5.25  6.40  ns  C18:3n3  0.57  0.57  0.64  0.52  0.36  0.12  0.61  0.11  0.64  0.57  0.16  0.60  ns  C20:5n3  5.94  6.17  5.86  5.84  5.57  5.91  5.51  6.04  5.89  5.74  5.79  5.06  ns  C22:5n3  0.89  0.92  0.87  0.91  0.82  0.92  0.80  0.88  0.82  0.86  0.89  0.76  ns  C22:6n3 Total n-3 H U F A *  3.80  4.04  3.79  4.06  4.23  3.98  4.17  4.20  4.83  3.59  3.40  2.95  ns  11.99  12.51  11.94  12.20  11.28  11.65  11.33  12.10  13.45  11.53  11.01  10.19  ns  Total P U F A  17.00  17.19  17.46  18.77  16.71  16.70  16.62  17.69  20.51  16.67  16.26  16.60  ns  Table 3.6b. Relative fatty acid composition (% of total fatty acids) for diets 1 through 12 on day 42 of storage trial. *Represents all fatty acids in this category; some are not shown. Means for a given parameter that share a common superscript letter were not significantly different (p<0.05). Values are means, ± standard deviation, of 3 replicate tanks for each diet. Not significant is represented by ns. Fatty  Acid  1 (icE) 2 (iCE)  3 (IcE)  4 (ICE)  5 (ice)  6 (iCe)  7 (Ice)  8 (iCe)  9 (ice+)  10 (iCe+)  11 (Ice+)  12 (ICe+)  SiR.  C14:0  5.43  4.92  5.81  5.08  5.64  8.66  5.26  5.39  4.97  5.00  6.35  5.09  ns  C15:0  0.42  0.38  0.41  0.41  0.42  0.56  0.44  0.46  0.40  0.41  0.50  0.41  ns  C16:0  17.24  16.15  17.39  16.34  17.58  20.35  17.93  18.49  16.51  16.76  20.15  17.01  ns  C18:0  2.72  2.60  2.58  2.61  2.70  2.41  2.92  2.91  2.71  2.75  3.15  2.79  ns  29.51  29.00  30.37  30.08  31.22  37.85  28.62  29.69  26.98  29.50  31.45  27.94  ns  7.81  7.90  8.16  7.67  6.66  8.16  8.37  8.47  8.79  6.33  7.20  6.32  ns  Total saturated* C16:l C18:ln9t  0.58  0.56  0.33  0.57  0.60  0.59  0.50  0.41  0.51  0.23  0.39  0.55  ns  C18:ln9c  22.92  22.65  22.59  22.83  23.18  21.40  25.43  24.91  23.63  24.00  26.33  24.07  ns  C18:ln7  4.09  3.71  5.18  3.50  4.15  3.89  3.60  3.74  3.39  3.07  4.43  3.61  ns  C20:l  8.59  8.55  8.27  8.49  8.74  6.58  9.60  9.84  9.14  9.29  9.59  9.60  ns  C22:ln9  7.65  9.32  7.44  8.92  8.19  6.16  10.34  9.05  10.04  9.43  10.66  9.97  ns  59.41  57.71  56.50  53.77  60.14  55.65  ns  52.49  53.90  53.28  53.14  52.34  47.94  Total dienes*  1.13  1.48  0.90  0.31  0.47  0.00  0.34  0.44  0.39  0.60  0.33  0.61  ns  C18:2n6t  0.41  0.55  1.09  0.54  0.44  0.35  0.14  0.08  0.62  0.05  0.35  0.04  ns  C18:2n6c  3.53  3.47  3.39  3.47  3.26  2.72  3.10  3.85  3.74  3.88  3.45  3.76  ns  C20:4n6  0.25  0.06  0.29  0.06  0.43  0.19  0.41  0.06  0.06  0.06  0.06  0.06  ns  Total n-6 P U F A *  4.50  4.72  5.05  4.39  4.54  5.74  3.88  5.23  4.87  4.70  4.68  4.46  ns  Total monoenes*  C18:3n3  0.59  0.59  0.41  0.59  0.70  0.32  0.67  0.92  0.67  0.66  0.71  0.76  ns  C20:5n3  6.00  5.64"  5.29  b  5.66  b  5.84  b  4.52  b  3.95  1.84  ab  5.78  5.79  b  1.35  5.85  b  p<0.05  C22:5n3  0.92  b  0.90  b  0.73  b  0.86  b  0.95  b  0.53  b  0.64  b  0.3 l  a b  0.94  b  0.24  0.93  b  p<0.05  C22:6n3 Total n-3 H U F A *  4.01  b  3.46  b  3.33  4.67  b  3.64  b  2.47  b  2.29  a  12.37  8.47  b  7.75  Total P U F A  16.87  b  b  b  b  10.59  b  10.39  15.3 l  b  15.45  b  b  12.07  b  11.44  16.47  b  15.98  b  b  14.20  b  b  3.70  11.62  b  b  0.90  b  3.89  a  3.63  b  3.39  a  b  8.08  a  3.67  ab  11.26  11.43  12.16  16.12  16.12  b  b  b  a  0.90"  b  6.93  b  b  b  p<0.05  b  11.35  b  p<0.05  15.80  b  p<0.05  Table 3.6c. Relative fatty acid composition (% of total fatty acids) for diets 1 through 12 on day 84 of storage trial. *Represents all fatty acids in this category; some are not shown. Means for a given parameter that share a common superscript letter were not significantly different (p<0.05). Values are means, ± standard deviation, of 3 replicate tanks for each diet. Not significant is represented by ns. Fatty  Acid  1 (icE)  2 (iCE)  3 (IcE)  4 (ICE)  5 (ice)  6 (iCe)  7 (Ice)  8 (iCe)  9 (ice+)  10 (iCe+)  11 (Ice+)  5.65  5.69  6.88  12 (ICe+) 8.20  Sig. ns  C14:0  5.55  5.39  5.43  6.33  5.59  5.51  6.95  6.76  C15:0  0.42  0.43  0.44  0.46  0.42  0.45  0.56  0.49  0.45  0.45  0.55  0.59  ns  C16:0  17.44  17.41  17.49  18.59  17.86  18.07  19.68  20.03  18.10  17.95  19.75  23.00  ns  C18:0  2.74  2.84  2.73  2.71  2.73  2.85  3.42  2.81  2.78  2.73  3.47  3.07  ns  28.44  28.43  28.79  30.91  30.71  29.50  32.61  32.04  29.37  30.76  34.09  37.39  ns  6.97  6.57  7.15  8.70  6.65  7.18  7.16  7.65  6.70  6.70  7.09  8.28  ns  0.53  0.58  0.57  0.36  0.31  ns  23.94  23.93  23.37  25.58  25.42  ns  5.27  ns  8.26  ns  Total saturated* C16:l C18:ln9t  0.00  0.54  0.61  0.55  0.61  0.60  0.40  C18:ln9c  23.50  24.88  23.11  23.14  23.27  23.70  25.92  C18:ln7  3.92  3.43  4.50  4.38  4.02  3.97  4.29  4.62  4.50  4.73  4.92  C20:l  8.72  7.50  7.56  7.90  8.59  8.90  9.48  8.05  9.05  8.60  9.40  8.44  9.95  8.14  7.96  7.94  8.50  11.02  7.67  8.61  9.07  9.94  7.29  ns  52.93  54.41  52.72  53.97  51.95  53.67  59.98  53.66  54.12  53.78  58.82  56.06  ns  Total dienes*  1.13  0.47  0.37  0.27  0.31  0.35  0.57  0.26  0.34  0.42  0.40  0.51  ns  C18:2n6t  0.08  0.04  0.34  0.37  0.36  0.35  0.63  0.43  0.08  0.08  0.48  0.41  ns  C18:2n6c  3.86  3.87  3.92  3.36  3.86  3.62  2.71  3.78  3.87  3.66  3.41  3.19  ns  C20:4n6  0.05  0.06  1.19  0.00  0.45  0.44  0.06  0.23  0.44  0.41  0.04  0.00  ns  4.65  3.61  4.44  4.71  4.45  4.11  4.01  ns  C22:ln9 Total monoenes*  Total n-6 P U F A *  4.61  4.52  5.76  3.93  5.02  C18:3n3  0.63  0.64  0.79  0.57  0.63  0.65  0.69  0.57  0.67  0.64  0.70  0.58  ns  C20:5n3  6.35  c  6.26  c  6.04  c  5.94  c  b  1.22  5.06°  5.78°  5.48°  0.94  a  0.69  a  p<0.05  C22:5n3  0.93  c  0.94  0.91  c  0.52  b  0.71°  0.89°  0.82°  0.16  b  0.18  a  p<0.05  C22:6n3 Total n-3 H U F A *  4.65  c  4.14  4.05  c  0.61  b  2.95°  3.80°  3.41°  0.53  b  0.28  12.88°  11.83  3.23  b  9.61°  11.45°  10.59°  2.58  b  2.03  Total P U F A  17.49  16.48°  6.83"  14.05°  16.17°  15.04°  6.69  c  6.04°  5.41  c  0.92  c  0.74°  0.93  c  c  4.27  c  3.31  4.10  c  c  12.17  c  16.69  c  c  12.36°  10.91  c  12.01  18.12  14.85°  17.03  c  c  c  c  a  a  a  6.04  a  p<0.05 p<0.05 p<0.05  Table 3.6d. Relative fatty acid composition (% of total fatty acids) for diets 1 through 12 on day 126 (end) of storage trial. *Represents all fatty acids in this category; some are not shown. Means for a given parameter that share a common superscript letter were not significantly different (p<0.05). Values are means, ± standard deviation, of 3 replicate tanks for each diet. Not significant is represented by ns. 3 (IcE)  4 (ICE)  5 (ice)  6 (iCe)  7 (Ice)  8 (iCe)  9 (ice+)  10 (iCe+)  11 (Ice+)  12 (ICe+)  Sig.  C14:0  5.12  4.95  4.85  5.09  5.19  5.83  6.42  6.55  5.30  5.65  6.99  6.11  ns  C15:0  0.00  0.40  0.39  0.41  0.42  0.46  0.52  0.52  0.42  0.42  0.53  0.48  ns  17.89  17.71  23.61  20.70  ns  Fatty  Acid  1 (icE) 2 (iCE)  C16:0  17.21  16.46  16.32  17.09  17.25  16.73  21.13  21.82  C18:0  2.72  2.68  2.66  2.91  2.89  3.07  3.36  3.62  2.80  2.66  3.81  3.65  ns  26.36  26.16  26.66  27.52  28.29  27.32  33.18  34.16  28.88  29.55  36.47  32.50  ns  8.03  7.84  7.77  6.19  6.38  7.06  7.48  7.47  6.49  6.60  7.28  7.46  ns  C18:ln9t  0.49  0.48  0.45  0.47  0.52  0.57  0.35  0.33  0.59  0.76  0.29  0.02  ns  C18:ln9c  24.72  24.51  24.42  26.41  25.16  23.42  27.52  26.90  23.81  22.88  25.07  28.49  ns  3.95  4.41  4.71  3.68  5.16  4.92  3.79  ns  8.83  10.03  9.90  8.98  8.47  9.03  10.37  ns  Total saturated* C16:l  C18:ln7  3.17  3.01  2.96  2.75  3.44  C20:l  9.05  8.63  9.15  7.48  8.90  9.64  9.74  10.04  10.35  10.06  9.58  10.54  9.76  9.59  8.97  9.44  11.03  ns  56.56  55.74  56.33  55.25  56.09  55.13  61.91  59.59  54.03  53.72  57.39  61.64  ns  Total dienes*  1.13  1.89  1.75  0.41  0.35  1.56  0.28  0.31  0.45  0.90  0.15  0.34  ns  C18:2n6t  0.15  0.51  0.07  0.04  0.51  0.37  0.09  0.06  0.35  0.33  0.35  0.15  ns  C18:2n6c  3.45  3.39  3.26  3.66  3.15  3.56  2.43  3.24  4.42  3.84  3.24  3.15  ns  C22:ln9 Total monoenes*  C20:4n6  0.06  0.25  0.02  0.04  0.05  0.07  0.26  0.21  0.44  0.54  0.14  0.22  ns  Total n-6 P U F A *  4.17  4.66  3.87  4.55  4.01  4.46  2.87  3.52  5.48  5.11  3.73  3.59  ns  C18:3n3  0.62  0.61  0.13  C20:5n3  6.08  C22:5n3 C22:6n3 Total n-3 H U F A *  11.77  11.55  Total P U F A  15.95"  16.22"  b  0.93  b  3.94  b  b  0.62  0.72  5.96  b  5.87  b  6.22  0.91  b  0.97  b  3.87  b  3.75 b  b  11.40  b  15.27"  0.13 b  6.09  0.98  b  0.92  4.08  b  3.91  b  b  0.72  0.15  0.63  0.71  6.13  0.80  a  0.79  a  5.90  b  5.40"  0.97  b  a  0.14  0.12  a  0.90  b  4.03  b  4.11  b  12.27  11.26  16.82"  15.27"  b  0.15 b  b  a  0.42  0.45  a  11.53  a  1.76  2.42  a  15.98"  4.63  b  b  a  5.94  a  0.16  ns  0.72  a  1.08  p<0.01  0.80"  0.1 l  a  0.13  a  p<0.02  3.68"  0.40  a  0.35  a  11.16"  10.72"  2.25  a  1.93  p<0.01  16.65"  15.83"  5.99  a  5.52"  p<0.02  a  a  p<0.01  Table 3.7. Relative fatty acid composition (% of total fatty acids) of muscle tissue for fish fed diets 1 through 12. The data only represent day 126 of the growth trial. *Represents all fatty acids in this category; some are not shown. Means for a given parameter that share a common superscript letter were not significantly different (p<0.05). Values are means, ± standard deviation, of 3 replicate tanks for each diet. Not significant is represented by ns. 1-icE  2-iCE  3-IcE  4-ICE  5-ice  6-iCe  7-Ice  8-ICe  9-ice+  10-iCe+  ll-Ice+  12-ICe+  Signif.  C14:0  4.01  3.98  4.36  3.81  3.50  4.14  4.04  3.29  4.04  4.35  5.57  3.96  ns  C15:0  0.35  0.35  0.32  0.34  0.20  0.14  0.34  0.25  0.30  0.36  0.22  0.20  ns  C16:0  17.16  17.10  18.57  17.51  17.97  20.35  18.21  17.61  17.28  18.31  20.03  17.65  ns  C18:0  3.65  3.65  4.03  3.77  4.22  4.11  4.12  4.29  3.95  3.89  4.23  4.36  ns  27.61  ns  6.13  ns  Fatty  Acid  (Relative  %)  27.32  27.38  29.03  27.46  27.65  29.96  28.5  26.99  27.43  28.87  31.48  C16:l  5.63  5.69  5.62  5.38  4.93  6.43  5.88  5.52  5.48  5.13  7.42  C18:ln9t  0.34  0.38  0.23  0.16  21.31°  25.22  Total saturated*  C18:ln9c  2 3  ab  99  24.08  0.37 ab  0.38  24.78  b  22.98  ab  0.29 b  0.18  25.22  b  0.36  0.30  24.47  ab  23.05  ab  24.75  0.24 ab  29.06  ns  0.31 c  28.85  c  p<0.05  C18:ln7  4.43  4.51  4.25  4.38  4.32  4.46  3.55  4.40  3.85  4.43  4.82  4.83  ns  C20:l  7.82  7.75  6.88  7.06  6.02  5.59  7.28  6.52  7.06  7.37  6.13  7.10  ns  6.88  6.78  5.46  6.06  4.64  3.55  4.04  4.60  5.69  6.33  4.22  5.01  ns  49.84  49.95  47.96  46.89  42.13  45.61  46.77  46.08  45.98  49.02  52.25  52.68  ns  Total dienes*  0.60  0.54  0.36  0.45  0.25  0.13  0.29  0.23  0.44  0.44  0.24  0.18  ns  C18:2n6t  0.34  0.37  0.27  0.35  0.14  0.10  0.26  0.22  0.23  0.30  0.19  0.26  ns  C18:2n6c  3.60  3.41  3.47  3.24  3.12  3.34  3.52  3.54  3.77  3.33  3.35  3.69  ns  0.00  0.00  0.01  ns  C22:ln9 Total monoenes*  C20:4n6  0.02  0.01  0.04  0.01  0.06  0.00  0.00  0.01  0.03  Total n-6 P U F A *  4.29  4.16  4.04  3.92  3.66  3.86  4.44  4.69  4.28  4.02  3.73  4.54  ns  C18:3n3  0.58  0.59  0.44  0.46  0.36  0.30  0.55  0.39  0.35  0.48  0.29  0.42  C20:5n3  5.06  4.94  4.74  5.67  b  6.64  5.46"  4.62  5.22  b  5.62  5.13  2.60  2.94  ns p<0.05  C22:5n3  2.05  bc  2.50  1.41  1.88  abc  2.36  C22:6n3 Total n-3 H U F A * Total P U F A  22.24  b  bc  2.03  b  bc  1.78  b  abc  11.20  10.08  ab  10.26  ab  17.95  bc  17.97  bc  18.42  bc  22.13  bc  22.46  abc  bc  bc  2.07  abc  b  c  ab  16.44  13.08  21.04  c  26.32  c  24.96  c  29.97  c  1 2 59  c  abc  b  abc  11.14  abc  20.43  bc  20.00  bc  24.30  bc  24.44  bc  1.88  14.02  bc  21.73 26.42  c  bc  b  1.86  c  abc  1 3 43  21.87 26.16  c  bc  b  abc  10.06  ab  17.66 21.68  a  ab  p<0.05  a b  p<0.05  1.15"  1.36  8.21  9.8 l  a  a  bc  12.29"  14.99  ab  bc  16.02  19.53  ab  3  p<0.05 p<0.05  Table 3.8. Vitamin E (Vit. E) concentrations in the diets at the beginning (day 0) and end (day 126) of the storage trial. Vitamin E concentrations of the muscle and liver tissues of the fish fed diets 1 through 12 at the end of the 126-day growth trial are also shown. Values are means, ± standard deviation, of 3 replicate tanks for each diet. Means, within each column, with common superscript letters were not significantly different (p<0.05). Diet  Vit. E - Diet (day 0) (IU/kg)  Vit. E - Diet (day 126) (IU/kg)  Vit. E - Muscle (IU/kg)  Vit. E-Liver (IU/kg)  1 (icE) 2 (iCE) 3 (IcE) 4 (ICE) 5 (ice) 6 (iCe) 7 (Ice) 8 (ICe) 9 (ice+) 10(iCe+) 11 (Ice+) 12 (ICe+)  32.61 31.80 31.57 31.04 31.09 30.77 31.19 30.81 31.09 30.47 30.71 30.90  28.11 31.26 9.94 8.81  9.27 9.03 8.05 8.69 6.63  22.07 20.83 14.09 19.76 17.16 19.80 16.70 15.50 23.20 19.27 11.16 21.21  6  f ab  a  1  2  0 7  abc  14.10 13.97 13.73 12.90 16.08 14.41 13.79  cd  cd  cd  bcd  d  d  cd  f  5  9 4  ef  def  ef  cdef  bcde  3.87 1.62 6.51 5.4 3.38 1.83  abc  a  cdef  bcd  ab  a  77  4.0  GENERAL CONCLUSION  Factors influencing dietary lipid peroxidation, including the sources and concentrations of fish oil and fish meal, nutrient composition, improper storage conditions and excessive presence of pro-oxidants, can decrease the nutritional quality of the fed salmon, and ultimately lower its market value. Atlantic salmon are currently the main farmed finfish species on both coasts of Canada and the farming of this species contributes significantly to the economies of these areas. Feed represents the single largest operational expense of fish farming. The cost of feed can be influenced by fluctuations in ingredient price, availability and acceptability, pelleting and handling requirements, and levels of anti-nutritional factors. Proper maintenance of quality assurance programs can ensure nutritional requirements are met in farmed fish thereby minimizing operational production costs due to nutritional deficiencies or overloads.  This study has demonstrated that iron-catalyzed dietary lipid peroxidation influence the growth performance and nutritional quality of post-juvenile Atlantic salmon. High dietary iron concentration, in the absence of ethoxyquin, appeared to be the most influential factor in promoting dietary lipid peroxidation levels that could elicit a toxicological impact in the experimental fish; however, the level of iron itself was not found to be toxic under the conditions of the study. Lipid peroxidation can influence many components of a feed such as its digestibility, safety, and nutritional value associated with vitamins, proteins and fatty acid composition. Salmon growth depression was influenced by protein modification, n-3 fatty acid deficiency and/or vitamin E deficiency. Further, the data implies that the factors influenced by dietary lipid oxidation decreased the nutritional quality of the fed salmon. The findings of this 78  study do not show a direct relationship between consumption of oxidized dietary lipids and in vivo oxidative stress.  The experimental diets, fish and conditions in this study were aimed to reflect a controlled commercial operation. However, without compromising the original intention of a controlled commercial operation, changing factors such as the life history stage of the experimental fish, dietary oil source, and length of the salmon growth and feed storage trials likely would have yielded more dramatic results. Each life history stage of Atlantic salmon has different nutritional and physiological requirements that must be met. These requirements are important factors to consider when examining how diet influences experimental fish, especially if the chemical composition of the diet can change over time.  Crude, instead of refined, fish oil was used as the lipid source in the present study. Crude fish oil can contain substances, such as proteins and anti-oxidants, that can influence its oxidative stability. The stability of a lipid source is influenced by its fatty acid composition since n-3 H U F A are susceptible to lipid peroxidation. Moreover, the lipid to protein ratio present in oil may exert an influence since proteins can interfere with propagation reaction. Further, the types and concentrations of anti-oxidants can differ between oil sources. Therefore, i f a refined oil source had been used, free of proteins and other potentially interfering compounds, the nature and level of the peroxidation products generated may have produced a more dramatic effect.  The diet storage trial and the growth trial, which ran concurrently, lasted only 126 days. If this experiment had continued for another 126 days, it is expected that the dietary effects would have been more dramatic. In theory, if the fish had continued to be fed pro-oxidant catalyzed lipid peroxidation there would have been (1) an increase in mortalities due to the extreme deficiencies 79  of vitamin E and n-3 H U F A in diets 7 (Ice), 8 (ICe), 11 (Ice+) and 12 (ICe+), (2) an indication of in situ oxidative stress in fish fed diets 7 (Ice), 8 (ICe), 11 (Ice+) and 12 (ICe+) due to deficiencies in dietary anti-oxidants, and (3) the development of liver lipoid disease in fish fed diets 7 (Ice), 8 (ICe), 11 (Ice+) and 12 (ICe+) due to the absorption of lipid oxidation products produced in the diet.  As mentioned earlier, lipid peroxidation causes beneficial and detrimental changes in the biochemical, sensory and nutritional properties of food. Several studies have linked the consumption of oxidized dietary lipids to the development of disease. However, the current literature fails to address in depth how important factors such as bioavailability, uptake, distribution and mechanisms of how lipid peroxidation or its products can influence health.  In general, future research should (1) further examine the toxicokinetics of specific lipid peroxidation products, (2) be designed to produce a better understanding of the biochemistry surrounding in vivo anti-oxidant defense systems, (3) define the role and possible risk of oxidized dietary lipids to human and animal health, and (4) establish how nutritional and flesh quality parameters such as flavour, aroma, colour and texture are influenced by the consumption of oxidized dietary lipids in cultured salmon.  80  5.0  REFERENCES  Addis, P.B., 1986. Occurrence of lipid oxidation-products in foods. Food and Chemical Toxicology 24, 1021-1030. Andersen, F., Lygren, B., Maage, A., and Waagbo, R., 1998. Interactions between two dietary levels of iron and two forms of ascorbic acid and the effect on growth, anti-oxidant status and some non-specific immune parameters in Atlantic salmon (Salmo salar) smolts. Aquaculture 161, 437-451. Anderson, J.S., Higgs, D.A., Beames, R . M . , and Rowshandeli, M . , 1997. Fishmeal quality assessment for Atlantic salmon reared in sea water. Aquaculture Nutrition 3, 25-38. A O A C , 2000. Official Methods of Analysis. Association of the Official Analytical Chemists, Gaithersburg. Baker, R.T.M., and Davies, S.J., 1997. Modulation of tissue alpha-tocopherol in African catfish, Clarias gariepinus (Burchell), fed oxidised oils, and the compensatory effect of supplemental dietary vitamin E. Aquaculture Nutrition 3, 91-98. Balfry, S.K., Sutton, J.S., Higgs, D.A., Skura, B.J., 2004. Cellular and humoral immune response of post-smolt Atlantic salmon {Salmo salar) fed actively oxidizing dietary lipids. Unpublished data. Bell, J.G., McEvoy, J., Tocher, D.R., McGhee, F., Campbell, P.J., Sargent, J.R., 2001. Replacement offish oil with rapeseed oil in diets of Atlantic salmon (Salmo salar) affects tissue lipid compositions and hepatocyte fatty acid metabolism. Journal of Nutrition 131, 1535-1543. Bell, J.G., Tocher, D.R., Henderson, R.J., Dick, J.R., Crampton, V.O., 2003. Altered fatty acid compositions in Atlantic salmon (Salmo salar) fed diets containing linseed and rapeseed oils can be partially restored by a subsequent fish oil finishing diet. Journal of Nutrition 133,2793-2801. Benzie, I.F.F., 1996. Lipid peroxidation: A review of causes, consequences, measurement and dietary influences. International Journal of Food Sciences and Nutrition 47, 233-261. Berntssen, M.H.G., Hylland, K., Bonga, S.E.W., and Maage, A., 1999. Toxic levels of dietary copper in Atlantic salmon (Salmo salar L.) parr. Aquatic Toxicology 46, 87-99. Billek, G., 2000. Health aspects of thermoxidized oils and fats. European Journal of Lipid Science and Technology 102, 587-593. Bjerregaard, P., Jorgensen, M.E., Borch-Johnsen, K., 2004. Serum lipids of Greenland Inuit in relation to Inuit genetic heritage, westernization and migration. Atherosclerosis 174(2), 391-398. Bligh, E.G., and Dyer, W.J., 1959. A rapid method of total lipid extraction and purification. Canadian Journal of Biochemistry and Physiology 37, 911-917. Bors, W., Erben-Russ, M . , Michel, C , and Saran, M . , 1990. Radical mechanisms in fatty acid and lipid peroxidation. In: Craste de Paulet, A . (Ed.), Free radicals, lipoproteins and membrane lipids. Plenum Press, New York, pp. 1-16. Bowry, V.W., Ingold, K . U . , and Stacker, R., 1992. Vitamin E in human low density lipoprotein. When and how this anti-oxidant becomes a prooxidant. Biochemical Journal 288, 341344. 81  Boyd, E . M . , 1973. Toxicity of Pure Foods. CRC Press, Cleveland, pp. 71-111. Brown, A.J., and Jessup, W., 1999. Oxysterols and atherosclerosis. Atherosclerosis 142, 1-28. Bureau, D.P., Kaushik, S.J., and Cho, C.Y., 2002. Bioenergetics. In: Halver, J.E., Hardy, R.W. (Eds.), Fish Nutrition. Academic Press, New York, pp. 1-59. Bury, N.R., Walker, P.A., and Glover, C.N., 2003. Nutritive metal uptake in teleost fish. Journal of Experimental Biology 206, 11-23. Chaiyapechara, S., Casten, M.T., Hardy, R.W., and Dong, F . M . , 2003. Fish performance, fillet characteristics, and health assessment index of rainbow trout (Oncorhynchus mykiss) fed diets containing adequate and high concentrations of lipid and vitamin E. Aquaculture 219,715-738. Cheeseman, K . H . , 1990. Methods of measuring lipid peroxidation in biological systems: A n overview. In: Craste de Paulet, A. (Ed.), Free radicals, lipoproteins and membrane lipids. Plenum Press, New York, pp. 143-152. Cheeseman, K . H . , and Slater, T.F., 1993. A n introduction to free-radical biochemistry. British Medical Bulletin 49, 481-493. Chisolm, G . M . , and Steinberg, D., 2000. The oxidative modification hypothesis of atherogenesis: An overview. Free Radical Biology and Medicine 28, 1815-1826. Cho, C.Y., and Kaushik, S.J., 1985. Effects of protein intake on metabolizable and net energy values of fish diets'. In: Cowey, C.B., A . M . Mackie, and J.G. Bell (Ed.), Nutrition and feeding in fish. Academic Press, London, pp. 95-117. Clark, W.L., and Serbia, G.W., 1991. Safety aspects of frying fats and oils. Food Technology 45 (4): 84-86. Corongiu, F.P., and Banni, S., 1994. Detection of conjugated dienes by second derivative ultraviolet spectrophotometry. Methods in Enzymology 233, 303-310. Cowey, C.B., Adron, J.W., and Youngson, A., 1983. The vitamin E requirement of rainbow trout (Salmon gairdneri) given diets containing polyunsaturated fatty acids derived from fish oil. Aquaculture 20, 85-93. Dabrowski, K., and Guderley, H., 2002. Intermediary Metabolism. In: Halver, J.E., Hardy, R. (Eds.), Fish Nutrition. Academic Press, New York, pp. 309-365. de Koning, A.J., 2002. The anti-oxidant ethoxyquin and its analogues: A review. International Journal of Food Properties 5, 451-461. Desjardins, L . M . , Hicks, B.D., and Hilton, J.W., 1987. Iron catalyzed oxidation of trout diets and its effect on the growth and physiological response of rainbow trout. Fish Physiology and Biochemistry 3, 173-182. Din, J.N., Newby, D.E., Flapan, A.D., 2004. Omega 3 fatty acids and cardiovascular diseasefishing for a natural treatment. British Medical Journal 328, 30-35. Dobarganes, C , 1998. Formation and analysis of high molecular-weight compounds in frying fats and oils. Ocl-Oleagineux Corps Gras Lipides 5, 41-47. Dobarganes, C , and Marquez-Ruiz, G., 2003. Oxidized fats in foods. Current Opinion in Clinical Nutrition and Metabolic Care 6, 157-163. Dobarganes, M.C., and Velasco, J., 2002. Analysis of lipid hydroperoxides. European Journal of Lipid Science and Technology 104, 420-428. Dunkley, W.L., Franke, A . A . , and Low, E., 1968. Compounds in milk accompanying feeding of ethoxyquin. Journal of Dairy Science 51, 1215-1218. Eberhardt, M . K . , 2001. Reactive oxygen metabolites: Chemistry and medical consequences. C R C Press, New York, pp. 13-30, 33-98. Esterbauer, H., Schaur, R.J., and Zollner, H., 1991. Chemistry and biochemistry of 4hydroxynonenal, malondialdehyde and related aldehydes. Free Radical Biology and Medicine 11, 81-128. 82  F A O , 2002. Overview of fish production, utilization and trade; based on 2001data. ftp ://ftp. fao.org/fi/document/trends/overview/2001 /commodit/2001 fi sheryoverview.pdf. April 15,2004. Fenaille, F., Guy, P.A., and Tabet, J.C., 2003. Study of protein modification by 4-hydroxy-2nonenal and other short chain aldehydes analyzed by electrospray ionization tandem mass spectrometry. Journal of the American Society for Mass Spectrometry 14, 215-226. Firestone, D., Horwitz, W., Friedman, L., and Shue, G . M . , 1961. Heated Fats I. Studies of the effects of heating on the chemical nature of cottonseed oil. Journal of the American Oil Chemists Society 38, 253-257. Forster, I., 1987. Fat oxidation and vitamin E and growth, M.Sc. Thesis, Animal Science. University of British Columbia, Vancouver, pp. 100. Forster, I., Higgs, D.A., Bell, G.R., Dosanjh, B.S., and March, B.E., 1988. Effect of diets containing herring oil oxidized to different degrees on growth and immunocompetence of juvenile coho salmon (Oncorhynchus-kisutcli). Canadian Journal of Fisheries and Aquatic Sciences 45,2187-2194. Frankel, E.N., 1998. Lipid Oxidation. The Oily Press, Dundee, pp. 300. Fridovich, I., 1993. Superoxide and superoxide dimutases. Free Radical Biology and Medicine, 15(5): 472. Geode, R.W., and Barton, B.A., 1990. Organismic indices and an autopsy based assessment as inidcators of health and condition in fish. American Fisheries Society Syposium 8:93108. Gershon, D., 1988. Oxidative and peroxidative damage and its role in the aging process. In: Quintanilla, A . (Ed.), Reactive Oxygen Species in Chemistry, Biology and Medicine. Plenum, New York, pp. 211-219. Gill, N . , 2002. Effect of varying concentrations of partially dehulled and extruded sunflower meal on growth performance and sensory attributes of post-juvenile Atlantic salmon (Salmo salar), M.Sc. Thesis, Food Science. University of British Columbia, Vancouver, pp. 186. Gray, J.L., and Monahan, F.J., 1992. Measurement of lipid oxidation in meat and meat products. Trends in Food Science & Technology 3, 315-319. Gunstone, F., 1996. Fatty acid and lipid chemistry. Blackie Academic and Professional, London, pp.1-33. Gutteridge, J.M.C., and Halliwell, B., 1988. The anti-oxidant proteins of extracellular fluids. In: Chow, C.K. (Ed.), Cellular Anti-oxidant Defense Mechanisms. C R C Press, Inc., Boca Raton, pp. 1-23. Halliwell, B., and Gutteridge, J.M., 1984. Oxygen toxicity, oxygen radicals, transition metals and disease. Journal of Biochemistry 219, 1-4. Halliwell, B., and Gutteridge, J.M.C., 1989. Free radical biology and medicine. Clarendon Press, Oxford, pp. 3-85. Halliwell, B., and Gutteridge, J.M.C., 1990. Role of free radicals and catalytic metal ions in human disease: an overview. Methods in Enzymology 186, 1-85. Halliwell, B., and Chirico, S., 1993. Lipid-peroxidation - Its mechanism, measurement, and significance. American Journal of Clinical Nutrition 57, S715-S725. Halver, J.E., 2002. The Vitamins. In: Halver, J.E., Hardy, R.W. (Eds.), Fish Nutrition. Academic Press, New York, pp. 62-141. Hamre, K., Kolas, K., Sandnes, K., Julshamn, K., and Kiessling, A . , 2001. Feed intake and absorption of lipid oxidation products in Atlantic salmon (Salmo salar) fed diets coated with oxidised fish oil. Fish Physiology and Biochemistry 25, 209-219. 83  Hamre, K., Christiansen, R., Waagbo, R., Maage, A., Torstensen, B.E., Lygren, B., Lie, 0., Wathne, E., and Albrektsen, S., 2004. Anti-oxidant vitamins, minerals and lipid levels in diets for Atlantic salmon (Salmo salar, L.)\ effects on growth performance and fillet quality. Aquaculture Nutrition 10, 113-123. Hara, S., and Totani, Y . , 1988. A highly sensitive method for the microdetermination of lipid hydroperoxides by potentionmetry. Journal of the American Oil Chemists Society 65, 1948-1950. Hardy, R., and Roley, D.D., 2000. Lipid Oxidation and Anti-oxidants. In: Stickney, R.R. (Ed.), Encyclopedia of Aquaculture. John Wiley and Sons, Inc., New York, pp. 470-476. Hardy, R.W., and Barrows, F.T., 2002. Diet formulation and manufacture. In: Halver, J.E., Hardy, R.W. (Eds.), Fish Nutrition. Academic Press, London, pp. 505-600. Hardy, R.W., Higgs, D.A., Lall, S.P., and Tacon, A.G.J., 2001. Alternative dietary protein and lipid sources for sustainable production of salmonids, Fishen Og Havet NR. 8. Institute of Marine Research, pp. 53. Haumann, B.F., 1993. Health implications of lipid oxidation. International News on Fats, Oils, and Related Materials 4, 800-810. He, P., 1998. Anti-oxidants in fish meals, fish feeds and farmed fish, M.Sc. Thesis. Proquest Digital Dissertations, Halifax, Nova Scotia. He, P., and Ackman, R.G., 2000. Residues of ethoxyquin and ethoxyquin dimer in ocean-farmed salmonids determined by high-pressure liquid chromatography. Journal of Food Science 65, 1312-1314. Helland, S., Storebakken, T., and Grisdale-Helland, B., 1991. Atlantic Salmon, Salmo salar. In: Wilson, R.P. (Ed.), Handbook of Nutrient Requirements of Finfish. C R C Press, London, pp.13-22. Henderson, R.J., and Tocher, D.R., 1987. The lipid composition and biochemistry of fresh-water fish. Progress in Lipid Research 26, 281-347. Hessler, J.R., Morel, D.W., Lewis, L.J., and Chisholm, G . M . , 1983. Lipoprotein oxidation and lipoprotein-induced cytotoxicity. Arteriosclerosis 3, 215-222. Higgs, D.A., and Dong, F . M . , 2000. Lipids and fatty acids. In: Stickney, R. (Ed.), Encyclopedia of Aquaculture. Wiley-Interscience, London, pp. 476-496. Higgs, D.A., Markert, J.R., MacQuarrie, D.W., McBride, J.W., Dosanjh, B.S., Nichols, C , and Hoskins, G., 1979. Development of practical dry diets for coho salmon, Oncorhynchus kisutch, using poultry-by-product meal, feather meal, soybean meal and rapeseed meal as major protein sources. In: Halver, J.E., Tiews, K . (Eds.), Finfish nutrition and fishfeed technology. Heenemann Verlagsgesellschaft mbh, Berlin, pp. 191-218. Higgs, D.A., Dosanjh, B.S., Prendergast, A.F., Beames, R . M . , Hardy, R.W., Riley, W., and Deacon, G., 1995. Use of rapeseed/Canola protein products in finfish diets'. In: Lim, C , D. Sessa (Ed.), Nutrition and utilization technology in Aquaculture. AOCS Press, Champaign, IL, pp. 130-156. Hung, S.S.O., Cho, C.Y., and Slinger, S.J., 1981. Effect of Oxidized Fish Oil, Dl-AlphaTocopheryl Acetate and Ethoxyquin Supplementation on the Vitamin-E Nutrition of Rainbow-Trout (Salmo-Gairdneri) Fed Practical Diets. Journal of Nutrition 111, 648-657. Itakura, K., Furuhata, A., Shibata, N . , Kobayashi, M . , and Uchida, K., 2003. Maillard reactionlike lysine modification by a lipid peroxidation product: immunochemical detection of protein-bound 2-hydroxyheptanal in vivo. Biochemical and Biophysical Research Communications 308, 452-457. Izquierdo, M.S., Obach, A., Arantzamendi, L., Montero, D., Robaina, L., Rosenlund, G., 2003. Dietary lipid sources for seabream and seabass: growth performance, tissue composition and flesh quality. Aquaculture Nutrition 9, 397-407. 84  Kanazawa, A., Sawa, T., Akaike, T., and Maeda, H., 2002. Dietary lipid peroxidation products and D N A damage in colon carcinogenesis. European Journal of Lipid Science and Technology 104, 439-447. Kanazawa, K . , and Ashida, H., 1998a. Catabolic fate of dietary trilinoleoylglycerol hydroperoxides in rat gastrointestines. Biochimica et Biophysica Acta-Lipids and Lipid Metabolism 1393, 336-348. Kanazawa, K., and Ashida, H . , 1998b. Dietary hydroperoxides of linoleic acid decompose to aldehydes in stomach before being absorbed into the body. Biochimica et Biophysica Acta-Lipids and Lipid Metabolism 1393, 349-361. Kanner, J., and Rosenthal, I., 1992. A n assessment of lipid oxidation in foods - Technical Report. Pure and Applied Chemistry 64, 1959-1964. Kanner, J., and Lapidot, T., 2001. The stomach as a bioreactor: Dietary lipid peroxidation in the gastric fluid and the effects of plant-derived anti-oxidants. Free Radical Biology and Medicine 31, 1388-1395. Kidd, P . M . , 1997. Glutathione: Systemic protectants against oxidative and free radical damage. Alternative Medicine Review 2, 155-176. Koshio, S., Ackman, R.G., and Lall, S.P., 1994. Effects of oxidized herring and canola oils in diets on growth, survival and flavour of Atlantic salmon, Salmo salar. Journal of Agricultural and Food Chemistry 42, 1164-1169. Kubow, S., 1992. Routes of formation and toxic consequences of lipid oxidation-products in foods. Free Radical Biology and Medicine 12, 63-81. Lall, S., 2002. The minerals. In: Halver, J.E., Hardy, R.W. (Eds.), Fish Nutrition. Academic Press, Amsterdam, pp. 260-308. Lang, K . , Billek, G., Fuhr, J., Henschel, J., von Jan, E., Kracht, J., Scharmann, H., and Waibel, H.-J., 1978. Ernahrungsphysiolgische Eigneschaften von Fritierfetten. Z. Ernahrungswiss Supplement 21, 1-61. Leaver, J., Law, A.J.R., Brechany, E.Y., and McCrae, C.H., 1999. Chemical changes in betalactoglobulin structure during ageing of protein-stabilized emulsions. International Journal of Food Science and Technology 34, 503-508. Lemon, D.W., 1975. A n improved T B A test for rancidity, New Series Circular no. 51. Department of Fisheries and Oceans, Halifax, pp. 1-4. Lezerovich, A., 1985. Determination of peroxide value by conventional difference and difference-derivative spectrophotometry. Journal of the American Oil Chemists Society 62, 1495-1500. L i , S.J., and King, A.J., 1999. Structural changes of rabbit myosin subfragment 1 altered by malondialdehyde, a byproduct of lipid oxidation. Journal of Agricultural and Food Chemistry 47, 3124-3129. Liu, J.F., Lee, Y . W . , and Chang, F.C., 2000. Effect of oxidized frying oil and vitamin C levels on the hepatic xenobiotic-metabolizing enzyme system of guinea pigs. Journal of Nutritional Science and Vitaminology 46, 137-140. Lovell, T., 1998. Nutrition and feeding of fish. Kluwer Academic Press, Boston, 267 pp. Luquet, P., and Watanabe, T., 1986. Interaction "nutrition-reproduction" in fish. Fish Physiology and Biochemistry 2, 121-129. Lygren, B., Hamre, K., and Waagbo, R., 1999. Effects of dietary pro- and anti-oxidants on some protective mechanisms and health parameters in Atlantic Salmon. Journal of Aquatic Animal Health 11,211-221. Makoto, M . , Nakamura, H., Diplock, A.T., and Kayden, H.J., 1993. Vitamin E: Its usefulness in health and in curing diseases. Japan Scientific Societies Press, Tokyo, pp. 3-12. 85  Mamett, L.J., 1987. The involvement of peroxyl free radicals in tumor initiation and promotion. In: Cerutti, P.A., Nygaard, O.F., Simic, M . G . (Eds.), Anticarcinogenesis and radiation protection. Plenum Press, New York, pp. 71-80. Martin-Polvillo, M . , Marquez-Ruiz, G., Jorge, N . , Ruiz-Mendez, M . V . , and Dobarganes, M.C., 1996. Evolution of oxidation during storage of crisps and french fries prepared with sunflower oil and high oleic sunflower oil. Grasas Y Aceites 47, 54-58. Meister, A., 1976. Glutathione metabolism and transport. In: Nygaard, O.F., Simic, M . G . (Eds.), Radioprotectors and anticarcinogens. Academic Press, New York, pp. 121-151. Meyer, W., 1978. Letter to the editor. Journal of the American Oil Chemists Society 55, 503. Milner, J.A., 2002. Strategies for cancer prevention: the role of diet. British Journal of Nutrition 87, S265-S272. Mourente, G., Diaz-Salvago, E., Bell, J.G., and Tocher, D.R., 2002. Increased activities of hepatic anti-oxidant defence enzymes in juvenile gilthead sea bream (Spams aurata L.) fed dietary oxidised oil: attenuation by dietary vitamin E. Aquaculture 214, 343-361. Mukai, K., 1993. Synthesis and kinetic study of anti-oxidant and prooxidant actions of vitamin E derivatives. In: Packer, L., Fuchs, J. (Eds.), Vitamin E in Health and Disease. Marcel Dekker Inc., New York, pp. 97-119. Nakamura, T., Tanaka, R., Higo, Y . , Taira, K., and Takeda, T., 1998. Lipid peroxide levels in tissues of live fish. Fisheries Science 64, 617-620. Naylor, R.L., Goldburg, R.J., Primavera, J.H., Kautsky, N . , Beveridge, M . C . M . , Clay, J., Folke, C , Lubchenco, J., Mooney, H., and Troell, M . , 2000. Effect of aquaculture on world fish supplies. Nature 405, 1017-1024. N R C , 1993. Nutrient Requirements of Fish. National Academy Press, Washington, DC, pp. 1-15. Opstvedt, J., and Aksnes, A., Hope, B., Pike, I.H., 2003. Efficiency of feed utilization in Atlantic salmon (Salmo salar L.) fed diets with increasing substitution of fish meal with vegetable proteins. Aquaculture 221, 365-379. Ouzounidou G, Ilias I, Tranopoulou H , Karataglis S., 1998. Amelioration of copper toxicity by iron on spinach physiology. Journal of Plant Nutrition 21(10),2089-2101. Ozben, T., 1998. Free radicals, oxidative stress and anti-oxidants: Pathological significance, Series A : Life Sciences. Plenum Press, New York and London, pp. 1-14. Packer, L., and Fuchs, J., 1993. Vitamin E in Health and Disease. Marcel Dekker Inc., New York, pp. 85-95. Pauly, D., Alder, J., Bennett, E., Christensen, V., Tyedmers, P., and Watson, R., 2003. The future for fisheries. Science 302, 1359-1361. Perjesi, P., Pinter, Z., Gyongyi, Z., and Ember, I., 2002. Effect of rancid corn oil on some onco/suppressor gene expressions in vivo. A short-term study. Anticancer Research 22, 225-230. Piriou, A., Tallineau, C , Chahboun, S., Pontcharraud, R., and Guillard, O., 1987. Copperinduced lipid-peroxidation and hemolysis in whole-blood - evidence for a lack of correlation. Toxicology 47, 351-361. Porter, N.A., Caldwell, S.E., and Mills, K . A . , 1995. Mechanisms of free-radical oxidation of unsaturated lipids. Lipids 30, 277-290. Rasmussen, R.S. 2001. Quality of farmed salmonids with emphasis on proximate composition, yield and sensory characteristics. Aquaculture Research, 32 (10): 767-786. Riedemann, N.C., Ward, P. A., 2002. Oxidized lipid protects against sepsis. Nature Medicine 8, 1084-1085. Rivers, J.P.W., Sinclair, A.J., and Crawford, M.J., 1975. Inability of the cat to desaturate essential fatty acids. Nature 285, 171. 86  Robards, K., Kerr, A.F., and Patsalides, E., 1988. Rancidity and its measurement in edible oils and snack foods. A review. Analyst 113, 213-224. Roffo, A . H . , 1939. Krebserzeugende Wirkung des aus Cholesterin geowonnenen Phenanthrenderivate. Z. Kerbsforch 49, 341-347. Rust, M . B . , 2002. Nutritional physiology, Fish Nutrition. Academic Press, London, pp. 367-452. Sargent, J., Bell, G., McEvoy, L., Tocher, D., and Estevez, A., 1999. Recent developments in the essential fatty acid nutrition of fish. Aquaculture 177, 191-199. Sargent, J.R., Tocher, D.R., and Bell, J.G., 2002. The Lipids. In: Halver, J.E., Hardy, R.W. (Eds.), Fish Nutrition. Academic Press, Amsterdam, pp. 181-257. Schaefer, E.J., and Brousseau, M.E., 1998. Diet, lipoproteins, and coronary heart disease. Endocrinology and Metabolism Clinics of North America 27, 711-732. Schaich, K . M . , 1992. Metals and lipid oxidation. Contemporary issues. Lipids 27, 209-218. Sergent O., Morel I., Cillard J., 1999 Involvement of metal ions in lipid peroxidation: Biological implications. In: Sigel A. and Sigel, H. Metal ions in biological systems. Marcel Dekker Inc., New York-Basel, pp. 251-277. Shahidi, F., and Wanasundara, U.N., 2002. Methods for measuring oxidative rancidity in fats and oils. In: Akoh, C.C., Minh, D.B. (Eds.), Food Lipids. Marcel Deker, Inc., New York, pp. 465-487. Shue, G . M . , Douglas, C D . , Firestone, D., Friedman, L., and Sage, J.S., 1968. Acute physiological effects of feeding rats non-urea adducting fatty acids (urea filtrate). Journal of Nutrition 94, 171-177. Sigma, 2003. Glutathione Peroxidase Cellular Assay Kit, pp. 1-3. Skaare, J.U., and Nafstad, I., 1979. The distribution of C-ethoxyquin in rat. Acta Pharmacologics et Toxicologica 44, 303-307. Staprans, I., Hardman, D.A., Pan, X . M . , and Feingold, K.R., 1999. Effect of oxidized lipids in the diet on oxidized lipid levels in postprandial serum chylomicrons of diabetic patients. Diabetes Care 22, 300-306. Staprans, I., Pan, X . M . , Rapp, J.H., and Feingold, K.R., 2003. Oxidized cholesterol in the diet is a source of oxidized lipoproteins in human serum. Journal of Lipid Research 44, 705-715. Staprans, I., Rapp, J.H., Pan, X . M . , Hardman, D.A., and Feingold, K.R., 1996. Oxidized lipids in the diet accelerate the development of fatty streaks in cholesterol-fed rabbits. Arteriosclerosis Thrombosis and Vascular Biology 16, 533-538. Staprans, I., Pan, X . M . , Rapp, J.H., Grunfeld, C , and Feingold, K.R., 2000. Oxidized cholesterol in the diet accelerates the development of atherosclerosis in L D L receptor- and apolipoprotein E-deficient mice. Arteriosclerosis Thrombosis and Vascular Biology 20, 708-714. Stokstad, E., 2004. Salmon survey stokes debate about farmed fish. Science 303, 154-155. Tacon, A.G.J., 1996. Lipid nutritional pathology in farmed fish. Archives of Animal NutritionArchiv Fur Tierernahrung 49, 33-39. Takeuchi, T., Watanabe, T., and Nose, T., 1979. Requirement for essential fatty acids of chum salmon (Oncorhynchus keta) in freshwater environment. Bulletin of the Japanese Society of Scientific Fisheries 45, 1319-1323. Tangney, C.C., McNair, H . M . , and Driskell, J. A., 1981. Quantitation of Individual Tocopherols in Plasma, Platelets, Lipids, and Livers by High-Performance Liquid-Chromatography. Journal of Chromatography 224, 389-397. Tarladgis, B.G., Watts, B . M . , and Younathan, M.T., 1960. A distillation method for the quantitative determination of malondialdehyde in rancid foods. Journal of the American Oil Chemists Society 37, 44-48. 14  87  ter Meulen, U . , Ende, M . , Hunneman, D.H., Remberg, G., and Walker, R., 1980. Metabolic studies on the anti-oxidant ethoxyquin. Journal of Animal Physiology 43, 164-170. Toborek, M . , Lee, Y.W., Garrido, R., Kaiser, S., and Hennig, B., 2002. Unsaturated fatty acids selectively induce an inflammatory environment in human endothelial cells. American Journal of Clinical Nutrition 75, 119-125. Tocher, D.R., Mourente, G., Van der Eecken, A., Evjemo, J.O., Diaz, E., Wille, M . , Bell, J.G., and Olsen, Y . , 2003. Comparative study of anti-oxidant defence mechanisms in marine fish fed variable levels of oxidised oil and vitamin E. Aquaculture International 11, 195216. Turek, J.J., Watkins, B.A., Schoenlein, I.A., Allen, K.G.D., Hayek, M . G . , and Aldrich, C.G., 2003. Oxidized lipid depresses canine growth, immune function, and bone formation. Journal of Nutritional Biochemistry 14, 24-31. Ursini, F., Zamburlini, A., Cazzolato, G., Maiorino, M . , Bon, G.B., and Sevanian, A., 1998. Postprandial plasma lipid hydroperoxides: A possible link between diet and atherosclerosis. Free Radical Biology and Medicine 25, 250-252. US EPA, 1994. E P A Method 200.15, Determination of metals and trace elements in water by ultrasonic nebulization inductively coupled plasma-atomic emission spectrometry. Methods for the determination of metals in environmental samples - Supplement I, Rev 1.2. EPA/600/R-94-111. Vangen, B., and Hemre, G.I., 2003. Dietary carbohydrate, iron and zinc interactions in Atlantic salmon (Salmo salar). Aquaculture 219, 597-611. Velasco, J., and Dobarganes, C , 2002. Oxidative stability of virgin olive oil. European Journal of Lipid Science and Technology 104, 661-676. Venugopal, B., and Luckey, T.D., 1978. Metal toxicity in mammals. Plenum Press, New York, pp. 24-32. Verleyen, T., Kamal-Eldin, A., Dobarganes, C , Verhe, R., Dewettinck, K., and Huyghebaert, A., 2001. Modeling of alpha-tocopherol loss and oxidation products formed during thermoxidation in triolein and tripalmitin mixtures. Lipids 36, 719-726. Voet, D., Voet, J.G., Pratt, C.W., 2002. Fundamentals of biochemistry. Wiley, New York, pp. 149-153. Waagbo, R., 1997. The impact of nutritional factors on the immune system in Atlantic salmon, Salmon salar L.: A review. Aquatic Fisheries Management 25, 175-197. Watanabe, T., 1982. Lipid nutrition in fish. Comparative Biochemistry and Physiology 73B, 315. Wendel, A., 1980. Enzymatic basis of detoxification. Academic Press, New York, pp. 333-353. White, A.R., Barnham, K.J., Huang, X . , Voltakis, I., Beyreuther, K., Masters, C.L., Cherny, R.A., Bush, A.I., and Cappai, R., 2004. Iron inhibits neurotoxicity induced by trace copper and biological reductants. Journal of Biological and Inorganic Chemistry 9, 269280. Williams, M.J.A., Sutherland, W.H.F., McCormick, M.P., de Jong, S.A., Walker, R.J., and Wilkins, G.T., 1999. Impaired endothelial function following a meal rich in used cooking fat. Journal of the American College of Cardiology 33, 1050-1055. Wilson, R.H., Thomas, J.O., Thompson, C.R., Launer, H.F., and Kohler, G.O., 1981. Absorption, metabolism, excretion of the anti-oxidant, 6-ethoxyl,2-dihydro-2,2,4trimethylaquinoline. Agriculture and Food Chemistry 7, 206-209. Wilson, R.P., 2002. Amino acids and proteins. In: Halver, J.E., Hardy, R.W. (Eds.), Fish Nutrition. Academic Press, Amsterdam, pp. 144-179. Yang, G., 1992. Detection of lipid hydroperoxides by high performance liquid chromatography coupled with post-column reaction. Trends in Food Science and Technology 3, 15-18. 88  Yildiz, G., Wehling, R.L., and Cuppett, S.L., 2003. Comparison of four analytical methods for the determination of peroxide value in oxidized soybean oils. American Oil Chemists Society 80, 103-107. Zar, J.H., 1996. Biostatistical analysis. Prentice-Hall, Englewood Cliffs, NJ., pp. 43-45.  89  Appendix 1. Data represent total mortalities for the fish fed diets 1-12 at each sampling interval and for the entire 126-day growth trial. *Mortalities due to technical problem with air line to tank. DT-1 Interval  -  2  2  1  1  DT-5  DT-6  DT-7  DT-8  DT-9  DT-10  DT-11  DT-12  Signif.  2  -—  2  -—  ••—  -----  1  -—  ns  —  2  1  -—  -—  4  -—  4  ns  3 - Day 84 to 126  Mortality  Entire  DT-4  2 - Day 42 to 84  Mortality Interval  DT-3  1 - Day 0 to 42  Mortality Interval  DT-2  Trial  Mortality  — -  -—  -—  1  2  -----  3  2  2  4  -  -  11*  -—  ns  -—  16  -—  ns  - Day 0 to 126  -—  1  Appendix 2. In situ measures of oxidative stress. Values are means, ± standard deviation, of 3 replicate tanks for each diet.  Diet  Conjugated Diene - Muscle (ug/kg)  Se-Gpox - Muscle (umol/min/ml)  Se-GPox - Liver (umol/min/ml)  1 2 3 4 5 6 7 8 9 10 11 12 Signif.*  54.50 58.33 55.83 66.44 51.50 56.44 65.83 51.11 57.72 62.94 57.28 66.06 ns  24.44 33.56 36.00 38.44  28.28 27.56 24.61 41.72  35.44 44.56  37.83 43.61  38.89 40.67 ns  35.17 53.22 ns  91  Appendix 3. General health observations recorded throughout the 126-day growth trial. Symbols represent; An-anemia, Fh-fin hemorrhage, Gp-pale gills, Lh-liver hemorrhage, Ly-yellow liver, Se-enlarged spleen, Sb-black spleen, Od- uncharacteristic odour and (#)-# fish >1. Day  Diet 1  Diet 2  43  Diet 3  Diet 4  Diet 5  Diet 6  Fh  48 51 83 Fh, Lh  84 85 86 93  Fh, S b  101 101  Fh, Lh  Fh, Lh  124 125 126  Day  Diet 7  43  Fh  48  An  51  Gp, Fh, L h , Ly, S e  83  An, Fh  Diet 8  Diet 9  Diet 10  Diet 11  Diet 12  G p , F h , L h , L y (5)  Fh, G p  Ly A n , L y ( 2 ) ; A n (3)  84 85  An,  Fh, Lh, Ly (2)  Lh, Ly, O d (2)  Fh F h , L h , L y ( 2 ) ; F h , L h (3)  86 93  An,  101  L h , Ly (9), L h , Ly (2)  101 124 125 126  An A n (2) An, Fh (5),  An ( 4 )  An An  Fh  A n (5)  An  (2)  

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.831.1-0091636/manifest

Comment

Related Items