Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The effect of phytic acid removal on the acceptability of canola and rapeseed feedstuffs as partial and… Prendergast, Angela Fay 1999

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

Item Metadata

Download

Media
831-ubc_1999-0311.pdf [ 11.02MB ]
Metadata
JSON: 831-1.0089028.json
JSON-LD: 831-1.0089028-ld.json
RDF/XML (Pretty): 831-1.0089028-rdf.xml
RDF/JSON: 831-1.0089028-rdf.json
Turtle: 831-1.0089028-turtle.txt
N-Triples: 831-1.0089028-rdf-ntriples.txt
Original Record: 831-1.0089028-source.json
Full Text
831-1.0089028-fulltext.txt
Citation
831-1.0089028.ris

Full Text

THE EFFECT OF PHYTIC ACID REMOVAL ON THE ACCEPTABILITY OF CANOLA AND RAPESEED FEEDSTUFFS AS PARTIAL AND COMPLETE REPLACEMENTS FOR FISH MEAL IN SALMONID DIETS by ANGELA FAY PRENDERGAST B.Sc, The University of Alberta, 1984 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE FACULTY OF GRADUATE STUDIES Department of Animal Science We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA January, 1999 © Angela Fay Prendergast, 1999 I n p r e s e n t i n g t h i s t h e s i s i n p a r t i a l f u l f i l m e n t o f t h e r e q u i r e m e n t s f o r a n a d v a n c e d d e g r e e at t h e U n i v e r s i t y o f B r i t i s h 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 sha l l m a k e i t 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 . 1 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 t h i s 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 t h i s t h e s i s f o r f i n a n c i a l g a i n sha 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 . D e p a r t m e n t o f /~Li/>'7#/> Scj£7UTf~ T h e U n i v e r s i t y o f B r i t i s h C o l u m b i a V a n c o u v e r , C a n a d a D E - 6 ( 2 / 8 8 ) Abstract Two experiments were conducted with juvenile rainbow trout {Oncorhynchus mykiss) (< 5g) in ambient well water, to investigate the influence of phytic acid removal on the feeding value of rapeseed and canola protein feedstuffs used as substitutes for fish meal and other animal protein in the diets. The first investigation examined whether phytic acid removal by commercial pretreatment of rapeseed protein concentrate (RPC) with microbial phytase, with /without essential amino acid supplementation or cation-anion balancing to mimic the chemical profile of a control diet based on fish meal (which provided 59% of the dietary protein), could improve the nutritional value of RPC for trout sufficiently to allow it to replace all of the fish meal without compromising performance. Results indicated that commercial dephytinization significantly enhanced the nutritional value of RPC to the extent that, with the addition of FINNSTIM™ (1.5 % DMB; palatability enhancer), it could replace all of the fish meal without compromising performance. Essential amino acid supplementation and cation-anion balancing provided no additional nutritional benefit. The second investigation, examined whether commercial dephytinization or dephytinization by oral administration of a low dose (1,000 PU/ g dry diet) or high dose (4,000 PTJ/g dry diet) of microbial phytase to diets in which fish meal provided 70% of the protein and canola meal (CM) provided 20%, or animal protein-free diets in which RPC provided 95% of the protein, could enhance the nutritional value of the diets for trout. Also, the effect of fibre reduction of CM with or without the addition of a high dose of phytase was examined. Findings indicated that in diets containing 1.5% FINNSTIM™ , commercial dephytinization as well as direct application of either dose of phytase significantly improved the nutritional value of C M for trout. By contrast, direct application of either dose of phytase did not improve the nutritional value of RPC for trout. Although commercial dephytinization did improve RPC nutritional value, the enhancement was not sufficient to make the diets nutritionally equal to the control fish meal diet. Fibre reduction with or without direct phytase addition did not nutritionally improve the C M diets to levels comparable to the control diets. iii TABLE OF CONTENTS Abstract ii Table of Contents iii List of Tables vii List of Figures x Acknowledgements xi Preface xii CHAPTER ONE 1.0 General introduction 1 CHAPTER TWO 2.0 Literature Review 7 2.1 Rapeseed and Canola 7 2.1.1 Taxonomy of rapeseed and canola 8 2.1.2 From rapeseed to canola; development of a crop 8 2.1.3 Classification of oilseed protein products 11 2.1.4 Processing of rapeseed and canola protein products 12 2.1.4.1 Oil removal and meal production 13 2.1.4.2 Upgraded meals/flours 13 2.1.4.3 Concentrates 1 6 2.1.4.4 Isolates 1 7 2.1.5 Nutritional properties of rapeseed and canola protein products 20 2.1.5.1 Protein and essential amino acids 2 2 2.1.5.2 Carbohydrates 2 4 2.1.5.3 Energy 2 7 2.1.5.4 Minerals 2 8 2.1.5.5 Vitamins 2 9 2.1.5.6 Lipids 3 0 2.1.5.7 Antinutritional factors 30 2.1.5.7.1 ErucicAcid 30 2.1.5.7.2 Glucosinolates 31 2.1.5.7.3 Saponins 36 2.1.5.7.4 Phenolics 36 5.1.5.7.5 Lectins 39 2.1.5.7.6 Phytic acid 40 2.1.5.7.6.1 Phytic acid : structure, properties, occurrence and function 40 I V 2.1.5.7.6.2 Phytic acid interactions 42 2.1.5.7.6.2.1 Phytic acid and minerals 43 2.1.5.7.6.2.1.1 Phytic acid and phosphorus 45 2.1.5.7.6.2.1.2 Phytic acid and zinc 46 2.1.5.7.6.2.1.3 Phytic acid and calcium 47 2.1.5.7.6.2.1.4 Phytic acid and iron 48 2.1.5.7.6.2.2 Phytic acid and protein 49 2.1.5.7.6.2.3 Phytic acid and starch 51 2.1.6 Reduction and removal of phytic acid 52 2.1.6.1 Mechanical means of phytic acid removal 52 2.1.6.2 Germination 52 2.1.6.3 Water extraction and differential solubility 54 2.1.6.4 Membrane separation 55 2.1.6.5 Ion exchange processing 56 2.1.6.6 Gene transfer 56 2.1.6.7 Enzyme treatment 57 2.1.7 Phytase: properties, occurrence and function 57 2.1.7.1 Use of extrinsic sources of phytase for phytic acid removal 60 2.1.7.2 Use of intrinsic dietary sources of phytase for phytic acid removal 60 2.1.7.3 Effects of dietary phytase addition: swine 63 2.1.7.4 Effects of dietary phytase addition: poultry 64 2.1.7.5 Effects of dietary phytase addition: fish (salmonids) 64 2.2 The role of chemoreception, chemoattractants and stimulants in diet utilization 65 CHAPTER THREE 3.0 General materials and methods 70 3.1 Culture facility and experimental tanks 70 3.2 Fish grading and distribution 71 3.3 Sampling procedures o 71 3.4 Proximate and energy analyses 72 3.5 Plasma samples 73 3.6 Liver samples 73 3.7 Determination of whole-body mineral composition 73 3.8 Preparation of feedstuffs for diet incorporation 73 3.9 Vitamin supplements 75 3.10 Mineral supplements 75 V 3.11 Amino acid supplements 76 3.12 Preparation of mash and pellets 76 3.13 Diet allocation 77 3.14 Feeding protocol 77 3.15 General maintenance 78 CHAPTER FOUR 4.0 EXPERIMENT I. Acceptability of dephytinized and undephytinized rapeseed protein concentrate as complete replacements for fish meal in diets formulated for rainbow trout (Oncorhynchus mykiss). 79 4.1 Abstract. 79 4.2 Introduction 81 4.3 Materials and methods 85 4.3.1 Fish and experimental conditions 85 4.3.2 Diet preparation and allocation 86 4.3.3 Commercial dephytinization 91 4.3.4 Feeding protocol 91 4.3.5 Sampling 9 1 4.3.6 Chemical analyses 1°° 4.3.7 Data calculation and analyses 1 0 0 4.3.8 Statistical analyses 1 0 3 4.4 Results 103 4.4.1 Assessed composition of experimental diets 114 4.4.2 Influence of diet treatment on fish growth, appetite, feed utilization and protein utilization 115 4.4.3 Influence of diet treatment on body composition 115 4.4.4 Influence of diet treatment on thyroid and liver status 116 4.4.5 Influence of diet treatment on general health and behaviour 117 4.5 Discussion 117 4.5.1 Fish growth and physiological response to dietary treatments 117 4.5.2 Influence of diet treatment on body composition 133 4.5.3 Influence of diet treatment on general health and behaviour 134 4.6 Conclusion 134 4.7 Acknowledgements 135 VI CHAPTER FIVE 5.0 EXPERIMENT II. Performance of rainbow trout (Oncorhynchus mykiss) fed diets in which undephytinized and dephytinized rapeseed protein concentrate were used to replace all animal protein and dephytinized and undephytinized commercial canola meal and experimental low-fibre, high protein canola meal were used to partially replace fish meal. 137 5.1 Abstract 1 3 7 5.2 Introduction 139 5.3 Materials and methods 142 5.3.1 Fish and experimental conditions 142 5.3.2 Diet formulation, composition and preparation 151 5.3.3 Commercial dephytinization: phytase pretreatment 151 5.3.4 Feeding protocol 152 5.3.5 Faeces collection 152 5.3.6 Fish weighing and sampling 152 5.3.7 Chemical analyses 153 5.3.8 Response criteria: data calculation and analysis 154 5.3.9 Statistical Analyses 154 5.4 Results 155 5.4.1 Assessed composition of experimental diets 155 5.4.2 Influence of diet treatment on fish growth performance, feed digestibility and utilization 160 5.4.3 Influence of diet treatment on body composition 172 5.4.4 Influence of diet treatment on thyroid and liver status 173 5.4.5 Influence of diet treatment on general health and behaviour 174 5.5 Discussion 182 5.5.1 Assessed composition of diets 182 5.5.2 Influence of diet treatment on fish growth and feed digestibility and utilization 183 5.5.3 Influence of diet treatment on general fish health and behaviour 189 5.6 Conclusion 1 8 9 5.7 Acknowledgements I 9 2 CHAPTER SIX 6.0 Summary, conclusions and recommendations I 9 3 BIBLIOGRAPHY 1 9 9 List of Tables Table 2.1: Phytase: selected sources and associated activators, inhibitors, and temperature and pH optima 61 Table 2.2: Associated taste types of arnino acids known to act as chemoattractants and stimulants in fish. 69 Table 4.1: Proximate composition and inositol phosphate (IP) concentration of protein sources used in experimantal diets fed to juvenile rainbow trout for 83 days. 88 Table 4.2: Descriptions and abbreviations for experimental diets fed to rainbow trout for 83 days. 89 Table 4.3: Composition and energy content of experimental diets fed to juvenile rainbow trout for 83 days. .....92 Table 4.4: Composition of amino acid supplements used to equalize essential amino acid indices of selected diets fed to rainbow trout for 83 days. 95 Table 4.5 : Composition of mineral supplements employed in experimental diets fed to juvenile rainbow trout for 83 days 96 Table 4.6 : Analyzed levels of selected cations and anions present in protein sources used in experimental diets fed to juvenile rainbow trout for 83 days 98 Table 4.7: Analyzed levels of dietary cation-anion balance as milliequivalents (mEQ) and levels of selected cations and anions present in experimental diets fed to juvenile rainbow trout for 83 days ; 99 Table 4.8: Analyzed levels of proximate constituents and gross energy in experimental diets fed to juvenile rainbow trout for 83 days 104 Table 4.9: Analyzed and formulated levels of essential amino acids (in relation to rainbow trout requirements), in experimental diets fed to juvenile rainbow trout for 83 days 105 Table 4.10: Essential amino acid indices (EAAI) and analyzed and expected essential amino acid ratios (A/E) of experimental diets fed to juvenile rainbow trout for 83 days 108 Table 4.11: Calculated (expected) levels of selected minerals in experimental diets fed to juvenile rainbow trout for 83 days I l l Table 4.12: Analyzed levels of selected minerals in experimental diets fed to juvenile rainbow trout for 83 days 112 Table 4.13: Analyzed levels of dietary cation-anion balance as milliequivalents (mEq) and levels of selected cations and anions present in experimental diets fed to juvenile rainbow trout for 83 days 113 Table 4.14: Analyzed levels of glucosinolates in diets fed to juvenile rainbow trout for 83 days 118 List of Tables (continued) Table 4.15: Analyzed levels of proximate constituents in whole carcass of juvenile rainbow trout fed experimental diets for 83 days 119 Table 4.16: Analyzed levels of selected minerals in whole carcass of juvenile rainbow trout fed experimental diets for 83 days 120 Table 4.17: Weight gain (WG), specific growth rate (SGR), total dry feed intake (DFI), feed efficiency (FE), protein efficiency ratio (PER), and percent mortality of juvenile rainbow trout fed experimental diets for 83 days 121 Table 4.18: Results of orthogonal contrasts of performance parameters presented in Table 4.17 122 Table 4.19: Terminal plasma titre of 3,5,3' triicdo-L-thyronine (T3), liver 5' monodeiodinase (5'D) activity and average mean weights of livers and whole carcasses of juvenile rainbow trout fed selected experimental diets for 83 days 125 Table 4.20: Results of orthogonal contrasts of performance parameters presented in Table 4.19 126 Table 5.1: Descriptions and abbreviations of experimental diets fed to rainbow trout for 84 days. 144 Table 5.2: Proximate composition and inositol phosphate (IP) concentration of protein sources used in experimental diets fed to juvenile trout for 84 days. 145 Table 5.3: Composition and energy content of test diets fed to juvenile rainbow trout for 84 days. 146 Table 5.4: Composition of mineral supplements employed in experimental diets fed to juvenile rainbow trout for 84 days 149 Table 5.5: Analyzed levels of proximate constituents and gross energy in experimental diets fed to rainbow trout for 84 days 156 Table 5.6: Analyzed levels of selected minerals in diets fed to juvenile rainbow trout for 84 days 157 Table 5.7: Requirement levels and analyzed levels of essential amino acids in experimental diets fed to rainbow trout for 84 days 158 Table 5.8: Analyzed levels of glucosinolates and inositol phosphates (IP) in diets fed to rainbow trout for 84 days 159 Table 5.9: Weight gain (WG), specific growth rate (SGR), total dry feed intake (DFI), mean daily dry feed intake (MDDFI), appetite (APP), feed efficiency (FE), protein efficiency ratio (PER), percent protein deposited (PPD) and percent mortality of rainbow trout fed the experimental diets for 84 days. 162 Table 5.10: Results of orthogonal contrasts of the performance parameters presented in Table 5.9 163 List of Tables (continued) Table 5.11: Digestibility parameters: apparent protein digestibility coefficients (ADCO), apparent digestible energy coefficients (ADECO), apparent digestible energy levels (ADE), digestible protein levels (DPDD) and digestible protein/digestible energy ratio (DP/DE) for the experimental diets fed to rainbow trout for 84 days 166 Table 5.12: Results of orthogonal contrasts of digestibility parameters presented in table 5.11 167 Table 5.13: Analyzed levels of proximate constituents in whole carcass of rainbow trout fed experimental diets for 84 days 175 Table 5.14: Analyzed levels of selected minerals in whole carcass of rainbow trout fed experimental diets for 84 days 176 Table 5.15: Results of orthogonal contrasts of means of rainbow trout whole carcass proximate constituents and selected minerals presented in Tables 5.13 and 5.14 177 Table 5.16: Tenninal plasma titres of L-thyroxine (Tt) and 3,5,3'; triiodo-L-thyronine (T3), liver 5' monodeiodinase (5'D) activity and mean weights of livers and whole carcasses of rainbow trout fed selected experimental diets for 84 days 180 Table 5.17: Results of orthogonal contrasts of performance parameters presented in Table 5.16 181 X L is t of Figures Fig. 2.1 Pre-press solvent extraction process for the production of oil and meal from rapeseed and canola seed (Youngs, 1991; Daun et al., 1993b; Canola Council of Canada, 1994). 15 Fig. 2.2 Food Research Institute (FRI)-71 process for the production of rapeseed protein concentrate (Jones, 1979; Youngs, 1991). 19 Fig. 2.3 The Anderson model of phytic acid (Richardson, 1986; Reddy et al., 1989). 42 xi Acknowledgements There are many people I would like to thank and acknowledge for their support of myself and this thesis from its beginning to its completion. Many of you I cannot thank enough. To my supervisory and exarnining committee, Dr. Beames, Mr. Dosanjh, Dr. Higgs, Dr. Shelford, Dr. Vanderstoep and Dr. Xu, I would like to extend my thanks for the knowledge and advice you willingly shared. In particular, Dr. Beames, your support and advice about school and life I am truly mankful for. I feel honoured to have known you, for your life and conduct is a testament to character, integrity, and grace, as well as humour. Dr. Keith Downey and Dr. Holmes, thank you for your time and willingness to discuss canola with me. Dr. J.K. Daun, thank you for your slides, advice and your interest in the investigations described in this thesis. For assistance with sampling and/or feeding I thank John Quinton, Dr. Satoh, Stewart Anderson, Mr. and Mrs. Rowshandeli, Andy Lamb,Walter Ernst, Michell and Nicole. For friendship and understanding thank you Clarie, Mary, Gerry, Vikki, Jennifer, Mrs. Hobbs, Kanji, Brian, Leeanne, Angela, Jeanette, Pam and M.H. Last but not least, I thank my mother and father, Trudy and Rob, Paul and Marie, Charlene and Doug, Rhonda, Aunt Ethline and Sage for their support, humour, belief, love, trust and acceptance. Preface Material has been published in relation to findings from this thesis. The relevant references are listed below. General results and trends observed primarily at the midpoint of the studies were presented. General information concerning final results, as would appear in an abstract were also described. Participants involved in the papers/articles were academic supervisors, and mnding agency representatives, who offered editorial comments and participated in the choice of format for the completion of the articles. Writing responsibilities were shared. Prendergast, A.F., Higgs, D.A., Beames, R.M., Dosanjh, B.S., Deacon, G., 1994. Canola: searching for substitutes. Northern Aquaculture 10(3): 15-20 Higgs, D.A., Prendergast, A.F., Dosanjh, B.S., Beames, R.M., Deacon, G., Riley, W. 1994. Canola protein offers hope for efficient salmon production In: High Performance Fish, Fish Physiology Association, D.D. Mackinlay (editor). Vancouver, Canada, pp. 337-382 Higgs, D.A., A.F. Prendergast, R.M. Beames, B.S. Dosanjh, S. Satoh, S.A. Mwaichireya, and Deacon, G. 1995. Potential for reducing the cost of salmon production by dietary inclusion of novel rapeseed/canola protein products. In: Rapeseed Today and Tomorrow, Proceedings of the 9th International Rapeseed Congress, Cambridge, U.K. 4 to 7 July 1995, vol. 1 of 4 pp. 133-138 Higgs, D.A., Dosanjh, B.S., Prendergast, A.F., Beames, R.M., Hardy, R.W., Riley, W., Deacon, G. 1995. Use of rapeseed/canola protein products in finfish diets. In: Lim, E.E., Sessa, D.J. (editors). Nutrition and Utilization Technology Aquaculture. pp 130-156 1 CHAPTER ONE 1.0 General introduction In recent decades much nutritional research within the aquaculture industry has been directed toward the identification and evaluation of cheaper alternative proteins to fish meal protein, which has long been the preferred protein source in formulated aquatic feeds. Characteristically cheaper than animal proteins on a protein basis, plant protein products such as those from rapeseed and canola are being assessed as potential substitutes for fish meal protein in feeds. Since the 1940's, fish meal has been a major source of protein in non-raminant feeds, particularly aquatic feeds formulated for carnivorous fish and shrimp which require relatively high concentrations of protein in their diets. As a principal feed component, fish meal typically comprises 30-35 %, 50 % and 30-60% of dietary protein in the formulated feeds of shrimp, eels and salmonids respectively (New and Wijkstrom, 1990). Notably, with respect to salmonid diets, especially those based on European or Scandinavian formulations, fish meal may also be used as the sole protein source, furrushing 100% of total dietary protein (40-60% CP (Dry matter basis; DMB)) (Tiews et al., 1979; Mundheim and Opstvedt, 1989). High quality fish meal is the preferred protein source because of its nutritional profile. Usually derived from fresh whole fish, it is a nutritionally dense, highly palatable, high protein (68-80% CP (DMB)) feedstuff rich in minerals and essential amino acids, and high in digestible protein and energy content (Cho, 1990; Pike et al., 1990; Webster and Tidwell, 1992; Higgs et al., 1995). It is a good source of available lysine and metiuonine, (which are often the amino acids most deficient in plant protein sources) (Higgs et al., 1988; Higgs et al., 1990; Webster and Tidwell, 1992), and its well balanced amino acid profile has been shown to closely resemble the amino acid requirement pattern for salmonids (NRC, 1993). Additionally, its general lack of fibre and low concentration of carbohydrates enhance its suitability for the digestive capabilities of salmonids. Unlike terrestrial ariimals which readily utilize carbohydrates as an energy source, salmonids preferentially utilize protein and lipids as energy sources, having no demonstrated requirement for carbohydrates (NRC, 1993; Wilson, 1994), and accordingly, having a very limited capacity to effectively utilize them (Cho, 1990; Pike et al., 1990). Until the early part of the 1970's, fish meal was readily abundant and relatively inexpensive (Webster and Tidwell, 1992). However, following a 1973 ban imposed on the Peruvian anchovy exports, a worldwide shortage of fish meal was initiated (Spinelli et al., 1979). Fish meal more than tripled in price (Spinelli et al., 1979), and to an aquaculture industry strongly reliant on its use, the shortage and the dramatic price increases were devastating. By 1975, prices had lowered significantly (Spinelli et al., 1979), but since that time, world fish meal supply and price have continued to show considerable variability. Currently, world fish meal production levels hover around 6-7 million metric tonnes annually (Asgard, 1988; Pike et al., 1990; Rumsey, 1993; Messerich, 1994) of which > 80% is used in feeds for terrestrial non-ruminants (Messerich, 1994), and merely 15% is used for aquaculrural feeds, with this being shared between the salmon industry and other aquaculture industries such as shrimp, eel and other finfish. Yet, due in part to reduced harvests thought to be caused by overfishing, improved harvest technologies, and reclassification of industrial-use fish species such as mackerel, menhaden and Alaskan pollock from "trash fish" traditionally used for fish meal production to food use fish (Spinelli et al., 1979; 1981; New and Wijkstrom, 1990) now well utilized in the rapidly growing surimi industry, (Hardy and Shearer, 1993), fish meal production is forecasted to decline by 5 % by the year 2000 (IAFMM, 1991; New, 1991). Over the same time period, world aquaculture production is projected to increase by more than 165% from a 1990 level of 12.1 million metric tonnes to a level of 20 million metric tonnes by the year 2000 (Chamberlain, 1993). Given that demand for fish meal is projected to increase whereas production levels are forecasted to decline, the future adequacy offish meal supply and quality is being questioned. Additionally, inasmuch as the cost offish meal is inextricably linked to supply and demand, the future affordability of fish meal is also questionable. With a reduction in harvestable fish, some investigators anticipate that a greater proportion of fish meal in the future will be derived from fish offal or cuttings rather than from whole fish, resulting in nutritionally poorer meals known as white fish meals (Bimbo, 1990; Hardy and Shearer, 1993). Lower crude protein and higher in bone and scale content, white fish meals are excessively high in ash and in 3 minerals such as calcium and phosphorus and with extensive use have been shown to be cataractogenic in salmonids (Ogino and Yang, 1978; Ketola, 1979). When fed to salmonids, such meals are more poorly digested by smaller than by larger fish (Kitamikado et al., 1964), and are associated with overall poor growth when fed at high levels (Watanabe et al., 1980; Asgard, 1988; Satoh et a l , 1989; NRC, 1993), unless careful attention is paid to the balancing of minerals. Fish meal like most protein sources of animal origin is considered to be expensive (unit protein basis). Already, with fish meal as the primary source of protein in typical salmonid diets used in North America, the cost of protein may represent as much as 67% of total feed cost (Higgs et al., 1995). As feed costs in turn represent 40-60 % of total farm operating costs, and profitability is characteristically marginal, any future increase in fish meal cost may seriously erode any profitability that may now exist. Due to concerns for the continued adequacy of fish meal supply and quality, and concerns for the continued affordability of the fish meal resource, nutritionists and feed manufacturers alike have increased their research efforts directed toward the identification and evaluation of cheaper alternative protein sources. Much attention has been given to cereals and oilseeds, which are readily available. However, because cereals typically contain only 10-15 % crude protein they are most often regarded as energy sources, whereas oilseed meals which typically contain 35-50% CP are primarily regarded as high protein products (Hatje, 1989). Dominant in terms of world production (Hatje, 1989), and claimed to have an amino acid profile well suited for meeting the essential amino acid requirements offish and other aquatic species (Wilson, 1989; Webster and Tidwell, 1992), soybean has emerged as the oilseed receiving the most attention (Webster and Tidwell, 1992) as a source of protein. Yet, soybean protein products are not the only obvious choices as replacements for fish meal, for although receiving comparatively less attention, rapeseed and canola oilseeds, (ranked third in terms of world production levels) (Downey, 1990; Salunkhe et al., 1992; Luhs and Friedt, 1994), have nutritional profiles which also appear to favour their potential for replacing fish meal in formulated feeds. Currently, rapeseed and canola are grown primarily for the oil content of their seeds which accounts for approximately 80 % of the agronomic value of the crop (Downey and Robbelen, 1989). 4 Although considered as by-products, the meals (solid residue fractions after oil extraction from the crushed seed), are valuable feedstuffs that are widely utilized as relatively inexpensive high protein replacements for soybean meal, and to a lesser extent, animal protein in formulated diets (Salunkhe, et al., 1992). They have a crude protein content of 35-44% (Downey and Robbelen, 1989) and have an amino acid profile that is well balanced (Gillberg and Tornell, 1976). Protein quality has been detenriined to be equivalent to that of whole herring meal and superior to that of soybean meal for respresentative cold water and warm water fish species (Higgs et al., 1988; 1989; 1995). However, the full expression of the quality of protein found within rapeseed and canola meals appears to be suppressed both by a low energy content (which in large part may be attributed to high fibre content (~ 12 % dry matter basis; DMB), and an oil-industry-dictated maximum residual oil level of 4% DMB), and the meal's content of antinutritional factors (ANFs), namely glucosinolates, phenolic compounds, fibre and other digestible carbohydrates and phytic acid. Collectively these ANFs are associated with, and responsible for poor meal palatability and low feed intake as well as goitrogenic and non-goitrogenic impairment of thyroid function, digestive enzyme inhibition, decreased digestibility, absorption and utilization of proteins, vitamins and minerals, and reduced growth (van Etten and Tookey, 1983; Cheeke and Shull, 1985; Jensen et al., 1990; Bell, 1993; Higgs et al., 1995; NRC, 1993). Due to the nature and potential severity of the impairment of digestive functions and other physiological functions that may be induced by these ANFs, recommendations have been made for maximum dietary inclusion levels of such meals in monogastric feeds. Typically, recommended dietary inclusion levels for monogastrics are restricted to 5-15% for rapeseed meal and 10 -25% for canola meal (Robblee et al., 1986; Miller, 1988; Higgs et al., 1995a). Permissable dietary inclusion levels differ for canola and traditional rapeseed because they are nutritionally distinct products. Canola is a nutritionally superior form of rapeseed, containing lower amounts of ANFs. Specifically, canola refers to Brassica napus and B. rapa {campestris) rapeseed varieties which have been genetically selected to contain < 2% erucic acid in the oil and < 30 umol of aliphatic (alkenyl) glucosinolates per gram of air-dried meal (Miller, 1988). By comparison, traditional rapeseed varieties, (high erucic acid rapeseed varieties; HEAR) contain 20-50% erucic acid in the oil and 50-100 umol/g of glucosinolates in the air-dried, defatted meal (Salunkhe et al., 1992; Bell, 1993). 5 Erucic acid (C221© 9) is a fatty acid that occurs in and is considered to be an ANF of the oil of rapeseed and canola. It is generally not considered to be of much nutritional consequence to the meal because the oil content of the meal is characteristically low, being less than, or equal to 4 % (DMB). Consequently, the improved nutritional value of canola meal relative to rapeseed meal is largely attributed to the reduction in the glucosinolate content, wherein the specific adverse effects attributed to the occurrence of glucosinolates, namely poor palatabilty, lowered feed intake, impaired thyroid function and depressed growth (van Etten and Tookey, 1983; Cheeke and Shull, 1985; Jensen et al., 1990; NRC, 1993) are much decreased. Specialized processing of rapeseed and canola meals and seeds may also be undertaken to reduce ANF content. This results in protein products that have been referred to as upgraded meals, concentrates and isolates. These contain > 50% CP, > 60% CP and > 90% CP respectively (Youngs, 1991). Lower in glucosinolate and phenolic content, these products are of improved nutritional value, and have been shown to be better tolerated in the diets of rainbow trout and Pacific salmon than are standard meals (Hajen et al., 1993a, 1993b; Higgs et al., 1995). However, due to the nature of the processing methods used in the further refined protein products, appreciable amounts of fibre can remain, effectively functioning as diluents of the protein content of the products. Phytic acid also is not readily lost through such processing steps and in fact, shows a disturbing tendency to concentrate, being found in higher concentration in upgraded rapeseed and canola meals and concentrates than in conventional meals. Due to its ability to complex with a variety of nutrients within the feedstuff, phytic acid is thought to account for the "masking " (restricted utilization) of the high quality protein found in rapeseed and canola protein products. At physiological pH and pH levels encountered in most feed, phytic acid is strongly negatively charged and readily complexes with enzymes, other proteins and multivalent cationic minerals, rendering them partially or totally unavailable to monogastric animals, which lack the enzyme necessary for phytic acid hydrolysis (Spinelli et al., 1983; Richardson, et al., 1985; Campbell and Bedford, 1992). Consequently, growth, survival of the animals, and utilization of feed and protein are decreased. A variety of methods (mechanical, biological and chemical) have been employed with varying degrees of success for the reduction of phytic acid content, one of which is the application of phytase, an 6 enzyme that hydrolyzes phytic acid (Spinelli et al., 1979; Irving,1980; Pointillart et al., 1987; Ward et al., 1991; Lei et al., 1993; Khan and Cole, 1995). Removal or reduction of phytic acid should improve nutrient utilization. However, rapeseed and canola meals would still have a substantially lower protein and energy content than typical high quality fish meals, and thus would not be suitable as complete replacements for fish meal in diets formulated for salmonids. It is, however, highly possible that concentrates and isolates, which have levels of protein and energy which are comparable to levels found in fish meals, may function as complete replacements for fish meal in salmonid diets, should phytic acid removal increase the bioavailabilty of nutrients found within the protein products. This thesis was designed to detennine whether the nutritional value of some rapeseed and canola protein products (standard meals, upgraded meals and concentrates), could be significantly improved by the removal of phytic acid. It was also designed to investigate whether, with the removal of phytic acid, rapeseed protein concentrates could be used to completely replace fish meal protein in diets for juvenile rainbow trout, without compromising fish performance. 7 CHAPTER TWO 2.0 L i t e r a t u r e R e v i e w The following literature review describes the history, development, and nutritional quality of rapeseed and canola protein products. Limited to a consideration of monogastric animals, particularly salmonids, the review describes the significance of the antinutritive factors intrinsic to rapeseed and canola protein products, particularly the significance of phytic acid. In addition, information on the various chemical, biological and physical methods which may be employed to enhance the nutritional quality and, consequently the biological value of rapeseed and canola protein products is presented. Mention is also made of the use of chemostimulants and chemoattractants in salmonid diets. 2.1 R a p e s e e d a n d C a n o l a Rapeseed and canola oilseed crops are grown and processed primarily for the oil content of their seeds. The meals, which are residues of the oil extraction process, are most often used as fertilizers in Asia. Alternatively, in Europe and North America they are valued as high quality protein supplements for animal feeds, and in particular, are widely utilized as relatively inexpensive replacements for soybean meal, and to a lesser extent, animal protein in formulated diets (Salunkhe et al., 1992). Although oilseeds usually contain a variety of antinutritional factors (ANFs) of varying toxicities, the presence of erucic acid (C22 co 9 ; m-13-docosenoic acid) and glucosinolates in canola and rapeseed distinguish these crops from other major oilseed crops (Luhs and Friedt, 1994). Notably, it is largely on the basis of the levels of these two ANFs that canola and rapeseed themselves are distinguished from each other, wherein canola is a genetic variation of rapeseed, specifically selected for lower erucic acid and glucosinolate content and improved nutritional qualities (Canola Council of Canada, 1994). 8 2.1.1 Taxonomy of rapeseed and canola Belonging to the family Cruciferae (Brassicaceae), rapeseed and canola are oil-yielding varieties of the Brassica genus (Bengtsson et al., 1972; Holmes, 1980; Shahidi, 1990; Salunkhe et al., 1992). Other members of this genus include horticultural crops such as broccoli, turnips, radishes, kale, cabbage, cauliflower and brussel sprouts (Downey and Robbelen, 1989; Salunkhe et al., 1992; Liihs and Friedt, 1994), for which seed oil content is of relatively little agronomic consequence. In Canada and Northern Europe, B. napus (Argentinian rapeseed, Swede rape, rape) and B. rapa (B. campestris) (Polish rapeseed, turnip rape) are the predominant rapeseed and canola species grown. In China and the Indian subcontinent B. napus, B. rapa (campestris) as well as B. juncea (Brown mustard, Indian mustard, leaf mustard, Oriental mustard) (Holmes, 1980; Salunkhe et al., 1992; Luhs and Friedt, 1994) are commonly grown (Shahidi, 1990). It is important to note that B. rapa and B. campestris are synonomous. However, following a recent review of the writings of Carl Linnaeus concerning the Brassica species, systematists now recommend the elimination of the use of B. campestris in favour the older term B. rapa (Thomas, 1994). 2.1.2 From rapeseed to canola: development of a crop The cultivation of rapeseed dates back centuries and has been documented in ancient literature such as the 4,000 yer old Indian Sanskrit literature which mentions the growth and utilization of B. rapa (Boulter, 1983; Luhs and Friedt, 1994), the fourteenth century literature from the Netherlands which reports the cultivation of B. napus (Luhs and Friedt, 1994), as well as eighteenth century writings of the Swedish scientist Carl Linnaeus which clearly describe the cultivation of rapeseed for the express purpose of oil extraction (Appelqvist, 1972). Historically, rapeseed oil was used for cooking and soapmaking, but was primarily utilized for illumination (Liihs and Friedt, 1994; Thomas, 1994). By the latter part of the Middle Ages, rapeseed was the most important source of lamp oil and the most important oilseed crop of Western Europe (Ohlson, 1972; Liihs and Friedt, 1994). However, European cultivation and production of rapeseed suffered a sharp decline following an increase in the use of petroleum based mineral oils aroUnd the year 1860 (Ohlson, 1972; Boulter, 1983; Luhs and Friedt, 1994). Demand for, and 9 production of rapeseed was not to rise again until the advent of the steam age, wherein rapeseed oil proved to be a superb lubricant for steam driven engines, having a great propensity for adhering to moisture-laden machine parts (Boulter, 1983; Shahidi, 1990; Canola Council of Canada, 1994). With the start of World War II however, worldwide trade routes from Asia and Europe to North America were cut off, leading to a shortage of imported food and industrial products, including rapeseed oil. Both the United States and Canada were forced to find alternative sources of oil. Throughout World War II, most rapeseed oil produced in Canada was directed for industrial use as a lubricant in merchant and naval vessels involved in the Allied war effort, but shortages in offshore supplies of edible oil necessitated the diversion of some rapeseed oil for use in the edible oil market to help meet demand (Shahidi, 1990). The development of canola is chronicled by this transition of primary sales of Canadian commercial rapeseed oil from industrial to food use markets. In 1948, post-war proposals were made for the establishment of an edible rapeseed oil industry (Lips et al., 1948). Following necessary refinements to industrial processing techniques, the first Canadian edible rapeseed oil was obtained in 1956 (Boulter, 1983; Shahidi, 1990; Canola Council of Canada, 1994). However, ready acceptance of the oil for human consumption was impeded by human health concerns, which arose following the release of experimental results concerning rats fed diets containing rapeseed oil, which was known to contain high amounts of eicosenoic acid (C2016) 9), and erucic acid (C221 o 9) at levels of 10-11% and, 20-55% respectively (Salunkhe et al., 1992). Research showed that rats fed diets containing rapeseed oil developed cardiac lipidosis and myocardial necrosis (Abdellatif and Vies, 1970; Beare-Rogers, 1970; Mattson, 1973; Salunkhe et al., 1992). Concerns were also being raised about rapeseed meal as a protein supplement (Canola Council of Canada, 1994), wherein within the livestock industry, the presence of antinutritive glucosinolates in the meal appeared responsible for reduced palatability, and thus lowered feed intake (Canola Council of Canada, 1994). Although later research was to show that erucic acid and eicosenoic acid posed more of a health threat to rats than to humans, (refer to section 2.1.7.6.1) glucosinolates were indeed causing palatabilty problems, and the nutritional concerns about the oil and the meal, once voiced, helped initiate research which led to nutritionally improved varieties of rapeseed (Mattson, 1973; Vaisey et al., 1973; 1974; McDonald, 1983). 10 In response to nutritional concerns, Canadian plant breeders, through genetic selection, developed and then released the first low erucic acid (< 5% in the oil) B. napus, variety of rapeseed, Oro, in 1968 (Canola Council of Canada, 1994). Three years later the first low erucic acid B. rapa rapeseed variety, Span, was released (Canola Council of Canada, 1994). Oil produced from the single low, (low in erucic acid but still high in glucosinolate content), varieties was termed canbra oil. In 1974, as a result of research initiated in 1958 by Dr. B. Stefansson and Dr. R.K. Downey, the first rapeseed variety low in both erucic acid (< 2% in the oil) and glucosinolate content (< 30 umol/g of defatted, air dried meal), was released (Canola Council of Canada, 1994). This new double-low (double-zero, 00) rapeseed variety was known as Tower and was derived from B. napus (Canola Council of Canada, 1994). Later, in 1977, Candle, the first double-low variety of B. rapa was released (Shahidi, 1990; Canola Council of Canada, 1994). (Candle is sometimes referred to as triple-low, (triple-zero, 000), because it is also inherently low in fibre (Bell, 1984; Shahidi, 1990)). The neologism chosen to describe the nutritionally improved double-low B.napus and B.rapa rapeseed varieties was canola (canola). The official definition of canola is the following: Brassica napus and Brassica rapa seed, oil, meal, hulls and protein extraction products from or of varieties with < 2% erucic acid in the oil and < 30 umol/g of one or more aliphatic (alkenyl) glucosinolates in the defatted, air-dried meal (Miller, 1988; Downey, 1990). An acronym for Canada oil-low acid, canola became a registered trademark in 1978 (Miller, 1988; Neshem, 1990). In 1987, it was designated as an official grade under the Canadian Grain Act and, canola seed was defined in the Seeds Act (Canola Council of Canada, 1994). Back in 1979, following the granting of trademark status for canola, Agriculture Canada, Health and Welfare Canada, and various Canadian scientists and administrators applied to the United States Food and Drug Administration (FDA), requesting GRAS (generally regarded as safe), designation for canola oil (Neshem, 1990). Around 1984, the FDA ruled that limited GRAS status would be granted. The term canola could be used, but only in association with rapeseed and erucic acid. All crude or refined canola oil was required to be labeled as low erucic acid rapeseed oil, LEAR oil, LEAR (canola) oil or low erucic acid (canola) oil (Neshem, 1990). In December 1988, canola oil was accorded full GRAS status (Neshem, 1990; Canola Council of Canada, 1994). After nine years of vigorous opposition by oilseed groups in the 11 United States, canola won recognition by the FDA as being synonomous with low erucic acid rapeseed (LEAR), and so could be used independently on food labels (Neshem, 1990; Canola Council of Canada, 1994). However, the nine year struggle for GRAS status seemed inordinately long when one considers that oil from traditional high erucic acid varieties of rapeseed (HEAR), was used for edible purposes in North America during the war, and is and was used for centuries as an edible oil in Asian countries such as India and China (Holmes, 1980; Liihs and Friedt, 1994). It is of some interest to note that in 1986, during the pursuit of GRAS standing, canola lost its trademark status. What amounted to an oversight or omission in the wording of an amendment made to the trademark by the Tradesmark Branch of Consumer and Corporate Affairs, was the cause (Canola Council of Canada, 1994). Despite the loss of trademark status however, the term canola remains registered to the Canola Council of Canada. Only countries mamtaining a license with the Council, (22 countries now registered), may use the word canola or the canola flower logo on any packaged product meeting canola standards (Canola Council of Canada, 1994). However, canola is now a generic term of sorts, and may be used to refer to other rapeseed species and varieties which are canola-like, or of canola quality (Canola Council of Canada, 1994), such as the recently developed low glucosinolate, low erucic acid variety of B. juncea which is of canola quality (Love et al., 1990; Luhs and Friedt, 1994; Slorninski et al., 1995). Traditional rapeseed varieties continue to be grown throughout the world, but in Canada they are grown in relatively small amounts, primarily under contract for industrial use (Canola Council of Canada, 1994). Comparatively, 98.5 % of the former rapeseed crop on an acreage basis was sown with canola by 1976 (Canola Council of Canada, 1994). In Western Europe a similar conversion to canola was made by 1978 (Niewiadomski, 1990). 2.1.3 Classification of oilseed protein In large part, the definitions, classification schemes and nomenclature used for the description of rapeseed and canola protein products are "borrowed" from the long established soybean industry. (Youngs, 1991; Daun et al., 1993b). Products may be separated simply on the basis of protein level. However, in general, categorization is based on the particular type and degree of processing employed (Youngs 1991). 12 Accordingly, three main categories are used, viz. meals, concentrates and isolates. Meals have the lowest protein content, and may be described simply as the residues of defatted oilseeds. Crude protein levels for rapeseed and canola meals are defined as < 50% but typically range between 35-44 % CP (dry matter basis, (DMB)) (Downey and Robbelen, 1989; Youngs, 1991). However, although most commercially produced rapeseed and canola meals may be described in these terms, a broader definition includes "upgraded meals" or "flours" which are derived from either standard meals or seeds which have been defatted and dehulled (Youngs, 1991). Hull removal or reduction yields a product which is higher in protein (> 50% crude protein (DMB)) and lower in hull and fibre content than standard commercial meals (Youngs, 1991; Daun et al., 1993b). The upgraded meals are also more finely ground (will pass through a 100-mesh screen) (Aref, 1970) than standard meals. It is important to note however, that although the term flour is generally considered to be synonomous with upgraded meals throughout the literature, flour has on occasion been used to describe concentrates which are also "flour-like" in appearance. Concentrates are residues of dehulled and defatted oilseeds which have undergone further extraction procedures in order to concentrate the protein by means of reduction and removal of non-protein constituents and, in particular, seed components which adversely affect flavour, colour and nutritional value (Youngs, 1991). In practice, a crude protein level of >60% (DMB) is characteristic of rapeseed and canola concentrates. Protein isolates are the protein products which result from the extraction of protein in a relatively pure form (> 90% crude protein (DMB)), from defatted hulled or dehulled seed (Youngs, 1991). 2.1.4 Processing of rapeseed and canola protein A variety of processing methods have been used for the production of rapeseed and canola protein products. Many methods have only been tested on a laboratory scale, but some methods have also been used commercially. Some details of the various processes are presented here but a more thorough examination of rapeseed and canola processing techniques may be found within the excellent review of Youngs (1991). 13 2.1.4.1 Oil removal and meal production Expeller pressing, pre-press solvent extraction and direct solvent extraction processing are the three main processing methods used for oil extraction of rapeseed and canola seeds (Daun et al., 1993b). Expeller pressing was the primary processing method employed in North America during the 1940's and 1950's but in recent decades has declined in popular use (Daun et al., 1993b). A simple screw press is used to extract the oil, but due to the tremendous pressure and high temperatures that are generated, effective removal of oil is difficult without adversely affecting the oil and meal quality (Daun et al., 1993b). Direct solvent extraction is the process by which oil is removed exclusively with a solvent (e.g. hexane). The process is fairly simple, but relatively large amounts of solvent and energy inputs are required, and substantially large amounts of residual oil may remain in the meal (Daun et al., 1993b). Pre-press solvent extraction is the primary method now being used for oil extraction from rapeseed and canola seeds. The process proceeds in seven main stages: seed cleaning, pre-conditioning, flaking, cooking, pressing, solvent extraction, and desolventizing and drying. These stages are depicted in Figure 2.1 (Unger, 1990; Daun et al., 1993b). 2.1.4.2 Upgraded meals/flours Upgraded meals are typically higher in protein and lower in glucosinolates, phenolics and fibre than standard meals. They are meals which have been nutritionally improved or "upgraded" by the reduction or removal of hulls, which contribute the majority of the antinutritive fibre (hemicellulose, cellulose, lignin) and phenolic compounds (tarinins, sinapine) commonly present in rapeseed and canola standard meals. Additionally, glucosinolates and other non-protein constituents may be reduced by the incorporation of water or solvent extraction stages. Due to a particularly small seed size («l-2 mm diameter), de-hulling of rapeseed and canola seeds is a relatively daunting processing challenge. In order to produce good quality meals, hulls must be removed with a minimal loss of oil or meat (cotyledon and hypocotyl). For this purpose water soaking (Eapen et al., 1969), extensive sieving (McCurdy and March, 1992), drying and/or air classification (Jones and Holme, 1979; Youngs, 1991) steps may be incorporated into the meal production process to facilitate 14 hull removal. Hull removal may either precede oil extraction, (front-end dehulling), or may follow oil extraction (tail-end de-hulling), but independent of when hulls are removed, oil is still pre-press solvent extracted from the meats fraction in a manner similar to processes used for the commercial production of standard meals. A large variety of front-end de-hulling processes have been used on a laboratory scale for the processing of rapeseeds and canola seeds (Eapen et al., 1969; Jones and Holme, 1979; Youngs, 1991; McCurdy and March, 1992). Although none of the front-end dehulling methods have been employed commercially, according to Youngs, (1991) upgraded meals/flours have been produced under industrial conditions at the pilot plant level by the French "Centre Technique Interprofessionnel des Oleagineux Metropolitains" (CETIOM) in collaboration with the Oleagri Recherches et Developpement group. The method employed involves seed cracking in a centrifugal impact dehuller, followed by the separation of hulls, kernels and dust in a fluidised bed sorting unit by means of air diffusion and air suction (Youngs, 1991). Resultant meal fractions contained approximately 5% hulls and hull fractions contained approximately 3.5-5% kernels (Youngs, 1991). Comparatively, standard meals contain 27-30% hulls (DMB) (Bell and Shires. 1982; Shahidi, 1990). Possible disadvantages to the commercial production of upgraded meals/flours by front-end de-hulling include the fact that most rapeseed/canola crushing plants are equipped for tail-end de-hulling, and re-fitting to accomodate front-end de-hulling processes would likely involve considerable expense. Additionally, because the hull fraction would not be included in defatting processes, up to 10% losses of oil might be encountered (Youngs, 1991). Such losses might not be readily accepted by the present oil-crushing community. However, it is possible that some of the opposition may be offset by the prospective sales of hull fractions that would have improved feeding value due to increased energy and protein content. Tail-end rapeseed and canola seed de-hulling processes have received less attention than front-end de-hulling processes, but have also been practiced at both the laboratory and pilot plant scale level, although they too have not been commercially produced. As with front end de-hulling methods, air classification may be the primary method used for de-hulling (Youngs, 1991). However, the starting material is usually commercially produced standard meals rather than whole seeds. Typically, standard D O C K A G E S E E D C L E A N I N G -aspiration -screen separation o f over and undersized particles -dockage reduced to < 2 .5% 15 S C R E E N I N G S P R E C O N D I T I O N I N G -seeds preheated to 30 -40°C -moisture adjustment V - P E X ™ screwpress F L A K I N G C O O K I N G - inactivation o f thioglucoside glucohydrolase by heating to 75-85°C for 20-40 minutes S C R E W P R E S S -60-70% o i l removal -cake formation D E S O L V E N T I Z E R T O A S T E R / D R Y E R - sparge steam - solvent extraction with heated hexane (50-60°C) - o i l content reduced to 1% -complete enzyme inactivation M E A L Further refinement F i g . 2 .1 : Pre-press solvent extraction process for the production o f o i l and meal from rapeseed and canola seed ( Y o u n g s 1991; Daun et al., 1993b; Canola Counc i l o f Canada, 1994). 16 meals are water extracted to remove glucosinolates, then are dried, powdered and screened (Youngs, 1991). Alternatively de-hulling may be accomplished with the use of liquid cyclones. Water extracted meal is dried, finely ground, suspended in hexane and then is separated into hull and flour fractions in the liquid cyclone (Y oungs, 1991). It is this method of tail-end dehulling that has been employed at the pilot plant level (Youngs, 1991). Advantages of tail-end de-hulling processes relate to the fact that the "starting material" is standard pre-press solvent extracted meals. Therefore potentially expensive equipment or procedural changes to existing crushing plants are not anticipated should commercial scale production be initiated. One possible disadvantage to tail-end dehulled meals however, is that most of the oil content has been removed, an thus only residual amounts or oil typically remain, accounting for the low available energy of the meal. 2.1.4.3 Concentrates Rapeseed and canola concentrates may be produced from whole seeds, dehulled seeds, standard meals or upgraded meals. Specific processing methods vary but all are designed to concentrate protein by means of the reduction and/or removal of non-protein constituents, particularly ANFs such as phenolic compounds (e.g. tannins, sinapine), and soluble problem sugars (e.g. stachyose and raffinose). Additionally, all methods include steps for the deactivation of thioglucoside glucohydrolase and the reduction of levels of glucosinolates in order to prevent the hydrolysis of glucosinolates to toxic aglycones (refer to section 2.2.7.6.2). Separation of protein from non-protein constituents is based on their differential solubilities in solvents such as water (Jones, 1979; Youngs, 1991), and alcohol (e.g. ethanol, methanol, isopropyl alcohol) used singly or in combination with water, sodium hydroxide or ammonia (Bhatty and Sosulski, 1972; Anjou et al., 1976; Rubin et al., 1984; Youngs, 1991). (If water is used as a solvent, heat must also be applied in order to inactivate thioglucosidase glucohydrolase, whereas if alcohols are used heat application is not necessary as the enzyme is denatured by the action of the alcohol itself (Youngs, 1991)). Following solvent extraction, oil removal, drying and grinding, the residue that remains is referred to as a "protein concentrate" which primarily consists of protein and insoluble polysaccharides 17 (Daun etal., 1993b). Rapeseed/canola concentrates have been produced on a laboratory scale. They have also been produced on a pilot plant scale by AB Karlshamns Oljefabrikers and Alpha Laval, AB in Karlshamn, Sweden (Youngs, 1991). A pilot plant was constructed expressly for the production of rapeseed protein concentrate with the intention of eventual commercial production. Despite the production of a concentrate with a protein content of 67% (DMB) commercial production was not attempted, due in part to a domestic market which was apparently determined to be too small to absorb the product generated by the plant (Youngs, 1991). Rapeseed protein concentrate has also been produced at the pilot plant level according to processes developed at the Food Research Institute (FRI) in Ottawa, Canada. The institute developed a number of methodologies for concentrate production, however the FRI-71 process, developed in 1971 is the only FRI process presented in detail here because concentrates produced by this methodology have been repeatedly evaluated as a protein source in the diets of salmonids (Yurkowski et al., 1978; Higgs et al. 1982; Hajen et al., 1993b; Higgs et al., 1991; Teskeredzic et al., 1995). As depicted in Figure 2.2, the dehulled seed fraction is subjected to boiling water treatment to inactivate thioglucoside glucohydrolase, leached with water to extract water-soluble components (e.g. glucosinolates, phenolic compounds, raffinose, stachyose), and extracted with hexane to remove oil and to concentrate the protein (Jones, 1979). Crude protein content usually is between 65-70% (DMB). However, appreciable amounts of crude fibre remain (4.9 - 6.0 %) (Jones, personal letters; Youngs, 1991) and phytic acid (3.6-7.5 %) is invariably concentrated within the protein residue (Youngs, 1991). 2.1.4.4 Isolates Like other rapeseed/canola protein products, isolates have been produced on a laboratory and on a plant scale. They have been prepared from whole seeds, but are most often produced from standard meals or upgraded meals (Youngs, 1991). Unlike the preparation of concentrates where non-protein constituents are extracted, leaving a concentration of protein, isolates are produced through the extraction of protein itself, in a relatively pure form. 18 Aqueous alkali extraction followed by acid precipitation is a common procedure for the preparation of vegetable protein isolates such as soybean isolates however, this methodology is unacceptable for the production of rapeseed or canola isolates because of the presence of appreciable quantities of phenolic compounds (Youngs, 1991). The phenolic compounds readily complex with proteins and promote the development of adverse flavours and colours under alkaline conditions, typically resulting in isolates which are bitter and dark brown in colour (Ramachandra et al., 1977; Sosulski, 1979; Thompson et al., 1982; Serraino and Thompson, 1984; Reddy et al., 1989; Youngs, 1991). (Refer to section 2.1.7.6.4). Low protein recoveries (« 50% protein recovered), also result when rapeseed and canola are processed in this way (Youngs, 1991). The addition of Na2S20s during extraction may improve the colour, and washing with alcohol may also improve the colour as well as. improve the flavour of the protein product, however, the lower protein recoveries are not improved (Thompson et a l , 1982; Youngs, 1991). According to Youngs, (1991), however, El Nockrashy, (1976) and El Nockrashy et al., (1977) were able to get high protein recoveries (>90 % of meal protein) using a procedure that involved alkali (NaOH) extraction. Later publications from the same SEED CLEANING CRACKING A I R C L A S S I F I C A T I O N -•HULLS meats I N A C T I V A T I O N O F T f f l O G L U C O S I D A S E G L U C O H Y D R O L A S E B Y I M M E R S I O N O F M E A T S I N 100° C W A T E R F O R T W O M I N U T E S • W A T E R G R I N D I N G u P R O T E I N C O N C E N T R A T E F i g . 2.2 F o o d Research Institute (FRI)-71 process for the production o f rapeseed protein concentrate (Jones, 1979). 20 laboratory however were less promising (Youngs, 1991) As alkaline conditions tend to adversely affect the quality of rapeseed or canola protein products, acidic conditions are preferred for the production of isolates. Typically extraction and precipitation is performed at acidic pH of 4.2-4.5 and results in isolates of high protein content, and good (light) colour (Thompson et al., 1982; Youngs, 1991). Acidic polymers such as sodium hexametaphosphate (SHMP), carboxymethyl-cellulose, polygalacturonic acid, and alginate may be used for both the extraction and precipitation stages and have been shown to improve protein recoveries (Gillberg and Tornell, 1976; Thompson et al., 1977; Youngs, 1991). Dialysis and ultrafiltration methods used singly or in combination with serial isoelectric precipitation are also methods that have been used to produce canola and rapeseed isolates. Given the existing methods of isolate preparation, commercial production is at present unlikely because the associated high costs of equipment and/or the large volumes of solvents which would be required to meet the conditions necessary for the production of isolates with good colour and high yields are prohibitive (Youngs, 1991). 2.1.5 Nutritional properties of rapeseed and canola protein The nutritional value of rapeseed and canola protein products as dietary constituents of monogastric feeds is dependent on the type of rapeseed variety from which the product is derived (e.g. canola, double low or traditional rapeseed varieties) and the degree and type of processing used to produce the product (e.g. standard meals, or upgraded meals, concentrates or isolates). The acceptability of the various products within monogastric diets is also a function of the species, size and/or age of the animals under consideration. The ANFs intrinsic to rapeseed and canola seeds, namely glucosinolates, saponins, phenolic compounds (e.g. sinapine, tannins, lignin), phytic acid, problem sugars (such as stachyose and raffinose) and fibre, are responsible for the lowered palatability, lowered protein and energy digestibility and lowered bioavailability of minerals and vitamins of protein products derived from these seeds. Impaired thyroid function and overall poor growth performance of animals fed protein products derived from rapeseed or 21 canola seeds are also ascribed to the physiological effects of these ANFs. Lower in glucosinolate content than rapeseed protein products are, canola protein products are nutritionally superior to comparative protein products derived from traditional rapeseed varieties. This nutritional difference between rapeseed and canola is reflected in the recommended maximum inclusion rates of rapeseed and canola meal for monogastric diets. Commercially, traditional rapeseed meal is limited to a maximum of 5 % of the dry diet of laying chickens, and starting pigs, 8 % for growing or finishing pigs, and 15 % for broilers (Canola Council of Canada, 1994). This compares to recommended maximum dietary inclusion levels of canola meal at 10 %, 12 %, 18%, and 20 % dry diet, respectively (Canola Council of Canada, 1994). Similarly, greater amounts of canola meal than rapeseed meals are tolerated in the diets of rainbow trout and Pacific salmon (Higgs et al., 1995). The seed coat colour of rapeseed and canola ranges from yellow to brown-black both within and between species. Although this variation in colour is of some esthetic value, it is also of some nutritional consequence because the content of carbohydrate, phenolic compounds (e.g. tannin) and protein and levels of digestible protein and energy found within the seed have been shown to vary with coat colour (Downey and Bell, 1990; Bell, 1993; Youngs, 1991). As previously stated, the nutritional value of rapeseed and canola protein products is affected by the degree and type of processing. With further refining the seeds and/or standard meals of rapeseed and canola may be used to produce upgraded meals, concentrates, and isolates that are higher in protein and lower in most ANFs (glucosinolates, saponins, phenolic constituents). These products are of greater nutritional value, and are better tolerated in the diets of monogastrics. However, because residual amounts of certain ANFs remain, the feeding value is still compromised. Although all the ANFs that persist through processing play a role in decreasing the feeding value of the various protein products, research indicates that glucosinolate content and low available energy content are the primary limiting factors in the use of canola and rapeseed meals and flours in salmonid diets (Yurkowski et al., 1978; Higgs et al., 1983; Fagerlund et al., 1987; Leatherland et al., 1987; Higgs et al., 1990; Bell, 1993). By contrast, phytic acid has been identified as the primary limiting factor in the use of rapeseed/canola concentrates (Youngs, 1991). 22 The acceptability of, or tolerance for, various rapeseed and canola protein products within the diets of monogastric animals as a function of species, size and/or age, is clearly evident with fish (Higgs et al., 1990) wherein warm water fish species such as carp and tilapia tend to show a greater tolerance for rapeseed and canola meals in their diets (28 % dry diet or 28% dietary protein), than do cold water fish such as coho and chinook salmon (16-23% dry diet, 13-22 % dietary protein) or rainbow trout (<13.5-20% dry diet, < 13.3- 18% dietary protein) (Higgs et al., 1990; Higgs et al., 1995). Similarly, as indicated above, juvenile Pacific salmon (e.g. coho and chinook) tend to show more tolerance for dietary rapeseed and canola protein products than do rainbow trout (Higgs et al., 1995), and it appears that rainbow trout are more sensitive to antinutritive constituents (e.g. glucosinolates) found in rapeseed and canola protein products than are salmon (Higgs et al., 1983; Hilton and Slinger, 1986). 2.1.5.1 Protein and essential amino acids The protein content of rapeseed and canola seed is inversely correlated to both its oil and fibre content (Downey and Robbelen, 1989; Niewiadomski, 1990). Total seed crude protein content shows some variability from year to year (Bell, 1993), due in part to different growing conditions, such as high soil nitrogen which serves to increase protein content while reducing oil content (Stefansson, 1983). Protein content also differs between cultivars (Niewiadomski, 1990), but protein levels typically range between 21-30% (dry weight basis) in the seed (Robblee et al., 1986; Jensen et al.j 1990). Comparatively, crude protein content usually ranges between 35-45% (DMB) in solvent extracted meals (Yurkowski et al., 1978; Serraino and Thompson, 1984; Downey and Robbelen, 1989; Bell, 1990; Downey, 1990), and is >50 % for upgraded meals, > 60 % for concentrates and > 90% for isolates. It is important to note however that, as is typical of many of the crude protein values given in literature regarding protein feedstuffs, a factor of 6.25 has been used to convert the nitrogen content, (deterrnined by the Kjeldahl digestion method), to an estimate of crude protein content. Although this is the factor that is routinely used for biological materials (Norton, 1989), it is believed to provide overestimates of crude protein values for rapeseed and canola (Norton, 1989). Because rapeseed and canola seeds contain substantial amounts of non-protein nitrogen (e.g. soluble amino acids, nucleic acids, phospholipids), lower factors in the order of 5.53 (Tkachuk, 1981), 23 5.67 (Norton, 1989), 5.7 (Bell, 1990) or 5.8 (Niewiadomski, 1990) have been suggested to be more accurate, although official acceptance, and thus adoption by international nutritional associations has not yet been accorded. Rapeseed and canola protein has a well-balanced essential amino acid composition and is considered to be a protein of good nutritional quality (Gillberg and Tornell, 1976; Serraino and Thompson, 1984; Jensen et al., 1990; Shahidi andNaczk, 1990; Bell, 1993). The amino acid profile compares favorably with soybean protein (Yurkowski et al., 1978; Jones, 1979) and the FAOAVHO reference protein (Jones, 1979), and based on the essential amino acid index (EAAI) approach of Oser (1959), has even been determined to be equivalent to that of whole herring meal and superior to that of soybean meal (Higgs et al., 1988, 1989). The amino acid balance found within canola protein concentrate has also been described as "approaching that of human milk" (Youngs, 1991). Additionally, the amino acid composition of rapeseed/canola protein concentrate closely resembles the amino acid composition of natural prey organisms of wild salmon (Higgs et al., 1990; 1994a). The digestibility and quality of rapeseed and canola protein however is compromised by the presence of fibre, glucosinolates, phenolics and phytic acid (Nwolkolo and Bragg, 1977; Ramachandra et al., 1977; Sosulski, 1979; Reddy et al., 1989; Bell, 1990; Bell, 1993). Rapeseed meal has less available protein than canola meal, while both meals have lower available protein than soybean meal or fish meal (Higgs et al., 1990). However, the protein present in protein products which have been nutritionally upgraded (e.g. ungraded meals, concentrates, isolates) to contain lower amounts of intrinsic fibre and/or antinutritional factors, is more available. Indeed, as measured by protein efficiency ratio (PER), rapeseed and canola concentrates have been consistently shown to be equivalent or comparable to soy protein concentrates and methionine- supplemented casein when incorporated into monogastric diets (Youngs, 1991). Similarly, canola/rapeseed flours and concentrates have yielded protein efficiency ratios with rats exceeding those obtained with soybean meal (Bell, 1990). Likewise, higher protein availabilities of further refined canola and rapeseed protein products for fish have also been demonstrated. In fact, protein digestibility coefficients for rapeseed protein concentrate in rainbow trout have been determined to be > 89.0 %, which is essentially equivalent to the average protein 24 digestibility coefficient for fish meal in Pacific salmon (i.e. 88.1 %) (Higgs et al., 1995). Comparatively, the protein digestibility coefficients for standard rapeseed meal and canola meal in trout may be as low as 63.8 % and 83.2 %, respectively. Similarly, a protein digestibilty coefficient found for rapeseed protein concentrate in chinook salmon (Oncorhynchus tshawytcha) has been determined to be as high as 95.6 %; which is substantially higher than the 84.5 % protein digestibility coefficient found for commercial canola meal in the same species (Hajen et al., 1993b). Additionally, it has been shown in pigs that yellOw rapeseed hulls have substantially higher protein digestibility (20 %) values than dark hulls (0.0 %) (Downey and Bell, 1990; Bell, 1993). Although the digestibility of yellow and dark colour rapeseed/canola hulls have not been evaluated in fish, based on the findings with pigs, evaluation of the digestible energy value of meals derived from yellow coloured seeds for fish is worthy of further investigation. 2.1.5.2 Carbohydrates The major carbohydrates found in rapeseed and canola meals may be divided into soluble carbohydrates or oligosaccharides and insoluble carbohydrates or polysaccharides. The soluble carbohydrates include the sugars sucrose, raffinose, stachyose, galactinol and digalactosyl glycerol which individually represent 7.0-7.8%, 0.33%, 2.5%, 0.1% and 0.1% DMB, respectively (Naczk and Shahidi, 1990; Bell, 1993). Together they constitute approximately 10.03-10.43 % DMB of the meal (Nackz and Shahidi, 1990), however estimates have been known to vary as high as 13.56 % DMB depending on the method of analysis used (Rao and Clandinin, 1972; Naczk and Shahidi, 1990). Of the soluble carbohydrates present in rapeseed and canola, sucrose has been repeatedly shown to vary in concentration with seed colour. Positively correlated (r = 0.66) with the incidence of yellow seed coats in canola varieties (Love et a l , 1990), sucrose concentrations on average are higher in meals derived from yellow seeded varieties (e.g. 8.7 %) than in meals derived from brown seeded varieties (e.g. 7.5%) of B. rapa, B. juncea or B. carinata (Love et al., 1990; Slorninski et al., 1995). Unlike sucrose, stachyose and raffinose, also referred to as "problem sugars", are soluble carbohydrates which are neither digested nor absorbed by monogastric animals due to a lack of the alpha-25 galactosidase enzyme, which is necessary for their hydrolysis (Lim and Akiyama, 1992). Passed into the large intestine, microbial fermentation converts the sugars to carbon dioxide and hydrogen, causing flatulence and discomfort (Lim and Akiyama, 1992). The insoluble carbohydrates that are present in rapeseed and canola are primarily comprised of structural high molecular weight carbohydrates which include cellulose, pectins, hemicellulose and starch at levels of 4-5%, 4-5%, 3 %, and < 1% DMB respectively (Naczk and Shahidi, 1990). Most of these and other insoluble carbohydrates found in rapeseed and canola are of little nutritional value to monogastrics such as salmonoids. Collectively, cellulose, hemicellulose, lignin, pentosans and other complex carbohydrates constitute dietary fibre (NRC, 1993). In rapeseed and canola meals crude fibre represents approximately 12 -15% DMB (Bell, 1993; Youngs, 1991), which is considerably higher than the fibre content of soybean meal (7 %) (Downey and Bell, 1990). More refined protein products of rapeseed and canola however have crude fibre contents that are lower, being in the order of of 7.10 % DMB for upgraded meals and 5.0 - 9.0% in concentrates (Jones, 1979; Youngs, 1991). Most of the fibre in canola and rapeseed protein products is derived from the hulls which represent 10-20% of the seed weight (DMB) (Luhs and Friedt, 1994) but constitute 27-30% of the oil-extracted meal (DMB) (Bell and Shires, 1982; Shahidi, 1990). Processing methods such as screening or air classification have been effective in reducing the fibre content of meals. Selective breeding may also provide a means for decreasing the fibre content of rapeseed and canola, for like sucrose, total fibre content appears to be associated with seed coat colour, wherein meals produced from lighter colour seeds are lower in fibre than are meals produced from darkly coloured seeds (Downey and Bell, 1990; Love et al., 1990; Slominski et al., 1995). Substantial amounts of fibre are also found within the embryo of rapeseed and canola seeds (Downey and Bell, 1990). Consequently, in terms of the evaluation of the fibre content of canola and rapeseed, neutral detergent fibre (NDF), which accounts for this fibre, is considered to be a more useful indicator of fibre content than are acid detergent fibre (ADF) or crude fibre detenninations (Bell and Keith, 1989; Downey and Bell, 1990). NDF values correlate better with apparent digestibility of protein (Bell and 26 Keith, 1989; Downey and Bell, 1990), and have been shown in pigs fed rapeseed meal, to be inversely related to both protein and energy digestibility (Downey and Bell, 1990). Furthermore, it has been said that for oilseeds there is no direct relation between crude fibre or ADF and the digestibility of nutrients (Vohra, 1989). In terms of utilization by salmonids and other monogastric animals, fibre is indigestible in the absence of bacterial action (NRC, 1993), and is known to decrease the bioavailability of minerals by binding minerals intrinsic and exogenous to the meal (Nwolkola and Bragg, 1977). It decreases protein digestibility and, known to have depressive effects on metabolizable and digestible energy (Bell, 1990), it is often described as the major factor restricting the available energy of the meals for monogastric animals (Bell, 1990;1993). Additionally, fibre is known to affect gastric emptying wherein insoluble fibres (e.g. cellulose, most hemicellulose and lignin) appear to increase gastric flow rates, leading to decreased nutrient absorption (Krogdahl, 1989). Conversely, soluble fibre (e.g. some hemicellulose, pectins, mucilages, oligosaccharides) which has been shown to inhibit digestive enzymes, tends to reduce the rate, leading to decreased feed intake (Krogdahl, 1989). In salmonids, the varied negative effects of fibre are evident at dietary levels exceeding 10%, and may include reduced growth, feed conversion and diet digestibility (Higgs et al., 1988; 1989; 1994a). Crude starch is also poorly digested by salmonids, and its presence is known to decrease the digestibility of other carbohydrates. However extrusion has been shown to improve starch digestibility (Pfeffer, 1994). Gelatinization is also recognized to improve starch digestibility (Bergot and Breque, 1983; Pfeffer, 1994) and this has been shown by Cho and Slinger, (1979) who found that gelatinized starch . had 75% higher digestibility for rainbow trout than did raw starch. It has been proposed that the improved digestibility of gelatinized starch is due to a difference in behaviour of amylase, whereby amylase binds to raw starch but is merely adsorbed onto gelatinized starch (Spannhof and Plantikow, 1983). Overall, salmonids are particulary sensitive to dietary carbohydrate. Unlike terrestrial animals which utilize carbohydrates as primary sources of metabolic energy, fish preferentially utilize protein and lipids as energy sources, having no demonstrated requirement for carbohydrates (NRC, 1993; Wilson, 1994). Inasmuch as carbohydrates are poorly utilized by all fish, coldwater fish such as salmonids appear 27 to digest carbohydrates less efficiently than warmwater fish (Wilson, 1994), a difference which is thought to be related to amylase activity which can be 10-30 times greater in fish such as catfish, than in fish such as rainbow trout (Hofer and Strumbauer, 1985; Wilson, 1994). Maximum tolerable levels of digestible carbohydrates vary among salmonid species between 15-30% (Bergot, 1979), with excessive levels being known to cause abnormally high levels of glycogen in the liver, depressed growth and increased mortality (Palmer and Ryman, 1972; Spannhof and Plantikow, 1983; Cho, 1990). The relative intolerance of salmonids for carbohydrates within the diet is thought to resemble the human condition of non-insulin dependent diabetes mellitus (NRC, 1993). In additon to a low amylase activity, the salmonids' poor utilization of carbohydrates is also attributed to relatively low hexokinase activity and an apparent lack of an inducible glucokinase (NRC, 1993; Wilson, 1994). The effective utilization of carbohydrates by salmonids also appears to be a function of the nature and complexity of the carbohydrate (Bergot, 1979; Hilton and Atkinson, 1982; Spannhof and Plantikow, 1983; Wilson, 1994), wherein alpha-linked polymers, free glucose (Pfeffer, 1994), and monosaccharides in general appear to be the carbohydrates which are most readily digested by salmonids. Considering the limited capacity of salmonids to readily digest complex carbohydrates and metabolically utilize high dietary levels of digestible carbohydrates, reductions in overall carbohydrate content, or more specifically, reductions in fibre or starch content of rapeseed or canola meals would be anticipated to increase the feeding value, and consequently the tolerable dietary incorporation levels of such meals for salmonids. Because phytic acid, lectins, and tannins are known to be responsible for malabsorption of carbohydrates (Ramachandra et al., 1977; Sosulski, 1979; Jaffe, 1983; Thompson, 1986), careful consideration of, or a reduction in, the content of these constituents would be prudent and would also be anticipated to increase the feeding value of a diet based on rapeseed and canola protein products. 2.1.5.3 Energy The low available energy values found for rapeseed and canola meals can be ascribed to their high fibre content, low oil content and the presence of glucosinolates. 28 Fibre content is thought to be the primary factor reducing the available energy level of the meals for monogastric animals, which are unable to digest fibre (Bell, 1990). (Refer to section 2.1.5.2). The oil content of rapeseed and canola meals is deliberately kept low (~ 4 % DMB) by the oilseed crashing industry which produces the meal simply as a by-product of oil extraction processes. Available energy is further restricted by the presence of glucosinolates and products of glucosinolate hydrolysis which have been shown to inhibit digestive functions in monogastric animals (Bell, 1993). Therefore, because canola meals are lower in glucosinolates than are traditional rapeseed meals, they are recognized to be higher in available energy energy (Bell, 1990). Similarly, further refined products such as upgraded meals, concentrates and isolates of rapeseed and canola which are all lower in fibre and/or glucosinolates, and higher in protein, are also higher in available energy content. Repeatedly it has been shown that the metabolizable energy value of canola or rapeseed meal for swine or chickens is lower than that supplied by soybean meal (e.g. 2900 and 1900-2000 kcal/kg vs 3300 and 2249 kcal/kg, respectively) (Downey and Bell, 1990). The digestible energy values for rapeseed and canola protein products that have been determined for trout and salmon show considerable variability (e.g. 6.0 MJ/kg - 18.2 MJ/kg) but are found to be higher for products which have been processed further (e.g. protein concentrates) (Higgs et al., 1995). Additionally, it must be noted that yellow hulls, which were found to have significantly higher energy digestibility values than darker hulls for pigs (30.0 % vs 3.0 % respectively) (Downey and Bell, 1990; Bell, 1993), may also be shown to have higher digestibility of energy in fish than darker hulls, although this has yet to be tested. 2.1.5.4 Minerals The mineral content of rapeseed /canola meal generally exceeds that of soybean meal, with the concentrations of zinc (Zn) and iron (Fe) being slightly higher than, and the concentrations of phosphorus (P), calcium (Ca), magnesium (Mg) and manganese (Mn) being about twice the concentrations found in soybean meal (Bell, 1990). Additionally, selenium (Se) concentrations are also higher in rapeseed and canola protein products, reaching almost ten times the concentration of Se found in soybean meal (Bell, 29 1990). However mineral bioavailability is low due to the presence of fibre and phytic acid/phytate which readily bind to the minerals present in rapeseed/canola meal. Consequently, mineral bioavailability is higher in protein products that have decreased levels of fibre and/or phytic acid. 2.1.5.5 Vitamins Phenolic compounds present within rapeseed and canola protein products are known to depress vitamin digestibility, (Ramachandra et al., 1977; Sosulski, 1979), although to the extent to which digestibility may be impaired is poorly understood. Nonetheless, vitamin concentrations found within the rapeseed and canola protein products show a tendency to be higher than the concentrations found in soybean meals. The content of two vitamins, namely pantothenic acid and tluamine are often comparable, but may also be lower in concentration, in rapeseed and canola protein products than found in soybean meal. However, concentrations of niacin, riboflavin, folic acid, and pyridoxine are generally higher in the former products than in the latter (Bell, 1990; Downey and Bell, 1990). Similarly, concentrations of vitamin E, an antioxidant, and biotin are also substantially higher than the concentrations found in soybean meal (Bell, 1990; Downey and Bell, 1990). Additionally, choline is consistently found in higher concentrations in rapeseed/canola meal (approximately 6700 mg/kg dry matter basis) than in soybean meal (approximately 2800 mg/kg dry matter basis) (Bell, 1990; Crimson et al., 1991; Canola council of Canada, 1994). As choline supplementation represents a significant expense in commercial production of poultry (estimated requirement for broilers 1000-1500 mg/kg), canola meal use within formulated poultry diets could be a prudent choice (Crimson et al., 1991). However, the bioavailability of total choline in canola meal has not yet been accurately established (Crimson et al., 1991). Additionally, some of the choline found within rapeseed/canola protein products may be in the form of sinapine, the choline ester of sinapic acid, which has been associated with egg-taint in brown egg shell laying hens (Blair and Reichert, 1984), and with reduced palatability of diets containing these products in salmonids (McCurdy and March, 1992). (Refer to section 2.2.7.6.3). Therefore the potential benefit or availability of choline content should be established. 30 2.1.5.6 Lipids The oil content of rapeseed and canola seeds represents approximately 80% of the agronomic value of the crop (Downey arid Robbelen, 1989). Typically, seed oil yield is approximately 40 % (Bell, 1990; Shahidi, 1990), although yields as low as 32% (from plants grown at a high temperature of 26.5°C) (Stefansson, 1983), and as high as 52% (from plants grown at a low temperature of 10°C) (Stefansson, 1983) have been recorded. However, as mentioned previously, the oil content of rapeseed and canola meals is restricted by the oil crashing industry to 4 %. Consequently, the contribution of the oil component of rapeseed and canola protein products to fish diets is very low, hence the energy and fatty acids contribution is also low. There is therefore no need to discuss these factors further. 2.1.5.7 Antinutritional factors Antinutritiorial factors (ANFs) have been identified as compounds that have a depressing effect on nutrient digestion and/or utilization of feedstuffs and thus on the biological performance of animals (Huisman and Tolman, 1992). It is important to note, however, that the ANFs present within feedstuffs may have specific beneficial functions within the plant. As with many plant-derived feedstuffs, rapeseed and cariola protein products contain many compounds which, in sufficient quantities, may be considered to be anitinutritive. However, the primary ANFs which are thought to be significant in adversely affecting the feeding value of rapeseed and canola protein and oil products are: erucic acid, glucosinolates, saponins, tannins, sinapic acid, and phytic acid. Fibre could also be considered an antinutritive factor, but its main adverse effect is as a diluent, limiting the concentration of nutrients, particularly energy and protein. There is extensive literature on the ANFs in rapeseed and canola. This review will concentrate mainly on phytic acid, but will first present a brief summary of the other ANFs. 2.1.5.7.1 Erucic Acid Erucic acid (C22i;9; c/s-13-docosenoic acid) (Salunkhe, et al., 1992) is a fatty acid found within the seeds of rapeseed and canola (Salunkhe et al., 1992; Bell, 1993). Its concentration is controlled by the 31 genotype of the developing seed (Grice and Heggtveit, 1983), pod location and plant growing conditions. Reduced concentrations of erucic acid are found in seeds from pods located lower rather than higher on the plant (Bechyne and Kondra, 1970). Depressed concentrations may also be found in seeds taken from plants subjected to low light intensities, high temperatures and/or drought (Stefansson, 1983). Unevenly distributed within the seed (Hofsten, 1970; Niewiadomski, 1990), erucic acid is found at a level of approximately 30% in the seed coat, 30% in the hypocotyl, and > 40% in the cotyledon (Appelqvist, 1971). Erucic acid is also found within the oil of traditional rapeseed as well as in the oil of canola and LEAR rapeseed varieties at levels of 20-60% (Niewiadomski, 1990; Salunkhe et al., 1992; Bell, 1993) and < 2 %, respectively. Despite the virtual elimination of erucic acid in LEAR or canola varieties, erucic acid may still be present in substantial amounts within processed canola and double low rapeseed oils due to the common occurrence of the seeds of stinkweed (Thlaspi arvenese), (42% concentration of erucic acid within the seeds), as dockage within the canola or LEAR crop harvest (Daun, 1983). However, because of the low levels of residual oil in meals and concentrates of rapeseed and canola, erucic acid is not a nutritional concern in these products. 2.1.5.7.2 Glucosinolates Glucosinolates (formerly called thioglucosides) (van Etten and Tookey, 1983), are sulphur rich glycosides which are associated with the sharp, "hot" or biting tastes of condiments and vegetables such as mustard (Sinapis sp.), horseradish (Amoracia lapathifolia), radish (Rapanus sativus), water cress (Nasturtium officinalis), rutabaga (B. nigra) and pak choi (B. chinensis) as well as oilseeds such as crambe (Crambe abyssinica), rapeseed and canola (B. rapa, B. napus, and B. juncea) (Cheeke and Shull, 1985). They are strongly acidic and strongly hydrophilic anions which are relatively unstable at extreme pH values, particularly alkaline extremes (Cheeke and Shull, 1985; Sorensen, 1990). Within the intact seed or plant, glucosinolates commonly occur as potassium salts (Cheeke and Shull, 1985; Niewiadomski, 1990). They are more highly concentrated in seeds rather than vegetative tissue and steadily increase in total concentration up to the time of harvest (Cheeke and Shull, 1985; Niewiadomski, 1990). Found dispersed throughout the parenchymous tissue, or neatly packaged within 32 specialized cells of the vegetative tissue known as idioblasts (Langer, 1983), glucosinolates co-occur, but are spatially separated from isoenzymes of thioglucoside glucohydrolase EC 3.2.3.1 (thioglucosidase glucohydrolase; thioglucosidase; formerly myrosinase) (Langer, 1983; van Etten and Tookey, 1983; Sorensen, 1990), which are capable of glucosinolate hydrolysis (van Etten and Tookey, 1983; Cheeke and Shull, 1985; Sorensen, 1990). Glucosinolate hydrolysis may also be initiated by non-enzymatic means however, and whether generated by enzymatic or non-enzymatic means, the types of hydrolytic products of glucosinolates that result are in part determined by the types of glucosinolates present, the treatment given to plant material prior to hydrolysis, as well as the conditions of hydrolysis (Langer, 1983). Thought to be activated in part by vitamin C (Langer, 1983), thioglucoside glucohydrolase is most active above 50 °C at moisture levels above 6 % (Beach, 1983). It is readily deactivated by moist (>6 %) heat administered at high temperatures (>75° C) but under low moisture conditions (< 6 %) it has been shown to resist deactivation, and retain activity at temperatures as high as 88 °C-93 °C (Beach, 1983). Upon freezing, crushing or mastication of plant or seed tissue, thioglucoside glucohydrolase is released, hydrolyzing the relatively innocuous intact glucosinolates to toxic degradation products, toxic products which are evident in both the oil and protein products of rapeseed and canola (Jensen et al., 1990). The degradation products that are produced by enzymatic hydrolysis are glucose, the acid sulfate ion, and thiocyanates, isothiocyanates or nitriles (van Etten and Tookey, 1983; Cheeke and Shull, 1985; NRC, 1993). Other enzymatic products include goitrin (5-vinyloxazolidine-2-thione; VOZT; VOT), progoitrin and epigoitrin which may be converted to goitrin (Cheeke and Shull, 1985). Non-enymatic hydrolysis of glucosinolates may be induced by the manipulation of pH whereby low, acidic pH promotes the production of nitriles and a higher pH (> pH 7.0) promotes the production of isothiocyanates (Langer, 1983; Bell, 1993). Nitriles are also known to be more readily generated in plants and seeds that have been freshly crushed rather than after the tissues have been heated (van Etten and Tookey, 1983). Thiocyanates and isothiocyanates yield the thiocyanate ion (SCN), which liberates accumulated icdine within the thyroid and inhibits further thyroidal uptake of iodine, ultimately resulting in depressed growth rate, hyperplasia, and hypertrophy of the thyroid (van Etten and Tookey, 1983; Niewiadomski, 33 1990). These goitrogenic effects are most pronounced when dietary iodine is low (Cheeke and Shull, 1985), but may be reduced and overcome with the addition of iodine to the diet (van Etten and Tookey, 1983; NRC, 1993). Nitriles also cause poor growth, and are also responsible for the development of liver and kidney lesions and liver necrosis that cannot be corrected by supplementation of the diet with iodine (Cheeke and Shull, 1985). Goitrin acts by inhibiting the incorporation of iodine into precursors of thyroxine (Tt), and thus interferes with T» synthesis and secretion (van Etten and Tookey, 1983; Cheeke and Shull, 1985). In an attempt to respond by secreting more T t , the gland becomes enlarged, forming what is referred to as a goitre (Church and Pond, 1974). The toxic effects are irreversible and are manifested as reduced growth rate as well as hyperplasia and hypertrophy of the thyroid (Cheeke and Shull, 1985). These effects have been demonstrated in poultry (Bell, 1993) and swine (Bell, 1993). The thyrotoxic effects of glucosinolates have also been demonstrated in trout (Yurkowski et al., 1978; Hardy and Sullivan, 1983; Hilton and Slinger, 1986), and salmon (Higgs et al., 1979; 1982; 1983), who as salmonids, appear to be especially sensitive to glucosinolate content. In this regard rainbow trout (Oncorhynchus mykiss) appear to be much more sensitive to glucosinolate content than other salmonids such as chinook salmon (O. tshawytschd) (Higgs et al., 1983), having an estimated maximum tolerance for total glucosinolate levels below 164 ug/g feed (Hilton, 1991), or 172 ug/g 3-butenyl isothiocyanate (Hilton and Slinger, 1986) (158 u/g air -dry basis) (< 1.40 pjmol/g) (Higgs et al., 1995) compared to a tolerance of 300 ug/g of 3-butenyl isothiocyanate for chinook (< 2.65 umol/g dry diet) (where 3-butenyl isothyocyanate equivalents measured in grams is the sum of 3-butenyl, 4-butenyl, 2-hydroxy-3-butenyl and 2-hydroxy-4-pentenyl glucosinolates x 113) (Hardy and Sullivan, 1983). Of the more than one hundred glucosinolates that have been identified (S0rensen, 1990; Bell, 1993), approximately thirty have been detected in rapeseed varieties (Sorensen, 1990). Of these, the five that are usually measured are sinigrin, gluconapin, glucobracissicanapin, epigoitrin/goitrin and napoleiferin, (allyl-, 3-butenyl-, 4-pentenyl-, 2-hydroxy-3-butenyl- and 2-hydroxy-4-petenyl- glucosinolates respectively). Due to the potential toxicities of degradation products of glucosinolate hydrolysis, low 34 glucosinolate rapeseed and canola varieties were developed through plant breeding. In addition, heat treatment, moisture adjustment to deactivate thioglucoside glucohydrolase, and water extraction steps to remove soluble glucosinolates, were incorporated into processing schemes used to produce protein products (refer to section 2.2.6). Rapeseed varieties low in glucosinolate content have been produced by genetically blocking the biosynthetic pathway of the production of aliphatic glucosinolates from methionine (Downey, 1990). In North America, canola and double low rapeseed varieties were originally developed from the Bronowski cultivar of Brassica napus, which was shown to contain as little as 10% of the glucosinolate level found in other traditional rapeseed cultivars (Stefansson, 1983). Through genetic selection, glucosinolate content in canola and other double low rapeseed varieties, was reduced to < 30 umoles aliphatic glucosinolates in the air-dried, defatted meal, representing a substantial reduction from the 110-150 umol aliphatic glucosinolates/ g commonly found in air-dried, defatted meal derived from traditional rapeseed varieties. Due the residual amounts of glucosinolates remaining in low glucosinolate rapeseed and canola varieties, thioglucosidase glucohydrolase must be inactivated during processing, in order to decrease the occurence of antinutritive and goitrogenic degradation products which could still form in the meal, concentrate or isolate (refer to section 2.2.6). The structural variation of glucosinolates is due mainly to the different side chains found in the molecule, which may exhibit more or less of a resemblance to their biosynthetic amino acid precursors (Sorensen, 1990). Quantitatively, the predorninant glucosinolates found in Brassica sp. are biosynthetically derived from methionine, phenylalanine and tryptophan, with the majority being derived from methionine (Sorensen, 1990). However, as they relate to rapeseed and canola, the glucosinolates are usually divided into aliphatic (alkenyl) glucosinolates and indole (indoyl) glucosinolates, which are primarily derived from metmonine and tryptophan, respectively. Due to their high relative concentration and strong goitrogenic and other antinutritional effects, aliphatic glucosinolates are often considered to be of greater nutritional consequence than are indole glucosinolates. In fact, the genetic methods which were used for the development of canola from traditional rapeseed varieties, were focused on the reduction of aliphatic glucosinolates (Bell, 1990). In particular, the 35 glucosinolates which have been markedly reduced are primarily those having butenyl and pentenyl side chains (Paik and Robblee, 1980; Cheeke and Shull, 1985), namely gluconapin, progoitrin, glucobrassicanapin and napoleiferin (Shahidi, 1990). By definition, canola contains less than 30 umoles of any one, or combination of these aliphatic glucosinolates (Shahidi, 1990). However, low levels of 10-12 umol aliphatic glucosinolates/ g of air-dried, defatted canola meal are considered to be typical of most varieties, representing at least an 80-90% reduction and thus improvement in the level of aliphatic glucosinolates (110-150 umol/ g of air-dried, defatted meal) found in traditional rapeseed (Bell, 1993). Through further genetic manipulation and selective breeding, canola varieties which contain less than 2 umol of aliphatic glucosinolates/g air-dried, defatted meal have also been developed (Downey, 1990). However, despite the reduction in aliphatic glucosinolates, indole glucosinolates have not been reduced. Consequently, their relative concentrations in low glucosinolate rapeseed or canola have increased, and now comprise about one third (Sorensen, 1990) to one half (Darroch and Bell, 1991) of total glucosinolate content. The toxicological effects of indoyl glucosinolates are not well understood, but they may produce thiocyanates and nitriles (e.g. acetonitriles) upon hydrolysis, depending on the pH of the medium, and the presence or absence of excess copper and iron ions which may promote acetonitrile production (Bell, 1993). In addition, thioglucosidase glucohydrolase is known to be produced by the microflora found within the gastrointestinal tract of poultry and rats (Bell, 1993). Therefore any glucosinolates remaining in the protein product have the potential for toxicity once they are subjected to hydrolysis. Perhaps it is the recognition of the potential toxicity of indole glucosinolates which has caused Europeans to include indole glucosinolates in their definition for double low rapeseed, and has encouraged them to adopt a maximum allowable level of 20 umoles of total glucosinolates/ g of air-dried defatted meal destined for feed use (Sorensen, 1990). As yet North Americans have not followed suit. Nonetheless, it is well recognized that the potential toxicity of enzymatic and non-enzymatic hydrolysis of glucosinolates is severe. Consequently, processing methodologies for the production of rapeseed and canola protein products necessarily include solvent washings (e.g. water, alcohols) and moist heat application for the reduction of glucosinolate content and the deactivation of thioglucoside glucohydrolase. (Refer to section 2.2.6). 36 2.1.5.7.3 Saponins Saponins are glycosides (Cheeke and Shull, 1985) that are valued industrially for their foaming properties. They are readily found in common plant feedstuffs such as alfalfa, rapeseed and canola, and are believed to function as a means of protection from insect pests (Cheeke and Shull, 1985). Characteristically bitter, they are often implicated in reduced feed palatability and thus feed intake but are well known for their hemolytic effects, and their ability to bind to and form insoluble complexes with cholesterol (Cooper-Driver, 1983; Cheeke and Shull, 1985). They bind bile salts that are needed for cholesterol absorption, and may actually cause bile salts to bind to polysaccharides in fibre. Saponins may also irritate and damage the intestinal mucosa (Cheeke and Shull, 1985) leading to gastrointestinal upset when ingested in large quantities (Cooper-Driver, 1983). They are also well recognized to have growth depressing effects in poultry and swine (Cheeke and Shull, 1985). Fish are known to be extremely sensitive to the detergent properties of saponins wherein upon exposure, paralysis may be induced (Cheeke and Shull, 1985). However in regard to the adverse effects of their ingestion, little is mentioned in the literature. It is likely that saponins act to reduce the palatability of rapeseed and canola protein products for fish. 2.1.5.7.4 Phenolics Described as water soluble compounds containing one or more phenol groups, plant phenolics are widely distributed throughout the plant, occurring anywhere from roots to seeds (Sosulski, 1979; Cheeke and Shull, 1985) to fruits (Cheeke and Shull, 1985). As components of feedstuffs, they have been implicated in the development of adverse flavours (bitterness and astringency), colours (darkened protein products), and odours of feeds as well as reduced digestibility and/or utilization of feed proteins and vitamins (Ramachandra et al., 1977; Sosulski, 1979). Phenolics, include well recognized compounds such as gossypol, tannins, and sinapine, (Cheeke and Shull, 1985), and may be divided into three major groups: flavenoids, phenolic acids and polyphenols (Deshpande et al., 1984; Kozlowska et al., 1990). Flavenoids, include pigments such as yellow favones and flavonols (Deshpande et a l , 1984). They are not considered 37 to be of particular significance to the feeding value of rapeseed or canola products, however. In contrast, phenolic acids and polyphenols are known to be significant antinutritive components of rapeseed and canola. Phenolic acids are found either free or esterified, or in insoluble-bound conjugated forms. The free and esterified forms predominate (Krygier et al., 1982; Deshpande et al., 1984). Free phenolic acids are antinutritive because their enzymatic oxidation yields quinones which reduce protein digestibility and utilization by covalently bonding to amino, thiol and methylene groups of amino acids in protein (Sosulski, 1979; Cheeke and Shull, 1985). Sinapic acids are some of the better known examples of free phenolics (Deshpande et al., 1984). Esterified phenolic acids include sinapine which is the choline ester of sinapic acid (3,5-dimemoxy-4-hyd^oxycimamic acid). Sinapine is found at a concentration of approximately 1% in rapeseed meal (Durkee and Thivierge, 1975; Cheeke and Shull, 1985), 1.0-2.0% in rapeseed flour (Lacroix et al., 1988), and 2.5-3.0% in canola meal (Blair and Reichert, 1984). It has been associated with reduced palatability of rapeseed and canola meals in salmonids (McCurdy and March, 1992) and other monogastric animals, but is particularly recognized for "egg taint" effects demonstrated in poultry. When included in the diet of layers as a constituent of rapeseed meal, sinapine has been implicated in the development of a fishy odour and/or flavour in the eggs of brown shelled laying breeds (Blair et al., 1975; Cheeke and Shull, 1985). The egg taint results from caecal bacterial conversion of choline released from sinapine into trimemylamine (TMA), which is deposited directly into the egg by brown shelled laying breeds (Blair et a l , 1975; Goh et a l , 1979; Cheeke and Shull, 1985). Due to a genetic anomally, such breeds lack the liver enzyme trimemylamine oxidase (Bell, 1993), leaving them with the inability to oxidize TMA to trimemylamine oxide (TMAO), which may subsequently be excreted (Blair et al., 1975; Goh et a l , 1979; Cheeke and Shull, 1985). Polyphenolic compounds considered to be of antinutritive significance include tannins, which although are perhaps better recognized as ANFs of sorghum are also significant ANFs of rapeseed and canola. Within the plant, tannins are believed to play a major role in protecting the plant from attacks by vertebrate herbivores, insects,- bacteria and viruses due to their astringency and ability to inactivate protein enzymes of insects and bacteria (Cheeke and Shull, 1985). Within the seed, tannins occur in the hull and 38 are in higher concentrations in darker hulls than in lighter coloured hulls (White, 1957; Bell, 1993). Like phytic acid, tannins are relatively heat stable (Thompson, 1986), and therefore survive common processing methods used for meal production. Tannins occur in concentrations of approximately 1.5-3.0% within rapeseed and canola meals (Leung et al., 1979; Lacroix et al., 1988; Bell, 1993). Weighing 500-6000 daltons, tannins may be subdivided into hydrolysable and condensed tannins, the former of which are more reactive (White, 1957; Alisone, 1971; Sosulski, 1979; Bell, 1993). The antinutritive effects of tannins include reduced feed and protein digestibility, depressed protein utilization and reduced weight gain (Despande et al., 1984; Krogdahl, 1989). These, adverse effects are in large part due to the strong ability of tarinins to bind to digestive enzymes (Thompson, 1986), inhibiting their activity, and to bind to dietary protein and starch (Despande et a l , 1984; Thompson, 1986), thereby preventing their digestion (Cheeke and Shull, 1985). Several functional groups within tannins enable them to precipitate proteins from aqueous solution (Cheeke and Shull, 1985). As with phenolic acid, oxidized derivatives of tannins known as quinones nonenzymatically react to polymerize or covalently link with amino, thiol and methylene groups of amino acids and arginine in protein, and through polymerization, form protem-tannin complexes which are relatively resistant to enzymatic and microbial digestion (Alisone, 1971; Cheeke and Shull, 1985). Similarly, tarinins may bind directly to digestive enzymes such as trypsin and amylase (Thompson, 1986), thereby inhibiting their activity (Cheek and Shull, 1985). Further, tannins are also known to have a strong affinity for protein rich in proline, such as collagen from animal tissues and the alcohol-soluble storage proteins of cereals (Thompson, 1993). However, as these proteins are characteristically deficient in essential amino acids, the adverse effects of these particular binding properties of tannins on protein digestion are considered to be relatively low (Thompson, 1993). Additionaly, antinutritive effects may also result from reduced feed palatability and consequently reduced feed intake due to the bitter, astringent sensation caused by the reaction of tannins with proteins or glycoproteins within the mouth (Cheeke and Shull, 1985). The nutritional importance of dietary tannin content to fish is not well documented, seemingly limited to references to tannins' contribution to the reduced palatability of oilseed meals in fish diets (Hossain and Jauncey, 1989). However tannins are believed to provoke many of the same responses in 39 fish as those found in other monogastric animals. Another well recognized polyphenolic compound is lignin. Lignin is not digested by monogastrics and, as a component of fibre, effectively acts as a diluent in protein products. Like tannins, lignin is found in significantly higher concentrations in darker coloured hulls than in yellow or lighter coloured hulls (Love etal., 1990; Bell, 1993). Total phenolics may be reduced or eliminated upon extraction with methanol, methanol and water, acetone and water (Mitaru et al., 1982), or methanol, acetone and water, which act by increasing the solubilities and thus extractability of phenolic components (Lacroix et al., 1988). Additionally, the solvent combinations and processing methods currently used for rapeseed protein concentrate production reduce total phenolic content by 60-83% (Dabrowske and Sosulski, 1983). In rats, removal of phenolics from rapeseed flour has been shown to to increase the digestion and utilization of protein nitrogen to levels comparable to that obtained with casein (Lacroix et al., 1988). Similarly, the reduction in phenolic content of rapeseed and canola protein products has been shown to improve protein digestion and/or utilization in poultry, swine, and salmonids. Selective breeding of yellow seed varieties of rapeseed and canola may also serve to reduce the polyphenolic content of meals, as meals derived from yellow seeded varieties have been shown to have a lower polyphenolic content than meals derived from darker seeded varieties (Love et al., 1990). 2.1.5.7.5 Lectins Lectins are common constituents of legumes (e.g. peas, beans), nuts (e.g. peanuts) and oilseeds (e.g. soybeans, rapeseed and canola) (Jaffe, 1983). Their name, derived from the word legere which means to choose, relates to their relatively specific affinity for bmding sugars, a specificity that is often described as akin to the bmding of an antibody with its antigen (Jaffe, 1983). Lectins may bind to free sugars or free or bound sugar residues found within polysaccharides, glycoproteins or glycolipids which may occur in cell membranes (Jaffe, 1983). Thus they may bind to cells of the intestinal tract and interfere with the absorption of nutrients. Lectins are also thought to negatively affect amylase activity (Thompson, 1986). Readily destroyed by moist heat, lectins are relatively resistant to dry heat (Cheeke and Shull, 1985). 40 Consequently, lectins, are not thought to be of any real nutritional consequence with regard to rapeseed or canola products, due to the fact that moist heat application is a standard practice of the rapeseed and canola oil and meal production industry. Lectins may, however, be of some nutritional concern in the feeding of whole rapeseed or canola seeds which is sometimes practiced with the feeding of swine (Bell, 1989). 2.1.5.7.6 Phytic acid Phytic acid is the major storage form of phosphorus in cereals, oilseeds and legumes (Atkinson and Morton, 1960; Reddy et al., 1989). It has a great propensity for binding to, and forming complexes with minerals, carbohydrates, and proteins, including enzymes (Reddy et al., 1989). Within the plant or seed, this propensity contributes to its ability to play a role in the initiation of dormancy (Sobolev and Rodionova, 1966; Reddy et al., 1989; Cosgrove, 1980), and serve as a source of energy (Atkinson and Morton, 1960), cations, and myoinositol (Reddy et al., 1989). However, under physiological conditions within most monogastrics that lack the phytic/phytate hydrolyzing enzyme phytase, phytic acid is generally unavailable for absorption (Reddy et al., 1989), and its propensity for binding and forming complexes with minerals, carbohydrates and proteins, negatively affects the bioavailability of such nutrients. Consequently, animal growth, survival, utilization of feed and protein, and thyroid function are compromised. In salmon, high dietary levels of phytic acid have also been associated with the induction of pyloric cecae structural anomalies, and the development of zinc deficiencies that have resulted in cataract formation under conditions of high calcium and phosphoms concentrations (Richardson et al., 1985). Clearly, phytic acid is an antinutrient of considerable significance. 2.1.5.7.6.1 Phytic acid: structure, properties, occurrence and function As the primary storage form of phosphoms in plants such as rapeseed and canola, phytic acid (syn. myoinositol hexaphosphate; myoinositol-l,2,3,5/4,6-hexakis (dihydrogen phosphate); C6H18O24P6 ; IP6) may account for 60-90 % of phosphoms stores (Serraino and Thompson, 1984; Reddy et al., 1989; Thompson, 1990). Its biosynthesis is strongly associated with the ripening period of seeds or grains, as is evinced by rapid accumulation of phytic acid during this time (Raboy and Dickinson, 1987; Reddy et al., 41 1989; Niewiadomski, 1990). Similar to other oilseeds and legumes, phytic acid in rapeseed and canola is found mainly in the embryo (Bell, 1993), within small (0.5-2.8p), crystalline globoids located within the protein bodies of the cells of the radicle and primary cotyledon (Yiu et al., 1983; Niewiadomski, 1990; Thompson, 1990). This is unlike the situation in most cereals (eg. wheat, rice, triticale, rye), where phytic acid is largely found within the protein bodies of the aleurone layers, (pericarp) (Reddy et al., 1982; Thompson, 1990). This is also unlike the situation in soybeans for which there is apparently no specific location for phytic acid deposition (Cheryan, 1980). In addition, phytic acid localization in rapeseed and canola is distinctly different from its distribution in com, where 88-90% is concentrated in the germ portion of the kernel (Cheryan, 1980; Reddy et al., 1989) and a relatively small amount (« 3.2%) is found in the endosperm (Reddy et al., 1989). The location of phytic acid within the seed/grain may play an important role in determining how readily phytic acid is retained or lost during seed/grain processing (O'Dell et al., 1972; Ressurrenccion et al., 1979; Reddy et al., 1989). Within unprocessed whole rapeseeds/canola seeds, phytic acid is found in concentrations of 2.0-4.0% (Aman, 1970; Thompson, 1990). Unfortunately however, phytic acid is not readily lost through seed processing, except in some protocols involved in the production of isolates. In fact, with most of the methodologies now available, its concentration appears to increase upon further processing, showing a disturbing tendency to be found in higher concentrations in many of the protein products which have been "nutritionally enhanced" to contain higher concentrations of protein. Typically phytic acid is found at levels of 2.0-5.0% in meals (Nwokolo and Bragg, 1977; Alii and Houde, 1986; Tzeng, 1987; Tzeng et al., 1988a; 1988b; 1990; Thompson, 1990), * 5.5 % in upgraded meals (Erdman, 1979; McCurdy and March, 1992), 3.6-7.5% in concentrates (Shah et al., 1976; Jones, 1979; Thompson, 1990; Youngs, 1991) and <1.0-9.8% in isolates (Gillberg and Tornell. 1976; Liu et al., 1982; Thompson et al., 1982; Serraino and Thompson, 1984; Thompson and Cho, 1984; Thompson, 1987; 1990; Tzeng et al., 1988a, 1988b; 1990). Following many decades of vigorous scientific debate about the true structure of phytic acid, the "Anderson model", depicted in Figure 2.3 , has emerged as the proposed structure most favored in the literature (Reddy et al., 1989). The model is believed to better account for the known physiochemical 42 properties, interactions and nutritional effects of the molecule (Cheryan, 1980; Reddy et al., 1989). Fig. 2.3 The Anderson model of phytic acid (Richardson, 1986; Reddy et al., 1989). However, another structure known as the "Neuberg model", deserves mention. It differs from the Anderson model by having three water molecules attached to adjacent pairs of phosphate groups by P-O-P linkages (C6H18O24P6 3 H 2 O ) (Cheryan, 1980; Reddy et al., 1989). Supporters of this model propose that not only does it exist, but that the Anderson model is simply a degradation product of the water-laden Neuberg model (Brown et al., 1961; Cheryan, 1980). Nonetheless, at present, the Anderson model is more widely accepted, and thus will be accepted as the true structure of phytic acid for this thesis. 2.1.5.6.6.2 Phytic acid interactions Phytic acid's high propensity for binding to and forming complexes with multivalent cations, proteins and enzymes is related to the wide range of dissociation constants of the twelve dissociable protons of the six phosphate groups of the molecule. As shown by Costello et al. (1976), six protons may dissociate in strong acid pHs (pKa 1.1-2.1), one dissociates in the weak acid range (pKa 5.70), two dissociate at pKa 6.80-7.60, and three dissociate within an extremely weak acid range, (pKa 10.0-12.0) (Costello et al., 1976; Reddy et al., 1989). As a consequence, the phytic acid molecule is strongly negatively charged over a wide range of pH values, enabling it to readily interact with positively charged moieties (Spinelli et al., 1983; Reddy et al, 1989; NRC, 1993). Phytic acid may exist as a free acid, salt or mixed salt, and accordingly is known as phytic acid, 43 phytate and phytin, respectively, terms which make specific reference to the particular state of the molecule. Although specifically defined, the terms are not always used in a consistent manner. Phytic acid is the term correctly used to refer the the free acid form of phytic acid (Reddy et al., 1989). Phytate is the term which should be used to refer to the stable, dry salt form of the molecule (Reddy et al., 1989). Phytin, has been recognized to refer specifically to the mixed calcium and magnesium salt of the phytic acid molecule (Wise, 1983; Reddy et al., 1989). However, repeatedly, the terms have been used interchangeably, most often with phytic acid and phytate being used in each other's place. Similarly, phytin is not always used as defined, for in some instances it has been found to refer to non-specific mixed salts of the molecule, rather than the specified calcium and magnesium salt (Reddy et al., 1989; Niewadomski, 1990). Consequently, some care must be exercised when discussing phytic acid and its interactions. 2.1.5.7.6.2.1 Phytic acid and minerals Dietary fibre, polysaccharides and polyphenolic compounds play major roles in the bioavailability of minerals in plant feedstuffs to fish and other non-ruminants, but phytic acid is also well recognized to play a primary role, due to its tendency to complex with minerals over a wide range of pH (Reddy et al., 1989). Each molecule is capable of binding up to six multivalent cations, but it may also be bound to two or more phytic acid molecules via cationic bridges (Graf and Eaton, 1990). The extent of bmding of phytic acid to minerals, and the solubilities of the resultant complexes are influenced by the pH of the medium (e.g. maximum solubility of calcium and magnesium salts is reached at pH 5.8 (Serraino and Thompson, 1984; Maddaiah et al., 1964; Reddy et al., 1989), the concentration and nature of phytic acid and phytate complexes, and the type and concentration of cations present (Serraino and Thompson, 1984; Reddy et al., 1989). In terms of general trends, equimolar concentrations of catiomphytic acid are found to be highly soluble over a wide range of pH (Graf and Eaton, 1984), and have even been said to be highly soluble at any pH (Graf, 1986). By contrast, wider molar ratios potentiate insolubility (Graf and Eaton, 1984; Graf, 1986; Reddy et al., 1989). It is also generally recognized that mono- and di-cationic phytates tend to be more soluble than polycationic phytates (Graf and Eaton, 1990). Consistency in the order of stability of phytate complexes, as a function of the particular type of 44 mineral that is bound does exist. This has been demonstrated by Vohra et al., (1965), (as cited by Reddy et al., 1989) who found the order of stability to be: Cu 2 + > Zn 2 + > N i 2 + > Co 2 + > Mn 2 + > Fe 3 + > Ca 2 + at pH 7.45 (Reddy et al., 1989; Graf and Eaton, 1990). Similarly, Maddaiah et al., (1964), although not * 2"t~ 3"h 2~h 2"h evaluating Ni or Fe , found the order to be the same except for the reversal in order of Zn and Cu . Although absent from the above comparisons, magnesium is recognized to readily bind to phytic acid, forming the phytin complex. Similarly, potassium is readily bound by phytic acid in the calcium-magnesium-potassium phytate complex commonly found in the aleurone layer of wheat and rice (Lott and Ockenden,1986; Graf and Eaton, 1990). Additionally, phytic acid is known to adsorb to mineralized tissues such as hydroxyapatite (Ca5(P04)30H) (Graf, 1986; Kaufman, 1986; Graf and Eaton, 1990), to which lower phosphoric esters of myoinositol (mono-, di-, tri-, tetra- and penta-) also readily bind (Graf, 1986). The formation of the relatively insoluble mineral complexes contributes to the reduced bioavailability of minerals in phytate-containing feedstuffs, for unlike ruminants, most monogastrics lack phytase or phytase-producing microflora in the gastrointestinal tract. Consequently, there is negligible hydrolysis of phytic acid or dissolution of phytate complexes, preventing subsequent absorption or utilization of bound minerals. Regular ICP (inductively coupled plasma emission spectrophotometry) analysis of feedstuffs makes no distinction between free mineral content or mineral-phytate complexes. Therefore, there can be great disparity between analyzed mineral content and biologically available mineral content. Feedstuffs must be specifically analyzed for inositol phosphate content (IP) in order to provide some indication of the potential problem of reduced mineral bioavailability, either due to the presence of phytate-mineral complexes within the feedstuff prior to ingestion, or due to post ingestive formation of complexes within the intestinal tract. When interpreting studies concerned with the estimation of mineral bioavailability to non-ruminants in phytic acid-containing feedstuffs, consideration must also be given to the actual source of phytic acid, for added sodium phytate reacts differently from naturally occurring phytate (Cheryan, 1980; Thompson, 1986). Added sodium phytate is highly soluble and ionizable over a wide range of pH, and is 45 free to complex with minerals under intestinal conditions, rendering them biologically unavailable (Cheryan, 1980). By contrast, naturally occuring phytic acid may already be present largely as mineral-phytic acid complexes and thus is less available for further complexing with dietary minerals (Cheryan, 1980). It must be noted that the binding of phytic acid to minerals may be diminished through the addition of chelating agents such as EDTA (ethyldiaminetetraacetic acid), to which cations preferentially bind in the presence of phytic acid (Cheryan, 1980). Mineral bioavailability is improved because the formation of mineral-EDTA complexes interferes with further mineral-phytic acid complex formation. Additionally, because the mineral-EDTA complex may be absorbed directly or may release the mineral just prior to absorption, mineral bioavailability is also improved (O'Dell, 1969; Cheryan, 1980; Reddy et al., 1989). It is also important to note that in the presence of phytic acid, Ca 2 + exhibits a bimodal effect on the solubility of Zn 2 + , Cu 2 + , Fe 2 +, Fe3+, Mg 2 + , Mn 2 + , Co 2 +, N i 2 + , Pb2 +, Sr2 +, Hg 2 + as well as other cationic minerals which have higher affinities for phytic acid than calcium (Graf and Eaton, 1984; Graf, 1986), whereby low concentrations of calcium increase the solubility of the other cations and high concentrations potentiate their precipitation (Graf, 1986). 2.1.5.7.6.2.1.1 Phytic acid and phosphorus As is characteristic of oilseed meals, much of the phosphorus present in canola meal is phytate bound, and in the absence of phytase, biologically unavailable to monogastrics. Typically, total phosphorus content of canola meal is 1.22%, of which 53% is phytate phosphorus (Bell, 1993). The remaining phosphorus is present as inorganic constituents and is also found in phospholipids and nucleic compounds (Niewiadomski, 1990). Although phytate phosphorus is partially available for swine, rats and chickens that have intestinal phytase activity, for most other monogastrics, salmonids included, phytate phosphorus is totally unavailable. Phosphorus availability from canola meal has been estimated to be in the order of 35 % in the absence of phytase (Youts, 1990). By comparison, total phosphorus content of soybean meal is only 0.66% of which 38% is phytate-bound (Bell, 1993). It is estimated that approximately 21% of the phosphorus content of soybean meal is available to monogastrics lacking 46 phytase activity (Youts, 1990). Hence it is clear that canola meal provides substantially more non-phytate phosphorus than soybean meal (Bell, 1993) and, has the potential to provide additional phosphorus upon phytase hydrolysis of the phytate content. Similar to the situation with zinc, calcium is known to affect the availability of phytic phosphorus. High concentrations of calcium have been shown to enhance the binding of phytic phosphorus to proteins (Okubo et al., 1976), decreasing the bioavailability of both the protein and the phosphorus. High calcium concentrations have also been related to the development of rickets in association with low availability of phytate phosphorus (Taylor, 1979). Additionally, high dietary levels of calcium (5.1 %) have been shown to exacerbate the negative effects of high dietary phytate, on zinc bioavailability and thereby enhance the development of bilateral lens cataracts in salmonids (Richardson et al., 1985; Spinelli et al., 1983). However, the presence, or addition of vitamin D, (whose primary function is to promote calcium absorption), is suggested to enhance the utilization of phytate-phosphorus (Taylor, 1979; Irving, 1980). 2.1.5.7.6.2.1.2 Phytic acid and zinc It is well established that zinc bioavailability is affected by the presence of phytic acid. In studies in which diets have been formulated to just meet zinc requirements, the presence of phytic acid has been associated with the induced deficiency of, and/or the reduced bioavailability of zinc, in poultry (O'dell et al., 1964; Nwokolo and Bragg, 1977), swine (Smith et al., 1962), rats (Davies and Nightingale, 1975; House et al., 1982; Flanagan, 1984), trout (Spinelli et al., 1983) and salmon (Richardson et al., 1985; NRC, 1993). Zinc bioavailability in the presence of phytic acid is influenced by the presence and relative concentration of calcium (Graf, 1986). At relatively equivalent, or close Zn:Ca molar ratios (e.g. 1:1, or 2:1 ), the complexing of zinc with phytic acid is diminished, possibly through the competition of zinc and calcium for positions on the phytic acid molecule (Byrd and Matrone, 1965; Cheryan, 1980; Reddy et al., 1989). However, zinc bioavailability is further influenced by the presence of high concentrations of calcium, which potentiates coprecipitation of zinc and calcium as a Zn-Ca-phytate complex (Serraino and Thompson, 1984; Graf, 1986; Reddy et al., 1989). The resultant complex is apparently more insoluble than either Zn-phytic acid or Ca-phytic acid complexes (Reddy et al., 1989). As the reduced bioavailability 47 of zinc in phytate containing feedstuffs is further aggravated or enhanced by high dietary calcium (Fordyce et al., 1987; Reddy et al., 1989), molar ratios (moles/kg), of (phytate) (calcium)/(zinc) are suggested to be predictive of zinc bioavailability (Davies et al., 1985; Reddy et al., 1989). As demonstrated with rats, (phytate) (calcium)/(zinc) molar ratios exceeding 3.5 moles/kg facilitate zinc-related reductions in growth (Davies et al., 1985). Although thought to be of less predictive value, phytic acid/zinc molar ratios greater than 10 (Davies and Olpin, 1979; Thompson, 1990) or 12-15 (Reddy et al., 1989) have also been associated with zinc deficiency. However, the predictive value of molar ratios is limited to similarly processed products (Fordyce et al., 1987; Reddy et al., 1989), because processing protocols alter the concentration and complexing of phytate, mineral, and other feed constituents (Ketelsen et al., 1984). The presence of magnesium may also accentuate phytate's effect on zinc bioavailability (Forbes et al, 1984; Reddy et al., 1989). Phytate's negative effect on zinc bioavailability may be diminished through the pretreatment of protein products with the chelating agent EDTA (emylenectiaminetetraacetate), to which cations preferentially bind in the presence of phytic acid (Cheryan, 1980). The formation of Zn-EDTA complexes (O'Dell et al., 1964; Reddy et al., 1989) makes zinc more available for absorption either through the release of zinc from the complex prior to absorption, or through the direct absorption of the relatively soluble Zn-EDTA complex (O'Dell, 1969; Cheryan, 1980; Reddy et al, 1989). 2.1.5.7.6.2.1.3 Phytic acid and calcium The complexing of phytic acid with calcium has been associated with reduced calcium bioavailability, induced calcium deficiency (Nwokolo and Bragg, 1977; Taylor, 1979; Reddy et al, 1989) and enhanced binding of phytic acid/phytate to other minerals (e.g. phosphoms, zinc) as well as to proteins (Cheryan, 1980; Irving, 1980; Graf, 1986; Reddy et al, 1989). The solubilities of calcium-phytic acid complexes vary as a function of pH (Gifford-Steffen and Clydesdale, 1993), tending to reach a maximum at, or around pH 5.15 (Serraino and Tornell, 1984). Solubility also varies as a function of calcium:phytate ratios (Gifford-Steffen and Clydesdale, 1993). The depressive effects of dietary phytic acid levels on growth, feed conversion, protein utilization 48 and protein digestibility have been shown to be exacerbated by the presence of high calcium concentrations in fish (Hossain and Jauncey, 1993) as well as in other monogastric animals. In particular, a high concentration of calcium (5.1%) in a diet containing 2.58 % phytic acid was noted to reduce zinc bioavailability and to promote the subsequent development of cataracts in chinook salmon (Richardson et al., 1985). Additionally, high dietary calcium concentrations have been associated with reduced calcium bioavailability (Gifford-Steffen and Clydesdale, 1993). The formation of highly insoluble hexa-, penta-, tetra- and tri-calcium complexes (Gifford-Steffen and Clydesdale, 1993), with or without the incorporation of protein (Hossain and Jauncey, 1993) is thought to be the primary causative factor of calcium's ability to enhance the adverse effects of "natural" or synthetic dietary phytic acid (Hossain and Jauncey, 1993). Mono-and di-calcium phytic acid complexes also form but, by contrast, are relatively soluble (Gifford-Steffen and Clydesdale, 1993). Under conditions of low calcium, intake calcium deficiency and the development of rickets are often evident since much of the calcium is rendered biologically unavailable due to phytic acid-calcium complex formation (Taylor, 1979). As with phytic-phosphorus, these effects may be tempered with vitamin D, which is thought to enhance the utilization of phytate bound calcium (Taylor, 1979). 2.1.5.7.6.2.1.4 P h y t i c a c i d a n d i r o n Some researchers have found that the presence of phytic acid or the addition of phytic acid to feedstuffs significantly reduces the bioavailability of iron, (Davies and Nightingale, 1975; Reddy et al., 1989). Alternatively, other researchers have found that high dietary levels of phytic acid had little or no effect on iron availability (Graf and Eaton, 1984; Reddy et al., 1989). Such differences in results may be attributed to considerable variation between studies in experimental conditions and animal species, and the presence or absence of intestinal phytase (Refer to section 2.2.9) (Reddy et al., 1989). In addition, the relatively unknown solubilities of the various iron-phytic acid complexes which form as a result of phytic acid's unique chelating abilities, may also, in part, account for the different findings. The binding of phytic acid to iron is of particular significance in regard to anti-oxidant functions within physiological systems. Existing in two different valency states, Fe2 + and the ferric ion Fe3 +, iron 49 often participates in the production of oxidative damage to biological systems (Graf, 1986). In what is known as the Haber-Weiss cycle, iron mediates the eventual production of the highly reactive oxy-radical OH, which mmscriminately attacks most biomolecules (Graf, 1986). Unlike most chelating agents, when phytic acid binds to iron it occupies all coordination sites on iron, and displaces water in aqueous solution. This action effectively makes iron catalytically inert, and maintains iron in the Fe 3 + state. Consequently hydrogen peroxide is unable to bind to Fe2+-phytate, and there is no generation of the hydroxyl radical (Graf, 1983, 1986; Graf et al., 1987; Graf and Eaton, 1990). Indeed, because iron-catalyzed generation of the oxy-radical requires the availability of at least one reactive iron coordination site (Graf and Eaton, 1990), the binding of phytic acid with iron inhibits not only the formation of the OH radical, but also the peroxidation of lipids and, furthermore, the binding lowers the oxidative-reduction potential of iron (Graf, 1986). Additionally, phytic acid has what is described as a unique ability to remove oxygen without concomitant generation of deleterious oxy-radicals (Graf, 1986). 2.1.5.7.6.2.2 Phytic acid and protein Within rapeseed, phytic acid-protein complexes are reported to occur naturally but may also form during protein extraction (Lott, 1985; Thompson, 1990). Due to the formation of protein-phytate complexes, protein bioavailability is reduced because the protein complexes are less subject to proteolytic digestion than the same protein is alone (O'Dell and de Boland, 1976; Reddy et al., 1989). Additionally, the binding of phytic acid/phytate to enzymes such as trypsin (Graf, 1986), and pepsin and amylase (Sharma et al., 1978; Thompson, 1986) reduces the effectiveness of such enzymes. The degree of bmding between phytic acid and protein is largely pH dependent, but is also a function of both the type and concentration of multivalent cations present (Serraino and Thompson, 1984), and the nature of the particular proteins under consideration (Reddy et al., 1989). Because phytic acid is negatively charged over a wide range of pH, and rapeseed proteins have wide ranging isoelectric points, protein-phytic acid binding occurs over much of the pH scale. Approximately 20-40% of rapeseed proteins have isoelectric points close to pH 11, and the remainder have isoelectric points spread between pH 4-8 (Gillberg and Tornell, 1976). (Refer to section 2.2.7.1). 50 Phytic acid's binding to protein may affect protein solubility, charge and structure (Thompson, 1987; Reddy et al., 1989). In this regard, it has been proposed that upon phytic acid/phytate binding structural changes occur in the protein (Chen and Morr, 1985). These may result in the protein solubilities being reduced below their isoelectric point (de Rham and Jost, 1979; Chen and Morr, 1985; Reddy et al., 1989) because there is the uncovering of more positive charges which may act as sites for protein hydration (Chen and Morr, 1985). At low pH (< 3.5), below the isoelectric point of many rapeseed proteins, proteins are largely positively charged, whereas phytic acid is negatively charged (Serraino and Thompson, 1984). Under these conditions the binding of Nrh + groups of protein, (free amino terminal, epsilon (e)- amino group of lysine, imidazole group of histidine, or guanidyl group of arginine), to phosphate groups of phytic acid is facilitated (Cheryan, 1980; Thompson, 1986; Reddy et al., 1989). The binding may either be by van der Waal forces, or by ionic or hydrogen bonding (Cheryan, 1980; Reddy et al., 1989). However, should a high concentration of calcium be present at acidic pH, protein-phytate complexes may be disrupted due to possible competition between calcium and the cation groups of protein (Serraino and Thompson; 1984). At neutral and intermediate pH, provided that proteins have an isoelectric pH lower than the prevailing pH, both proteins and phytic acid are negatively charged (Serraino and Thompson, 1984; Graf and Eaton, 1990). Although phytic acid and protein may dissociate (Graf and Eaton, 1990), bmding still occurs (Serraino and Thompson, 1984). Under these conditions, binding may be by direct binding of phytic acid to the ct-NH2 terminal group and 8-NH2 groups of lysine which would remain protonated (Omasaiye and Cheryan, 1979; Cheryan, 1980; Thompson, 1986; Reddy et al., 1989). Binding may also occur between phytic acid and basic proteins (Reddy et al., 1989), or between phytic acid and any remaining protonated lysyl and arginyl protein side chains (Thompson, 1987; Reddy et al., 1989). In addition, binding may be mediated by a multivalent cation, whereby the cation forms bridges between the phosphate groups of phytic acid and the carboxyl or histidyl groups of the protein that remain protonated (Omasaiye and Cheryan, 1979; Cheryan, 1980; Reddy et al., 1989). At high pH (>9.0), interactions between phytic acid and protein occur at reduced rates, but the nature of the interactions are not well understood (Cheryan, 1980; Reddy et al., 1989). 51 As previously stated, the binding of phytic acid to protein is not just a function of pH, but also of the cationic environment (Serrainb and Thompson, 1984), whereby the solubility of protein-phytate complexes is affected by the type of mineral that is found within and without the complex. Calcium and magnesium appear to be of particular significance wherein experimentation with soy proteins suggests that protein-phytate complexes which include bound calcium and/or magnesium may be less soluble than protein-phytic acid/phytate complexes lacking these minerals (de Rham and Jost, 1979). Supportive evidence for this includes experiments with soy proteins which found that additions of EDTA at concentrations above those of calcium and magnesium, (which normally induces the dissociation of mineral-phytic acid complexes), failed to dissociate the soy protein-phytate complex (de Rham and Jost, 1979). That is, the cations which tend to preferentially bind to EDTA in the presence of phytic acid (Cheryan, 1980), whether individually or together, showed a greater affinity for the protein-phytate complex than for EDTA (de Rham and Jost, 1979). In addition to differential binding according to pH, there may be differential binding of phytic acid to different conformational forms of protein (Reddy et al., 1989). This has been demonstrated with soy proteins wherein the 1 IS dimer, glycinin, was observed to bind less phytate than the 7S dimer (Reddy et al., 1989). 2.1.5.7.6.2.3 Phytic acid and starch The presence of phytic acid is thought to reduce the digestion and absorption of starch (Thompson, 1986;1990; Reddy et al., 1989). Proposed mechanisms for this effect include: the reduction of amylase activity due to phytic acid's chelation of calcium (whose halogen salts serve as the enzyme's catalysts), the binding of phytic acid to the enzyme itself, which is a protein (Thompson, 1986), and/or the possible binding of phytic acid to starch (Thompson, 1990). Starch is proposed to be bound directly to protein which is, or then becomes, complexed with phytic acid (Thompson, 1990). However, although theoretically possible, phytic acid's binding to starch has not yet been confirmed (Thompson, 1990). In a rapeseed or canola diet the nutritional importance of starch bmding may not be great, for unlike cereals, relatively little starch is present. Although starch content may reach 50% in the immature 52 seed, starch is gradually converted to oil as the seed matures (Bell, 1993). Consequently, a relatively small amount of starch remains in canola meal (< 3%) (Bell, 1993). In a mixed diet however, potentially more starch from other dietary constituents could be available for binding, and the finding of phytic acid and starch could be of greater nutritional significance. 2.1.6 Reduction and removal of phytic acid The reduction or removal of phytic acid from plant feedstuffs has been associated with improved bioavailability of protein, carbohydrates and cationic minerals such as phosphorus and zinc, whose dietary deficiencies have often been associated with high phytic acid content. Methods for phytic acid reduction or removal are varied and include, mechanical processes, germination, selective extraction, membrane filtration, ion exchange methods and, enzymatic treatment methods which utilize extrinsic and intrinsic sources of phytase (Cheryan, 1980; Reddy et al., 1989). Processing conditions, extracting agents and the form and localization of phytate are factors which influence the effectiveness of the various dephytinizing procedures (O'Dell et al., 1972; Cheryan, 1980; 1982; Naczk et al., 1986; Reddy et al., 1989). 2.1.6.1 Mechanical means of phytic acid removal The phytic acid and phytate found in cereals is concentrated in the outermost layers of the kernel (bran) and may be effectively removed in significant amounts by milling (O'Dell et al., 1972; Reddy et al., 1989). The polishing of rice also removes appreciable amounts of phytic acid and phytate (Ressurrenccion et al., 1979; Reddy at al., 1989). However, such processing methods are relatively ineffectual in the removal of the phytic acid and phytate found in dry beans and rapeseed and canola, as concentrations are highest deep within the cotyledons (Reddy et al., 1989). In fact, the milling of these seeds may actually increase the concentration of phytate within the flours produced from the seeds (Reddy et al., 1989). 2.1.6.2 Germination During germination, the phytic acid/phytate content of seeds may be reduced through the hydrolytic action of the enzyme phytase. The extent and speed of hydrolysis is dependent on the germination 53 conditions and the amount of phytase activity inherent within the seed. Phytase content and activity is species dependent, and varies widely as illustrated by five day germination of soybeans arid Alaska peas which showed an approximate 3-fold and 3 8-fold increase in phytase activity respectively (Chen and Pan, 1977 according to Reddy et al., 1989). Activity is measured in phytase units (PIT), wherein each unit is defined as the activity that liberates 1 jumol of inorganic phosphorus from an excess of sodium phytate/ minute at 37° C and pH 5.5 (Pointillart, 1991). Classified according to the level of phytase activity, wheat (e.g. wheat bran 1-4 PU/g (Pointillart, 1991)) and rye (e.g rye bran 8 PU/g (Pointillart, 1991) fall into the category of high phytase activity. Barley, however, is considered to have moderate phytase activity (Reddy et al., 1989), and oats (Reddy et al., 1989; Bruce and Sundstol, 1994) and maize are described as having negligible phytase activity (Reddy et al., 1989). In addition, soybean, canola and rapeseed are also described as having negligible or zero phytase activity (Stone et al., 1984; Nair et al., 1991; Saridberg et. al., 1993). Once germination is initiated phytase activity increases, and maximum hydrolysis of phytate is reached in 3-10 days or more (Reddy et al., 1989). However, it is unclear whether the increased activity is due to the activation of pre-existing phytase and/or de-novo synthesis of phytase (Irving, 1980). According to Irving (1980), Kuvayeva and Kretovich (1978) managed to isolate two forms of phytase from germinating cotyledons of peas, but could only isolate one form from imgerminated dry cotyledons (Irving, 1980). During germination, amino acids labeled with 1 4 C were shown to be incorporated into the form of phytase that was exclusive to germinated cotyledons (Irving, 1980). This observation was said to be indicative, of de novo synthesis of phytase (Irving, 1980). However, more research is needed in order to identify to what degree de novo and/or stimulation of pre-existing phytase is responsible for phytate hydrolysis during germination. During germination there are other seed compositional changes which may accompany phytic acid reduction. Some of these are nutritionally beneficial while others are not. For example, Thompson and Serraino (1985) noted a 65% loss of phytic acid and a 10% increase in protein following germination of rapeseed seeds for 7 days. However, these benefis were accompanied by a rise in free fatty acid values from 1 to 120, because of fat hydrolysis; an undesirable result due to the development of rancid flavour in 54 the oil. Based on these results, germination is not likely to be a reasonably effective means of phytic acid removal in rapeseed or canola. 2.1.6.3 Water extraction and differential solubility Differences in the solubilities of proteins and phytic acid/phytates are the basis for the removal of phytic acid/phytate by water extraction. The pH is critical for the effective separation and recovery of phytic acid and undegraded protein. Additionally, water type, temperature, and cationic environment of extraction are also critical for efficient removal of phytic acid/phytate, and the recovery and prevention of the degradation of protein. Typically, water dispersions of the seeds or protein products are pH adjusted so that phytic acid, already in solution, is rendered insoluble, and this facilitates subsequent removal by centrifugation or filtration. Alternatively, pH adjustments may be used to maximize phytic acid solubility while decreasing protein solubility, facilitating the separation of phytic acid and protein. Gillberg and Tornell (1976) found that phytate solubility decreased at pH 4.5-8, and increased at pH 8-10. They also determined that minimum phytate solubility was reached at pH 11.1 (Gillberg and Tornell, 1976). Comparatively, it has been shown that approximately 20-40% of rapeseed proteins have isoelectric points close to pH 11 whereas the rest have isoelectric points spread between pH 4 and pH 8 (Serraino and Thompson, 1984). Consequently, acidification to a pH of 3.5, which is just below the isoelectric points of many of the rapeseed/canola proteins present, would help to niinimize protein solubility while mamtairiing the solubility of phytic acid. Additionally, extraction at acidified pH's of around 3.5-5 also serves to optimize glucosinolate extraction (Youngs, 1991). Basification has also been tried for the separation of proteins from phytic acid. Basification to pH above 11 renders phytic acid insoluble while proteins are found to be soluble. However, rapeseed and canola products acquire an undesirable dark colour as well as a strong flavour above pH 9, partially due to the presence of polyphenols and the formation of lysinoalanine which is nutritionally undesirable (Serraino and Thompson, 1984; Reddy et al., 1989). Hence, basification does not appear to be a reasonable choice for the dephytinizing of rapeseed and canola. Water is the solvent 55 commonly used for extraction, but not all water types nor water temperatures are equally effective, as demonstrated by the use of distilled and tap water for the extraction of phytic acid from rapeseed. When crushed rapeseed was extracted at 0°C, 64 % of the initial phytic acid content present was removed with the use of distilled water, but removal was completely eliminated when tap water was substituted (Jones, 1970). A change in the temperature of extraction also affected phytic acid losses whereby crushed rapeseeds subjected to boiling water for 5 minutes before extraction at 0 °C showed only a 24% phytic acid reduction whereas a 64% level of phytic acid removal was achieved when the boiling water treatment was omitted (Jones, 1970). A proposed reason for the observed difference was that the higher temperatures may have contributed to increasing the bonding strengths of phytic acid-mineral and/or phytic acid-protein complexes. Changing the cationic environment also alters the effectiveness of water extraction of phytic acid. Competitive chelators, (e.g. calcium) may be added to the water dispersion as a means of enhancing phytic acid removal, by disrupting either the formation of new phytic acid complexes or pre-existing complexes, and allowing ready separation of phytic acid from minerals and proteins. (Refer to section 2.2.7.6.7.2). 2.1.6.4 Membrane separation Dialysis and ultrafiltration membrane separation methods rely on molecular size differences and therefore may be used to separate rapeseed proteins (molecular weight of ~ 13,000) (Gillberg and Tornell, 1976; Serraino and Thompson, 1984) from phytic acid and its metal salts (molecular weights in the order of < 1,000 ) (Reddy et al., 1989). However, because much of the phytic acid is associated with protein and consequently is not readily removed by ultrafiltration (Serraino and Thompson, 1984), membrane separation is often paired with differential solubiltiy methods in order to improve separation. Similarly, solubility adjustments such as miriimizing nitrogen solubility and maximizing phytic acid solubility are desirable for efficient dephytinizing by dialysis (Serraino and Thompson, 1984). In general, phytic acid removal is enhanced by acidification (e.g. pH 3.5), and restricted by basification (e.g. pH 9.0). Additionally the potential for phytic acid removal may be diminished through heat treatment of protein products (e.g concentrates) wherein heat treatment likely strenghthens the 56 electrostatic interaction between protein and phytic acid. This serves to stabilize the protein-phytic acid complex, and effectively reduces the solubility of both nitrogen and phytic acid (Gillberg and Tornell, 1976; Serraino and Thompson, 1984). This strengthening of bonds may also reduce the ability of chelating agents such as EDTA to dissociate phytate-protein complexes (Serraino and Thompson, 1984). Finally, although dialysis is able to remove substantial amounts of phytate with or without pH adjustment, nitrogen losses are relatively large. In addition, dialysis is labour intensive and time consuming. Consequently, dialysis is unsuitable for commercial application of phytic acid/phytate removal whereas ultrafiltration may be more suitable. Since glucosinolates, phytates and phenolics are signficantly lower in molecular weight than are rapeseed/canola proteins (Diosady, et al., 1991), ultrafiltration is a reasonable procedure which can be employed for the separation of protein from phytic acid and other constituents. The pressure activated ultrafiltration process offers a relatively high rate of flux, mild processing conditions and selectivity, advantages which make it a highly promising method for commercial removal of phytic acid/phytate (Cheryan, 1980; Reddy et al., 1989). It has been used to successfully remove phytate (95% reduction) in soy extract adjusted to pH 6.7 (Omosaiye and Cheryan, 1979; Reddy et al., 1989), and could be coupled with low pH and high calcium or high pH and low calcium for optimization of phytate removal. 2.1.6.5 Ion exchange processing Cation and anion exchange resins may be used singly, together, and/or in combination with dialysis, for phytic acid /phytate removal over a range of pH (Brooks and Morr, 1982; Reddy et al., 1989). The passage of protein product dispersions through the resin column facilitates the disruption of protein-cation-phytic acid complexes, aiding the separation and removal of phytic acid (Brooks and Morr, 1982; Reddy et al., 1989). Phytate reduction of up to 99% from soy extracts has been achieved with this methodology (Brooks and Morr, 1982; Reddy et al., 1989). 2.1.6.6 Gene transfer Phytase-producing DNA from the Aspergillus niger fungii has been successfully transferred into 57 the genome of tobacco (Nicotiana tabacum cv. Petit Havana SRI) yielding a maximum phytase expression of 1 % of soluble protein in seeds and leaves (Pen et al., 1993). Seeds from the transgenic variety (MOG413.25) were tested in an unpelleted soybean, sunflower and sorghum based broiler diet, at a phytase concentration of 295 PU/kg diet (« 3.3g/kg plant phytate) (Pen et al., 1993). When broilers fed the transgenic seed containing diets were compared to broilers fed control diets supplemented with inorganic phosphorus or diets supplemented with a commercial phytase preparation (Natuphos® from Gist-brocades N.V., The Netherlands), at concentrations of 202 PU/kg diet and 404 PU/kg diet, no significant difference in growth performance was found, providing compelling evidence for the hydrolysis of phytic acid (Pen et al., 1993). Although transgenic phytase-producing rapeseed and canola varieties have not yet been developed, it is believed and hoped that there may be potential for repeated success in rapeseed and canola at some future time (Youngs, 1991). 2.1.6.7 Enzyme treatment Enzymatic hydrolysis of phytate is another avenue for improving the nutritional quality of phytate-containing feeds. Oral application of the enzyme phytase, which catalyzes phytate hydrolysis, may be in the form of crude or purified phytase preparations, or may be via the inclusion of seed or seed products which have intrinsic phytase activity (refer to section 2.1.8.2). The relative ease of phytase inclusion, as opposed to complex dephytinizing processing methodologies presently available, makes this approach to the improvement of the nutritional value of phytate-containing feedstuffs a favored one. 2.1.7 Phytase: properties, occurrence and function The enzyme phytase (myo-inositol hexakisphosphate phosphohydrolase) is capable of hydrolyzing myoinositol hexakisphosphate (phytic acid; IP6) to inorganic orthophosphate, a series of lower phosphoric esters of myoinositol and free myoinositol (Nayani and Markakis, 1986; Reddy et al., 1989), in a manner that is somewhat analagous to the digestion of carbohydrates to free sugars. Ubiquitous in its distribution, phytase has been detected in the seeds, pollen, roots and leaves of plants, as well as in the blood plasma of 58 various lower vertebrates (birds, reptiles, and fishes such as the Black Bass), and the erythrocytes of birds and sea turtles (Cosgrove, 1980; Irving, 1980; Reddy et al., 1989). Phytase has also been found in rumen microflora, the intestinal mucosa of rats, rabbits, chickens, guinea pigs, hamsters and man, as well as in yeast, soil and soil microorganisms, micorrhiza microorganisms, bacteria (e.g. Bacillus subtilis (natto) N -77), and most frequently in fungi, particularly Aspergillus sp. (Rapoport et al., 1941; Cosgrove, 1980; Irving, 1980). Varying in origin, phytases differ in substrate specificity as well as in modes of hydrolytic action (Reddy et al., 1989). Activation energies, temperature and pH optima, as well as chemical activators or inhibitors of phytase activity, also vary as a function of enzyme origin (Reddy et al., 1989; Reddy and Pierson, 1989). (Refer to Table 2.1). Substrate specificities, as well as temperature and pH optima of Aspergillus sp. phytases are of particular interest to the feed industry, as dietary phytase supplements are often derived from this source. Additionally, the effects of pelleting conditions on the retainment of phytase activity is of particular importance/relevance to salmonid diets. This is because, unlike poultry or swine diets, fish diets must be pelleted by virtue of the fact that fish live and feed in water. Pelleting effects on phytase are of particular relevance to swine diets produced in the Netherlands, where > 90 % of the swine diets are steam pelleted, primarily in an effort to reduce dust generation or the occurrence of Salmonella sp. (Jongbloed and Kemme, 1990). The activity of phytase derived from Aspergillus sp. may be reduced to zero once internal pellet temperatures reach 80°C (Jongbloed and Kemme, 1990) although Simons et al. (1990) found that phytase retained 46 % of its activity at 87°C. Categorized on the basis of pH optima, phytases may be separated into either five or four classes (Irving, 1980). The five classes are as follows: class I phytases have pH optima which range from 7.3-8.0 (e.g. rat organs), class II have pH optima of 5.3-5.6 (e.g. higher plants), class III have a pH optima of 4.0 (e.g. Aspergillus sp.), class IV have pH optima of 6.0-7.0 (e.g. plasma and erythrocytes of lower vertebrates) and class V have a pH optima of 2.2 (e.g. Eschericia coli) (Irving, 1980; Nayini and Markakis, 1986). Phytases may alternatively be grouped into four classes that are thought to be more accurate, wherein class I, II and IV of the previous scheme are left unchanged, and class III and V are 59 considered to be one distinct group (Irving, 1980). This latter classification scheme takes into account the wide ranging, and occasional bimodal pH optima of Aspergillus sp. (e.g. A. flcuum NRRL3135 with pH optima 2.5 and 5.3 as well as 2.0 and 5.5; A. niger NRRL65 with an optima of 2.5; A. terreus No. 9A-1 with a pH optima of 4.5) (Irving, 1980; Reddy and Pierson, 1989). Separated on the basis of mode of hydrolytic action, only two phytases are recognized by The International Union of Pure and Applied Chemistry and the Union of Biochemistry (IUPAC-UIB), namely 3-phytase and 6-phytase (Reddy etal., 1989). The enzyme 3-phytase EC 3.1.3.8 (myo-inositol-hexakisphosphate-3-phosphohydrolase), is recognized to initiate hydrolysis at the ester bond at the 3-position. Dephosphorylation then proceeds in stepwise fashion with the removal of one phosphate at a time. By contrast, the enzyme 6-phytase EC 3.1.3.2.6 (myo, inositol-hexakisphosphate phosphohydrolase), is recognized to begin hydrolysis at the 6-position according to the following reaction. Dephosphorylation then continues in stepwise fashion with the ordered removal of phosphates from position 5, 4 ,3 or 1, then 1 or 3 (Reddy et al., 1989). The final product myoinositol-2-phosphate is then hydrolyzed by a myoinositol phosphatase to free myoinositol and orthophosphate (Reddy et al., 1989). According to Irving (1980) however, the above nomenclature does not include the F2 fraction of wheat-bran phytase, referred to as 5-phytase which begins hydrolysis at the ester bond positioned at the 5-position, initially yielding myoinositol-l,2,3,4,6-pentakisphosphate, which is then dephosphorylated in stepwise fashion to myoinositol-1-phosphate as the final product (Reddy et al., 1989). Alternatively, based on their studies of germinating mung beans (Phasolus aureus), De and Biswas (1979) have proposed the existence of another enzyme system wherein a phytate-ADP-phosphotransferase specifically removes phosphate from carbon 2 of phytate, after which an enzyme like wheat-bran F2 phytase phosphorylates the resultant myoinositol 1,3,4,5,6 pentaphosphate to myoinositol-1-phosphate as a final product (De and Biswas, 1979; Reddy et al., 1989). This proposed action may be the basis for the occasions in the literature where the F2 wheat-bran phytase is referred to as 2-phytase (Reddy et al., 1989). Nonetheless, only 3-phytase and 6-phytase are presently recognized. In general 3-phytase is considered to be characteristic of the phytase found in microorganisms, and the phytase shown to be present in the intestinal mucosa of animals, whereas 6-phytase is thought to be characteristic of seeds and vegetative parts of higher 60 plants (Cosgrove, 1980; Nayini and Markakis, 1986; Reddy et al., 1989). 2.1.7.1 Use of extrinsic sources of phytase for phytic acid removal Crude (Han, 1989) or purified (Lei et al., 1993) preparations of phytase are used to supplement diets with extrinsic, or rather, exogenous sources of the enzyme. Commercial preparations are primarily derived from microbial sources of phytase which exhibit particular substrate specificity, and pH optima which are compatible with gastric conditions (Campbell and Bedford, 1992). For feed application, phytase derived from fungal Aspergillus sp. is particularly effective. Aiko Biotechnology Ltd., (Rajamaki, Finland) produces commercial phytase preparations referred to as Finase™ which is derived from this species. Derived from A niger (yax.ficuum, 3-phytase), Finase™ is usually sold with an activity of approximately 500,000 PU/g (Lei et al., 1993). One such preparation, Finase-F is know to remain active up to 60° C (Nasi, 1990). Commercial phytase preparations derived from Aspergillus sp. are also available from GIST-Brocades (Delft) (Simons and Versteegh, 1991). Commercial wheat phytase preparations are also available from Sigma Chemical Co., (St., Louis, M.O.), with activty in the order of 0.05 PU/mg solid (Han, 1989). Such preparations may be used to pretreat, (dephytinize) high-phytate protein sources before diet formulation, or may be added to the diet in order to facilitate in vivo hydrolysis of phytic acid /phytate. Additionally, as demonstrated by Nair et al., (1991), using Aspergillus ficuum NRRL 3135 and canola meal, the phytic acid/phytate content of a high phytate feedstuff may be reduced to zero by means of growth of a phytase producing organism directly on the feedstuff. As found by Nair et al., (1991) a 10% increase in crude protein content of the protein meal may accompany the dephytinization. 2.1.7.2 Use of intrinsic dietary sources of phytase for phytic acid removal Another means of supplying dietary sources of phytase is through the dietary inclusion of seeds or other feedstuffs with intrinsic concentrations of phytase. Wheat (Sandberg et al., 1993), wheat bran, rye, rye bran (Pointillart, 1991) and transgenic tobacco seeds are feedstuffs which have been T a b i d . 1: Phytase: selected sources and associated activators, inhibitors, and temperature and pH oj Source Soybean White mustard (Sinalpis sp.) Corn Wheat meal Wheat bran Wheat flour Rat (Intestinal mucosa) Chicken (Intestinal mucosa) Yeast (Sarcomyces sp) Yeast (Pichia farinosa) Fungi {Aspergillus ficuum) NRRL3135 Fungi (Aspergillus niger) NRRL65 Fungi (Aspergillus terreus) 9A 1 Activators Inhibitors pH Temperature Optima Optima CC) Reference Fe 2+ C a 2 + F 4 M g 2 + , C a 2 + F ,CN" , Z n 2 + , C u 2 NaN, A g + ,Hg + oxalates M g 2 + , C a 2 + F , Z n 2 + , A s 0 4 , C u 2 + A g + , H g + M g 2 + , Z n 2 + 2+ M g F e 2 + , C u 2 + E D T A , oxalates, citrate, tartrate inorganic P inorganic P inorganic P 4 8 5.2 4.8 5.15 5.0, 4.5-5.0 5.15 7.0, 7.2, 7.5,7.8 7.2,7.2-7.4, 8.3 4.6 2.2 2.5&5.3 2.0 & 5.5 5.0 2.5 4.5 60 50 50 55 55 60 1,8 2,8 1,2,3, 8 2,3,8 1,3,4,8 1 2,3 2,3, 2,3 2,3 2,3,5,6 • 2,3 2,3 Bacteria (Pseudomonas sp) Bacteria (Bacillus subtilis) Bacteria (Escherichia coli) M g 2 + , C a 2 + C a 2 + E D T A , Z n 2 + , C d 2 + B a 2 + , C u 2 + , F e 2 + , Al 3+ 5.5 75, 6.0-6.5 2.2 4.5 & 5.4 2,3,7,8 1,3 62 Table 2.1 (CONTINUED): Phytase: selected sources and associated activators, inhibitors and temperature and pH optima 1 Reddy et al., 1989 2 Nayani and Markakis, 1986 3 Irving, 1980 4 Sandberg and Svanberg, 1991 5 Jongbloed et al., 1992 6 Nair etal., 1991 7 Simons etal., 1990 8 Irving and Cosgrove, 1971 9 - = not provided 63 employed in this way. 2.1.7.3 Effects of dietary phytase addition: swine Phytase is present within the intestinal mucosa of pigs; however wide disagreement exists regarding the extent to which the intestinal phytase is able to affect phytate hydrolysis. Some investigators report that mucosal phytase concentration is normally too low to affect the digestion of phytate in pigs (Pointillart, 1988; Bruce and Sundstol, 1994). Conversely, others report that due to mucosal phytase, phytate phosphorus availability to swine ranges between 20-60% and averages 33% for pigs between 50-90 pounds in weight (Besecker et al., 1967; Noland et al., 1968; Bayley and Thompson, 1969; Reddy et al., 1989). Although this controversy has yet to be resolved, there is common agreement that dietary supplementation of microbial phytase significantly reduces the anitinutritive effects of phytic acid. Supplemented primarily to com and soybean or maize and soybean based pig diets, microbial phytase has repeatedly been shown to improve the absorption (Paullauf et al., 1992), digestibility (Beudeker et al., 1990; Khan and Cole, 1995) retention, and thus utilization of phytate phosphorus (Lei et al., 1992) even in the absence of phosphate supplements (Khan and Cole, 1995). Additionally, improved feed conversion and total feed digestibility has also been observed (Lei et al., 1992). The application of phytase from Aspergillus niger was also able to improve calcium digestibility and retention (Nasi, 1990). Furthermore, the specific supplementation of phytase in the form of the commercial preparation Finase™ has been shown to significantly improve the apparent absorption of Mg, Fe, Cu and Mn (Pallauf et al., 1992) as well as the bioavailability of phytate phosphoms and zinc (Cromwell et al., 1993) The use of feedstuffs rich in endogenous phytase (e.g. rye bran) has also been successful in improving the absorption and retention of phytate phosphoms and dietary calcium in com, soybean and rapeseed meal based diets, although no differences in calcium absorption were observed between diets (Pointillart, 1991). Characteristic of either intrinsic or extrinsic phytase supplementation of diets, an improvement in phosphorus absorption is often accompanied by a decrease in faecal phosphoms output (in the order or 35%-56%) (Pallauf et al., 1992; Simons et al., 1990) and an increase in urinary phosphorus excretion 64 (Khan and Cole, 1995). 2.1.7.4 Effects of dietary phytase addition: poultry Microbial phytase supplementation of diets formulated for chickens has been shown to increase phytate phosphorus availability (Nelson et al., 1971; Classen and Bedford, 1991), which is often accompanied by a decrease of as much as 50% in faecal phosphorus output (Beudeker et al., 1990; Simons et al., 1990). Additionally, phytase supplementation serves to increase phytate phosphorus digestibility (Beudeker et al., 1990) and retention (Campbell and Bedford, 1992), and improve feed conversions of chickens fed high-phytate diets (Campbell and Bedford, 1992). Similarly, the dietary use of feedstuffs with intrinsic phytase activy, (e.g. transgenic phytase-containing tobacco seeds) (Pen et al., 1993) (Refer to section 2.1.8.6) has been shown to improve the utilization of phytate phosphorus. It should be noted however, that phytase is present in the intestinal mucosa of chickens. Consequently, some of the increased utilization of phytate phosphorus in phytase supplemented diets may be partly attributable to the action of endogenous phytase. However, the extent to which endogenous phytase participates in dietary phytate hydrolysis is uncertain, because the level of effectiveness and/or activity of endogenous phytase has yet to be established. 2.1.7.5 Effects of dietary phytase addition: fish (salmonids) Salmonids lack intestinal phytase. Consequently, the biological availability of phytate phosphorus is zero in the absence of dietary supplementation of phytase. With dietary microbial phytase supplementation however, the phytate phosphorus from salmonid diets based on soybean meal (Spinelli et al., 1979; Riche, 1993; Cain and Garling, 1995), canola meal, and/or cottonseed meal (Riche, 1993), is rendered biologically available and fish growth rates and feed conversions are significantly increased to levels that are often comparable to commercial diet preparations (Cain and Garling, 1995). In addition, in the absence of inorganic phosphorus, faecal phosphorus excretion by fish fed phytase supplemented diets may be lower than faecal phosphorus excretion of fish fed commercial diets, wherein Cain and Garling, (1995) demonstrated that the phosphorus content in the effluent of rainbow trout fed phytase supplemented 65 soybean-based diets was 65-88 % less than levels found in the effluent of fish fed a commercial diet. Commercial phytase preparations have also been used successfully to enhance the feeding value of phytate-rich salmonid feeds. Recently Riche, (1993) prepared soybean meal, canola meal, and cottonseed meal based diets with and without the addition of the microbial phytase preparation Finase™ (Aiko, Ltd., Rajamaki, Finland) at a concentration of 3,750 PU/g dry diet. Rainbow trout fed the phytase treated diets were shown to have significantly (P < 0.05) higher apparent phosphoms absorption than fish fed the same diets without phytase supplementation (Riche, 1993). Pretreatment of feedstuffs with phytase before dietary incorporation has also been used successfully to enhance the feeding value of phytate-rich feeds for salmonids. One such example is the study by Spinelli et al., (1979) whereby samples of soybean meal were incubated with a phytase preparation obtained from a commercial source and/or derived from wheat. Once the phytic acid content was eliminated OMP-type diets (Oregon moist pellet) were then prepared in which 50% of the fish meal was replaced with either undephytinized or dephytinized soybean meal (Spinelli et al., 1979). Diets were fed to rainbow trout (-7.0 g initial weight) for 210 days (Spinelli et al., 1979). Both feed conversion and growth were found to be greater for fish fed the dephytinized soybean diet than for fish fed the diet with undephytinized soybean meal. Phytase pretreatment of phytate-rich feedstuffs, and phytase supplementation of diets high in phytate clearly renders phytate phosphoms biologically available to salmonids. However, although feed conversion and fish growth are increased by phytate hydrolysis, information concerning the specific changes in protein digestibility, protein retention and overall mineral availability that may accompany phytate hydrolysis is sadly lacking, but is certainly worthy of further investigation. 2.2 The role of chemoreception, chemoattractants and stimulants in diet utilization Fish generally eat to satisfy their energy needs, but whether fish will eat a particular diet is dependent on: the nutritional history and current status of the fish, fish health and reproductive state, water temperature and quality, and the appearance, texture, consistency, smell and taste of the feed. Accordingly, there are occasions when fish will readily consume their feed and when they will not, and it is in these latter 66 circumstances that chemoattractants and/or stimulants may be utilized in order to help initiate, maintain, and/or enhance feeding behaviour. In intensive salmonid production systems, feed acceptance and intake is typically low for diseased and/or stressed fish, which characteristically exhibit poor appetite. Additionally, alevins often show poor acceptance of formulated diets when they are first introduced to exogenous feeding (Ward, 1991). It has also been repeatedly shown that diets high in vegetable proteins (e.g. soybean, rapeseed and canola proteins) can be unpalatable. Consequently, feed intake is generally low, as such diets are not readily accepted by salmonids (Spinelli et a l , 1979; Fowler, 1980; Higgs et a l , 1983; Hilton and Slinger, 1986; Dabrowski et a l , 1989; Rumsey et a l , 1993; Rumsey, 1994). Chemoattractants may be utilized in these instances to improve feed acceptability. In the natural environment, as in the culture environment, acceptability of prey or feed items is a function of many factors, including the fish's capability of food detection. Salmonids rely on olfaction and vision for the detection and selection of prey or feed items. Appetite and mouth size or gape necessarily play a role in what is taken into the mouth, but taste (gustatory sensitivity) is a primary determinative factor for what food particles are ultimately ingested or rejected (Sutterlin and Sutterlin, 1970; Adron and Mackie, 1978; Hyatt, 1979; Rumsey, 1988; Bres, 1989; NRC, 1993; Higgs et a l , 1994a;b). The palantine organ of salmon is located in the roof of the mouth, and consists of folds of dermis stretched between dermal teeth (Sutterlin and Sutterlin, 1970). The area is scattered with taste buds and is innervated by the palantine nerve (Sutterlin and Sutterlin, 1970). Gustatory sensitivity of this area and olfactory response to chemical cues derived from the composition of food items help to define palatability, and accordingly help to determine feeding behavior (Hyatt, 1979; Rumsey, 1988; NRC, 1993). Compounds that cause an animal to respond by orienting toward the apparent source are termed attractants (Mackie, 1982). Compounds that promote ingestion and continuation of feeding are termed stimulants (Mackie, 1982). Examples of such substances are various amino acids and betaine, used singly or in combination (Carr, 1982; Hara, 1982; Mackie, 1982; Mackie and Mitchell, 1985; Ward, 1991; Clarke et al,1994). Amino acids have been described in terms of associated "tastes", according to human perceptions 67 (Table 2.2 ; Ward, 1991). However, as there is no way of determining specific "tastes" per se of fish, only degrees and modes of response to chemical cues may be determined. The degree to which amino acids may solicit a positive feeding response, that is, their effectiveness as feed stimulants or attractants for fish varies with isomerization state (Hara, 1973; Hara, 1977; Carr, 1982; Mackie, 1982; Mackie and Mitchell, 1985; Ward, 1991) and also with fish species (NRC; 1993), age and life stage (Shparkovsky et al., 1983; Meams, 1986). Generally, L-isomers of amino acids are more effective at eliciting positive feeding responses than D-isomers (Hara, 1973; Hara, 1977; Carr, 1982; Mackie, 1982; Mackie and Mitchell, 1985; Ward, 1991). Studies that indicate species differences in reponse tend to group carnivorous species together and herbivorous species together, wherein stronger responses are associated with groupings of amino acids that best reflect the chemical composition of their natural prey and feed items (Mackie, 1982; NRC, 1993). Carnivorous fish show greater responses to basic and neutral amino acids such as glycine, proline, taurine, valine and betaine (trimethylglycine; glycine betaine, (CHO3NCH2COOH) (Carr, 1982; NRC, 1993), the latter of which is known to be widely distributed in marine insects and elasmobranchs, but absent from teleosts (Love, 1970; Mackie, 1982). Herbivorous fish tend to respond more favorably to acidic amino acids such as aspartic acid and glutamic acid. Studies by Shparkovsky et al. (1983) and Meams (1986) are suggestive of age-related differences in response. L-leucine was found to evoke greater responses in two year old Atlantic salmon than in fry and yearlings (Shparkovsky et al., 1983). Likewise, L-alanine as well as L-proline were more effective stimulants in younger rather than older fish (Mearns, 1986). It is also generally accepted that mixtures of amino acids may be more effective stimulants than individual amino acids (Ward, 1993). An amino acid mixture is included in the commercial betaine preparation FINNSTIM™ expressly for that purpose, to increase palatability of the product (Clarke et al., 1994). FINNSTIM™ is a product of Finnsugar Bioproducts, Finland, that includes 90-94% anhydrous betaine (Virtanen et al., 1989; Clarke et al., 1994) and 3% of an L-amino acid mixture primarily comprised of alanine, serine, valine, glycine, leucine and isoleucine (Virtanen et al., 1989; Clarke et al., 1994). Its predominant use is in investigative studies for osmoregulation in fish, where at a dietary concentration of 1.0-1.5% its betaine content has been demonstrated to enhance the hypo-osmoregulatory capacity and the 68 feeding behaviour of Atlantic salmon (Virtanen et al., 1989; Virtanen, 1992), chinook salmon (Virtanen, 1992; Clarke et al., 1994), Arctic charr (Virtanen, 1992) and rainbow trout (Virtanen, 1992). Although Clark et al. (1994) discount the possibility that the betaine and L-amino acids in FINNSTIM™ behaved as feed stimulants and thus could have accounted in part for observed stimulation of growth in salt water challenged yearling chinook, FINNSTIM™ has been employed specifically as a feed stimulant in rainbow trout diets high in plant protein sources (Riley et al., 1993; Teskeredzic et al., 1995) with good success. Additionally, betaine has been successfully utilized in rainbow trout diets to maintain dietary intake of soybean-based diets (Rumsey et al., 1994). 69 Table 2.2: Associated taste types of amino acids known to act as chemoattractants and stimulants in fish Amino acid Taste type1 Reference L-histidine bitter Ward, 1991 L-arginine l l L-phenylalanine II L-alanine sweet Solmes, 1969; Ward, 1991 L-proline tt L-threonine M D-tryptophan II D-phenylalanine II D-histidine It D-leucine tt L-aspartate sour Ward, 1991 L-asparagine tt L-glutamate flat Ward, 1991 L-leucine chocolate (sweet) Ward, 1991 L-valine it L-isoleucine 1 taste according to human gustatory sensitivities 70 CHAPTER THREE 3.0 General materials and methods General materials and methods common to both experiments presented in this thesis are described in the following sections. Specific, detailed methodologies for individual experiments are described in the materials and methods sections of chapters four and five. 3.1 Culture facility and experimental tanks Two experiments were conducted at the Department of Fisheries and Ocean's West Vancouver Laboratory located in West Vancouver, British Columbia, Canada. Durations of experiments I and II were 83 days and 84 days, respectively. Experimental tanks were housed within an indoor aquarium where lighting conditions simulated a natural photoperiod due to the use of Vitalite, Durotest 40W fluorescent lights regulated by an astronomical timing cell. Both experiments were conducted in well water supplied at ambient temperature. Water temperatures remained relatively constant during experiment I (10.1 - 10.8°C) as well as during experiment II (9.8 - 10.9°C). However, twice during experiment II the well water delivery system was shut down, and dechlorinated city water had to be used as an alternate water source. The use of dechlorinated city water for 2 and 6 hour durations caused water temperatures to reach highs of 13.5°C and 14.2 °C and dissolved oxygen content to drop to 9.4 ppm. and 9.6 ppm. respectively, for the length of time that well water was unavailable. For both experiments water was aerated by means of pressurized air delivered to each tank through polyvinylchloride tubing which terminated in weighted, highly perforated rubber rings, (sections of garden soaker hoses, Water Wick Inc., High River, Alberta, Canada), submersed below the surface. Dissolved oxygen concentrations were monitored daily (Dissolved oxygen/ BOD meter, model 7932, L.G. Nester Co., Melville, New Jersey, USA.), and kept between 9.4 ppm -10.0 ppm. All tanks were fitted with 71 hinged plexiglass lids which were clear over the front third and opaque over the back two thirds in order to provide areas of decreased light intensity, affording the fish some cover. 3.2 Fish grading and distribution Immature rainbow trout of initial weight 1-4 g were used as the experimental animal. Because physiological stress associated with crowding, netting, or other routine capture or handling procedures of fish culture production systems (Barton et al., 1980; Specker and Schreck, 1980; Matthews et al., 1986) as measured by plasma C o r t i s o l level, has recently been shown to be greatly reduced by pre-exposure of fish to the tranquilizer metomidate (Kreiberg, 1992), a commercial preparation of metomidate hydrochloride (Marinil™) was utilized prior to fish handling. Upon receipt, fish were lightly sedated by adding Marinil™ at a concentration of 0.25mg/l to the water. Small numbers of fish were then netted into an anesthetic bath of tricaine methane sulfonate (TMS; MS222), (60 mg/1) buffered with an equal weight of sodium bicarbonate (NaHCCh). Once fish were fully anesthetized (belly-up), they were blotted dry then individually weighed to the nearest 0.0lg. The calculated average weight and standard deviation of the first 100 fish measured was then used to grade fish for experimental use (mean weight ± 1.0-1.1 S.D.). Appropriately sized fish were randomly distributed, five at a time, into the prepared experimental tanks. Fish not falling within the acceptable range were placed in a stock tank for other scientific use. 3.3 Fish sampling procedures Following an initial sample on day 0, fish were sampled every two weeks during experiment I and every three weeks during experiment II. Feed was withdrawn seventeen hours prior to weighing. During sampling Marinil™ and an anasthetic tricaine methane sulfonate (TMS) were used at concentrations of 0.25 mg/1 and 60 mg/1 respectively. TMS was always buffered with an equal concentration of sodium bicarbonate. Fifteen minutes prior to sampling, individual tank inflows were shut off and Marinil™ was 72 added. Once the fish reached a stupified state the water level in the tank was lowered and all fish were netted into a well aerated container filled with water dosed with Marinil™ Small groups of fish were then immersed in a well aerated, anesthetic bath of buffered TMS. Once fully anesthetized, individual fish were withdrawn, lightly blotted and weighed to the nearest 0.0lg. Once measured, fish were removed to a container filled with aerated Marinil™ dosed water and allowed to recover from a deeply sedated state. Fish taken for proximate analysis measurements were randomly withdrawn directly from water dosed with Maranil™ and killed by a quick cranial blow. Fish were then weighed, vacuum-packed (vacuum packer, Spiromac, model 350, Canada), and immediately placed on ice before final storage at -20°C. All fish taken for blood and liver extractions were killed, weighed and measured in the same manner. Blood and liver samples were stored at -70°C. After the entire population of an individual tank had been weighed, and samples had been taken, remaining fish were netted back into their respective experimental tanks, which had been scrubbed clean and refilled. Flows were then adjusted to the prescribed rate. 3.4 Proximate and energy analyses Representative samples of protein feedstuffs, diet samples and whole frozen fish were ground in a blender to a homogenous state from which aliquots were quantitatively removed for analysis. When not in use, homogenates were frozen at -20°C for short term storage and -40°C for long term storage. Moisture and ash were determined according to AO AC (1975) methods and crude lipid according to Bligh and Dyer (1959). Total nitrogen (Technicon AutoAnalyzer II, Kjeldahl, colorimetric industrial method No. 334-74W/B ) ( % N x 6.25) was used to determine crude protein. If replicates differed by >0.5% for crude protein or > 1% for other determinations, the analysis was repeated. Gross energy values for all diet and faecal samples were determined by means of a Gallenkamp adiabatic bomb calorimeter (Gallenkamp Co. Ltd., Loughborough, UK), wherein 0.5 and 0.2 g samples of diets and faeces were used respectively. All measurements were conducted in duplicate. 73 3.5 Plasma samples Blood was withdrawn from caudal vessels, then centrifUged to separate the plasma. After short term storage at -70 °C plasma samples were packaged and sent on dry ice to the Department of Zoology, University of Manitoba, where plasma titres of L-thyroxine (T4) and 3,5,3 1 triiodo-L-thyronine (T3) were determined according to the procedures of Omeljaniuk et al., (1984). For experiment I, the caudal peduncle was cut and the fish were bled into ammonium heparinized capillary tubes. For experiment II, heparinized lcc syringes were used for the withdrawal of blood. 3.6 Liver samples Frozen livers were shipped on dry ice to the Department of Zoology, University of Manitoba. Six pools of three livers each per replicate of dietary treatment were analyzed for 5' monodeiodinase (5 ' D) activity according to the procedures of Eales et al., (1992). 3.7 Determination of whole-body mineral composition Subsamples of whole ground fish were prepared as for moisture analysis according to AO AC (1990) procedures. Dried samples were then ground and sent to the Northwest Fisheries Center, Seattle, Washington, USA for mineral analysis. Acid digestion and determination of mineral concentrations by inductively coupled plasma emission spectrophotometry (ICP) were completed according to Shearer (1984). 3.8 Preparation of protein sources All protein feedstuffs were finely ground in a hammer mill (Fitzmill, model JT., CL 9001, type BW 204, Fitzpatrick Co., Elmhurst, Illinois, USA) fitted with a 590um metal screen die (particle size would pass through a U.S. no. 30 sieve), then mixed in a commercial Hobart mixer (model M802, Hobart Manufacturing Co., Troy Ohio, USA.) for twenty minutes. Feedstuffs were then analyzed for mineral and 74 amino acid composition and proximate constituents prior to diet formulation. Mineral assessment was conducted by plasma emission spectroscopy at Quanta Trace Laboratories Inc., Bumaby, B.C., Canada. Amino acid determinations were conducted at AAA Laboratories, Mercer Island, Washington, USA. (For experiment I, amino acid analyses of herring meal, Norse-LT94™ fish meal and undephytinized and dephytinized rapeseed protein concentrate were conducted by Northwest Fisheries Center according to similar procedures). Any feedstuff requiring pre-treatment with phytase was sent to Aiko Ltd., Biotechnology in Rajamaki, Finland for dephytinization and inositol phosphate determinations before being returned and analyzed for mineral and amino acid content and proximate constituents before diet incorporation. Mineral analysis of protein feedstuffs involved acid digestion of ground samples according to modified Environmental Protection Agency (EPA) method 3051(Quanta Trace Laboratories Inc., Bumaby, B.C. Canada) and analysis of the resulting solution by ICP-AES (inductively coupled plasma-atomic emission spectrometry) with ultrasonic nebulization. Means of two replicates were used for diet formulation. Analysis for alanine, arginine, aspartic acid, glutamic acid, glycine, histidine, isoleucine, leucine, lysine, methionine, phenylalanine, proline, sereine, threonine, tyrosine and valine was by 20 hour 6N-HC1/ 0.05 % mercaptoethanol hydrolysis at 115°C with one crystal of phenol added before acid hydrolysis. Serine was increased by 10 % and tJireonine by 5 % to compensate for destruction by acid. Cystine determinations involved performic acid oxidization prior to hydrolysis and calculation from cysteic/alanine ratios. Tryptophan determinations required 48-hour alkaline hydrolysis at 135 °C. Commercial dephytiniztion (phytase pretreatment) was conducted by Aiko Biotechnology Ltd., wherein each protein source, (commercial canola meal, low fibre canola meal or rapeseed protein concentrate) was slurried in tap water and adjusted to pH 5.0 with sulphuric acid. Finase™ S 40 phytase (microbial phytase preparation from Aspergillus niger) was added at the rate of 5000 PU/g of protein, (one phytase unit, PU, is the activity that liberates 1 nmole of inorganic phosphate from sodium phytate at pH 5 and 37 °C (Lei et al., 75 1993)). The mixture was then incubated at 55 °C for four hours, cooled to^approximately 8 °C, stored overnight at refrigeration temperatures (5-8 °C) and spray dried (air outlet temperature was 75 °C). Samples of undephytinized and dephytinized protein sources were then analyzed for inositol phosphates using the Aiko method C-120 Rev. A (Aiko Biotechnology Ltd.). 3.9 Vitamin supplements Vitamin premixes were identical for both experiments but were prepared individually immediately prior to the commencement of each experiment and the preparation of diets. Choline chloride, (which can decrease the stability of vitamin E, vitamin K and other vitamins in the prernix (Halver, 1972; NRC, 1993)), and ascorbic acid, (which is very labile; NRC, 1993) were not included in the prernix but were later added directly to the completed diet (see section 3.12). In preparation of the supplements, half of the prescribed amount of wheat starch carrier was added to the bowl of a bench Blakeslee® mixer (model B-12, G.S. Blakeslee and Co., Scarborough, Ontario, Canada.). Measured amounts of vitamins were added one at a time, with mixing being carried our after each addition. After all vitamin additions, the remainder of the wheat starch was added to the bowl, and the preparation was mixed for ten minutes. The prernix, which was quantitatively transferred to a "V"-shaped Twin Shell Dry Blender (Patterson-Kelly Co., Division of Harsco Corp., Pennsylvania, USA) was then mixed for a further fourty-five minutes, then carefully transferred to a plastic bag and stored for 1-2 days at ~ 4 - 6 °C until needed. 3.10 Mineral supplements To help facilitate ready mixing, individual mineral compounds were pre-ground to a fine powder in a coffee grinder. Required quantities were then weighed and added one at a time to the half of the prescribed alphacell carrier which had been placed in the bowl of a bench-top Blakeslee® mixer. Mixing was conducted between each mineral addition. After the addition of mineral compounds the reserved half of the carrier was added to the bowl and the preparation was mixed for fifteen minutes. For diets for which 76 the alphacell concentration in the mineral supplement was insufficient to act as an effective carrier, some of the extruded wheat from the basal diets was used as a carrier. For experiment II, chromium sesquioxide (CftOs) was added to all diets as an indigestible marker for digestibility evaluation. To facilitate even distribution throughout each diet a portion of the carrier used for the mineral supplement was reserved for pre-mixing with the chromic oxide. The carrier and the chromium sesquioxide were placed in a large mortar and then the mixture was ground with a pestle until all lumps had disintegrated. The mixture was then added to the mineral pre-mix and the preparation was mixed for a further ten minutes, bagged in a clear plastic bag then vigorously shaken for five additional minutes. 3.11 Amino acid supplements Amino acid supplements used in experirnnent I were prepared in the same manner as were the vitarnin pre-mixes, with the exceptions that wheat starch (Whetstar 3™) was used as the carrier and the supplement was mixed for thirty minutes. 3.12 Preparation of mash and pellets Test diets were prepared in October, 1992 (experiment I) and August, 1993 (experiment II). Air-dry dietary components were weighed then gradually mixed together with all pre-mixes in a large, commercial Hobart® mixer (Hobart Manufacturing Co., Troy, Ohio). Choline chloride and ascorbic acid were added separately, and the mash was mixed again for ten minutes. Ethoxyquin was added to a portion of the sardine oil before both were added to the mash and mixed for twenty minutes. Each diet mash was then steam pelleted in a CL-type 2 California laboratory pellet mill (California Pellet Mill Co., San Francisco, California), fitted with a dye size appropriate for the mouth gape of the experimental fish (Hilton and Slinger, 1989). Immediately following pelleting, pellets were spread to dry over wire mesh trays in an up-welling air dryer. Pellets were then screened to remove fines. Subsequently, reserved 77 sardine oil was broadcast over pellets (in a Monarch cement mixer with an electric paint sprayer in experiment I, or by use of a pump-action, garden plant fine spray-mister in experiment II). Once placed in clear plastic bags, pelleted diets were vigorously shaken to help ensure a good oil coating. Portions of pellets in use were kept in well labelled, sealed containers which were returned to refrigeration units (4°C) after the last feeding of each day. Reserved pellets and unpelleted mash were vacuum-packed and stored at -20 °C until they were needed. Representative pelleted samples of each experimental diet were ground in a blender then assessed for proximate composition, mineral content (Quanta Trace laboratories Inc., Bumaby, B.C., Canada), and amino acid content (AAA Laboratories, Mercer Island, Washington, USA.). Representative samples were also analyzed for glucosinolate content at the Agriculture Canada Research Station, Saskatoon, according to the procedures of Daun and Mcgregor (1981).. 3.13 Diet allocation For both experiments test diets were randomly assigned within each of two blocks of tanks located within the aquarium, such that there were two replicate tanks of fish per diet, and one of each test diet assigned per block. 3.14 Feeding protocol Using a spoon for distribution of pellets, fish were hand fed their assigned test diets to satiation three times daily. Exceptions were sampling days (early morning feeding session omitted), and days preceeding sampling days (fish fed only in the early morning). Satiation was judged to be the point at which fish showed very little or no apparent interest in feeding. During each feeding, individual tanks of fish were presented with their prescribed diets until feeding activity declined. Feed was then presented to successive tanks of fish until all tanks had been approached. Diets were presented once more to each group of fish until the fish appeared truly satiated. Every effort was made to ensure that there was little or no wastage of feed. Daily feed intakes per replicate 78 group of fish were recorded. 3.15 General maintenance All tanks were siphoned regularly to remove faeces and the iron oxide which was occasionally deposited from the well water. A transparent siphon was used so as to reduce any associated stress caused by the disturbance. Water temperature, flow rate and dissolved oxygen content were recorded daily and adjusted where possible. During each sampling day, tanks, lids and aeration rings were thoroughly cleaned. Detailed daily records concerning fish behavior and mortality were maintained. 79 CHAPTER FOUR 4.0 E X P E R I M E N T I . Acceptability of dephytinized and undephytinized rapeseed protein concentrate as complete replacements for fish meal in diets formulated for rainbow trout (Oncorhynchus mykiss). 4.1 Abstract A trial of 83-d. duration was conducted using juvenile rainbow trout (Oncorhynchus mykiss) held in ambient well water (10.1 -10.8 °C) to assess the effects of phytic acid removal on the acceptability of rapeseed protein concentrate as a complete replacement for the fish meal component (59 % of the crude protein (CP)) of a typical rainbow trout diet. The study was also undertaken to determine if the amino acid profile of rapeseed protein concentrate was adequate for optimal fish growth, and whether the feeding value of diets containing the concentrate could be improved through dietary balancing of cations (Na+ and K + ) and anions (Cf and S"2). Duplicate groups of juvenile (initial weight 3.9-4.6 g) trout were fed to satiation three times daily one of ten experimental diets formulated to be isonitrogenous (42 % , CP, dry matter basis (DMB)) and isoenergetic (metabolizable energy (ME) 17 MJ/kg, (DMB)), and to contain 17.5 % lipid (DMB) as well as 1.5% FINNSTIM™ (DMB) as a palatability enhancer. For each diet, 59% CP was provided by either high quality fish meal (B.C. whole herring meal, 72.58 % CP, or Norse- LT94™ (NFM), 76.22% CP), or rapeseed protein concentrate (RPC) that was dephytinized (DRPC) 63.11% CP) or undephytinized (URPC, 65.38%CP). The diet containing herring meal (HFM) served as the basal control diet for the study and that containing NFM as another control. N F M is recognized as the highest quality fishmeal in the world. Performance parameters used to compare growth and physiological response of trout to dietary treatment were weight gain (WG), specific growth rate (SGR), total dry feed intake (DFI), feed efficiency (FE), protein efficiency ratio (PER), whole carcass proximate and mineral compositions, plasma titres of 80 T/3, hepatic 5' monodeiodinase activity and average liver weight. Weight gains, specific growth rates, feed efficiencies, protein efficiency ratios and carcass proximate and mineral composition were not compromised when rapeseed protein concentrate pretreated with microbial phytase completely replaced the high quality herring meal protein of the control diet. Fish fed URPC diets had consistently inferior weight gains, specific growth rates, feed efficiencies and protein efficiency ratios relative to fish fed the fish meal (HFM and NFM) or DRPC diets. Besides a depression of growth, the URPC treatment groups also exhibited reduced thyroidal function. Neither essential amino acid supplementation nor cation-anion balancing improved fish performance when they received either the DRPC or URPC diets. Although no direct measurement of the effect of FINNSTIM™ on feed intake was performed, there was strong evidence that dietary inclusion of FINNSTIM™ helped maintain feed intake of the groups ingesting fish meal-free diets. No particular nutritional advantage was afforded by the use of low temperature Norse-LT94™ fish meal in place of high quality, domestically produced B.C. whole herring meal under the conditions of this investigation. 81 4.2 Introduction Feed costs are the single largest expenditure of salmonid culture and much of the cost of feed may be directly related to the cost of the fish meal component of the diet. Although most plant feedstuffs are typically cheaper than feedstuffs derived from animal sources, oilseed feedstuffs (e.g. soybean meal, cottonseed meal, rapeseed and canola meal) are particulary economically attractive due to their relatively high protein content, (typically 35-50% CP; Hatje, 1989), and their worldwide availability. Readily available on a worldwide scale, rapeseed and canola feedstuffs may well represent one of the best plant protein alternatives to fish meal protein. The level of protein found in rapeseed/canola concentrates, typically 65-70% CP, is comparable to the >68% CP levels found in fish meals. The protein is of good quality wherein the amino acid profile is well balanced, (Gillberg and Tornell, 1976), and closely resembles the essential amino acid profiles of FAO/WHO reference proteins (Jones, 1979) and natural prey organisms as well as the essential amino acid pattern of salmonids (Higgs et al., 1990; 1994a). Based on the essential amino acid index (EAAJ) approach of Oser (1959), protein quality has also been determined to be equivalent to that of whole herring meal and superior to that of soybean meal (Higgs, et al., 1988; 1989). However, with one exception (Kaushik et al., 1995) attempts to replace all of the fish meal in a modem salmonid diet with oilseed protein products have been unsuccessful, since typically, growth performance and nutrient utilization in these cases have been significantly reduced (Spinelli et al., 1979; Higgs et al., 1991; Teskeredzic et al., 1995). The limited success has often been ascribed to any one, or a combination of the following: a suboptimal amino acid balance, an improper balance or supply of available minerals, and the presence and physiotoxic effects of intrinsic antinutritional factors (ANFs) (Murai, 1992). ANFs are thought to be the factors limiting the nutritional value of rapeseed protein concentrate. As is common to all rapeseed and canola feedstuffs, rapeseed protein concentrates contain ANFs in the form of glucosinolates, saponins and phenolic compounds such as tanriins and sinapic acid. However, due to the nature of the processing methodology used for the production of RPC (e.g. the FRI-71 82 methodology of the Food Research Institute in Ottawa, Canada), these constituents are typically reduced to <1% (Youngs, 1991); consequently they are thought to exert rninimal effects on the utilization of RPC. Although the nutritive value of RPC may be negatively affected by the appreciable amounts of fibre (3.6-9.0% DMB; Jones, personal letters; Jones, 1979;Youngs, 1991) which remain in this product, it is the phytic acid component (3.6-7.5%) of RPC (Shah et a l , 1976; Jones, 1979; Thompson, 1990; Youngs,1991) that is believed to be the primary constraint to the full utilization of RPC/CPC in animal diets (Youngs, 1991). Strongly negatively charged over a wide range of pH, phytic acid readily binds multivalent cations, proteins and digestive enzymes (Cheryan, 1980; Irving, 1980; Spinelli et a l , 1983; Graf, 1986; Reddy et a l , 1989; NRC, 1993) such as trypsin (Graf, 1986), pepsin, and amylase (Sharma et a l , 1978; Thompson, 1986). The complexes that are formed are relatively insoluble. Hence, the bioavailability of protein and minerals can be reduced, and growth, feed conversion, and protein digestibility may be depressed. Additionally, phytic acid's chelating abilities may induce mineral deficiencies, as has been demonstrated in rainbow trout, whereby a phytic acid induced zinc deficiency resulted in the development of cataracts (Richardson et al., 1985). Structural anomalies in the pyloric cecae and histological evidence of depression of thyroid function have also been observed (Richardson et al.,1985). The enzyme phytase, which converts phytic acid to largely available forms of inorganic orthophosphates, and lower phosphoric esters of myoinositol (Nayani and Markakis, 1986; Reddy et a l , 1989) is not present in most fish (an exception is Black Bass (Cosgrove, 1980; Irving, 1980)). However it is produced by various microorganisms and plants and, is present within the intestinal mucosa of swine (Besecker et a l , 1967; Noland et al., 1968; Bayley and Thompson, 1969; Reddy et a l , 1989), and chicken. The supplementation of swine diets with commercial and crude preparations of microbial phytase (from fungal Aspergillus sp.) has been demonstrated to improve feed conversion (Lei et a l , 1992), and the absorption (Pallauf et a l , 1992), digestibility (Beudeker et a l , 1990; Khan and Cole, 1995) and retention of phytate phosphoms (Khan and Cole, 1995) in the presence and absence of inorganic phosphoms 83 supplements (Khan and Cole, 1995), as well as improve calcium digestibility and retention (Nasi, 1990). Furthermore, the use of a particular commercial Aspergillus phytase preparation known as Finase™ (from Aiko Biotechnology Inc., Rajamaki, Finland) has been reported to improve the apparent absorption of magnesium, iron, copper and manganese, (Pallauf et al., 1992) as well as the bioavailability of phytate phosphorus and zinc (Cromwell et al., 1993) when supplemented to swine diets. Additionally, through the dietary incorporation of plant feedstuffs with intrinsic phytase activity (e.g. rye bran) into com, soybean and rapeseed meal-based swine diets, Pointillart, (1991) demonstrated that the absorption and retention of phytate phosphorus and dietary calcium were improved. Although relatively few phytase investigations have been conducted with salmonids, there is evidence that the use of crude or commercial preparations of phytase may significantly improve the feeding value of phytate-rich feedstuffs for salmonids. Spinelli et al., (1979) found that pretreatment of soybean meal with microbial and/or wheat phytase prior to dietary incorporation, at a level of 50% fish meal substitution, improved growth and feed conversion of juvenile rainbow trout. Similarly, Cain and Garling (1995) supplemented soybean-based rainbow trout diets with microbial phytase and found that feed conversions were significantly improved to levels that were comparable to those obtained with commercial rainbow trout diets. Additionally, Riche (1993) supplemented soybean meal, canola meal and cottonseed meal-based rainbow trout diets with commercially available Finase™ phytase and found a significant improvement in apparent phosphorus absorption. Phytase supplementation of rapeseed protein concentrate, however, has so far not been found to fully enhance the nutritive value of this protein source for salmonids. In this regard, previous research found that undephytinized RPC could comprise approximately 22 % of dietary protein without compromising the performance of rainbow trout (Yurkowski et al., 1978) or juvenile chinook salmon (Higgs et al., 1982) and the upper dietary limit for inclusion of RPC was not established for either species. Recently however, Teskeredzic et al., (1995) conducted a study on trout in which RPC was pretreated with Finase S 2X™ and then this protein source was substituted for 33.3, 66.6 and 100% of the herring meal, (represented 19.6 %, 39 % or 59% of protein in a diet with 43% CP), in the 84 basal diet. The researchers also included 1.5% FINNSTIM™ in all diets, a compound known to have chemoattractant and chemostimulant properties, in an attempt to maintain equivalent feed intake in all groups. Although 100% fish meal replacement proved unsuccessful, Teskeredzic et al., (1995) found that both undephytinized and dephytinized RPC could replace 33.3 and 66.6 % of the fish meal in the basal diet without compromising the growth rate, feed intake, feed efficiency or protein and energy utilization of the trout. The lack of improved performance of the trout fed dephytinized RPC diets relative to the undephytinized diets was ascribed to the dephytinization procedure that was used which significantly reduced the quality of the RPC protein (Teskeredzic etal., 1995). Therefore, it remains to be established to what degree an improved dephytinization process may enhance the feeding value of RPC for salmonids. The supplementation of essential amino acids has been shown to significantly improve the feeding value of oilseed proteins (e.g. soybean) for fish (Dabrowska and Wojno, 1977). Although the essential amino acid profile of rapeseed and canola proteins appears to be well balanced, it has not yet been determined whether or not amino acid supplementation might improve feeding value. Additionally, there has been some suggestion that dietary cation-anion balance, in reference to Na +, K + and Cf with or without S~2, (Mongin, 1981; Block, 1992), (expressed as milliequivalents (mEq)/kg diet), may affect the utilization of amino acids by fish. Chiu et al. (1984) for example, found that whereas significant differences in growth and feed conversion were evident in rainbow trout fed diets balanced to 0 mEq/lOOg (with respect to Na +, K+and Cl"), no significant differences in growth or feed conversion were evident when the same diets were cation-anion balanced to - 20 mEq/lOOg (Chiu et al., 1984). However, a later study conducted by Chiu et al., (1987) using diets balanced to 0 mEq/kg or -200mEq/kg with respect to Na + , K+and Cl" and containing varying levels of lysine and arginine, showed that alterations in rainbow trout growth, feed efficiency and metabolism of amino acids varied as a function of the dietary content of these minerals rather than as a function of their overall cation-anion balance. These latter findings were in agreement with those of Wilson et al. (1985), who, instead of using amino acid test diets as in both studies of Chiii et al. (1984; 1987), utilized intact proteins and tested the effects of dietary electrolyte balances 85 ranging between (-9 and 63.8 mEq/lOOg) on growth and amino acid utilization in rainbow trout. Wilson et al. (1985) concluded that although dietary changes in electrolyte balance had little or no effect on growth or feed conversion of rainbow trout fed diets containing intact protein, their fmdings did not exclude the possibility of dietary electrolyte balance interactions with amino acid utilization in fish (Wilson et al., 1985). Therefore, the possibility remains that the cation-anion balance of diets may affect the feeding value of intact proteins such as those found in rapeseed protein concentrate for salmonids. Accordingly, this investigation was undertaken to: (1) re-examine if phytic acid removal or reduction significantly improves the feeding value/acceptability of rapeseed protein concentrate when it is used as a complete replacement for the fish meal component of rainbow trout diets, (2) determine if the amino acid profile of rapeseed protein concentrate is optimal for rainbow trout, and whether additional supplementation of essential amino acids is necessary, and (3) investigate whether dietary cation and anion balancing with respect to Na +, K + and Cl", and S"2 in the presence and absence of phytic acid in rapeseed protein concentrate would be beneficial for growth and/or feed efficiency. Juvenile rainbow trout were chosen as the experimental animal because they are generally more resistant to stress, and grow faster thatn salmon during periods of reduced day length. Also, small rainbow trout (~ 2 g), appear to be particularly sensitive to residual levels of glucosinolates in their diets, and they have been shown to be less tolerant than Pacific salmon of rapeseed and canola protein feedstuffs (Higgs et al., 1995). Therefore it was t hought that the findings with rainbow trout should be transferable to salmon, although this assumption requires confirmation. 4.3 Materials and methods Refer to chapter three for detailed descriptions of general methodologies. 4.3.1 Fish and experimental conditions In October, 1992 juvenile rainbow trout (Oncorhynchus mykiss) were obtained from Sun Valley 86 Trout Farms Ltd., Mission B.C.. At the time of receipt, the fish were selected for uniform weight (mean weight 2.18 g ± 1.1 S.D. (0.88 g)) and then randomly distributed into twenty 800 1 fiberglass tanks at a preliminary stocking density of 85 fish per tank (~ 0.25 g/1). Tanks were arranged in two rows (blocks) of ten tanks. Each tank was supplied with aerated well water at a relatively constant ambient temperature of 10.1-10.8° C. Flow rates were regulated to 8 1/min, and dissolved oxygen content was maintained at 10.1-10.2. ppm. A natural photoperiod was simulated with the use of daylight fluorescent lights (Vitalight, Durotest 40W), controlled by an astronomical timing cell. An 83-day growth study was conducted following a four week acclimation period. At the onset of the study, mean weights of fish were 3.9-4.6 g. Stocking densities were reduced to 75 fish per tank (~ 0.40g/l). 4.3.2 Diet preparation and allocation The experimental diets were prepared as described in chapter three, sections 3.8 and 3.12. Ten experimental dry diets were formulated to be isonitrogenous (42 % CP, DMB), isoenergetic (estimated metabolizable energy (ME) 17 MJ/kg, DMB), and equal in lipid content (17.5 % DMB). The protein feedstuffs utilized (Table 4.1) were: domestically produced B.C. whole herring meal (72.58% CP, DMB), undephytinized rapeseed protein concentrate (Bronowski FRI-73-5 (343)) (65.38 % CP, DMB, commercially dephytinized rapeseed protein concentrate (Bronowski FRI-73-5 (343)) (63.11 % CP, DMB), and Norse LT-94™ mixed-fish meal (76.22% CP, DMB), all of which supplied 59% of the dietary protein. The remainder (41% CP, DMB) of the dietary protein was supplied by commercially available poultry-by-product meal, blood flour (spray-dried), com gluten meal, extruded wheat and dried whey. Each diet was supplemented with FINNSTIM™ at a level of 1.5%. Rapeseed protein concentrate was obtained from Dr. J. Jones (deceased), formerly of the Food Research Centre (FRI) of Agriculture Canada, Central Experimental Farm in Ontario, Canada. The concentrate was prepared by the de-hulling and water and hexane extraction of seeds of the Bronowski 87 rapeseed variety, by the FRI-71 process. Prepared in 1974, the concentrate was kept in low humidity, non-refrigerated storage at the Food Research Centre, Ottawa until 1990 and then at the West Vancouver government laboratory until the commencement of this experiment. Diet compositions (Table 4.3) were developed after all protein sources were assessed for proximate (Table 4.1), amino acid and mineral composition (data not presented), and undephytinized and dephytinized rapeseed protein concentrate were analyzed for phytic acid (IP6), and lower phosphoric esters of inositol phosphate (IP5, IP4, IP3) to verify dephytinization (Table 4.1). The dietary treatments and abbreviations are presented in Table 4.2. The control diet (HFM) contained B.C. whole herring meal, which supplied 59% of dietary protein. The next four diets contained commercially dephytinized rapeseed protein concentrate (DRPC) which was isonitrogenously substituted for the fish meal. The first of the four diets was not modified further. The next diet was modified by the addition of an amino acid supplement that made dietary essential amino acid content mimic the essential amino acid profile of the control diet (DRPC + EAA). The next diet was supplemented with a mineral preparation that allowed the diet to rnirnic the levels of cations (as Na + and K + ) and anions (as Cl" and S"2) supplied by the herring meal control diet (DRPC + C/A). The last of the DRPC diets was modified by the addition of, both an amino acid supplement to mimic the essential amino acid profile of the control diet, and a mineral preparation that was cation/anion balanced in relation to the herring meal control diet (DRPC + EAA + C/A). The next four diets (URPC), (URPC + EAA), (URPC +C/A) and (URPC + EAA + C/A) were the same as the previous four diets, but the dephytinized RPC was replaced by the undephytinized RPC. The last diet (NFM) was essentially a second fish meal control wherein Norse LT-94™, a "low temperature" fish meal processed at (60-70°C), supplied 59% of dietary protein. This fish meal, has been reported to increase growth and improve feed conversion rate by at least 10% above the values obtained for salmonid fed diets with regularly processed fish meals (Pike et al., 1990), and thus was included in this 88 C/A + EAA C/A + + + CJ °j CJ CJ 2 Q 2 Q 2 Q 2 Q "8 oo 3 o is o "§ '3 T j <« o i s •c & <L> a o •a « i 1 1 CO e o •a a. •c u Q U 1 1 o o •a a <a a c o •a • & u CO u Q 1 5 o U J3 co •a •c 03 T J 1) C L .a si o o\ 1 i a .5 '3 £ "3 <D <U CO §• 2 •a <u N •a i C L u -a >, .o T j I a ••3 o ON 3 NT o\ •a o cl 73 c o ~5 o •a •S + I 1 o o c •a u OJ Si 2 T j i> N •a i o. u TJ TJ U I* e u o T3 cn ca 3 o o p. TJ 2 TJ 1) N •a i C L O TJ I T 3 C L § • a 3 o l-t Ou o T3 o OS TJ 73 a o "a o •a ca o + cl ii o CJ fl o o a o C L TJ 8-2 TJ CJ N •a i C L u •a TJ C L C L 9 CO g '53 o C L <L> ••a o o N OS cn •3 13 ° fl 151 o a e o o 1 1 s <a •«*• ON H o •a o >» x> T J "H, C L D CO fl u O L H C L <L> ••a SO 3 s ¥3 CJ + CJ 1 + CJ <: + CJ Ui Q >r 2 D + 0 y + O 1/1 Jo ° fl ca <a .-^  ca co g a c o fl 73 T3 M I M T J O 73 •a TJ U O •3 o N <L> o & T J n 01 C T f O rt <1> *"' "> 8 o «5 S3 a H o U w <u fl u s C L C L TJ o ca o c ••a & a u (J —J 73 •a A ca j 2 ? fl w 5: -2 90 study because it is known to be of the highest quality. The composition of the amino acid supplements used to equalize the dietary essential amino acid content of the rapeseed protein concentrate diets (DRPC + EAA), (DRPC + EAA + C/A), (URPC + EAA) and (URPC + EAA + C/A), with that of the herring meal control diet (HFM) are presented in Table 4.4. The compositions of the ten separate rnineral supplements that were used to equalize the dietary mineral supply of all diets and the cation and anion contents of diets (DRPC + C/A), (DRPC + EAA + C/A), (URPC +C/A) and (UPRC + EAA + C/A) with respect to the (HFM) control diet are presented in Table 4.5 and were based on the assessed mineral content (by analysis) and the cation (Na+ and K + ) and anion ( Cl" and S"2) content of the protein feedstuffs (Table 4.6). All diets were formulated to contain > 0.8 % inorganic phosphorus and >0.06 % magnesium from animal and inorganic sources. Completed diets were analyzed for proximate composition and, gross energy (Table 4.8), arnino acid and mineral content to confirm that the dietary supply met formulation expectations. Expected and assessed dietary arnino acid contents, expressed as % dietary protein, are presented in Table 4.9. Actual and expected dietary essential amino acid content, expressed as essential amino acid ratios (A/E) and essential amino acid indices (EAAI) are presented in Table 4.10.and expected dietary essential amino acid content, expressed as essential arnino acid ratios (A/E) and essential arnino acid indices (EAAI) are presented in Table 4.10.expected dietary essential arnino acid content, expressed as essential amino acid ratios (A/E) and essential amino acid indices (EAAI) are presented in Table 4. lO.dietary essential amino acid content, expressed as essential amino acid ratios (A/E) and essential amino acid indices (EAAI) are presented in Table 4.10.essential amino acid content, expressed as essential amino acid ratios (A/E) and essential amino acid indices (EAAI) are presented in Table 4.10.amino acid content, expressed as essential arnino acid ratios (A/E) and essential amino acid indices (EAAI) are presented in Table 4.10. Expected and analyzed dietary mineral content expressed as mg/kg dry diet are presented in Table 4.11 and Table 4.12, respectively. Expected and analyzed dietary cation and anion balance as milliequivalents (mEq) are presented in Table 4.7 and Table 4.13, respectively. 91 4.3.3 Commercial dephytinization Rapeseed protein concentrate was dephytinized by Aiko Ltd. Biotechnology, in Rajamaki, Finland, by means of pretreatement with a commercial phytase preparation (Finase™ 40) before dietary incorporation, according to the procedures described in chapter three (section 3.8). Samples of the initial undephytinized and dephytinized protein sources were then analyzed for inositol phosphates using Aiko method C-120 Rev. A, to confirm phytic acid removal (Table 4.1). 4.3.4 Feeding protocol Using a spoon for the distribution of pellets, fish were hand fed three times daily between 0830-1600 hrs.. During a four week acclimation period fish were fed a commercial diet at a rate of 1.5 % of body weight per day. During the experimental period, following the allocation of dietary treatments, and an initial sample (day 0), fish were fed their prescribed diet to satiation. Records of daily feed intake, fish mortality, fish behaviour, water temperature and dissolved oxygen content per replicate group were maintained. (Refer to section 3.14). 4.3.5 Sampling A detailed description of sampling protocol is provided in section 3.3. Sampling was conducted at least 17 hours post prandially. Fish were initially lightly sedated within their respective tanks with Marinil™ (metomidate hydrochloride) before being removed to an anaesthetic bath of tricaine methane sulfonate (TMS) buffered with sodium bicarbonate (NaHCOs), until they were fully anaesthetized. Fish taken for proximate determinations or blood and liver extraction were removed from the Marinil™ bath, killed with a quick cranial blow then blotted dry and weighed (g ± 0.01 g) before the extraction of blood and livers or before being vacuum packed and frozen for later determination of proximate composition. All other fish were removed from the buffered TMS bath, blotted dry then weighed individually (g ± 0.01 g). Fish were then allowed to recover in water containing Marinil™ before being returned to their respective tanks which had been cleaned and replenished with water in their absence. T a b l e 4 . 3 : C o m p o s i t i o n a n d e n e r g y c o n t e n t o f e x p e r i m e n t a l d ie ts f e d to j u v e n i l e r a i n b o w t r o u t f o r 83 days . I n g r e d i e n t D i e t ( g / k g DM)2 H F M D R P C D R P C D R P C D R P C + E A A + C / A + E A A + C / A B . C . w h o l e h e r r i n g m e a l ' 3 3 6 . 5 N o r w e g i a n N o r s e - L T 9 4 R 1 P o u l t r y b y - p r o d u c t m e a l 7 7 . 0 R a p e s e e d p r o t e i n c o n c e n t r a t e U n d e p h y t i n i z e d D e p h y t i n i z e d B l o o d flour; s p r a y - d r i e d 4 2 . 0 C o m g l u t e n m e a l 65 .5 E x t r u d e d w h e a t 191 .1 D r i e d w h e y 6 0 . 0 D L - m e t h i o n i n e 3 .54 L - l y s i n e H C 1 _ S a r d i n e o i l ( s t a b i l i z e d ) 3 119 .86 M i n e r a l s u p p l e m e n t 4 3 1 . 8 0 V i t a m i n s u p p l e m e n t 5 2 0 . 0 0 E t h o x y q u i n 0 . 0 1 P e r m a p e l l 6 9.72 A l p h a c e l l 18 .89 C a H P O < -C a C 0 3 -N a H 2 P 0 4 - H 2 0 2 . 1 1 K H 2 P O , -C a C l 2 -M g S 0 4 - 7 H 2 0 -K 2 S 0 4 -A m i n o a c i d s u p p l e m e n t 7 -C o n s t a n t c o m p o n e n t s 8 2 2 . 0 0 M e t a b o l i z a b l e e n e r g y 17.12 ( M J / k g ) 9 7 7 . 0 7 7 . 0 7 7 . 0 7 7 . 0 3 8 8 . 8 3 5 5 . 9 3 8 8 . 8 3 5 5 . 9 4 2 . 0 4 2 . 0 4 2 . 0 4 2 . 0 6 5 . 5 65 .5 6 5 . 5 6 5 . 5 138.4 138.4 138 .4 138 .4 6U.0 6 0 . 0 6 0 . 0 6 0 . 0 5 .09 5 .09 5 .09 5 .09 1.62 1.62 1.62 1.62 1 1 3 . 6 1 117 .01 1 1 3 . 6 1 1 1 7 . 0 1 3 1 . 8 0 3 1 . 8 0 3 1 . 8 0 3 1 . 8 0 2 0 . 0 0 2 0 . 0 0 2 0 . 0 0 2 0 . 0 0 0 .09 0 .09 0 .09 0 .09 9.73 9 .72 9.73 2 . 6 9 - 0 . 3 1 1.25 -- - - 1 6 . 0 1 16.64 15.26 12.24 14.43 7.73 7 .84 6 .78 _ - - 4 . 1 2 --30.5 -30 .5 2 2 . 0 0 2 2 . 0 0 2 2 . 0 0 2 2 . 0 0 16.83 16.93 16.83 16.93 93 T a b l e 4.3: ( C O N I T N U E D ) Composi t ion and energy content of experimental diets fed to juveni le rainbow trout for 83 days. Ingredient D i e t ( g / k g DM)2 U R P C U R P C U R P C U R P C N F M + E A A + C / A + E A A + C / A B . C . whole herr ing meal ' Norweg ian N o r s e - L T 9 4 R I Poultry by-product meal 77.0 Rapeseed protein concentrate Undephyt in ized 380.4 Dephyt in ized -B l o o d flour; spray-dried 42.0 C o m gluten meal 65.5 Ext ruded wheat 141.7 D r i e d whey 60.0 DL-meth ion ine 5.09 L- lys ine HC1 1.62 Sardine o i l (stabil ized) 3 M i n e r a l supplement 4 118.67 31.80 V i t a m i n supplement 5 20.00 E thoxyqu in 0.09 Pe rmape l l 6 5.48 A l p h a c e l l -C a H P 0 4 24.31 C a C 0 3 -N a H 2 P 0 4 H 2 0 4.32 K H 2 P 0 4 -C a C l 2 M g S 0 4 7 H 2 0 -K 2 S 0 4 -A m i n o acid supplement 7 -Constant components 8 22.00 Metabol izable energy 16.89 ( M J / k g ) 9 320.4 77.0 77.0 77.0 77.0 348.2 380.4 348.2 -42.0 42.0 42.0 42.0 65.5 65.5 65.5 65.5 138.4 138.4 138.4 191.1 60.0 60.0 60.0 60.0 5.09 5.09 5.09 3.54 1.62 1.62 1.62 _ 121.73 118.67 121.73 119.70 31.80 31.80 31.80 31.80 20.00 20.00 20.00 20.00 0.09 0.09 0.09 0.01 9.72 4.88 9.030 9.72 2.02 - - 37.23 24.61 25.43 25.60 -- 3.55 - -3.94 - - --3.65 - -- - 5.66 -26.3 - 26.3 -22.00 22.00 22.00 22.00 16.86 16.82 16.86 17.02 94 Table 4.3 (CONTINUED): Composition and energy content of experimental diets fed to juvenile rainbow trout for 83 days F i s h meal stabilized wi th 250 mg ethoxyquin/kg. H F M = Br i t i sh Columbian whole herring meal, conUol ; N F M = Norwegian N o r s e - L T 9 4 R fish meal, control; D R P C = dephytinized rapeseed protein concentrate; D R P C + E A A = dephytinized rapeseed protein concenUate supplemented w i t h essential amino acids to m i m i c amino acid profile of B . C . herring meal control diet; D R P C + C / A =dephytinized rapeseed protein concentrate balanced wi th respect to the cations N a + and K * and the anions C l " and S 0 4 " 2 ; D R P C + E A A + C / A = dephytinized rapeseed protein concenUate supplemented w i t h essential amino acids to m i m i c amino acid profile of B . C . herring meal control diet.and balanced wi th respect to the cations N a + and K + and the anions C l " and S O / 2 ; U R P C = undephytinized rapeseed protein concentrate; U R P C + E A A = undephytinized rapeseed protein concentrate supplemented w i t h essential amino acids to m i m i c amino acid profile o f B . C . whole herring meal control diet; U R P C + C / A = undephytinized rapeseed protein concentrate balanced wi th repect to the cations N a + and K + and the anions C l " and S O / 2 ; U R P C + E A A + C / A = undephytinized rapeseed protein concentrate supplemented wi th essential amino acids to mimic amino acid profile o f B . C . whole herring meal control diet, and balanced wi th respect to the cations N a + and K + and the anions C l ' and S 0 4 ' 2 . Stabil ized wi th 0 .05% ethoxyquin. The mineral supplements were distinct for each diet and are provided i n Table 4.5 The v i tamin supplement was prepared i n one batch and supplied the fol lowing levels o f nutrients/kg dry diet: D-ca lc ium pantothenate, 192.53 mg; pyridoxine HC1, 49.26 mg; riboflavin, 75 mg; folic acid, 18.75 mg; thiamine mononiUate, 55.99 mg; biotin, 75 mg; cyanocobalamine ( B 1 2 ) , 90 mg; menadione sodium bisulfite, 109.09 mg; DL-tocopheryl acetate (E), 300 I U ; cholecalciferol ( D 3 ) , 2400 I U ; v i tamin A acetate, 5,000 I U ; inosi tol , 400 mg; niacin, 304.57 mg; butylated hydroxytoluene, 22 mg; raw wheat starch (Whetstar 3 ^ , 17.993.01 mg. Permapcl l is a l ign in sulphonatc binder A m i n o acid supplement supplied as per Table 4.4 C o m m o n dietary components (g/kg dry diet) were as follows: ascorbic acid, 2; choline chloride (60%), 5; F i n n s t i m ™ , 15. Estimated by ascribing 0.0188 MJ/g protein, 0.0356 MJ/g l ip id , 0.0159 M J / g animal carbohydrate, and 0.00669 MJ/g raw starch (Beamish et a l . , 1986). Table 4.4: Composi t ion o f amino acid supplements used to equalize essential amino acid indices. Supplements A m i n o ac id (g/kg dry diet) I ' I I 2 argin ine 3 4.96 4.47 histidine - -isoleucine 2.10 2.10 leucine 4.10 3.80 lys ine 4 8.25 6.37 methionine 2.80 2.50 cysteine - -phenylalanine 1.50 1.20 tyrosine 2.10 2.00 threonine 2.50 1.80 tryptophan - 0.10 val ine 2.20 2.00 T h i s supplement was used to equalize the essential amino acid indices o f D R P C + E A A and D R P C + E A A + C / A wi th those o f the H F M control diet; the supplement was included at 30.51 g/kg o f the total diet formulation T h i s supplement was used to equalize the essential amino acid indices of U R P C + E A A and U R P C + E A A + C / A with those of the H F M control diet; the supplement was included at 26.34 g/kg of the total diet formulation A r g i n i n e was supplied i n the form of a rg in ine-HCl Lys ine was supplied i n the form of lys ine -HCl T a b l e 4 . 5 : C o m p o s i t i o n o f m i n e r a l s u p p l e m e n t s e m p l o y e d i n e x p e r i m e n t a l d ie ts f e d t o j u v e n i l e r a i n b o w t r o u t f o r 83 days, ( i n c l u d e s m a i n f o r m u l a s u p p l e m e n t a l m i n e r a l c o n t r i b u t i o n s ) D i e t ( m g / k g d r y d i e t ) ' I n g r e d i e n t H F M D R P C D R P C D R P C D R P C + E A A + C / A + E A A + C / A C a a s ( C a C 0 3 ) - 6 6 6 3 . 8 6 1 0 8 . 8 4 8 9 9 . 6 5 7 7 6 . 1 ( C a C l 2 ) - - - 1484 .8 8 6 . 6 ( C a H P 0 4 ) - 3 9 1 5 . 0 4 4 7 0 . 0 4 1 9 4 . 4 4 7 1 6 . 2 P as ( C a H P 0 4 ) _ 3 0 2 7 . 3 3 4 5 6 . 4 3 2 4 3 . 3 3 6 4 6 . 8 ( N a H 2 P 0 4 H 2 «0) 473 .3 - - - -( K H 2 P 0 4 ) 3 4 1 4 . 4 1758.6 1784 .0 1542.5 1 5 9 3 . 6 M g as ( M g S 0 4 7 H 2 Q ) 1077.5 3 5 0 . 0 3 5 0 . 0 3 5 0 . 0 3 5 0 . 0 C l as ( C o C l 2 6 H 2 0 ) 3.6 3.6 3.6 3.6 3.6 ( C a C l 2 ) - - - - 1 5 3 . 1 : ( N a C l ) 1726 .0 3905 .3 3 9 9 9 . 0 3 9 0 5 . 3 3 9 9 9 . 2 C u as ( C u S 0 4 5 H 2 0 ) 7.5 7.8 7.9 7.8 7 ,9 F e as ( F e S 0 4 7 H a O ) 105.7 90 .7 9 5 . 0 9 0 . 7 9 5 . 0 Z n as ( Z n S 0 4 7 H 2 0 ) 114.5 105.0 107.6 105 .0 107 .6 M n as ( M n S 0 4 H 2 0 ) 80 .3 47 .7 50 .8 4 7 . 7 5 0 . 8 N a as ( N a H 2 P 0 4 H 2 0 ) 351 .3 _ . _ _ ( N a F ) 5.5 5.5 5.5 5.5 5.5 ( N a C l ) 1119 .4 2 5 3 2 . 7 2 5 9 3 . 5 2 5 3 2 . 7 2 5 9 3 . 6 ( N a 2 S e 0 3 ) 0 .1 0 .1 0 .1 0 .1 0 . 1 K a s ( K H 2 P 0 4 ) 4 3 0 9 . 9 2 2 1 9 . 8 2 2 5 1 . 9 1947 .2 2 0 1 1 . 6 ( K I O j ) 1.5 1.5 1.5 1.5 1.5 ( K J ) 1.8 1.8 1.8 1.8 1.8 ( K 2 C 0 3 ) - 2 2 1 9 . 8 2 2 4 8 . 6 - -( K 2 S 0 4 ) - - - 2 4 9 2 . 5 2 4 9 2 . 2 l a s ( K I O j ) 5.0 5.0 5.0 5.0 5.0 ( K I ) 5.0 5.0 5.0 5.0 5.0 F as ( N a F ) 4.5 4.5 4.5 4.5 4 .5 C o as ( C o C l 2 6 H 2 0 ) 3.0 3.0 3.0 3.0 3.0 S e a s ( N a 2 S e 0 3 ) 0 .2 0.2 0.2 0.2 0 .2 S 0 42 - as ( C u S 0 4 5 H 2 0 ) 11.7 12.2 12.3 12.2 12.3 ( F e S 0 4 7 H 2 0 ) 181.9 156.0 163.4 156.0 163 .4 ( K 2 S 0 4 ) - - - 3 0 6 1 . 9 3 0 6 1 . 5 ( M g S 0 4 7 H 2 0 ) 4 2 9 1 . 2 1382.6 1382.6 1382 .6 1 3 8 2 . 6 ( M n S 0 4 H 2 0 ) 125 .1 6 8 . 0 73 .4 6 8 . 0 7 3 . 4 ( Z n S C y 7 H 2 0 ) 168.3 154.3 158.2 154.3 158 .2 A l p h a c e l l 1692 .2 3 4 7 5 . 9 1343 .8 8 9 4 . 6 7 6 9 3 . 8 R e f e r to T a b l e 4 .2 f o r e x p l a n a t i o n o f d ie t d e s i g n a t i o n s T a b l e 4 .5 ( C O N T I N U E D ) : C o m p o s i t i o n o f m i n e r a l s u p p l e m e n t s e m p l o y e d i n e x p e r i m e n t a l d ie t s f e d t o j u v e n i l e r a i n b o w t r o u t f o r 83 days, ( i n c l u d e s m a i n f o r m u l a s u p p l e m e n t a l m i n e r a l c o n t r i b u t i o n s ) D i e t ( m g / k g d r y d i e t ) ' I n g r e d i e n t U R P C U R P C U R P C U R P C N F M + E A A + C / A + E A A . + C / A C a a s ( C a C 0 3 ) ( C a C l 2 ) ( C a H P 0 4 ) 3 4 9 6 . 0 7 1 6 2 . 0 3 3 2 7 . 9 7 2 5 0 . 9 1 4 4 3 . 1 1 6 4 3 . 4 7 4 9 2 . 3 2 5 5 1 . 5 4 8 4 . 5 7 5 4 2 . 8 4 7 5 2 . 6 P as ( C a H P 0 4 ) ( N a H 2 P 0 4 H 2 0 ) ( K H 2 P 0 4 ) 5 5 3 8 . 0 9 6 9 . 1 1753 .9 5 6 0 6 . 8 8 8 3 . 4 1782 .2 5 7 9 3 . 4 969 .3 1 4 9 8 . 0 5 8 3 2 . 5 8 8 7 . 2 1 5 5 2 . 6 3 7 0 5 . 9 94 .3 5 7 1 . 4 M g a s ( M g S 0 4 7 H 2 0 ) 3 5 0 . 0 3 5 0 . 0 3 5 0 . 0 3 5 0 . 0 8 7 0 . 0 C l as ( C o C l 2 6 H 2 0 ) ( C a C l 2 ) ( N a C l ) 3.6 3 9 0 5 . 3 3.6 4 0 0 3 . 6 3.6 2 9 0 7 . 6 3 9 0 5 . 3 3.6 857 .3 3 9 9 9 . 2 3 .6 C u a s ( C u S 0 4 5 H 2 0 ) 8.0 8 .1 8.0 8.1 7 .8 F e a s ( F e S 0 4 7 H 2 0 ) 100.0 103.6 100.2 103 .6 1 0 4 . 1 Z n a s ( Z n S O < 7 H 2 0 ) 107.6 110 .1 107.7 110 .1 117 .8 M n a s ( M n S 0 4 H 2 0 ) 4 5 . 5 4 8 . 9 4 5 . 6 4 8 . 9 8 1 . 4 N a a s ( N a H 2 P 0 4 H 2 0 ) ( N a F ) ( N a C l ) ( N a 2 S e 0 3 ) 7 1 9 . 2 5.5 2 5 3 2 . 7 0 .1 6 5 5 . 6 5.5 2 5 9 6 . 4 0 .1 7 1 9 . 3 5.5 2 5 3 2 . 7 0 .1 6 5 8 . 4 5.5 2 5 9 3 . 6 0 .1 7 0 . 0 5.5 0 . 1 K as ( K H 2 P 0 4 ) ( K I O j ) ( K I ) ( K 2 C 0 3 ) ( K 2 S O < ) 2 2 1 3 . 9 1.5 1.8 2 2 1 3 . 9 2 2 4 9 . 6 1.5 1.8 2 2 4 9 . 6 1890 .9 1.5 1.8 2 5 4 4 . 0 1959 .9 1.5 1.8 2 5 3 9 . 4 7 2 1 . 3 6 1.5 1.8 I as ( K I 0 3 ) ( K I ) 5 .0 5.0 5.0 5.0 5.0 5.0 5.0 5.0 5.0 5.0 F a s ( N a F ) 4.5 4.5 4.5 4.5 4 .5 C o as ( C o C l 2 6 H 2 0 ) 3.0 3.0 3.0 3.0 3.0 S e a s ( N a 2 S e 0 3 ) 0.2 0.2 0.2 0.2 0.2 S 0 42 ' as ( C u S 0 4 5 H 2 0 ) ( F e S 0 4 7 H 2 0 ) ( K 2 S 0 4 ) ( M g S 0 4 7 H 2 0 ) ( M n S 0 4 H 2 0 ) ( Z n S 0 4 7 H 2 0 ) A l p h a c e l l 11.7 172.0 1382.6 6 4 . 2 158 .1 2 8 5 . 8 5.0 178.3 1382 .6 7 0 . 1 161.9 316 .7 12.5 172.3 3 1 2 5 . 2 1382 .6 6 4 . 4 158.3 3 0 5 9 . 6 12.7 178.3 3 1 1 9 . 4 1382 .6 7 0 . 1 161 .9 1952.3 12 .1 179 .1 3 4 7 1 . 8 127 .0 173 .1 2 4 3 1 . 0 R e f e r t o T a b l e 4 .2 f o r e x p l a n a t i o n o f d ie t d e s i g n a t i o n s 99 S3 o 1 3 X I I, 1 IS a I to 8 • 1^  IS i f 5 00 ao 1—1 • IO N O S «/-> oo 2 £ 3 oo O H £ ON cn ^ T ? < ; T * 2 oo 00 N © m 2 00 o § 2 oo i—i 1—1 '—' l-H «—1 © © © o o ON ON ON ON ON <s cs s + o Q + + u & Q 3 o •a u « ••a <<-< o a o •a CN T f ' 1 C M el 100 Following an initial sample on day 0, fish were sampled every 13-14 days. On day 0 and day 83 one and ten fish, respectively, were removed randomly from each replicate tank for the determination of final whole carcass proximate composition. The ten fish obtained on day 83 were also utilized for the determination of whole carcass mineral composition. During the terminal sampling period (day 83) an additional 18 fish were randomly withdrawn from each replicate tank. The caudal peduncle on each fish was cut and blood was withdrawn from the caudal vessels into ammonium heparinized capillary tubes which were sealed then centrifuged for the separation of plasma. Additionally, livers were excised from each of the eighteen fish and were weighed, foil-wrapped then quickly frozen on dry ice. Plasma and livers were stored at -70 °C until they were assayed for T 3 and hepatic 5D activity respectively, in order to assess thyroidal status. (Refer to sections 3.5 and 3.6) 4.3.6 Chemical analyses Detailed descriptions of all analytical procedures used in this investigation are presented in sections 3.4 -3.8 of chapter three. Proximate analyses, conducted according to the procedures outlined in Higgs et al., (1979), were performed in triplicate on protein feedstuffs, and in duplicate on diets with the exception that the protein analyses of each diet were performed in quadruplicate. Whole carcasses of fish withdrawn from each replicate tank on days 0 and 83 were pooled by fours and twos respectively, before each pool was homogenized, yielding 5 and 100 homogenized fish samples from each day. Mineral analyses of whole carcasses obtained at the terminal sample (day 83) were performed once on each homogenized sample such that there were 10 mineral values determined per dietary treatment. The plasma titres of T 3 reported for each dietary treatment, were determined according to the procedures of Omeljaniuk et al., (1984), and are the means of the 36 plasma samples obtained for each dietary treatment group (18 per replicate tank) at the termination sample (day 83). (Refer to section 3.5). The livers obtained at the terminal sample (day 83) were pooled by threes into 6 pools per dietary treatment for the determination of 5' monodeiodinase activity (measured as pmols of T 4 deiodinated to 101 T3/hr/mg microsomal protein) according to the procedures of Eales et al., (1992). (Refer to section 3.7). 4.3.7 Data calculation and analyses Essential amino acid balance was determined according to the procedures of Oser, (1959), using the following equation: EAAI= n "\j aai/AAi x zsnlAAi x x aan/AAn Where aa is the essential amino acid ratio (A/E) (Arai, 1981) of each essential amino acid, obtained as a percentage of the total essential amino acid content including cystine and tyrosine (Dy Penaflorida, 1989); AA= the A/E ratio in the reference protein (the HFM diet in this investigation); and n= the number of essential amino acids. Dietary cation (K + and Na+) and anion (Cf and S"2) balances were calculated according to the procedures of Mongin, (1981) and Dishington (1975) as follows: mEq/kg dry diet = [(dietary K + (mg/kg)/ atomic weight K + (g)) + (dietary Na + (mg/kg)/ atomic weight Na + (g))] -2 2 [ (dietary Cl ' (mg/kg)/ atomic weight Cf (g)) + (dietary S" (mg/kg)/ atomic weight S" (g)] Performance parameters used to measure the response to dietary treatments were weight gain (WG), specific growth rate (SGR), dry feed intake (DFI), feed efficiency (FE), and protein efficiency ratio (PER). The parameters were calculated as follows: Weight gain (WG) (g wet weight gain/fish) = (W2 - Wi) Where Wi is the initial mean wet weight of fish (g), and W2 is the final mean wet weight of fish (g) Specific growth rate (SGR) (% wet weight gain /day) = ((In W 2 - In Wi)/ T) x 100 102 Where W i is the initial mean wet weight of fish (g), W2 is the final wet weight of fish (g) , and T is the culture period in days Dry feed intake (DFI) (g dry feed intake/fish) = (total dry feed intake of fish in tank)(g) @ T / (number of fish in tank) @ T Where T is the culture period in days.T is the culture period in days.is the culture period in days.the culture period in days.culture period in days.period in days. DFI values per dietary treatment were obtained by averaging the DFI values for each replicate tank. Feed efficiency (FE) (g wet weight gain /g dry feed intake) = WG/ total dry feed intake of fish in tank) @ T Where T is the culture period in days. Protein efficiency ratio (PER) (g wet weight gain/g protein intake) = WG/ total intake of dietary protein @ T Where T is the culture period in days Additionally, because thyroidal hormones serve to regulate growth in fish (Donaldson et al., 1979; Cyr and Eales, 1988; Riley et al., 1993), and growth may be altered by the thyrotoxic effects of phytic acid (Richardson et al., 1985) as well as the hepato- and thyrotoxic effects of glucosinolates and hydrolytic products of glucosinolates (Yurkowski et al., 1978; Higgs et al., 1979), plasma titres of T3 and hepatic 5D activity were determined in order to assess liver and thyroidal status, (hepatic 5' monodeiodinase (5'D) is responsible for the conversion of thyroxine ( T 4 ) , to the cellularly active 3,5,3' triiodo-L-thyronine (T3) through the removal of one iodine from the 5' position of the phenolic ring of T 4 (Riley et al., 1993). 103 4.3.8 Statistical analyses Statistical analyses were performed using the computer package Systat 6.0 for Windows , (SPSSR Inc., Chicago Illinois, 1996). All statistics were tested at the 0.05 probability level. Following a Bartlett's test for homogeneity of variance, performance data were subjected to two-way anaylsis of variance (ANOVA) according to the general linear means model. Where F-values indicated significance, orthogonal contrasts were used to compare dietary treatment means. 4.4 Results 4.4.1 Assessed composition of experimental diets As presented in Table 4.1, phytic acid removal was not 100 % complete. Commercial dephytinization of rapeseed protein concentrate resulted in a 95.9% reduction in phytic acid (IP6) content (52.2 umol/g vs 2.1 umol/g) which was accompanied by a 3.5 % decrease in protein content. Although the completed diets were not assessed for phytic acid content and the contribution of phytic acid from extruded wheat is largely unknown, the total dietary concentrations of phytic acid (IP6), based on the analyzed values for IP6 in undephytinized and dephytinized-RPC before dietary incorporation, would be expected to be in the order of approximately 700 - 800 umol/g for the dephytinized RPC diets and approximately 18,000 - 20, 000 umol/g in the undephytinized RPC diets. Estimates of phytic acid content for the fish meal diets (HFM and NFM) were assumed to be in the order of 0.5%, which is the level of phytic acid typically found in commercial trout diets (Spinelli et al., 1983; Hilton, 1989). Proximate determinations and bomb calorific measurements of diets confirmed that all diets were isonitrogenous (42.16 - 43.59 % CP, DMB), isoenergetic (22.27 - 23.06 MJ/kg GE, DMB) and similar in lipid content (17.53 - 19.49 % lipid, DMB), closely approximating formulation estimates (Table 4.8). Diets also varied only slightly in dry matter (92.17 - 93.15 %) and ash (8.23 - 9.24 %, DMB) content. (Table 4.8). With the exception of lysine, the essential amino acid content of all diets closely approximated 104 a s s Sr1 ^ Q + Q 3 + (Ll 13 CL, CS r-OS 0 0 cs OS oo rs OS T f OS CS OS ro OS 00 r-4 OS OS SO rs OS 00 in CS OS OS <L> ts Q ro f S rs T f 0 0 CM CN T f SO T f T f 00 T f O rs ( N T f ro T f so uo T f OS rn T f ro OS rs T f ro T f Q O S CX •a 2 u ro rs SO 00 so rs cs CN T f ro rs rs t -cs — rs o Q SO O ro oo Os T f OS OS rs ro os o ro oo' ro od ro •—< Os os 1 3 CX u T J s SO OS rs rs o o ro rs © ro' rs so © CO cs OS 00 rs CN SO CN CN so CN CN so ts CN OS rs CN Q 60 ^ SB ts a o •a <L> •a •a (41 o ts o •a £ 2 CN T f (D •s CX u (1 u 3 1 u s (A u 3 > a o CX o tS o a cx 8 X u I T a b l e 4.9: A n a l y z e d a n d f o r m u l a t e d leve ls o f essent ia l a m i n o ac ids ( i n r e l a t i o n t o r a i n b o w t r o u t r e q u i r e m e n t s ) , i n e x p e r i m e n t a l d ie t s f e d t o j u v e n i l e r a i n b o w t r o u t f o r 83 days. D i e t ] P a r a m e t e r H F M D R P C D R P C D R P C D R P C + E A A + C / A + E A A + C / A A m i n o a c i d s 2 ( % o f d i e t a r y p r o t e i n ) R e q u i r e m e n t A r g 6 .0 6 . 7 6 2 6 .29 6 .63 6 .22 6 . 3 1 ( 6 . 1 5 ) ( 5 . 5 2 ) ( 6 . 0 8 ) ( 5 . 5 3 ) ( 6 . 0 8 ) H i s 1.8 3 .32 3 .35 2 .97 4 . 1 2 3 .34 ( 2 . 8 1 ) ( 3 . 0 6 ) ( 2 . 8 5 ) ( 3 . 0 7 ) ( 2 . 8 5 ) He 2 .2 4 . 2 1 4 .05 4 .13 3 .97 3 .86 ( 3 . 8 7 ) ( 3 . 5 6 ) ( 3 . 8 2 ) ( 3 . 5 8 ) ( 3 . 8 2 ) L e u 3.9 9 .20 8.93 9 .31 8.83 8 .79 ( 8 . 8 4 ) ( 8 . 2 6 ) ( 8 . 7 6 ) ( 8 . 2 8 ) ( 8 . 7 6 ) L y s 5.0 4 .09 3 .30 4 . 2 1 3 .44 3 . 5 1 ( 6 . 3 7 ) ( 5 . 0 8 ) ( 6 . 3 0 ) ( 5 . 1 0 ) ( 6 . 3 0 ) M e t & C y s " 4.5 ( 4 . 1 7 ) ( 3 . 9 8 ) ( 4 . 4 0 ) ( 3 . 9 9 ) ( 4 . 4 0 ) M e t 5 3.32 3 .20 3 .32 2.93 3 .19 ( 2 . 9 1 ) ( 2 . 3 5 ) ( 2 . 8 9 ) ( 2 . 3 6 ) ( 2 . 8 9 ) P h e & T y r 6 5.1 8.23 8.23 8 .42 8 .12 7.63 ( 7 . 6 4 ) ( 7 . 1 5 ) ( 7 . 5 5 ) ( 7 . 1 8 ) ( 7 . 5 5 ) T h r 2.2 3 .99 3 .77 4 .15 4 .05 4 . 1 6 ( 3 . 9 8 ) ( 3 . 5 8 ) ( 3 . 9 3 ) ( 3 . 5 9 ) ( 3 . 9 3 ) T r p 7 0.5 ( 1 . 0 0 ) ( 1 . 0 4 ) ( 0 . 9 7 ) ( 1 . 0 4 ) ( 0 . 9 7 ) V a l 3 .2 5 .56 5 .44 5 .50 5 .36 5 .25 ( 5 . 2 3 ) ( 4 . 9 6 ) ( 5 . 1 7 ) ( 4 . 9 8 ) ( 5 . 1 7 ) T a b l e 4 .9 ( C O N T I N U E D ) : A n a l y z e d a n d f o r m u l a t e d leve ls o f essent ia l a m i n o a c i d s ( i n r e l a t i o n t o r a i n b o w |Qg t r o u t r e q u i r e m e n t s ) , i n e x p e r i m e n t a l d ie ts f e d to Jeven i le r a i n b o w t r o u t f o r 83 days . D i e t j P a r a m e t e r 2 U R P C U R P C U R P C U R P C N F M + E A A + C / A + E A A + C / A A m i n o ac ids ( % o f d i e t a r y p r o t e i n ) R e q u i r e m e n t 3 A r g 6 .0 6 .48 6 . 0 1 6 . 4 1 7 . 3 4 6 .83 ( 5 . 6 2 ) ( 6 . 1 3 ) ( 5 . 6 2 ) ( 6 . 1 3 ) ( 5 . 8 6 ) H i s 1.8 3.73 3.05 3 .74 3 .89 2 . 8 4 ( 3 . 0 4 ) ( 2 . 8 6 ) ( 3 . 0 4 ) ( 2 . 8 6 ) ( 2 . 8 8 ) He 2.2 4 . 0 1 3 .91 3 .87 4 .55 4 . 1 0 ( 3 . 5 8 ) ( 3 . 8 6 ) ( 3 . 5 8 ) ( 3 . 8 6 ) ( 3 . 6 9 ) L e u 3.9 9 .05 8 .58 8 .99 10 .02 9 .20 ( 8 . 3 3 ) ( 8 . 8 3 ) ( 8 . 3 4 ) ( 8 . 8 3 ) ( 8 . 4 0 ) L y s 5.0 3 .50 3.33 3.43 4 .53 3 .92 ( 5 . 4 6 ) ( 6 . 3 5 ) ( 5 . 4 7 ) ( 6 . 3 5 ) ( 6 . 2 0 ) M e t & C y s 4 4.5 ( 4 . 0 5 ) ( 4 . 4 3 ) ( 4 . 0 4 ) ( 4 . 4 3 ) ( 4 . 0 0 ) M e t 5 2 .98 3 .21 2 .97 3 .58 3 . 3 1 ( 2 . 4 2 ) ( 2 . 9 0 ) ( 2 . 4 2 ) ( 2 . 9 0 ) ( 2 . 7 3 ) P h e & T y r 6 5 .1 7 .73 8 .15 7.73 9 .52 8 .35 ( 7 . 2 4 ) ( 7 . 6 1 ) ( 7 . 2 4 ) ( 7 . 6 1 ) ( 7 . 1 8 ) T h r 2 .2 4 . 2 2 3 .74 4 .23 4 . 6 0 4 . 3 4 ( 3 . 7 7 ) ( 3 . 9 8 ) ( 3 . 7 8 ) ( 3 . 9 8 ) ( 3 . 4 9 ) T r p 7 0.5 ( 1 . 0 4 ) ( 1 . 0 0 ) ( 1 . 0 4 ) ( 1 . 0 0 ) ( 1 . 0 0 ) V a l 3.2 5.53 5 .02 5 .30 5 .89 5 .49 ( 5 . 0 2 ) ( 5 . 2 2 ) ( 5 . 0 2 ) ( 5 . 2 2 ) ( 5 . 0 0 ) Table 4.9 ( C O N T I N U E D ) : Analyzed and formulated levels of essential amino acids ( in relation to rainbow trout requirements), i n experimental diets fed to juvenile rainbow trout for 83 days. Refer to Table 4.2 for explanation of diet designations. E a c h value represents the mean of two replicate samples; bracketed values represent the formulated values Required ranges or levels recommended i n literature (Higgs et a l . , 1994; Haard, 1995). Cystine can replace ha l f the methionine requirement Cyst ine analysis was unavailable at the time of dietary assessment. Meth ion ine estimates and the estimates of the dietary supply of other amino acids closely approximated actual dietary supply. Consequently, it is assumed that dietary requirement for methionine and cysteine was adequate. Tyrosine can spare ha l f the phenylalanine requirement Tryptophan analysis was unavailable at the time of dietary assessment T a b l e 4 . 1 0 : E s s e n t i a l a m i n o a c i d i n d i c e s ( E A A I ) a n d a n a l y z e d a n d e x p e c t e d essent ia l a m i n o a c i d r a t i o s ( A / E ) o f e x p e r i m e n t a l d ie ts f e d to j u v e n i l e r a i n b o w t r o u t f o r 83 days. D i e t 1 2 A m i n o a c i d s 3 H F M D R P C D R P C D R P C D R P C + E A A + C / A + E A A + C / A A r g 14 .90 14 .51 14.63 14 .10 14.73 ( 1 2 . 2 9 ) ( 1 2 . 2 1 ) ( 1 2 . 2 1 ) H i s 7 .32 7.73 6.55 9 .34 7 .79 ( 5 . 6 3 ) ( 5 . 7 2 ) ( 5 . 7 2 ) l i e 9 .28 9 .34 9 .11 9 .00 9 . 0 1 ( 7 . 1 7 ) ( 7 . 6 6 ) ( 7 . 6 6 ) L e u 2 0 . 2 8 2 0 . 6 0 2 0 . 5 4 2 0 . 0 2 2 0 . 5 1 ( 1 7 . 6 8 ) ( 1 7 . 5 7 ) ( 1 7 . 5 7 ) L y s 9 .02 7 . 6 1 9 .29 7 .80 8 .19 ( 1 2 . 7 2 ) ( 1 2 . 6 5 ) ( 1 2 . 6 5 ) M e t & C y s 4 - - -M e t 5 ( 8 . 3 2 ) ( 8 . 8 4 ) ( 8 . 8 4 ) P h e & T y r 6 18 .14 18.98 18.58 18 .41 17 .81 ( 1 5 . 2 7 ) ( 1 5 . 1 6 ) ( 1 5 . 1 6 ) T h r 8 .80 8 .69 9 .16 9 .18 9 .71 ( 7 . 9 4 ) ( 7 . 8 8 ) ( 7 . 8 8 ) T r p 7 . . . . . ( 1 . 9 9 ) ( 1 . 9 4 ) ( 1 . 9 4 ) V a l 12.26 12.55 12.14 12.15 12.25 ( 1 0 . 4 5 ) ( 1 0 . 3 8 ) ( 1 0 . 3 8 ) E A A I 8 100 9 9 . 4 1 99 .46 100 100 Table 4.10 (continued): Essential amino acid indices ( E A A I ) and analyzed and expected essential amino acid ratios ( A / E ) of experimental diets fed to juvenile rainbow trout for 83 days. Diet; A m i n o a c i d s 3 U R P C U R P C U R P C U R P C N F M + E A A + C / A + E A A + C / A Arg 14.64 His 8.43 He 9.06 Leu 20.45 Lys 7.91 Met & Cys4 Met5 Phe & Tyr6 17.47 Thr 9.54 Trp7 Val 12.50 14.38 14.67 14.58 15.15 (12.20) (12.22) 7.30 8.56 7.73 6.30 (5.69) (5.69) 9.36 8.86 9.04 9.10 (7.68) (7.68) 20.53 20.57 19.90 20.41 (17.57) (17.57) 7.97 7.85 9.00 8.70 (12.64) (12.64) (8.81) (8.81) 19.50 17.69 18.91 18.53 (15.13) (15.13) 8.95 9.68 9.14 9.63 (7.91) (7.91) (1.98) (1.98) 12.01 12.13 11.70 12.18 EAAI 100.79 99.23 100.72 100.29 99.23 (100.00) (100.00) 110 Table 4.10 ( C O O T I N U E D ) : Essential amino acid indices ( E A A I ) and essential amino acid ratios ( A / E ) o f experimental diets fed to juvenile rainbow trout for 83 days. 1 Refer to Table 4.2 for explanation of diet designations. 2 E a c h value represents the mean of two replicate samples; bracketed values represent the estimated/expected values 3 Values given represent the ratio of the (dietary content o f a particular essential amino acid/ total essential amino acid content o f diet) x 100 ; Because cystine and tryptophan analyses were unavailable at the time of assessment, the denominator o f this ratio considers only those essential amino acids for w h i c h there were analyses 4 Cyst ine can replace ha l f the methionine requirement 5 Cystine analysis was unavailable at the time of dietary assessment. Meth ionine estimates and the estimates of the dietary supply of amino acids closely approximated actual dietary supply. Consequently it is assumed that dietary requirement for methionine and cysteine was adequate. 6 Tyrosine can spare ha l f the phenylalanine requirement 7 Tryptophan analysis was unavailable at the time dietary assessment 8 Calculated according to the procedures of (Oser, 1959), using the ( H F M ) control diet as the reference protein, and calculated without the values for cystine and tryptophan which were unavailable at the time of dietary assessment. Ill 11»s So Q + PH < 2 W Q + -M e ca L H C3 OH CO t-H © m r -cs O O © ON p T t © © © NO o m CN CN 00 T t cn © m f~ CS T t ON o ON p m © © © NO ! © m t-H Tt' f N 00 T t cn © '/-> t> cs o ON p © © © © NO © T t cn NO 1—1 oo' T t cn i n t-H O m r~ cs T t ON o ON p m © © © NO © Tf' CS m 00 T t m 1—1 p •n t -f N m t-H © ON p © • © © o NO © T t ^ H m' NO i—1 00 T t CO p m r-fN T t ON o ON p T t © © © SD ' © m t-H T f fN fN 00 T t m p i n r-fN T t © ON o T t © © © NO © T t m CN 00 T t .0002 © T t T t T t ,0002 © CN r n t-H >rl .0002 © t~ f N cn m ,0002 m 00 T t ^H O N f N 0002 00 <n O N f N <n T t m T t m T t m t—i © © m © m f N m r-f N m fN © T t O N O N © © T t O N - H O © © O N © © T t CN © © © NO. © © T t © , NO fN 00 T t ^H © © T t © SD CN 00 T t , © ! £ • © © f N © © © f N © © © CN © © © fN © © © f N © © © © en p cn © c n ca X> VH 1 o s a so ca 5 PH ca c fl o c l fl .£? 'to i> T S •M •3 CM O fl o 0> CM OH T J « 73 I CD § •8 O N 00 ON 73 o .fl T3 S ca 1 6 eg 'co C3 6 N O © © ' A l co" a O 'Xl OH CO o XI a. o 6X1 VH O .fl X = 0s-00 © ' A l I .3 T J 73 c2 a -4—» ••5 73 112 2 3 < S a y Q + 2 w a + IS o cn CM CM' 0 0 cn cn N O co O T f ' cn cs O N CS o o CS o o CM © o cs o © cs © © cs © © cs © © CN © © CS © © CN © © © Tl o w-i O Tl © NO © © T f © © T f © cn © o cn 3 O N uo O N cs N O O N NO —< cs r-© cn CN O N cn cs T f N O O N cs © O N NO « n cs r-© N O V I CN t— 0 0 TI oo cs r-© TI cs © T f CN m N O © o O N r-© CN o o © O N © r-© O N r-© T f 0 0 © O N N O © O N N O © i n cn oo T f Tl m' cn cn cn i n m cs' © cs' m CM' T f CN T f N O N O r-O N N O CS cn NO' cn r-' oo N O ' O N T f N O ' o r-' O N cn >o' T f N O ' r-oo 0 0 o T f CN 0 0 T f T f CN CN I n CS T f CS © CS T f © O N cs 0 0 cs « -" cn O N r—I cn CN T f 0 0 O N T f 0 0 CN 0 0 © © © 0 0 t -V © r-' m © © O N © V © ' r-' i n © © 0 0 © N O V © ' N O m © © N O cn © NO_ V O N vi « n © © O N cn © O N V O N vi Tl © ©_ © 0 \ V © ' r-' Tl © ©_ CS © V ©' r-' Tl © Tl © © Tl V O N r-' Tl © © CS © v' oo' N O ' Tl © © O N © Tl V © ' vi 3 I? c o '3 c .£? 'w I . M r-a <<-> o e o •a OJ CN T f o 03 u 3 oj J? 2 e N OJ 0 0 113 3 , 3 •a oo O S CJ + a « .2 -a u D o WO 0 0 o o os O S C O O S o o SO »™H — i o 0 0 O S so T f r-0 0 cs o o O S T f SO O S C O 0 0 »—, VO 0 0 cs SO- so' cs cs cs 0 0 r-O S SO 0 0 o T f 0 0 T f 0 0 vo 3 + u s + + u Q a o •a OJ O a •1 cs 3 a! . 6 0 I 6 0 3 1 "3 3 ; f w j r oo os s ° w •-" C J 3 O J 3 N ^ u * y e •^ "l-S 5 s bo S — #a ^ « - a 8 SP — . a, o . 5 s « T l O 114 calculatedVexpected values and met or exceeded known dietary requirement levels (Table 4.9). Lysine levels varied across all diets from 3.30-4.53% dietary protein, regardless of whether or not the diets were supplemented with amino acids. Although all lysine levels were determined to be lower than the 5.0 % of dietary protein that they were formulated to contain, all levels were within the required dietary range reported for lysine (3.7-6.1 % protein; NRC, 1993; Haard, 1995) in the literature for trout (Table 4.9). As depicted in Table 4.10 dietary EAAI determinations were not distinctly different across diets (99.41 -100.0), regardless of the presence or absence of EAA supplementation (Table 4.1). It must be noted however, that cystine analysis and analysis for tryptophan were unavailable at the time of dietary assessment. Consequently, the dietary EAAI values were calculated on the basis of analyzed levels of nine essential amino acids (Table 4.10). Nonetheless, as shown in Table 4.9, considering that the dietary contents of essential amino acids closely approximated expected values, with the exception of lysine, the actual dietary cystine and tryptophan levels likely met expected levels. Actual dietary mineral content closely approximated calculated/estimated levels (Table 4.11 and Table 4.12). However, actual dietary cation-anion balance (Table 4.13) did not agree with formulated estimates (Table 4.7). Although differences between calculated and analyzed levels of cation-anion balance have also occurred in other investigations, (Wilson et al., 1985), reasons for the differences observed in this investigation are unclear. Actual K, Mn, Cl and Zn levels were somewhat lower than expected values, and S levels were higher than expected, but all other mineral levels were remarkably similar to the calculated values (Table 4.11 and 4.12). No alkenyl glucosinolates were detected in diets; only indoyl glucosinolates were present. Fish meal diets (HFM and NFM) were glucosinolate-free, and only the DRPC + EAA + C/A treatment of the dephytinized-RPC group of diets contained glucosinolates (0.2 umol/g). However, all undephytrnized-RPC diets contained glucosinolates at levels of 0.4- 0.5 umol/g. (Refer to Table 4.14). 4.4.2 Influence of diet treatment on fish growth, appetite, feed utilization and protein utilization The performance data are presented in Table 4.17 with orthogonal contrasts in Table 4.18. The performance of the Norse-LT94™ (NFM) treatment group and the dephytinized-RPC 115 treatment groups, were not significantly different (P > 0.05) from the performance of the control (HFM) group, in terms of weight gain (WG), specific growth rate (SGR), feed intake (DFI), feed efficiency (FE) and protein efficiency (PER). The exception was the fish fed the cation-anion balanced dephytinized RPC diet (DRPC + C/A) which had significantly lower weight gain (P < 0.05) relative to the control group but was otherwise not significantly different (P > 0.05) in terms of SGR, DFI, FE or PER. The weight gains, specific growth rates, feed efficiencies and protein effciency ratios of all undephytinized-RPC treatment groups were significantly inferior (P < 0.05) to those of the HFM and NFM treatments. Although mean feed intake values were also depressed, the results were not always significant, (e.g. URPC + EAA vs HFM). Relative to the NFM treatment groups, most groups ingesting the dephytinized RPC diets had significantly (P < 0.05) lower protein efficiency ratios but were otherwise not significantly different (P > 0.05) in terms of WG, SGR, DFI, and FE. The one exception was the DRPC treatment, which did not differ significantly from the NFM treatment in terms of any of these parameters. The only significant difference (P < 0.05) found among the dephytinized treatment groups, was a lower feed intake (DFI) for the DRPC + C/A group, relative to the DRPC + EAA + C/A group. Among the undephytinized treatment groups, the URPC + EAA + C/A group had significantly lower (P < 0.05) growth than the URPC group, and relative to the URPC + EAA treatment group, both the URPC + C/A and the UPRC + EAA + C/A groups grew and ate significantly less (P < 0.05). Across all treatments, with respect to weight gain, specific growth rate, feed intake, feed efficiency and protein efficiency ratio, the performances of the undephytinized treatment groups were significantly inferior to those of the dephytinized treatment groups. 4.4.3 Influence of diet treatment on body composition Dietary treatment did not significantly affect the terminal levels of dry matter (26.86% - 29.67%), protein, (40.90% - 48.34% DMB), lipid (36.36 - 40.69 % DMB), and ash (6.12% - 9.84% DMB) (Table 4.15). This was also true for various minerals i.e., Ca, Cr, Cu, Fe, Mg, Mn, P, K, Na and Zn when expressed on a wet weight basis (Table 4.16). The carcass content of minerals closely approximated "normal" mineral levels for rainbow trout < 1500g (Shearer, 1984). However, although not significant (P > 116 0.05), all levels of Cu (1.7- 8.8 umol/g wet weight) and Zn (28.4- 34.7 umol/g wet weight) were higher than "normal" Cu ( 0.74 - 1.66 umol/g wet weight) and Zn (23.4 - 26.6 umol/g wet weight) levels found in fish fed commercial salmonid diet preparations (Shearer, 1984). Additionally, although not statistically significant, carcasses offish from the (DRPC) and (DRPC + C/A) treatment groups had particularly higher Cu levels (8.8 and 5.6 umol/g wet weight, respectively) than the range of Cu levels observed for the other treatment groups (1.7-2.6 umol/g wet weight). 4.4.4 Influence of diet treatment on thyroid and liver status Data are presented in Tables 4.19 and 4.20. Fish fed the NFM diet had plasma T 3 titres (2.641 ± 0.281 ng T 3 /ml) and liver weights (0.51 ± 0.503 g) which were significantly lower (P < 0.05) than those obtained for the control group (HFM) (4.014 ± 0.310 ng T3/ml and 0.59 ± 0.587 g respectively), despite the fact that high quality fish meal was the primary protein source (59 % CP, DMB) for both dietary treatments. The 5'D activity levels were not significantly (P > 0.05) different, however. Relative to the controls (HFM), the livers were smaller and the 5'D activity levels were higher (P < 0.05) for fish from all of the dephytinized-RPC treatment groups. However, plasma titres of T 3 were for the most part not significantly different from those of the fish fed the control diet (HFM), with the exception that fish consuming the diet supplemented with essential amino acids, (DRPC + EAA) had plasma levels of T 3 which were significantly lower (P < 0.05) than control levels. All undephytinized -RPC treatment groups also had smaller livers relative to the control treatment group (HFM). However, 5'D activity levels were not significantly different, while plasma titres of T 3 were significantly reduced (P < 0.05). Compared to the Norse-LT94™ (NFM) treatment group, the other fish meal control, the dephytinized-RPC treatment group supplemented with essential amino acids and anion-cation balanced (DRPC + EAA + C/A) had significantly (P < 0.05) higher plasma levels of T 3 , whereas levels found in all other dephytinized groups were not significantly different. Additionally, with the exceptions that the 5'D activity levels of the DRPC + EAA treatment group and the average liver size for the DRPC treatment 117 group were not significantly different from those of the NFM group, all 5'D activity levels were significantly higher (P < 0.05) and liver weights were significantly lower than those of the N F M treatment group. With the exception that plasma titres of T 3 for the URPC + EAA + C/A were significantly lower (P < 0.05) relative to the NFM treatment group, all undephytinized treatment groups had plasma titres of T 3 and 5'D activity levels that were not significantly different (P > 0.05), and liver weights which were significantly smaller (P < 0.05), than those of the N F M treatment group. Among the dephytinized treatment groups, the lower plasma titre of T 3 of the DRPC + EAA treatment group relative to the DRPC + EAA + C/A treatment group was the only significant (P < 0.05) difference found. With the exception that liver weights for the URPC + EAA treatment group were significantly higher (P < 0.05) than those of the URPC + C/A treatment group, no other significant differences were detected among the undephytinized treatment groups. Mean plasma T 3 and 5'D activity levels for all undephytinized-RPC treatment groups were determined to be significantly lower than the respective mean levels for all dephytinized-RPC treatment groups, although liver size did not differ significantly (P > 0.05). 4.4.5 Influence of diet treatment on general health and behaviour There were no apparent differences in fish health and behaviour as a result of dietary treatment. All fish appeared to exhibit normal feeding behaviour and showed no signs of altered health status. Following two days of some spitting of feed pellets in the dephytinized and undephytinized-RPC treatment groups, possibly due to the increased hardness of the pellets relative to the fish-meal diets, fish readily accepted and consumed all diets. There were mortalities < 4 % of initial number of fish in each treatment; Table 4.17. However, mortalities were not a consequence of dietary treatment but occurred at the time of sampling. 4.5 Discussion 4.5.1 Fish growth and physiological response to dietary treatments The results of this study indicate that the replacement of the fish meal component (59 % dietary 118 •»-< co Q rV ^ Q + o^ <! Q + f3 OH c o •a •-H to o a S o U u •a o a, CO CO CO CO Z Z Z Z _ OS OS SO ro to T t <s SO CS SO ro cs es r- OS m CO Os o •o © ' © © ' © ' o t > © « n so T t CO ON so so 00 CM T t CO T t © SO Os CO SO cs r-' 00 so' SO* cs T t ro ro Os so © CS CO m 00 t - ' so" so cs T t CO o OS ro © CS T t t - T t 00 oo' SO 00 cS T t ro SO 00 T t I H T j " o 00 00 Os r-' Os' OS es T t CO cs o T t tn so Os so T t Os' o so so" CS T t ro cs OS ro o r- 00 OS 00° so f-' cs T t ro r- _ { OS SO 00 OS so Os SO © ' cs T t T t ro ro 00 SO SO CS ro © OS SO sd © cs T t T t 00 Os cs SO , - H r-OS © SO sb cs T t ro CS SO T t in SO Os ro CO T t ro 00 so CS m CO D Q s 2 u TJ H u Q s? 3 Q TJ '3. i <«H O IH g 1> * "8 tu 3 •s Xi s OH O H 1 "9 S, o ci o s to* u 3 1 I "3, I c l o « 6 © •B O % S a> to o 6 . u O H H3 (11 *3 H3 to « 13 ;a C! o, a to <u _ 3 •a i tkH '* O TJ 43 '3. 3 -3 <u 3 •a C H 3 CJ $ b i ) t w ca > O H •a e <J si ca y 120 3 © ^ -H SS cs as 60 CO U 2 u CJ Q + + CJ < Q + Q + OJ c3 a, OJ w O !> I <-> 60 |3 I & <3 NS NS NS NS NS NS NS NS NS NS 000'I o o o uo o T f rs 3 o 0 0 ro 0 0 uo T f ro 0 0 uo T f ro ro T f r~ 0 0 ro 000'I ~* d d d d d d d d uo 0 0 ro o ro O cs o oo o 0 0 T f rs r-o as vo «o © ro vq Os CS ro T f CS d d d d t -» ' 0 0 I—< d 4892.2 CS d rs 12.9 361.3 vq 4380.7 2885.6 1234.2 34.7 4200.2 co d vq cs 13.4 346.2 © rs' 3984.3 2182.7 1470.3 28.4 4546.3 cs d ro rs 14.2 348.5 ro rs 4113.4 2921.5 1268.5 31.7 4483.7 CS • d o rs 13.0 354.9 0 0 4156.6 2854.4 1252.2 31.7 4255.3 cs d ro cs' 13.0 338.3 Os 4021.0 2719.3 1352.9 32.5 4246.2 cs d rs 13.7 348.0 r-4003.0 2647.0 1340.2 31.0 4677.0 CO d vq uo" 13.9 355.4 vq 4298.0 2811.0 1288.9 33.2 4488.1 cs d 12.0 341.6 vq 4126.9 2975.8 1147.6 30.3 4458.2 ro d 0 0 0 0 15.5 347.0 OS 4111.4 2782.5 1311.4 33.0 4875.5 T f d ro rs' 14.4 357.2 r-4382.5 2777.5 1329.4 32.8 u 3 C J OJ It, 60 s cd N </> a o •a CU a C/5 CU T3 •«-• QJ xi o c o •a cd cj CS uo' u CS H 6 CJ •3 o o t CJ I •a CJ o o a, OJ 3 > •8 p O OJd § •H ^ CO © 1 — ' rs >, d l VO O = i T f -vo qf CO « i t c u 3 o a I '3 60 s cj" LL, 3" i-T U CM* CJ o VO § +" C uo o rs O VO M rs CL, ro c/i -Oj Os & T ? « > T f OJ uo t. OS © +1 S ° ° 0 0 OJ •3 J3 a 0 0 0 0 d - H a, 0J — 1 - H OJ 3 1 Xi 3 <-> I D - H ro ro ro oo" 3 3 ^ 3 CJ ~~ CJ 3 2 w cj Q + + 2 UJ o o 00 T t r-' T t r o c s o i n O N © rs N O 00 ©' rs T t c o e s o o o r s cs' — i 00 I S •a 6 0 U is w O 15 00 O N T t O N r s rs' m rs' 00 r s O d T t — o m d d o r o N O r s f - r -r o •-« t—' •a 6 0 a in © O N — 00 T t r s 00 T t d r s O N T t f - ' r s m >—i —I O N rs' r o r s c n r s N O — • —< O r s i n r s r o O N r s o o o r o O d T t 00 r s O N © O N ~ H . - H O TJ U M - H 6 0 UH Q r o O N o NO o O N o O N o o o N O © d O N N O O N O N 00 es' O r s o CS T t •u (L> C»-H l i 6 0 "a •a 6 0 CD o d o d N O rs ' O d o T t N O N O N O rs' OH 6 0 •a 6 0 0< & O s o •a CL) -a TJ < M o c o •a r s Tt" CD 2 o o "8 t—H o o O H 1) 3 ca > •a •8 6 0 V ) •a o •a 10 1 to o "BH o CD a •a TJ CD U 3 o -«-» <D 3 i ca CD a co CD •a o 6 121 122 123 i— "8 u 8. 8 8, o o o o O O 3 to a o u x> 0 0 f—I <u i—4 •a u ^ -5 <: Q D T + Q Q + 2 OJ <+-. o fl o •c 6 ' l<3 oo co oo oo 55 Z Z Z CO CO CO CO Z Z 55 55 CO CO CO CO Z Z Z Z C O z co Z CO z <o - f l •a oo t •a 6 0 u ••a c . o o u -29 O o CO £ 5 "8 •29 E Q T 3 0 0 "a •a 0 0 CO Z CO z CO Z e o D . 0 0 •a 6 0 Pi i* P H -—' O •a T 3 •w U =9 <<-4 o c o •a CN T * U I Pi o v\ I D C O a y ^ < Q T + <-> <; 2 0 Q + &3 Q + 2 D o •UI u a CL. o o o o CS T f N O N O T f V O O N O N O N m cs N O N O cs 0 0 o T f NO 0 0 NO o 0 0 0 0 © T f c m H <U| o Vi •a 3 cu cs C S o T f © co 0 0 © ' C O m O N © N O © O N C S © T f NO cs cs cs © © © © © © C O © © C O ro © r-ro co O N cs O N d cs © © N O cs r-ro rs ro oo od rs O N 0 0 rs O N 125 o •a •a *-* u ••a <« o a o •a Ui cs T f Ct. o b T3 <o 73 o CL, VI Ui xl U Vi > « N O N O C O C O <*H <ui o O a a 6 o •S •3 « c <u 1,1 w <u C J o © \/l CO « CD U X> u u s Ui O . CL o 3 3 73 73 > > xi X) o C J a CtJ W W CD 3 1 •8 X i p 73 cj •a V) •3 f3 to 127 o 5? 129 protein) of the HFM diet with dephytinized RPC, did not compromise the weight gain, specific growth rate, feed efficiency or protein conversion of rainbow trout, nor did it appear to alter feed intake (DFI), when FINNSTIM™ was included in the diet at 1.5 %. The findings also indicate that supplementation of the dephytinized RPC with essential amino acids to mimic the essential amino acid profile of the control diet, with or without additional cation-anion balancing, neither enhanced nor reduced the performance of the fish. The lack of appreciable enhancement of performance as a result of amino acid supplementation of the diets containing dephytinized RPC suggests that this protein source contains a comparable essential amino acid profile to that of high quality fish meal, and hence the essential amino acid profile in RPC appeared to be optimal for rainbow trout growth. However, the findings from this investigation do not entirely rule out the possibility that amino acid supplementation could provide some nutritional enhancement of rapeseed protein concentrate for rainbow trout. The findings of Murai et al., (1989), for example,suggest size specific reponsiveness of rainbow trout to arnino acid supplementation. Murai et al., (1989) specifically examined the effectiveness of amino acid supplementation of soybean flour diets on rainbow trout of different body size. Based on the results of three identical trials which utilized full siblings of different size classes, and common diet formulations, that were prepared at the same time under identical conditions, Murai et al. (1989) found that larger fish (« 8.9 g) showed a positive change in growth performance due to arnino acid supplementation, but smaller fish (< 5 g) showed no response to the amino acid supplementation. Although size specific responses to dietary treatments were not assessed under the conditions of the present investigation, they may be worthy of further study. Cation-anion balancing of dephytinized RPC diets (DRPC + C/A and DRPC + EAA + C/A) also did not appear to improve the nutritive value of the fish meal-free dephytinized-RPC-based diets for the trout. This apparent lack of nutritional benefit to dietary cation-anion balancing is consistent with the investigations of Wilson et al. (1985) which found that changes in the dietary electrolyte balance had little or no effect on growth or feed conversion of rainbow trout fed intact protein. However, based on the findings of the present investigation, wherein fish fed the DRPC + C/A diet had significantly lower weight gains relative to fish in the control (HFM) treatment group, there is some suggestion that cation-anion balancing may have reduced the feeding value of the dephytinized-RPC diet. The mechanisms and reasons 130 for the reduced weight gain of fish fed the DRPC + C/A are unclear as mineral composition of the diets met requirements and there were no significant differences between treatments in carcass mineral composition. Although the use of dephytinized-RPC in place of fish meal did not induce changes in weight gain, specific growth rate, feed intake, feed efficiency or the utilization of dietary protein for growth, there appeared to be some alteration in liver and thyroidal status. Liver weights were lower and 5'D activity levels were elevated relative to the controls, whereas plasma T 3 levels were maintained near control levels. These alterations in thyroidal status are not readily explained. The effects of rapeseed or canola protein products on thyroid hormone metabolism in trout is not well understood (Hilton, 1991). Additionally, the identification and mode of action of dietary constituents which might modify thyroid function, and/or the degree and type of responsiveness of the thyroid to nutritional state are not completely understood (Eales et al., 1992). What is known is that thyroid function, and particularly 5'D function in trout are stimulated by caloric intake, protein intake, and increased somatic growth (Eales and MacLatchy, 1989; Riley et al., 1993). However, 5'D activity is not always positively correlated with somatic growth, and may be elevated under conditions of compromised growth (Riley et al., 1993). Additionally, previous studies with salmonids have suggested that below maximum tolerable levels for thyrotoxic compounds (e.g. glucosinolates and/or their products of hydrolysis), the thyroid may be able to compensate, and counter suppression of function in order to maintain normal growth and normal levels of thyroid hormones (Yurkowski et al., 1978; Higgs et al., 1979; 1982; Hardy and Sullivan, 1983). Compensatory thyroidal response could possibly be evinced by elevated thyroid hormones and/or 5'D activity, indicating an increased conversion of T 4 to the highly active T 3 . However, increased hepatic generation of T 3 is not always associated with an increase in plasma T 3, as elevated uptake of T 3 in tissues may lead to increased plasma T 3 clearance (Eales et al., 1992). Given that the fish fed the dephytinized-RPC diets, which were virtually devoid of glucosinolates (refer to Table 4.13), had growth rates, weight gains and feed intakes which were not different from those of fish in the control group, it may be theorized that the elevated activity levels for 5'D were not associated with increased caloric or protein intake, nor increased growth. It may also be surmised that elevated 5'D activity levels did not occur in response to thyrotoxic effects induced by glucosinolates or their hydrolytic 131 products, as glucosinolates were not detected in three of the four dephytinized-RPC diets. It may be possible that the thyroidal response of fish was a compensatory response to some unknown factor within the diets, or to the residual amounts of phytic acid within the diets. However, the relationship between dietary phytic acid content and hepatic and/or thyroidal response in fish is ill-defined and largely unknown. Certainly further clarification of the mechanisms and explanations for the observed hepatic and thyroidal response of trout to dietary inclusion of high amounts of rapeseed protein concentrate is worthy of further investigation. Perhaps the monitoring of plasma T 4 levels, and histopathological examination of thyroidal and hepatic tissues would prove beneficial in this regard. In direct contrast to the performance of fish fed dephytinized-RPC, the performance of fish fed diets in which the rapeseed protein concentrate was not dephytinized was significantly inferior to the performance of fish in either of the fish meal groups (HFM and NFM) or any of the dephytinized treatment groups. Weight gain, specific growth rate, feed intake, feed efficiency and protein efficiency were significantly reduced. As with dephytinized dietary treatments, no appreciable nutritional benefits appeared to be conferred by essential amino acid supplementation, or the cation-anion balancing of the undephytinized-RPC diets. The plasma titres of T 3 and the liver size of fish fed undephytinized-RPC diets, in which glucosinolates were detected, were significantly reduced relative to that of the fish on the fish meal control (HFM), although 5'D activity levels were not significantly different. These findings are consistent with the depressed thyroidal function, depressed thyroid hormone titres, reduced conversion of T 4 to T 3 and the histopathological changes in the liver which are associated with the consequences of dietary intake of glucosinolates. Many of these adverse effects are usually manifested when dietary glucosinolate content exceeds 164 pg total glucosinolates/ g feed (Hilton, 1991). As tolerance limits are largely ascribed to the toxicological impact of alkenyl glucosinolates rather that indoly glucosinolates, the maximum tolerable dietary glucosinolate level for trout has also been described as 172 ug/g 3-butenyl isothiocyanate/g diet, which is an alkenyl glucosinolate (Hilton and Slinger, 1986). Yet in this investigation undephytinized-RPC diets and all the other diets were devoid of alkenyl glucosinolates. Only indoyl glucosinolate were detected. Indoyl glucosinolates are known to produce thiocyanates and nitriles (e.g. acetonitriles) upon 132 hydrolysis, depending on the pH of the medium, and the presence or absence of excess copper and iron ions may increase acetonitrile production (Bell, 1993). However, little is known about the particular toxicological effects and physiological response of trout to indoyl glucosinolates. Consequently, it is unclear to what extent dietary indoyl glucosinolates, alone or in combination with phytic acid or the other ANFs may have been responsible for the observed thyroidal response of fish fed the undephytinized-RPC diets. Considering the good performance of fish consuming the diets with dephytinized-RPC and the poor performance of fish fed the undephytinized-RPC diets, the findings suggest that phytic acid reduces potential weight gain, specific growth rate, feed efficiency and protein efficiency when rapeseed protein concentrate is used as a replacement for fish meal in a rainbow trout diet. Given that the weight gains of each dephytinized-RPC treament group were not statistically different (P > 0.05) from the weight gains of one or both fish meal treatment groups (HFM and NFM), whereas the weight gains of the undephytinized-RPC treatment groups represented 58 - 74 % of the weight gain achieved by the control groups (HFM and NFM), it may be said that the removal of phytic acid from rapeseed protein concentrate may facilitate at least a 26 % increase in weight gain when RPC is used to completely replace fish meal (59 % dietary CP) in a rainbow trout diet. Similarly, it may be stated that the removal of phytic acid from rapeseed protein concentrate, when RPC is used to completely replace fish meal (59 % dietary CP) may facilitate an 11 % increase in specific growth rate, an 18 % increase in feed efficiency, and a 17 % increase in protein efficiency ratio. Despite the fact that Norse-LT94™ is reputed to be one of the best high quality fish meals currently available, no significant differences (P > 0.05) were found in terms of weight gain (WG), specific growth rate (SGR), feed intake (DFI), feed efficiency (FE) or protein efficiency (PER) between fish fed the Norse-LT94™ fish meal diet (NFM), and fish fed the domestically available B.C. herring meal control diet (HFM). These findings contrast with previous reports that Norse-LT94™may support growth and feed conversion rates at levels at least 10% higher than feed conversion rates obtained with regularly processed fish meals (Pike et al., 1990). Instead these observations suggest that in terms of these performance parameters, under the conditions of this investigation, no particular advantage is afforded by the use of low 133 temperature Norse-LT94 in place of high quality, domestically produced B.C. whole herring meal, which is also produced at lower temperatures (< 90°C). However, whereas Norse-LT94™ is advertised to be relatively consistent in quality, no such assurance of quality is associated with the production of B.C. whole herring meal. Consequently, the quality of B.C. herring meal may vary between production batches and may not always produce fish performance equal to that achieved with Norse-LT94™. Therefore, the corollary to the finding that HFM and NFM may be equal in quality is that this may only be true when the quality of B.C. herring meal is as high as that of the meal used in this investigation. Significant differences (P < 0.05) between the Norse-LT94™ (NFM) and the herring meal (HFM) diet were found in terms of liver and thyroidal status. Fish fed the control (HFM) diet had significantly higher (P < 0.05) titres of T 3 (4.014 vs 2.641 ng T3/ml) and higher liver weights (0.59 vs 0.5 lg) than the fish fed the (NFM) diet. The 5'D activity levels were not significantly different (P > 0.05), however. As both diets were based on fish meal, and were of essentially the same formulation, supporting feed intakes and growth levels which differed only slightly, this difference in thyroidal status is difficult to explain. Both levels of plasma T 3 may be considered to be within a normal range, as plasma T 3 titres in fish fed the (NFM) diet were consistent with the range of T 3 titres that have been obtained for immature rainbow trout of < 65 g («2.0-3.0 ng Ts/ml) (Eales et al., 1992). By contrast the plasma T 3 offish fed the (HFM) diet approached the levels of « 6.0 ng T3/ml that have been observed in larger rainbow trout (i.e. > lOOg) used in other investigations (Hardy and Sullivan, 1983). 4.5.2 Influence of diet treatment on body composition i There was no significant modification of fish body composition due to the incorporation of undephytinized rapeseed protein concentrate as a complete replacement for fish meal, (Table 4.16) despite the fact that low mineral absorption is characteristically ascribed to feedstuffs of high phytate content. The levels of selected minerals in fish fed the URPC diets were comparable to the "normal" mineral composition values of rainbow trout (Shearer, 1984). Additionally, there was no apparent alteration in skin pigmentation which has sometimes been associated with the feeding of diets high in rapeseed/canola protein (Yurkowski et al., 1978). 134 4.5.3 Influence of diet treatment on general health and behaviour No alteration of fish health and behaviour in association with dietary treatment was apparent. 4.6 Conclusion The removal of phytic acid in rapeseed protein concentrate by pretreatment with microbial phytase was shown to facilitate the complete substitution of this protein source for high quality fish meal (59 % of the diet CP) in diets for trout. Indeed, this nutritional strategy did not compromise the growth performance, (as measured by weight gain, and specific growth rate, appetite and utilization of proteins for growth or the whole carcass proximate or mineral (Ca, Cr, Cu, Fe, Mg, Mn, P, K, Na, Zn) compositions, of the fish when the diets were concurrently supplemented with inorganic phosphorous and FINNSTIM™. No apparent nutritional advantage was conferred by amino acid supplementation of dephytinized or undephytinized -RPC diets to mimic the essential amino acid levels in the fish meal control diet (HFM), suggesting that the natural essential amino acid profile of rapeseed protein concentrate was already optimal. Similarly, no nutritional advantage was conferred by the cation-anion balancing of dephytinized or undephytinized-RPC diets. Although no direct measurement of the effect of FINNSTIM™ on feed intake was performed, there was evidence to suggest that dietary inclusion of FINNSTIM™ helped maintain feed intake in trout fed the fish meal-free diets. Additionally, the findings suggest that the dephytinization procedure employed by Aiko Biotechnology, Rajamaki, Finland in this study was improved relative to the procedure that was employed in the study by Teskeredzic et al., (1995), as significant differences in the nutritional quality of dephytininized and undephytinized RPC diets were clearly evident. Further, the findings suggest that domestically produced B.C. whole herring meal can be prepared to be of a similar nutritional value to Norse-LT94™ . This investigation confirms the feasibility of replacing the entire fish meal component of a typical rainbow trout diet with a plant protein source, namely, dephytinized rapeseed protein concentrate. It is believed that this is the first time that a plant protein product has successfully replaced all of the fish meal 135 component in a salmonid diet without compromising growth. The results also point to the intriguing possibility that, dephytinization of commercially available canola meal may enable enhancement of the dietary level of this protein source in trout and possibly other salmonids. The potential cost savings in the rearing of salmonids may be substantial. In 1992 the cost of protein concentrate production from canola, after considering shipping costs and FINNSTIM™ addition was estimated to be approximately 80 % of the cost of B.C. whole herring meal or approximately 66 % of the cost of premium fish meals (i.e. high quality low temperature meals), on a per kilogram protein basis (Dr. S. McCurdy, POS Pilot Plant Corp., Saskatoon, Saskatchewan). The additional cost of phytase application as Finase™ (Aiko Biotechnology) at approximately $203.42 (Cdn)/kg, (actitivty of 500,000 PU/g), at the commercial dose of 5,000 PU/g protein would add only approximately $0.002/kg protein to the protein concentrate. However, due to recent genetic research at Aiko Biotechnology whereby gene transfer of the phytase producing gene of Aspergillus niger into another organism has yielded a phytase source which has the capability of greatly increasing the rate of production, future costs of Finase™ are expected to decrease rather than increase (Timo Vaara, personal communication). Although no examination was made concerning the associated cost of commercial phytase application, this must be factored into the total cost of fish meal replacement with dephytinized rapeseed/canola protein concentrate. Nonetheless, potential cost benefits associated with the replacement of fish meal by dephytinized RPC are clearly evident. However, before these possibilities may be realized, identification of another more practical and economical method for dephytinization may be necessary as dephytinization procedures used in this investigation are not readily amenable to current feed manufacturing processes. Additionally, as the maximum dietary inclusion level of RPC was only 59 % of dietary protein, the upper limit to dietary inclusion of dephytinized rapeseed protein concentrate at the expense of animal protein in formulated salmonid diets remains to be determined. 4.7 Acknowledgements This study was supported by NRC/IRAP funding as adrninistered through Moore Clark Co. (Canada) Inc. and the Department of Fisheries and Oceans of Canada, West Vancouver Laboratory. 136 Thanks are extended to representatives of NRC/IRAP (Yvonne Jones, Dr. Warren Ngata and Dr. John Howard), representatives of Moore Clark (Greg Deacon and Jill Mills), and representatives of Aiko Ltd., Biotechnology, Finland (Timo Vaara, Marja Turuner and Dan Haglund). Thanks are also extended to Dr. R. Hardy and Dr. Shearer of Northwest Fisheries Centre, Seattle Wa., U.S.A., Dr. Eales of the Department of Zoology, University of Winnipeg, Manitoba, Dr. P. Reaney of Agriculture Canada Research Station, Saskatoon, Saskatchewan, as well as Dr. S. Satoh of the Department of Aquatic Biosciences, Tokyo University of Fisheries, Tokyo, Japan. 137 CHAPTER FIVE 5.0 EXPERIMENT II. Performance of rainbow trout (Oncorhynchus mykiss) fed diets in which undephytinized and dephytinized rapeseed protein concentrate is used to replace all animal protein and dephytinized and undephytinized commercial canola meal and low-fibre, high protein canola meal are used to partially replace fish meal in practical trout diets. 5.1 Abstract An 84-day study, which was partitioned into a growth portion (day 0-day 84) and a digestibility portion (day 74-day 83 inclusive), was conducted using juvenile rainbow trout (Oncorhynchus mykiss) held in ambient well water (9.8-10.9°C) to assess the effect of phytic acid removal on the acceptability of rapeseed protein concentrate (RPC) as a complete replacement for fish meal (95 % of the dietary protein), or commercial canola meal and low fibre canola meal as partial replacements for fish meal (20 % of the dietary protein) in a practical trout diet. Dephytinization was undertaken either by pretreatment of the preceding protein source with microbial phytase or supplemention of diets containting these sources with a low dose (l,000PU/g dry diet), or a high dose (4,000PU/g dry diet) of microbial phytase as Finase™ FP-500. Additionally, the effects of fibre reduction on the acceptability of canola meal in a practical trout diet were assessed. Juvenile rainbow trout were fed to satiation three times daily one of twelve isonitrogenous (45 % CP, DMB) and isoenergetic (18 MJ/kg, DMB) experimental diets that were formulated to contain equivalent levels of lipid (17%, DMB), and FINNSTIM™ (1.5%, DMB); a palatability enhancer. Diets also contained 0.5% Cr203 (chromium sesquioxide) as a marker for digestibility measurements. All experimental diets were variations of the control diet. Test proteins were Norse-LT94™ fish meal, undephytinized commercial (high fibre) canola meal, dephytinized commercial canola meal, undephytinized, fibre reduced canola meal, undephytinized rapeseed protein concentrate, and dephytinized rapeseed protein concentrate. Norse-LT94™ fish meal supplied 95 % of dietary protein in the control diet, and 75 % of the dietary protein in the canola meal diets, for which canola meal furnished 20% of dietary 138 protein. Al l RPC diets were devoid of animal protein and thus contained RPC at a level of 95 % of dietary protein. Extruded wheat supplied 5 % dietary protein in all diets. The results indicated that dephytinization of commercial canola meal (20% of the dietary protein) by pre-treatment with microbial phytase, or by dietary application of microbial phytase at concentrations of 1,000 PU/g DMB and 4,000 PU/g DMB enhanced the nutritional value of the meal for rainbow trout. The greatest nutritional enhancement was given by the direct application of a low phytase dose (1,000 PU/g dry diet). The findings also indicated that when canola meal provided 20% of the dietary protein, fibre reduction with or without microbial phytase addition of 4,000PU/ g dry diet was ineffectual in improving the nutritional value of the canola meal. In addition, under the conditions of this investigation, direct microbial phytase application (1,000 or 4,000 PU/g dry diet) did not enhance the nutritional value of RPC. By contrast, commericial dephytinization by pre-treatment of RPC with microbial phytase before diet incorporation did improve the nutritional value of RPC, although fish performance and diet energy digestibility were poorer than noted for trout fed the control NFM diet. The feed intake data suggested that diet palatabilty may have been reduced by microbial phytase addition directly to the diets and/or the high dietary inclusion of rapeseed protein concentrate. 139 5.2 Introduction The protein found in rapeseed/canola meals (< 50% crude protein (CP); typically 35-44% CP), upgraded meals (flours) (> 50% CP) and concentrates (>60 % CP; typically 65-70%CP ) is of good quality and has a well balanced amino acid profile (Gillberg and Tornell, 1976; Jones, 1979; Higgs et al., 1988; 1989; 1990; 1995a). However, antinutritional factors (ANFs) intrinsic to the feedstuffs limit feeding value. Bitter saponins (Cheeke and Shull, 1985) and phenolic compounds (e.g. tannins, sinapic acid) reduce palatability and in the case of phenolics, also reduce protein and vitamin utilization by forming insoluble complexes with digestive enzymes (e.g. trypsin) and other proteins (Ramachandra et al., 1977; Cheeke and Shull, 1985; Sosulski, 1979). Crude fibre, which is found at concentrations of ~ 12-15 % DMB in meals (Youngs, 1991; Bell, 1993), * 7.0 % DMB in upgraded meals (Youngs, 1991) and ~ 4.9-9.0 % DMB in concentrates (Jones, personal letters; Jones, 1979; Youngs, 1991) is also antinutritive in its effects, wherein it is indigestible in the absence of bacterial action, and may decrease the bioavailability of minerals by binding minerals intrinsic and exogenous to the meal (Nwokolo and Bragg, 1977). Fibre is also known to inhibit digestive enzymes (Krogdahl, 1989), decrease protein digestibility and have depressive effects on metabolizable and digestible energy (Bell, 1990). However, it is the glucosinolate content that has been identified as the primary constraint to the utilization of rapeseed/canola meal and upgraded meal by salmonids (Higgs et al., 1983; Fagerlund et al., 1987; Leatherland et al., 1987), and, it is the phytic acid content that is thought to be the primary constraint to the full utilization of rapeseed/canola protein concentrates (Youngs, 1991). Well-recognized as the compounds responsible for the sharp, biting tastes of condiments such as mustard and horseradish, glucosinolates are sulphur-rich glycosides which lower the palatability and thus feeding value of rapeseed and canola. Upon hydrolysis, glucosinolates yield physiotoxic thiocyanates, isothiocyanates, nitriles, and oxazolidine-2-thiones such as goitrin (VOZT; VOT), progoitrin and epigoitrin (goitrin precursors) which cause goitrogenic and non-goitrogenic impairment of thyroid function, depressed growth, and hepatic lesions and necrosis (van Etten and Tookey, 1983; Cheeke and Shull, 1985; Niewiadomski, 1990; NRC, 1993). Reduction of glucosinolate content improves feeding value. Consequently meals derived from canola ( < 30 umol of aliphatic (alkenyl) glucosinolates per gram of air-140 dried meal (Miller, 1988) are nutritionally superior to rapeseed meals which have a higher glucosinolate content (50-100 pmol/g of glucosinolates in the air-dried, defatted meal (Salunkhe et al., 1992; Bell, 1993)). Through further processing and the production of upgraded meals and concentrates, both rapeseed and canola feedstuffs may be nutritionally improved, and consequently better tolerated in salmonid diets. Due to their improved feeding value, upgraded canola meals may comprise as much as 40% of dietary protein in rainbow trout diets without compromising growth (McCurdy and March, 1992). In contrast, rainbow trout have a demonstrated tolerance for only 7.4 - 21.4 % of dietary protein from rapeseed meal (Higgs et al., 1995) and 18.0- 22 % of dietary protein from canola meal (Hardy and Sullivan, 1983; Higgs et al., 1995). Provided that diets containing canola meal are supplemented with triiodo-L-thyronine (T3), the acceptable level of canola meal may be extended to 24-27% of the dietary protein for rainbow trout, chinook (Oncorhynchus tshawytshcha) and coho (O. kisutch) salmon (Higgs et al., 1983; Fagerlund et al., 1987; Higgs et al., 1995). Rapeseed and canola protein concentrates are also better tolerated in salmonid diets than are typical meals but their acceptability and tolerance are primarily restriced by phytic acid content. Phytic acid, found at concentrations of 2.0-5.0% in defatted meals (Nwokolo and Bragg, 1977; Alii and Houde, 1987; Tzeng, 1987; Tseng et al., 1988a, b, 1990; Thompson, 1990), «5.5% in flours (Erdman, 1979; McCurdy and March, 1992) and 3.6-7.5% in concentrates (Shah et al., 1976; Jones, 1979; Thompson, 1990; Youngs, 1991), is strongly negatively charged at physiological pH, and readily binds multivalent cations, proteins and digestive enzymes (Cheryan, 1980; Irving, 1980; Spinelli et al., 1983; Graf, 1986; Reddy et al., 1989; NRC, 1993) forming relatively insoluble complexes. Additionally, phytic acid may induce mineral deficiencies leading to tissue anomalies in different organs (Richardson et al., 1985). Relatively little has been done to investigate the degree to which phytic acid restricts the utilization ofRPC/CPCor standard meals and upgraded meals of rapeseed and canola in fish. Still fewer investigations have examined the potential benefits that phytic acid reduction and removal may have on the feeding value of rapeseed and canola protein feedstuffs for salmonids. No phytase activity has been demonstrated in gastrointestinal tract of salmonids, and negligible phytase activity has been determined in rapeseed and canola seed, and most of this is thought to be 141 destroyed during the thermal treatment of seeds to deactivate thioglucoside glucohydrolase deactivation (Stone et al., 1984; Nair et al., 1991; Sandberg et a l , 1993). Consequently, there is negligible hydrolysis of phytic acid and/or the dissolution of phytate complexes. Thus the absorption and utilization of bound minerals and proteins is prevented. Commercial phytase preparations (most often derived from fungal Aspergillus sp.) are readily available, and tend to exhibit particular substrate specificity, and pH optima which are compatible with gastric conditions (Campbell and Bedford, 1992). Pretreatment of feedstuffs (e.g. soybean meal) or completed diets (e.g. soybean, canola or cottonseed meal based), with microbial phytase has been shown to significantly improve growth (Spinelli et al., 1979) and feed conversion (Spinelli et al., 1979; Cain and Garling, 1995), as well as improve phosphorous absorption (Riche, 1993) of salmonids. This experiment was designed to extend the findings of chapter 4, by examining the nutritional effects of phytic acid reduction and/or removal on the utilization of various rapeseed/canola feedstuffs incorporated into the diets of rainbow trout (O. mykiss), as partial or total replacements for fish meal and other animal protein. It was also designed to evaluate the effectiveness of a more practical means of dephytinization than used in chapter four. Specific objectives were the following: (1) to determine if dephytinization of commercial canola meal by commercial pretreatment with phytase significantly enhances its feeding value for rainbow trout when used as a partial replacement for fish meal in the diet, (2) to determine if dephytinization of rapeseed protein concentrate by commercial pretreatment with phytase significantly enhances the feeding value of concentrates to such an extent that enables the successful replacement of fish meal and other animal proteins (95% of the dietary protein) in a diet for rainbow"trout, (3) to determine if the addition of phytase to commercial canola meal and upgraded canola meals, significantly improves the feeding value of such meals for rainbow trout when used as a partial replacement for fish meal (4) to determine if the addition of phytase to diets containing, rapeseed protein concentrate enables this protein source to entirely replace all of the fish and other animal proteins (95% of the dietary protein) in a diet for rainbow trout and, (5) to determine if a fibre-reduced upgraded canola meal has a significantly enhanced feeding value for rainbow trout relative to standard canola meal. 142 5.3 Materials and methods 5.3.1 Fish and experimental conditions In August, 1993 rainbow trout (Oncorhynchus mykiss) were obtained from Spring Valley Trout Farms, Langley, B.C.. The fish were selected for uniform weight (mean weight 1.5g ± 0.6g) and were randomly distributed into twenty-four 150-litre fiberglass tanks at an initial stocking density of 70 fish per tank. Tanks were arranged in a single row, but were divided into two blocks such that there were 12 tanks per block. The tanks were part of a digestibility system that has been described by Hajen (1990). No filters were utilized, however, since the water supply was well water, and the drainage system was modified initially to prevent loss of the small fish. Each tank was supplied with running (4-6 1/min), aerated (dissolved oxygen (dCh) 10.1-10.5 ppm), well water at ambient temperature (9.8-10.9 "C).1 A natural photoperiod was maintained with daylight fluorescent lights, (Vitalite, Durotest 40W), controlled by an astronomical timing cell. The experiment was conducted over the course of 84 days and was partitioned into a growth study (day 0 - day 84) and a digestibility study (day 74 - day 83 inclusive). The same fish were used throughout. All tanks were siphoned daily until faecal collection started. Digestibility measurements were made according to the method described by Hajen (1990). 5.3.2 Diet formulation, composition and preparation Twelve experimental dry diets (Table 5.1) were formulated to contain similar levels of digestible protein, lipid and digestible energy content (45%, 17% and 18 MJ/kg DMB respectively). The test protein feedstuffs utilized in the diets, as presented in Table 5.2, were as follows: Norse-LT 94 fish meal (76.2 % CP), undephytinized high fibre canola meal (38.3 % CP), commercially dephytinized high fibre canola meal (38.7 % CP), undephytinized low fibre canola meal (45.9 % CP), undephytinized rapeseed protein concentrate (65.4 % CP), and commercially dephytinized rapeseed protein concentrate (63.1 % CP). All other protein stemmed from extruded wheat (16.3 % CP). 1 On two occasions well-water line shut downs forced the use of dechlorinated city water for 2-6 hour periods. 143 Diet compositions, as presented in Table 5.3, were developed after all protein sources were assessed for proximate (Table 5.2), amino acid and mineral composition (data not presented). Based on the amino acid assessments, methionine was necessarily supplemented to all diets, except those containing URPC, in order to meet dietary requirments and equalize dietary supply. All other essential amino acids were supplied in sufficient amounts to meet or exceed dietary requirements (Ogino, 1980; NRC, 1981; Cho, 1990). Similarly, based on the assessed mineral levels in the feedstuffs, twelve separate mineral supplements (Table 5.4) were prepared and supplemented to the experimental diets to ensure that the dietary mineral supply met the known mineral requirements of trout. All diets were formulated to contain > 0.8% inorganic phosphorous (P) and > 0.06% magnesium (Mg) from animal and inorganic sources (Lall, 1989). Also, all diets were formulated to contain 0.5% chromium sesquioxide (CftOa), an indigestible marker used in digestibility studies. Additionally, because low acceptance of diets containing high proportions of vegetable feedstuffs has been reported for fish (Moyano et al., 1992a, 1992b; Jackson et al., 1982), FINNSTIM™, a betaine-based commercial preparation that has been known to act as a feeding stimulant (Virtanen et al., 1989; Virtanen, 1992; Teskeredzic et al., 1995) was added at a level of 1.5% DMB to all diets, to help ensure similar acceptance of diets. The completed diets were analyzed for proximate composition and gross energy conteent as well as mineral and amino acid composition to ensure that the determined nutrient and energy levels met expected values. (Tables 5.5, 5.5, 5.6, and 5.7, respectively). Additionally, the completed diets were analyzed for glucosinolate and inositol phosphate content (Table 5.8). Descriptions and abbreviations of the dietary treatments are presented in Table 5.1. High quality Norse-LT 94™ fish meal comprised 95% of dietary protein in the control/ basal diet (NFM). The second diet (NFMHP) closely mimicked the control diet but was modified by the addition of a high dose (4000 PU/g diet), of a powdered Finase™ FP-500 phytase preparation, (activity * 410,000 PU/g) (One phytase unit, PU, is the activity that liberates 1 nmole of inorganic phosphate from sodium phytate at pH 5 and 37 °C; Lei et al., 1993). At the expense of the fish meal component, (reduced to « 75% of dietary As city and well-water mixed, temperatures reached 13.5 and 14.2°C and d02 reached 9.4 and 9.6 ppm. 144 "8 § § § -SP B o o o i e © © © "8 -1 E 6 75 C3 "3 3 73 B E CO Ui © © ON H •a 73, c '53 e O H ••a 73 ON ON H I 73< I c e ••a ON. ON * 1 0 s -•3 3 u a o I 8-"8 "3. O H 0s-O 'C • t f " t f N •a N •a i & -a § "8 & .5 .5 e O H I u =8 S O H =a N ? 0 s © CN <D ••a m ON ••a ON-m ON "8. a v x 2 a 53 a O H O H 5 S <U CD ••a -a 73 73 S O v O 6s- 0s-VI IO ON ON 145 0 0 3 •a 1 "J CD i I tu a 8 3 o to c '5 o t-i cx, o o •a tU c o t_> o X ! o 3 c o *3 'lo o CX a o •a o U l c u IN in' to 1 T f '&) M g 5a X i U i tU a 0 s X ! 3 T3 TJ tu TJ s U tu <u ti Q u 1 C 'u o U l CU 0 0 vi vq cn 0 0 © vo r-1 '5b o Z i n 0 0 © o —> d TJ-' vo O d i n VO co • — i Ov O N E TT OV H •J o Z TJ d TJ d Tj d ro •a a TJ d vo *-> cn ON T f CS VO vo vo' CS cn T f cn cn vo o oo q vo" cn 0 0 cn o OS cs cn vq v i cn cn vo' vd oo cn cs 0 0 vq m O N r-' T f vd <n T f 0 0 — i vi I—' O N O N 0 0 cs 0 0 cn vd ON 0 0 •8. tu 7 3 d TJ •a X ! Cu •s I a o NO & tu 3 T J tU 5 o. co o X ! p. 60 3 •a 3 o o 5 o XI CX, § NO & I Table 5.3: C o m p o s i t i o n a n d e n e r g y c o n t e n t o f test d ie ts f e d t o j u v e n i l e r a i n b o w t r o u t f o r 8 4 days . 146 Ingredient Diet (e/kg dry w e i g h t ) 1 3 N F M N F M H P U H F C M U H F C M L P U H F C M H P D H F C N N o r w e g i a n N o r s e - L T 9 4 R 2 5 6 0 . 9 5 6 0 . 9 4 4 2 . 8 4 4 2 . 8 4 4 2 . 8 4 4 2 . 8 R a p e s e e d p r o t e i n c o n c e n t r a t e U n d e p h y t i n i z e d D e p h y t i n i z e d : ; - - - -H i g h fibre c a n o l a m e a l U n d e p h y t i n i z e d D e p h y t i n i z e d - - 2 3 4 . 9 2 3 4 . 9 2 3 4 . 9 2 3 2 . 7 L o w f i b r e c a n o l a m e a l U n d e p h y t i n i z e d - - - - - -E x t r u d e d w h e a t 138.5 138.5 137 .1 137 .1 137.5 138 .5 S a r d i n e o i l ( s t a b i l i z e d ) 4 108.0 108.0 102 .6 102 .6 102 .6 100 .2 M i n e r a l s u p p l e m e n t 5 2 0 . 0 2 0 . 0 2 0 . 0 2 0 . 0 2 0 . 0 2 0 . 0 V i t a m i n s u p p l e m e n t 6 2 0 . 0 2 0 . 0 2 0 . 0 2 0 . 0 2 0 . 0 2 0 . 0 D L - M e t h i o n i n e 2.6 2.6 1.4 1.4 1.4 3.0 E t h o x y q u i n - - - - - -P e r m a p e l l 7 10.0 10.0 10.0 10.0 4 .5 10 .0 A l p h a c e l l 80 .4 7 0 . 8 4 .2 1.8 .- 5.8 P r e g e l a t i n i z e d w h e a t s t a r c h 3 2 . 7 32 .5 - - - -C a H P 0 4 - - - - - -N a C l - - - - - -K H 2 P 0 4 - - - - - -F i n a s e ™ F P - 5 0 0 p h y t a s e - 9.8 - 2.4 9.8 -C o n s t a n t c o m p o n e n t s 8 2 7 . 0 27 .0 2 7 . 0 2 7 . 0 2 7 . 0 2 7 . 0 D i g e s t i b l e e n e r g y ( M J / k g ) 9 17.9 17.9 17.9 17.9 17.9 17.9 147 T a b l e 5 . 3 ( C O O T T N U E D ) : C o m p o s i t i o n a n d energy con ten t o f test d ie ts f e d t o j u v e n i l e r a i n b o w t r o u t f o r 84 days. I n g r e d i e n t D i e t f g / k e d r y w e i g h t ) 1 3 U L F C M U L F C M H P U R P C U R P C L P U R P C H P D R P C N o r w e g i a n N o r s e - L T 9 4 R 2 4 4 2 . 8 4 4 2 . 8 - - - -Rapeseed p r o t e i n c o n c e n t r a t e U n d e p h y t i n i z e d D e p h y t i n i z e d - - 6 5 3 . 9 6 5 3 . 9 6 5 3 . 9 6 7 7 . 4 H i g h f i b r e c a n o l a m e a l U n d e p h y t i n i z e d D e p h y t i n i z e d - -_ _ - -L o w f i b r e c a n o l a m e a l U n d e p h y t i n i z e d 196 .0 196 .0 - - - -E x t r u d e d w h e a t 138.5 138.5 138.5 138.5 138 .5 138 .5 S a r d i n e o i l ( s t a b i l i z e d ) 4 108.2 107.8 9 6 . 1 9 6 . 1 9 6 . 1 9 5 . 7 M i n e r a l s u p p l e m e n t 5 2 0 . 0 2 0 . 0 20 .0 2 0 . 0 2 0 . 0 2 0 . 0 V i t a m i n s u p p l e m e n t 6 2 0 . 0 2 0 . 0 20 .0 2 0 . 0 2 0 . 0 2 0 . 0 D L - M e t h i o n i n e 1.2 1.2 - - - 0.8 E t h o x y q u i n - - 0.10 0 . 1 0 0 . 1 0 0.1C P e r m a p e l l 7 10.0 10.0 8.5 6 . 1 3 .9 0 .2 A l p h a c e l l 36 .4 27 .0 - - - -P r e g e l a t i n i z e d s t a r c h - - - - - -C a H P 0 4 - - 11.7 11.7 16.5 8 .8 N a C l - - 12.0 12.0 14 .2 11.6 K H 2 P 0 4 - - 12.27 12 .27 - -F i n a s e ™ F P - 5 0 0 p h y t a s e - 9.8 - 2.4 9 .8 -C o n s t a n t c o m p o n e n t s 8 2 7 . 0 2 7 . 0 27 .0 2 7 . 0 2 7 . 0 2 7 . 0 D i g e s t i b l e e n e r g y ( M J / k g ) 9 17.9 17.9 17.9 17.9 17 .9 17 .9 148 T a b l e 5.3: ( C O ^ ^ T N U E D ) : Composi t ion and energy content o f test diets fed to juveni le rainbow trout for 84 days. 1 Refer to Table 5.1 for explanation of diet designations 2 Norwegian N o r s e - L T 9 4 R fish meal stabilized wi th 250 mg ethoxyquin/kg. 3 N F M = Norwegian N o r s e - L T 9 4 R f ish meal, control; N F M H P = Norwegian N o r s e - L T 9 4 R f ish meal supplemented wi th a h igh dose o f F i n a s e ™ FP-500 phytase (4,000 P U / g diet), phytase control; U H F C M = undephytinized h igh fibre canola meal; U H F C M L P = undephytinized h igh fibre canola meal supplemented wi th a l ow dose of F i n a s e ™ FP-500 phytase (1,000 P U / g diet); U H F C M H P = undephytinized h i g h fibre canola meal supplemented wi th a h igh dose of F i n a s e ™ FP-500 phytase (4,000 P U / g diet); D H F C M = dephytinized h igh fibre canola meal; U L F C M = undephytinized experimental l ow fibre canola meal ; U L F C M H P = undephytinized low fibre canola meal supplemented wi th a h igh dose o f F i n a s e ™ FP-500 phytase (4,000 P U / g diet); U R P C = undephytinized rapeseed protein concentrate; U R P C L P = undephytinized rapeseed protein concentrate supplemented wi th a low dose of F i n a s e ™ FP-500 phytase (1,000 P U / g diet); U R P C H P = undephytinized rapeseed protein concentrate supplemented wi th a h igh dose o f F i n a s e ™ FP-500 phytase (4,000 P U / g diet); D R P C = dephytinized rapeseed protein concentrate. 4 Stabil ized wi th 0.05% ethoxyquin. • 5 The mineral supplements were distinct for each diet; see Table 5.4 for description o f ind iv idua l components. 6 The v i t amin supplement was prepared i n one batch and supplied the fo l lowing levels o f nutrients/kg dry diet: D - c a l c i u m pantothenate, 192.53 mg; pyridoxine HC1, 49.26 mg; riboflavin, 75 mg; fol ic acid , 18.75 m g ; thiamine mononitrate, 55.99 mg; biot in, 75 mg; cyanocobalariun (Bi 2 ) , 90 mg; menadione sodium bisulfite, 109.09 mg; DL-tocopheryl acetate (E), 300 I U ; cholecalciferol ( D 3 ) , 2400 I U ; v i t amin A acetate, 5,000 I U ; inosi tol , 400 mg; niacin, 304.57 mg; butylated hydroxytoluene, 22 mg; wheat starch, 17,993.01 mg. 7 Permapell is a l i gn in suphonate binder. 8 C o m m o n dietary components (g/kg dry diet) were as follows: ascrobic acid, 2; choline chloride (60%), 5; F i n n s t i m ™ , 15; chromic oxide, 5. 9 Calculated on the basis of the sum of individual digestible energy values (Cho and Kaush ik , 1991) T a b l e 5.4 C o m p o s i t i o n o f m i n e r a l s u p p l e m e n t s e m p l o y e d i n e x p e r i m e n t a l d ie t s f e d t o j u v e n i l e r a i n b o w t r o u t f o r 84 days . I n g r e d i e n t D i e t ( m g / k g d r y d i e t ) ' N F M N F M H P U H F C M U H F C M L P U H F C M H P D H F C M C a a s ( C a C 0 3 ) - - - - - -( C a H P 0 4 ) - - - - - -P as ( C a H P 0 4 ) - - - - - -( N a H 2 P 0 4 H 2 0 ) - - 2 2 8 . 7 2 2 8 . 7 - -M g as ( M g O ) - - - - - -C u a s ( C u S 0 4 5 H 2 0 ) 12.7 12.7 12 .1 12 .1 1 2 . 1 .11 .5 F e a s ( F e S 0 4 7 H 2 0 ) 2 0 7 . 3 2 0 7 . 3 169 .8 169 .8 169 .9 1 4 3 . 1 Z n a s ( Z n S 0 4 7 H 2 0 ) 150.7 150.7 142.5 142.5 142 .5 135.3 M n as ( M n S 0 4 H 2 0 ) 9 3 . 6 9 3 . 6 8 2 . 9 8 2 . 9 8 2 . 9 8 2 . 0 N a a s ( N a H 2 P 0 4 H 2 0 ) . - 169.8 169 .8 - -( N a F ) 5.5 5.5 5.5 5.5 5.5 5.5 ( N a C l ) 2 5 . 0 2 5 . 0 5 5 7 . 8 5 5 7 . 8 7 2 7 . 6 5 4 5 . 9 ( N a 2 S e 0 3 ) 0 .1 0 .1 0 .1 0 .1 0 . 1 0 . 1 K as ( K I 0 3 ) 1.5 1.5 1.5 1.5 1.5 1.5 ( K I ) 1.8 1.8 1.8 1.8 1.8 1.8 ( K 2 C 0 3 ) 51 .7 51 .7 - - - -( K 2 S 0 4 ) - - - - - -I as ( K I 0 3 ) 5 .0 5.0 5.0 5.0 5.0 5.0 ( K I ) 5:0 5.0 5.0 5.0 5.0 5.0 F a s ( N a F ) 4.5 4.5 4.5 4 .5 4^5 4 .5 C o as ( C o C l 2 6 H 2 0 ) 3.0 3.0 3.0 3.0 3 .0 3.0 S e a s ( N a 2 S e 0 3 ) 0 .2 0.2 0.2 0 .2 0 .2 . 0 .2 A l p h a c e l l 17774 .3 17774.3 15749 .6 1 5 7 4 9 . 6 1 6 3 3 7 . 3 1 6 9 6 8 . 7 1 R e f e r to T a b l e 5 .1 f o r e x p l a n a t i o n o f d ie t d e s i g n a t i o n s 150 T a b l e 5 . 4 ( C O N T T N U E D ) : C o m p o s i t i o n o f m i n e r a l s u p p l e m e n t s e m p l o y e d i n e x p e r i m e n t a l d ie t s f e d t o j u v e n i l e r a i n b o w t r o u t f o r 84days . I n g r e d i e n t D i e t ( m g / k g d r y d i e t ) 1 U L F C M U L F C M H P U R P C U R P C L P U R P C H P D R P C Ca as ( C a C O j ) - - - - 4 0 . 5 1 2 9 6 . 9 ( C a H P 0 4 ) - - 1 4 5 9 . 1 1 4 5 9 . 1 - 2 5 8 6 . 5 P as ( C a H P 0 4 ) - - 1128.3 1128.3 - 2 0 0 0 . 0 ( N a H 2 P 0 4 H 2 0 ) 2 2 8 . 9 - 1177 .2 2 7 9 1 . 8 - -M g as ( M g O ) - - 8 0 0 . 0 8 0 0 . 0 8 0 0 . 0 8 0 0 . 0 C u a s ( C u S 0 4 5 H 2 0 ) 10.5 10.5 13.0 13.0 13.0 11.8 F e a s ( F e S 0 4 7 H 2 0 ) 162.5 162.5 177 .8 177 .8 177 .8 146 .8 Z n a s ( Z n S 0 4 7 H 2 0 ) 133.8 133.8 136 .2 136.2 136 .2 11 .1 M n as ( M n S 0 4 H 2 0 ) 7 7 . 2 7 7 . 2 3 3 . 0 3 3 . 0 3 3 . 0 2 3 . 1 N a a s ( N a H 2 P 0 4 H 2 0 ) 169.8 _ 8 7 3 . 6 8 7 3 . 6 - -( N a F ) 5.5 5.5 5.5 5.5 5.5 5.5 ( N a C l ) 8 5 7 . 1 1026 .9 - ' - - -( N a 2 S e 0 3 ) 0 .1 0 .1 0 .1 0 .1 0 . 1 0 .1 K as ( K I 0 3 ) 1 5 1.5 1 5 1.5 1.5 1.5 ( K I ) 1.8 1.8 1.8 1.8 1.8 1.8 ( K 2 C 0 3 ) - - 3 0 2 9 . 9 3 0 2 9 . 9 3 0 2 9 . 9 3 0 5 5 . 5 ( K 2 S 0 4 ) - - - - 3 5 2 4 . 1 3 0 5 5 . 5 I as (KIO3) 5.0 5.0 5.0 5:0 5.0 5.0 ( K I ) 5.0 5.0 5.0 5.0 5.0 5.0 F as ( N a F ) 4.5 4.5 4.5 4 .5 4 .5 4.5 C o as ( C o C l 2 6 H 2 0 ) 3.0 3.0 3.0 3.0 3 .0 3.0 S e a s ( N a 2 S e 0 3 ) 0 .2 0.2 0.2 0 .2 0.2 0 .2 A l p h a c e l l 15087 .4 1 5 6 7 5 . 1 1371 .4 1371 .4 3 6 1 2 . 8 1769 .5 R e f e r t o T a b l e 5 .1 f o r e x p l a n a t i o n o f d ie t d e s i g n a t i o n s 151 protein), undephytinized commercial (high fibre) canola meal comprised 20% of dietary protein in the third, fourth and fifth diets. The third diet (UHFCM) contained no Finase™, the fourth diet (UFfFCMLP mimicked the third diet (UHFCMLP) but was supplemented with a low dose (1000 PU/g diet) of Finase™ and the fifth diet (UHFCMFTP) again rnimicked the third diet but was supplemented with a high dose (4000 PU/g diet) of Finase™. The sixth diet (DHFCM) also closely mimicked the third diet with the exception that commercially dephytinized high fibre canola meal comprised 20% of the dietary protein. Undephytinized experimental low fibre canola meal comprised 20% of the dietary protein in the seventh (ULFCM) and eighth (ULFCMHP) diets which were very similar in composition except that the eighth diet was supplemented with a high dose (4000 PU/g diet) of Finase™ (this experimental protein source was not received in time to be sent to Finland for commercial dephytinization). Undephytinized rapeseed protein concentrate comprised 95 % of the dietary protein for diets nine through eleven, (URPC, URPCLP, URPCHP, respectively). The ninth diet (URPC) contained no Finase™ whereas the tenth (URPCLP) and eleventh diet (URPCHP) were supplemented with a low and high dose of Finase™ phytase, respectively. Finally, the twelfth diet (DRPC) was formulated to contain commercially dephytinized rapeseed protein concentrate to provide 95% of the protein. Powdered Finase™ phytase was added to the NFMHP, UHFCMLP, UHFCMHP, ULFCMHP, URPCLP and URPCHP diets at the time that the external coating of fish oil was applied to the finished pellets. Finase™ phytase was sifted (size 40 mesh sieve), onto pre-weighed portions of dried pellets which had been placed in a metal pan. Oil was then applied to pellets by means of a plant spray mister and pellets were gently stirred. More Finase™ phytase was sifted over the pellets, more oil was applied and pellets were again mixed. This procedure was repeated until the prescribed amounts of Finase™ phytase and oil had been applied. Pellets were then placed in plastic bags and briskly shaken for ten minutes to help ensure uniform coating. 5.3.3 Commercial dephytinization: phytase pretreatment Rapeseed protein concentrate and commercial high fibre canola meal were dephytinized by Aiko Ltd. Biotechnology, Rajamaki, Finland, by means of pretreatment with a commercial phytase preparation 152 (Finase™ 40) before dietary incorporation, according to the procedures described in chapter three, section 3.8. Samples of the initial undephytinized and dephytinized protein sources were then analyzed for inositol phosphates using Aiko method C-120 Rev. A, (Table 5.2). 5.3.4 Feeding protocol Using a spoon for the distribution of pellets, fish were hand fed to satiation three times daily between 0830 hr and 1600 hr. During a week-long acclimation period, fish were fed the basal diet (NFM). Throughout the experimental period, following the allocation of dietary treatments and an initial sampling of fish for subsequent determination of initial whole body proximate composition (refer to section 5.3.6), fish were fed their prescribed diets. Daily records of behaviour, mortality, feed intake, water temperature and dissolved oxygen content were maintained. 5.3.5 Faeces collection Faeces collection commenced on day 74 and continued through day 83. The procedures of Hajen (1990) were followed. At the end of the collection period the frozen faeces from the daily collections were pooled for each tank and freeze-dried. They were then ground with a mortar and pestle and stored in sealed containers at 4-6 0 C pending analyses. 5.3.6 Fish weighing and sampling Every three weeks, all fish were removed from their respective tanks for weighing and sampling. Prior to each sampling, feed was withheld for at least 17 hrs.. All fish were anaesthetized with metomidate-HCL (0.25 mg/1) followed by tricaine methane sulfonate (60 mg/1) and individually weighed (nearest 0.0lg). Five fish were removed randomly from each tank at day 0 and killed. These fish, as well as ten fish randomly selected from each tank at the termination of the experiment (day 84), were killed, vacuum packed, frozen and stored at -20 °C for later determination of whole body proximate and mineral compositions. In addition, livers and blood were extracted from eighteen fish randomly selected from each 153 replicate tank assigned diets NFM, UHFCM, ULFCM , URPC and DRPC. Livers were removed, weighed, wrapped in labelled aluminum foil and immediately frozen in liquid nitrogen. Fish were bled from the caudal vessels into lcc heparinized syringes. Plasma was separated by centrifugation and placed on dry ice. Once all samples were collected, livers and plasma were vacuum packed and stored at -70°C, until plasma was assayed for T 3 and T 4 , and livers were assayed for 5' monodeiodinase activity in order to assess thyroidal status (Refer to section 3.5 and 3.6). 5.3.7 Chemical analyses Detailed descriptions of all analytical procedures used in this investigation are presented in sections 3.4- 3.8 of chapter three. Proximate analyses, conducted according to the procedures outlined in Higgs et al. (1979), were performed in triplicate on protein feedstuffs, and in duplicate on diets with the exception that protein analyses of each diet were performed in quadruplicate. Whole carcasses of fish withdrawn from each replicate tank on days 0 and 84 were pooled by fifteens and twos respectively, before each pool was homogenized. Mineral analyses of whole carcasses obtained at the termination sample (day 84) were performed once on each homogenized sample such that there were 10 mineral values determined per dietary treatment. Faecal moisture and crude protein content were determined by the same methods used for proximate evaluation of diets and feedstuffs. Dietary and faecal chromic oxide content were determined according to the spectrophotometry procedure of Fenton and Fenton (1979); (faecal ash determinations were made during these procedures). The plasma titres of T3 and T 4 reported for each dietary treatment, were determined according to the procedures of Omeljaniuk et al., (1984), and are the means of the 36 plasma samples obtained for each dietary treatment group (18 per replicate tank) at the termination sample (day 84). (Refer to section 3.5). The livers obtained at the terminal sample (day 84) were pooled by threes into 6 samples per dietary replicate for the determination of 5' monodeiodinase activity according to the procedures of Eales et al., (1992) (Refer to section 3.7). 154 Gross energy values for all diet and faecal samples were determined by means of a Gallenkamp adiabatic bomb calorimeter (Gallenkamp and Co. Ltd., Loughborough, UK). Amino acid and mineral compositions of feedstuffs and test diets were determined by AAA Laboratories, Mercer Island,WA, USA, and Quanta Trace Laboratories Inc., Burnaby, B.C. respectively. Mineral compositions of dried homogenized fish samples were determined by the Northwest Fisheries Science Center, Seattle, WA, USA.. Glucosinolate and phytic acid levels in the experimental diets were determined at the Agriculture Canada Research Station, Saskatoon, Sask, and Aiko Ltd., Biotechnology, Rajamaki, Finland, respectively. Plasma and liver samples were shipped on dry ice to Dr. J.G. Eales, Department of Zoology, University of Manitoba for assessment of plasma titres of T 4 and T 3 , and hepatic 5'D activity. 5.3.8 Response criteria: data calculation and analysis Weight gain (WG), specific growth rate (SGR), dry feed intake (DFI), mean daily dry feed intake per fish (MDDFI), feed efficiency (FE), protein efficiency ratio (PER) and percent protein deposited (by weight; PPD) were calculated according to the procedures of Higgs et al. (1991). Appetite (APP) (voluntary feed intake) was calculated with the following equation: APP = (mean daily dry feed intake per fish/ geometric mean weight of fish) x 100 Digestibility coefficients were determined according to the indicator method of Maynard and Loosli (1969) whereby, the apparent digestibility coefficients for dry matter, protein and energy for each diet were calculated according to the following equation: ADC (%) =100 x 1- % Cr 2Q 3 in feed x % nutrient in faeces % Cr203 in faeces x % nutrient in feed 5.3.9 Statistical Analyses Statistical analyses were performed using the computer package SystatR 6.0 for WindowsR, from SPSS Inc., Chicago Illinois, 1996. All statistics were tested at the 5 % probability level. Following a Bartlett's test for homogeneity of variance, the performance data were subjected to two-way analysis of 155 variance (ANOVA) according to the general linear means model. Where F-values indicated significance, orthogonal contrasts were used to compare dietary treatment means. 5.4 Results 5.4.1 Assessed composition of experimental diets All diets containing undephytinized canola meals contained approximately 8.1 umol/g IP6. In the undephytinized RPC diets this was substantially higher, at approximately 41pmol/g IP6. (Table 5.8) Commercial dephytinization of high fibre canola meal resulted in a 97 % decrease in the phytic acid content and negligible changes in protein content. Commercial dephytinization of rapeseed protein concentrate resulted in similar reduction in phytic acid content with a slightly greater (3.5 %) decrease in protein content (Tables 5.2). Proximate composition determinations and bomb calorific examinations of diets confirmed that all diets were isonitrogenous (43.6 - 45.2 % DMB), isoenergetic (22.1 - 22.7 % MJ/kg DMB) and relatively equal in lipid content (15.9 - 20.0 % DMB), and thus they closely approximated the formulated estimates. The diets were also relatively equal in dry matter (88.8 - 91.3 %) and ash content (8.1 - 9.4 % DMB) except for gross energy content which was slightly higher (Table 5.5). The essential amino acid levels in the diets were relatively uniform across all diets and, with the exception of the methionoine and cystine content, all amino acid levels met or exceeded required levels (Table 5.7). The methionine and cystine contents of the diets were lower than the expected values. However all levels exceeded the lower requirement for Met plus Cys (i.e. 2.2 % protein; Wilson, 1989; Haard, 1995) cited in literature. Analyzed levels of dietary minerals (Table 5.6 ) closely approximated calculated/estimated levels. No glucosinolates were detected in the NFM and NFMHP diets. Whereas only indoyl glucosinolates at a level of 0.05 umol/g DMB were present in the DRPC diet, all other diets contained both alkenyl and indoyl glucosinolates. Undephytinized high fibre canola meal diets contained 2.15-2.30 umol/g DMB alkenyl glucosinolates and 0.30-0.35 pmol/g DMB indoyl glucosinolates. Comparatively, the dephytinized high fibre canola meal diet (DHFCM) contained approximately half the amount of alkenyl Q oo 00 © Os <N —< OS iri oo' iri i n m rs 5 * 4 T t T t sq Tj- ' OS SO rs rs rs rs S ••a s C J it, C J © Os m OS rs OS © OS f - T f SO T f ' Os' so' T f — • T f OS T f Os' so' m rN © T f os r»' T f - H s q oo i n m' oo' so' T f rs rs rs rs rs rs so rs rs 13 So a o. CJ 8 73 > e • CD 8 P . lo u 71 © r-— ' T f OS T f © Os SO SO rs rs Cu u L H o a •a o (U m i n i— © Os © OS © Os O OS a 3 •a a. r3 OH s -a si •a s O H Q T f Os' T f so T f OS T f m T f oo m' T f m rs so X a '55 o iH OH u •a s r-od 00 00 sd m SO O rs © r-' J3 3 •a 'S, 53 &5 Ci to rs rs rs rs rs rs rs rs 6 0 WH fl u o •a <u X ) ••-» 0> TJ* o fl o •a <L> , o m u <H-t 3 6 1) to SH O H U V H CD 73 > - f l c_) 8 t / f D > '53 H - » s O H o c o •a o. CD O CD CD 3 •S T 3 •a N O vi <u s * 1 8 S M l ? S a l o o 00 i—i O o p 8350 NO" m cs O N cn V l m 5980. vi O o vo © O o o o 9440. ro cs' t—4 O N m 5820. vi o © O V I © o o 9810 O N ro rs rs' rs' o T f 6050. T f rs o T f V | 00 © © © 9910. T f ro cs r-' r-' r-m 5760. 00° © © O N r- T f o o © 1490C T f cs' l-H vd co 2440. 00° © p O N r~ 00 o o © 1650C I—H V I O N O N vd V O ro 2400. ro' © p T f ro o o rs 1590C V O V ) rs' N O ' vd V ) ro 1930. rs' 00 © O V I rs o o © 1670C O N 00 rs T f T f 00 ro 2410. © »—< © < N O N O N O o O N 1610C T f rs Tf" 00 r-co 2390. r-' O N © T f O r- V I o vq 1550C T f rs cs i—< rs' t -ro 2220. vd O N © N O ro v> p o V I 1670C O N cs cs' T f vi r-ro 1300. T f O N p o V I O N O o 00 1600C CS rs rs O N 00 ro 1280. rn 00 Ca Cr Co Cu Fe Mg Mn © © V I V l © © ON VO © © rs © © r-© rs © © r-cs © © O N m © © v > rs © ©' © 00 ro © o rs m © © v i o m © © c-rs © O CS rs © 00 © O N 00 00 © © ' N O 00 © s T f 00 © t-O N © o m © © © O N © © ' © O N © © © N O © © r-© o © oo © C~ V O 00 © cs' V © cs V © cs' V © cs' V © rs' V © rs V © rs' V © rs' V © rs' V o cs V © cs V o cs V © V I © © ' © V O © © ' O N m N O co V I rs © rn © m © © © ' O N V I — r - i m © O N © V O © © ' V I N O © ro cs N O © V O t-N O © N O © T f m vo © N O T f N O © CS O N rs cs V O rs O N ro rs © O N © rs V I CS c s N O O N 00 © cs T f CS cs e o •a <u OJ =3 <4-l o « o •a rs v i u 2 id OJ 0 0 CS Z e N u 8 2 U H O H o >—i O N i t - - . H ONi eoi cs ** r - ' c i r t r - ' <ri T t r - T t r-i v i •—i o T t p os t~- p ; oo cn T t t—' c i T t oo' v i cs' oo' T t —< w i p © T t Os Os r— O N t - , T t T t r-' c i T t t— v i c i r— T t —< m O O T t O N O N C S 00 00 T t T t t-' c i T t r— v i T t r— T t — ! v i u 0 I T t so m N O T t N O i n N O in N O m N O T t NO in NO o NO* c s r--cs' cs' SO cs' cs cs m T t o 00 o 00 N O * ~ ' T t OO -r -r - ' 00 r-' 00 t-' o T t N O C S m T t v> T t in T t in T t m T t C S C S o 00 o 00 © 00 © 00° o oo' O N c i 00 i - H r-' - * 00 r-' O N r-' cs 00 c n 00 © m ' o T t o T t cn T t o 00 00 O N c n od o \ r-° r - ' •oo r-" T t N O r-° O N C S T t —«' 00 cs T t —< 00 T t © v i d O N d O N - H T t -4 oo C N T t —«' 00 T t T t <-< O N •—< T t — ' m m bO -5 a 5 in - H T t v i •Ja-i n cu cs cs d cs c i 158 -=» l i CL) I H '•B o •8 e § 1 <D 6 u •S in <o if OH a u ! o 00 00 2 u oo TJ c r a 8 <u kH •3 c r Si S •a o a 6 a OH u a I O H s o •3 00 3 o i a u T 3 _g & cx in ex o e o . f l 'In O O _3 "5b Q 5 u it, C J u C J a 3 X i cd £•> "3 43 •a •§> ° o a 3 . m in T J © © O <o wi i n vo v© © ' o ' -*' wi O O r - r s m m o TT VO — ' o o ' —«' i n m m ON m i n o vo c i © m o m ~-> CN cn o m o m m oo ri 6 N m m ( N O N o o o m CN co i n T J T J T J e d d T J T J T J e e e ca CJ to O ^ w T f <n __. O VI VO CN m -a d . m T J 3 "* ^ d ^ « n ^ - T J VO ( N VO oo ©' d m rs vo od •-< ©' T J d _ m V0 0 - T J ~ ° VI C ON o r-' —< ON O r-' —«' m d vi. T J d m d ^ VI c • + + ^ oo o a _ m in • o o . VI Vt ^ in in £ o o ° V I V I a o 5 fl P •!2 C o u fl w o I 8 60 jg 1*- ^  i fl o S h 3 oj xl xi '53 CX —. 60 § 1 159 CJ cd o I fe vo m T J - cn & & & & cn o xi CX •s o a g Q VO 3 T J •5. OJ s ^ CU o o 'I f 2 2 4 "a T J 2 5 « CO fl_> •3 60 y C s -a fl o Jto OJ CX V} o X i </) CX 'Jo — OJ .3 5 En ^ I § u ? a .5 VD fl • Eb cn cj & ccj T J OJ OJ T J +-» O fl T J fl + 160 glucosinolates (1.15 umol/g DMB) and about one third less indoyl glucosinolates (0.20 umol/g DMB). Fibre reduced canola meal diets (ULFCM and ULFCMHP) contained 1.65 and 1.55 umol/g DMB alkenyl glucosinolates, respectively and 0.35 and 0.40 umol/g DMB indolyl glucosinolates. Relative to diets containing canola meal, the diets containing undephytinized RPC contained lower levels of alkenyl glucosinolates (0.45 - 0.50 pmol/g DMB) and higher levels of indoyl glucosinolates (0.65 - 0.70 pmol/g DMB). 5.4.2 Influence of diets on fish growth and feed digestibility and utilization Dietary inclusion of any of the canola meal test proteins (20% dietary protein) at the expense of fish meal, with or without commerical dephytinization or supplementation with microbial phytase, did not significantly alter (P > 0.05) appetite, feed efficiency, protein efficiency or level of protein deposition, relative to values obtained for fish fed the control diet (NFM). Other measured parameters were significantly affected by dietary treatment, however. Partial substitution of fish meal with undephytinized, unsupplemented high fibre canola meal (UHFCM) yielded weight gains, specific growth rates, feed intakes (DFI and MDDFI), available digestible protein (DPDD) and digestible protein to digestible energy ratios (DP/DE) that were not significantly different (P > 0.05), from control (NFM) levels. However, the apparent digestibility coefficients for protein and energy as well as the digestible energy content of the diet were significantly reduced (P < 0.05), relative to controls (NFM). Upon supplementation of the high fibre canola meal diet with a low (1,000 PU/g diet) dose of phytase, the weight gain and feed intake (DFI and MDDFI) of the trout were significantly (P < 0.05) increased and rose above the levels obtained for the fish on the unsupplemented, undephytinized high fibre canola meal diet (UHFCM), and the control (NFM) diet (Table 5.9). The digestibility of protein and energy of the diet were also improved by dephytinization so that the values were not significantly different from those obtained for the control diet. Overall this diet treatment yielded the highest values for weight gain, specific growth rate, and feed intake, and yielded feed efficiency, protein efficiency and percent protein deposited (PPD) values which were almost identical to those noted for fish receiving the control 161 (NFM) treatment. Supplementation of the high fibre canola meal (20% dietary protein) with a high dose (4,000 PU/g diet) of phytase (UHFCMHP) also improved diet digestibility relative to the unsupplemented, undephytinized high fibre canola meal treatment (UHFCM), wherein diet digestibility was equivalent (P > 0.05) to that of the control diet (NFM). Partial replacement of fish meal with undephytinized, unsupplemented canola meal that had undergone fibre reduction before dietary incorporation (ULFCM) (20 % dietary protein) yielded digestibility measures and fish performance values which were statistically equivalent (P > 0.05) to those given by the undephytinized, unsupplemented high fibre canola meal (UHFCM) treatment group, and also equivalent to the control (NFM) group, with the exception that the apparent digestible energy content of the diet (ADE) was significantly lower (P < 0.05) than that of the control. With supplementation of the fibre reduced diet with a high dose (4,000 PU/g diet) of phytase (ULFCMHP), the apparent digestible energy level of the diet was raised to a level equivalent (P > 0.05) to that of the control diet. The digestible protein to digestible energy ratio (DP/DE) of the diet was elevated above that of the control diet (NFM), but no other parameters were significantly changed. Partial replacement of fish meal with high fibre canola meal which had been commercially dephytinized by pretreatment with microbial phytase before dietary incorporation (DHFCM) (20% dietary protein), yielded an apparent digestible energy coefficient and an apparent digestible energy value for the diet which were significantly lower than those of the control diet. However, in terms of fish performance, weight gain and total feed consumption were significantly higher (P < 0.05) than values obtained for the control group (NFM). The performance of the commercially dephytinized, high fibre canola meal treatment group (DHFCM) was equal (P > 0.05) to the performance of the undephytinized, unsupplemented treatment group (UHFCM) in all respects except for weight gain, which was determined to be significantly greater for the dephytinized, high fibre canola meal (DHFCM) treatment group. Relative to the high fibre canola meal treatment groups supplemented with microbial phytase (UHFCMLP and UHFCMHP), no significant difference (P < 0.05) between the dephytinized treatment group (DHFCM) and C / J I l l Q 1 i TJ CfH 3 o •§ •a >, o " ' * _ l ^ 3 T J O a 'O 5 3^ S "o a *•* < ^ a-a OJ S 1 i •a c - TJ " c2 «3 ^ c x is O 'O _ CJ 00 .-S3 " ' to OJ o 2 & 5 | oo s-- 2-y £ •a .§ OJ OJ '5 e OJ «"> s u o j o U l C L . ON a OJ CO o fe s Q o o o ©' t s 0 0 vi 0 0 0 0 0 0 cn cs T f 0 0 vO O N vn o o o © CS ©-© ' cs CS CS T f v q v q vo 0 0 cs Ov r » — ' vo T f CS ON VO V I T f cs cs 0 0 —< r~ v i cn cs o Os v i cn IB •a 60 CJ oo O V I cs 00 cn ~ - cs O T f ~ - C S T f O — . T f o o o d oo vo Tf cs cn T f ON o T f ON ~a •a 6 0 I • ' Pi o O0 CS v i T f vo NO T f CS NO VO NO T f cn T f cn O o o V I o o VO d cs •—4 © CS d T f © 0 0 © 0 0 d oo ©' ON 0 0 r~ —' T J CJ .OJ 33 I—I fcu a © cs vo NO © VO © © © © © cs cn © © cs cn T f vo NO V I CS CS CS o C S V l cs © C S © © © cs CS o O N V I — © cn r-d t CO 43 _ T J .8 1* H 60 o x> vo -< © cs 0 0 © V O 0 0 ON 0 0 0 0 0 0 —< — c ON —' d V I 0 0 © ON CN ON © © V I o © © cn cn V I VO © V I V I oo T f ON V I © V I Ov 60 "a •a 60 60 • a W T J £ 3 NO © o ON ON ON o CN 60 ~a • a 60 NO O © V I N O t -d cs V ) v i cs V l cs ON uo cn cs 0 0 0 0 0 0 r-l © cs oo' cs C O VI ©' cn r-C N O N V I © C O 00 ON 0 0 C N © C O o" C O CN CN si 60 3 S 2 ON © ' cn T J OJ a CJ T J a ' .§ o U l Q a, cu T f T f CN CS T f T f T f T f NO T f o cn ON T f V I T f W0 o 2 o •a 60 'lo OJ T J ••a Cui O 3 O •a CJ V I OJ OJ Pi 162 p d A CU II a a s 1 V I © © v/l CU a OJ 3 1 •8 X i o OJ > J J OJ •s OJ X I ? co OJ 3 •a 3 OJ oj £ JJ CO OJ CJ es OJ f I •a 3 oo 163 E • £ » « # » » * » » » co * » Z co co Z Z co Z co Z ~5 •a co Z CO Z CO CO Z Z CO Z CO z CO Z co Z t "a •a eo co Z * 5 % 0 s •-—' Pi a CO co Z co co co co z z z z i .4) t E a CO Z CO Z CO Z CO CO Z Z oo -3 E !9 Q co co z z co co z z co Z co Z CO Z co Z ••—' fr 5| CO z CO CO Z Z CO Z CO Z CO Z 60 60 11 co Z co Z CO Z CO z CO Z co Z co Z 60 "a a 60 s i 3.S 5 Ha CO Z CO Z CO Z CO z CO z co Z co Z TJ O Cu •s .3 S Cu b Q O H CL, 2 O •a T 3 ••a 13 a o •a 4> >/-> ,<u 2 i3 O C J «/-> C J §5 164 co Z CO Z CO Z CO Z CO Z CO z 1 •a bo CD is * — ' co Z CO Z CO Z CO z co Z co Z CO Z 00 % 0 s -o CO co Z CO Z CO Z CO z CO Z CO z CO z co Z 5 TJ CD ,CD Q co Z co Z CO Z CO Z CO Z CO Z CO z co Z co co Z Z l i t CO Z CO z CO z CO z CO z CO Z co Z CO z I i t Xl •a e fr 1 5! CO Z CO Z CO z CO Z CO Z CO z CO Z co Z co Z 00 n 60 -CO Z CO Z CO z CO Z CO z CO z CO Z CO Z CO Z oo bo co Z CO Z CO Z CO Z CO z CO Z CO Z CO Z CO Z 4) "1 !$ Q bo c a 2 a, cu "8 En O O H CD TJ c 01 -«—» o a< 0 s -a O H O H S o •a CD TJ 4—' CD ••a <*H o Ci o •a CD »-H •s oi C/D 5 2 Q I C J 1> 15 c j CJ c j U H u c/l WH <L> w <U a O H • • * * i n «/-> « n © o r s r s o © o o o o © © ' d d d © ' v> O N 00 r~-T t N O N O O N N O m r s O 00 ~* © d d r s © ' f-H N O c n VO f - r s 00 T f T f 00 r-' c s O N 00 r s T f c s 00 m O N 00 </-> © p r n CS r s cs' r s 00 O N r s O N 00 o T f r s O N m O N O N © r - m N O m m T f r n 00 d r s O N 00 1-H c s r s T f T f «n m ^ H SO cn N O CS O N <n c n 00° cs' O N 00 1—1 r s T f CS O N r~ •n 00 <n 00 c s N O I-H ?-H <n O N N O —'t O N 00 I—I © CS T f T f CS © 00 cn N O 00 00 »—H —H O N c n 00 r-' 1—1 00 oo O N r s m T f 00 m r s N O T f <n N O T f N O oo' cs' 00 vi © 00 oo 00 CS m cn N O t~ m N O N O O N T f - H O N T f 00° © —H* 00 00 © CS T f T f T f © O N c s 00 <n © m 00 O N T f O N i n © ' 00 O O O N m c s T f 00 N O c s cn O N i n •n © —H r-' c s 00 cs' ~ H 00 00 I—H O N CS cn © i n r s T f c s c n © © O N N O O N d •-t O N oo Tf CS N O m r - 00 m c n © r s N O 1—1 vi O N O N © O N 00 i — ' O N c s o u o w 6 0 ID •a 4? 6 0 3, D 6 0 Q 0H Q e o t, T -1 •a •a x> ca x> 2 cu 166 168 OO oo * Z Z oo oo oo oo » z z z z oo « oo oo oo z z z z oo z 00 * Z O O 00 00 00 00 Z Z Z Z Z 00 00 00 00 00 Z Z Z Z Z oo oo oo oo » Z Z Z Z oo * oo oo oo z z z z c o •a oo oo oo oo oo z z z z z oo oo * z z 3 •a u T3 *3 e o •a & ,3 l l (J .P e O u o o o w t>0 D 5 SI § I 8 I t/i as oo Z in w jo w w 55 Z Z 55 Z OO CO oo oo oo Z 85 Z Z Z i 8 3 8 O W W I 3 n 1 i I g 4 o 170 the low phytase treatment group (UHFCM) in all respects except for weight gain, which was determined to be significantly greater for the dephytinized, high fibre canola meal (DHFCM) treatment group. Relative to the high fibre canola meal treatment groups supplemented with microbial phytase (UHFCMLP and UHFCMHP), no significant difference (P < 0.05) between the dephytinized treatment group (DHFCM) and the low phytase treatment group (UHFCMLP) were found in terms of fish performance or diet digestibility. However, relative to the high phytase diet (UHFCMHP) significant differences were found with respect to digestibility measurements, wherein, the apparent digestible energy coefficient and the digestible protein content, of the dephytinized high fibre canola meal diet (DHFCM) were significantly lower (P < 0.05). The replacement of the entire fish meal component of the control diet with any of the rapeseed protein concentrate test proteins (95 % dietary protein), (undephytinized or dephytinized irrespective of procedure), yielded appetites which were significantly (P < 0.05) higher than, and weight gains, feed efficiencies, protein efficiencies, and levels of protein deposited as flesh which were significantly (P < 0.05) lower than those observed for the control (NFM) treatment group. Additionally, the apparent digestible energy contents of the diets (ADE) and dietary ratios of digestible protein to digestible energy (DP/DE) were also significantly (P < 0.05) lower and higher, respectively than corresponding control (NFM) levels. Al l other measured parameters varied as a function of the particular RPC dietary treatment employed. Replacement of the entire fish meal component with undephytinized, unsupplemented RPC (URPC) yielded an apparent digestible protein coefficient, a digestible protein level and a digestible protein to digestible energy ratio (DP/DE) which were significantly higher than those found for the control diet (NFM). Also, specific growth rate and total dry feed intake (DFI) were significantly reduced (P < 0.05) relative to controls. However, daily feed intake and the apparent digestible energy coefficient of the diet were not significantly different from respective control (NFM) values. When the same diets were supplemented with a low dose (1,000 PU/g diet) of microbial phytase (URPCLP), appetite and the ratio of digestible protein to digestible energy (DP/DE) of the diet were significantly (P < 0.05) lower than corresponding values obtained for the unsupplemented, undephytinized-RPC treatment group (URPC). However, compared to the control (NFM) treatment, digestibility 171 parameters and the performance of fish fed the low phytase RPC diet (URPCLP) followed the same pattern as that noted for the undephytinized, unsupplemented RPC treatment group (URPC) in all respects, except the fish fed the URPCLP diet consumed significantly less (P < 0.05) feed daily (MDDFI) than the control group (NFM), whereas daily feed consumption of the undephytinized, unsupplemented RPC group (URPC) was the same as the control level (NFM). Supplementation of the undephytinized-RPC diet with a high level of phytase (4,000PU/g diet) (URPCHP) resulted in significantly lower (P < 0.05) values for weight gain, and specific growth rate than found for fish fed diet URPC. This was also true for the apparent digestible protein coefficient, digestible protein content and digestible protein to digestible energy ratio (DP/DE) noted for this diet relative to values obtained for the unsupplemented, undephytinized RPC (URPC) treatment. Relative to the control diet (NFM), the apparent digestible protein coefficient of the URPCHP diet and the digestible protein level of the diet (DPDD) were not significantly (P > 0.05) different. However, the specific growth rate, and feed consumption and of the fish fed the URPCHP diet and the apparent digestible energy coefficient for the high phytase-RPC diet were significantly lower (P < 0.05) than respective control values. Replacement of the entire fish meal component of the diet with commercially dephytinized-RPC (95 % dietary protein) significantly (P < 0.05) increased the weight gain and growth rate of the trout and total feed consumption, but decreased appetite relative to fish ingesting the undephytinized, unsupplemented-RPC diet. Relative to fish fed the low phytase-RPC diet (URPCLP), weight gain, specific growth rate, and total and daily feed intake (DFI and MDDFI) were significantly (P < 0.05) increased when the trout were fed the diet containing commercially dephytinized RPC, but other parameters were unchanged. Relative to the high phytase-RPC diet, commercial dephytinization of RPC did not significantly (P < 0.05) change the apparent digestibility of protein and the available energy level of the diet. However, appetite was significantly (P < 0.05) decreased, in fish fed the latter diet whereas all other measured parameters were significantly (P < 0.05) increased relative to those of fish fed diet URPCHP. The trout fed the control diet (NFM) and diet DRPC had comparable specific growth rates and feed consumption, and the apparent digestibility coefficients noted for protein and energy for these two diets 172 were also not found to be significantly different (P > 0.05). However, the level of digestible protein in diet DRPC was higher (P < 0.05) than that found for the control diet (NFM). As a general comment, all groups fed the diets in which rapeseed protein concentrate was used to replace all of the fish meal (95 % of dietary protein), exhibited significantly lower weight gain, feed efficiency, protein efficiency, percent protein deposited as flesh (PPD), and there were reduced levels of apparent digestible protein in the diets (DPDD), and they also had higher appetite as well as dietary ratios of digestible protein to digestible energy (DP/DE) than respective values found for the control group (NFM). It must be noted however, that all DP/DE ratios, irrespective of dietary treatment, were within the range of 22-26 MJ/kg diet which Cho (1990) recommended as optimal for growth of salmonids. The fish meal diet which was supplemented with a high dose of phytase (4,000 PU/g diet) in order to highlight the possible nutritive effects associated with the phytase product itself, had a significantly higher (P < 0.05) digestible protein to digestible energy ratio (DP/DE) than did the control (NFM) diet. The performance of fish fed these diets (NFMHP and NFM) however, did not significantly differ (P > 0.05) with respect to any of the measured performance parameters and there were no other differences found in the measured digestibility paramenters (Refer to Table 5.10 and Table 5.12). Some differences were found in carcass composition. 5.4.3 Influence of diet treatment on body composition As shown in Tables 5.13, 5.14 and 5.15, dietary treatment did not significantly influence terminal whole carcass levels of protein (46.5- 54.2 % DMB), lipid (37.1- 43.0 % DMB ), ash (4.9- 7.8 % DMB) or levels of Cr, Na, K, Fe or Cu . Significant differences were, however, found for whole carcass dry matter as well as Mn, Mg, Ca, P and Zn content, but mostly in relation to the dietary inclusion of rapeseed protein concentrate test proteins (95 % dietary protein), and the high phytase fish meal treatment group (NFMHP). Yet, the variations in carcass dry matter, Ca, Mg, Mn, and P contents were considerable and not clearly related to dietary treatment, making interpretation difficult. The significant differences in mineral content that were of particular note were as follows: the carcasses from all RPC treatment groups were significantly (P < 0.05) higher in Mn and Zn content than 173 values found for trout fed the control diet (NFM), although carcass Zn content did not significantly (P > 0.05) differ within RPC diet treatment groups. Additionally, some of the the mineral levels in the carcasses from the undephytinized, high phytase-RPC treatment group (URPCFfP) appeared to reflect the overall poor performance of fish fed this diet, wherein the levels of Ca, Mg and P detected in the carcass of these fish were significantly lower than those found in the control fish (NFM), or those found in the undephytinized, unsupplemented-RPC diet treatment group (URPC). A significantly (P < 0.05) higher Mg content in the carcasses of the high-phytase fish meal treatment group (NFMHP), relative to control levels was also observed. There was no modification of carcass mineral content as a result of dietary inclusion of any of the canola meal test proteins at a level of 20% of the dietary protein, (undephytinized, or undephytimzed, and with or without phytase supplementation, or fibre reduction). In relation to the proposed "normal" levels of minerals in the bodies of rainbow trout (Shearer, 1984), Ca, Fe, Mg, Mn, P, Na were generally within normal ranges for all dietary treatments, Cr, Cu, were generally higher than normal ranges, and K levels were generally lower than normal ranges. The Zn levels in the carcasses were close to or lower than "normal" levels in trout fed the fish meal and canola meal-based diets but were higher than "normal" levels in the fish meal-free RPC treatment groups (Refer to Table 5.14). 5.4.4 Influence of diet treatment on thyroid and liver status Only samples taken from the following dietary treatment groups were assessed for T 3 , T 4 , hepatic 5'D activity and liver size: NFM, UHFCM, ULFCM, URPC and DRPC (Refer to Tables 5.16 and 5.17). Plasma titres of T 3 and T 4 , hepatic 5'D activity levels and liver sizes of UHFCM and ULFCM treatment groups did not differ significantly (P > 0.05) from those of the control group (NFM). Plasma levels of T 4 from URPC and DRPC treatment groups also did not differ significantly (P > 0.05) from control levels. However significant differences in plasma titres of T 3 , activity levels of 5'D and liver size were evident. Plasma titres of T 3 were significantly lower (P < 0.05) (1.35 ± 0.167 vs 2.43 ± 0.213 ng/ml) and livers were significantly (P < 0.05) smaller (0.20 ± 0.024 vs 0.29 i 0.024 g) in the URPC treatment group than 174 in the control group (NFM), but 5'D activity levels were not significantly (P > 0.05) different. Comparatively, plasma titres of T 3 and average liver size in the commercially dephytinized-RPC treatment group (DRPC) were not significantly (P > 0.05) different from values for the controls (NFM). However, 5'D activity levels were significantly higher (P < 0.05) in the dephytinized-RPC treatment group (DRPC) than they were in the control (NFM) group (6.72 vs 1.54 ng/ml). When the response of the undephytinized high fibre and undephytinized low fibre canola meal treatments were compared to each other (UHFCM vs ULFCM), significant differences were evident only with respect to plasma levels of T 3 which were higher (P < 0.05) in fish receiving the ULFCM treatment than fish from the UHFCM treatment, (2.94 vs 2.05 ng/ml respectively). 5.4.5 Influence of diet treatment on general fish health and behaviour Fish fed any of the rapeseed protein concentrate diets (URPC, URPCLP, URPCHP and DRPC), showed an initially strong aversion to pellets. Fish repeatedly mouthed then violently expelled the pellets, sometimes trying the same pellet as much as seven times before ultimate ingestion or final rejection of the pellet. Wasted pellets were counted, and based on the average weight of individual pellets per diet, the amount wasted was subtracted from the amount fed. The observed reluctance of fish to ingest the pellets was no longer evident after two days in fish fed the DRPC diet. However, fish fed the other three diets still rejected pellets with some consistency for a period of approximately 10 days following the commencement of the growth trial. Faecal casts were evident from day 7-10 in the faecal settling columns of the tanks of the fish fed these diets (URPC, URPCLP, URPCHP). Thereafter, the fish agressively ingested these diets and there was no evidence of faecal casts being found in the respective faecal settling columns that were affixed to each tank. Mortalities resulted when fish jumped out of their respective tanks through a small opening located near the water inlet spout. Once the opening was covered with mesh no more "jumpers" were evident. No mortalities resulted from dietary treatment effects. fe » CO CO co 55 55 55 _ VO VO C l o r- 00 ON © © o 1 © ' © © ' © 175 OS T t r s <o © r s m © vo © ' © ' © ' © ' T f © OS r s Os T t r-' c i T t © r s Os 00 VO r s c i T t T t Os i - H OS 00 r-' r s c i VO OS c i VO r s T t C l t-' r s T t m r-' 00° C l c n Os 00 00 r s rs' « n r-' r-' m © T t r s T t oo' r s © ' m r s T t r - •n m T f oo' r s © ' m r-' T f r o <n T t m Os' r s VO T t vd i n T f T t m 00 r s t - ' T f © v> VO VO OS r s © ' m vd r s T f T t vq « n 00 r s IO r~ © ' T f Os © T t © r-' r s r—1 i n t - ' © ' T f 00 m r s 00 r s i n m os' © r s g <D TJ a & 1 TJ i l o C3 o •a C M 8 CD •s •gs CD I o CD c o i ; 8 1° ' 0S- Q S S "5? xi 3 "2 '8. Oi in "8 •a X l o 176 oo 5 vi IS 3 o j o % s 1 § • NS NS NS » • o ro vo NO NO T f V O o o © o o © ©' d d d d so ro T f rs o rs ON r-T f o T f O ro ro © T f ' d d d d <n r- rs r- r s 00 5157 ro - ro' ro oo' T f ro m <n 00 <n p r-1462 T f rs' vo' r-rs rs T f NO r~ ro NO 4950 T f rs rs' T f T f ro cs' ro 00 r - 00 NO 5460 T f ro vo <n vo ro rs r- T f NO ON T f 4978 vd ro' T f ON V O ro 00 m T f ro ON 5540 T f rs rs' t - ' m ro p O T f 4445 vo' vo' T f d T f ro i—< © rs 00 vo r- co 4448 00 rs ro ro' ro ro rs O N r- ro T f 4803 oo' ro T f NO in ro 00 00 r- vo 5095 00 T f r-' vo' vo ro © vo ro ro ON NO 5582 ON ro' 00° T f ON ro r-00° T f O m m 00 T f id T f T f d m ro rs Ca U 3 u u Mg Mn o o d oo oo Z Z T f r-•—i r o ro t - NO rs cs NO T f cs T f ro O V O 00 ro' T f cs' 00 T f vo ro vo T f rs ON T f ON r-' I—* T f C O r-NO C O vo rs C S "-1 o — ON NO d NO rs o ro T f T f T f C S ~" _ _ 00 d ro' ON 00 C O cs vo ro VO T f cs ON 00 ON 00* oo' 00 NO ON ro NO ro T f cs T f ro' 00 C S ON 00 00 <n T f T f T f C S p T f T f T f cs' d T f VO vo rs VO cs T f cs ON C O 00* vo' d ON cs T f o T f T f T f cs C O cs 00 m' co' cs 00 r~ T f vo ro T f cs '-H o rt ON T f d ro ON ON T f T f T f T f cs NO T f ON ON cs r-' rs 00 O ON ON vo T f C S rt o o ON d T f T f vo 00 ro m vo T f cs o d ro vo d 8 VO r-' cs vo cs cs CS C S C S vo C S 3 o •a CO 3 M "lo C J T J . «"> I-a o a o •a cs NO cs | V 0 O J l l | H o <a c C J B u g O J T J I O J 8 O H O J 3 1 ^3 •s vo T f 0 ' T ? -H op OJ © | -H oo, s i +• 2 V O TJ a ^ a « z oo . -H. °* a bp is 2 1 OJ 2.S o NO vo o cs fl •=•>" fl o O J ~ Ui ro e -H cs s ro C J O N | -H V 5 T f C J vo ro G O N o -H 5 o ° 00 O J 00 6 °°. O J o 2 "H I C v ^ o j 1 - 1 -H ro ro ro oo" ro' -H 00O J ! XJ o e N 178 co Z Z oo 55 CO 55 oo 55 oo Z r/i Z oo Z .S3 03 • o .sr 2 05 ? l i t S O s J Q oo Z oo Z oo Z oo Z oo Z oo Z oo Z oo Z co oo Z Z oo oo Z Z co co Z Z oo oo Z Z oo oo Z Z oo oo Z Z 00 C/3 z z oo oo z z co -z * oo Z oo Z co Z oo Z oo Z oo Z CO Z oo Z co Z oo Z oo Z co Z co Z co Z co Z oo Z 00 Z oo Z i-a Ma a i-a rs i) IS I - H rs 5 N 179 55 55 co co 55 55 CO 55 CO 55 CO 55 p 4 180 © ©' A CU II in * IQ co • • • Z m ~ © © vo © © © m © o © © ' © ' © © o 00 o o OV <o Ov cn »-* m © © © © ' © 00 T t CN © cn I m T t cn »—H «T> T f OV Ov CN CN 00 © cn CN vo cn r~ Tf_ oi CN m I ng/ml iOf 13 CN VD VO cn cn 00 T t cn vi CN o CN r-CN cn © ov CN © VO CN VO 00 © Ov ON t - ' o •a 1 I O o a <n 6 2 i 1 <o o ©' A l CU «-» ca * is o 8 •a I 1 C O is i PH I I 1 8 8> 1 o ii • • H D to 55 55 co 55 co 55 co 55 co 55 co 55 co 55 co 55 to 55 182 5.5 Discussion 5.5.1 Assessed composition of diets Commercial pretreatment of canola meal and rapeseed protein concentrate with microbial phytase, successfully reduced phytic acid content by at least 95 % but not totally, which is similar to the findings in experiment I. As was concluded in Chapter 4, the lack of complete efficiency of phytic acid removal is likely attributable to strengthening of phytate-protein and phytate-mineral bonds, wherein heat treatment, such as that routinely used in rapeseed and canola processing for the deactivation of thioglucoside glucohydrolase, is believed to strengthen the electrostatic interaction between phytic acid and protein and stabilize the protein-phytic acid complex. This effectively reduces the solubility of both nitrogen and phytic acid (Gillberg and Tornell, 1976; Serraino and Thompson, 1984). The phytic acid content of all diets, including the control diet (NFM) (0.7 pmol/g IP6 contributed from extruded wheat) exceeded 0.5umol/g IP6 which is the level typically found in commercial salmonid feeds (Spinelli et al., 1983). It is also a level which has previously been demonstrated to reduce growth and feed conversion of rainbow trout by as much as 10%, relative to phytic acid-free diets (Spinelli et all., 1983). Consequently, in the absence of complete phytic acid removal, reduced growth and feed conversion might be expected in all dietary treatment groups. However the antinutritive effects of phytic acid might be expected to be most severe in the undephytinized-RPC treatment groups for which phytic acid content was at least 60 times greater than that found in the control diet, has previously been demonstrated to reduce growth and feed conversion of rainbow trout by as much as 10%, relative to phytic acid-free diets (Spinelli et all., 1983). Consequently, in the absence of complete phytic acid removal, reduced growth and feed conversion might be expected in all dietary treatment groups. However the antinutritive effects of phytic acid might be expected to be most severe in the undephytinized-RPC treatment groups for which phytic acid content was at least 60 times greater than that found in the control diet, previously been demonstrated to reduce growth and feed conversion of rainbow trout by as much as 10%, relative to phytic acid-free diets (Spinelli et all., 1983). Consequently, in the absence of complete phytic acid removal, reduced growth and feed conversion might be expected in all dietary treatment groups. However the antinutritive effects of phytic acid might be expected to be most severe in the undephytinized-RPC treatment groups for which 183 phytic acid content was at least 60 times greater than that found in the control diet, been demonstrated to reduce growth and feed conversion of rainbow trout by as much as 10%, relative to phytic acid-free diets (Spinelli et all., 1983). Consequently, in the absence of complete phytic acid removal, reduced growth and feed conversion might be expected in all dietary treatment groups. However the antinutritive effects of phytic acid might be expected to be most severe in the undephytinized-RPC treatment groups for which phytic acid content was at least 60 times greater than that found in the control diet, demonstrated to reduce growth and feed conversion of rainbow trout by as much as 10%, relative to phytic acid-free diets (Spinelli et all., 1983). Consequently, in the absence of complete phytic acid removal, reduced growth and feed conversion might be expected in all dietary treatment groups. However the antinutritive effects of phytic acid might be expected to be most severe in the undephytinized-RPC treatment groups for which phytic acid content was at least 60 times greater than that found in the control diet, to reduce growth and feed conversion of rainbow trout by as much as 10%, relative to phytic acid-free diets (Spinelli et all., 1983). Consequently, in the absence of complete phytic acid removal, reduced growth and feed conversion might be expected in all dietary treatment groups. However the antinutritive effects of phytic acid might be expected to be most severe in the undephytinized-RPC treatment groups for which phytic acid content was at least 60 times greater than that found in the control diet. The glucosinolate content of all diets containing canola meal exceeded 1.40 pmol total alkenyl glucosinolates, (measured as the sum of 3-butenyl, 4-butenyl, 2-hydroxyl-3-butenyl and 2-hydroxy-4-pentenyl /g dry diet (Hardy and Sullivan, 1983)), which has been described as the level at which depressed growth performance, reduced feed efficiency and impaired thyroidal and hepatic function have been observed in rainbow trout. By comparison the RPC diets had glucosinolate contents which were well below maximum tolerable levels for alkenyl glucosinolates. 5.5.2 Influence of diet treatment on fish growth and feed digestibility and utilization Weight gain, growth rate, feed consumption, appetite, feed efficiency, protein efficiency, protein deposition, carcass mineral and proximate composition, as well as hepatic and thyroidal response were not significantly altered by partial substitution of fish meal with undephytinized high fibre (commercial) canola 184 meal, at a inclusion level of 20% dietary protein. Diet digestibility was significantly reduced, however. Microbial phytase supplementation of undephytinized high fibre canola meal diets effectively improved growth response and feed digestibility and utilization, indicating that there was some degree of in vivo dephytinization of feed, allowing for improved nutrient utilization. These findings corroborate the findings of Spinelli et al. (1979) and Cain and Garling (1995), who found that the addition of microbial phytase to plant-protein based diets significantly improved trout growth rates. Both phytase doses used in this investigation improved the feeding value of high fibre canola meal diets, however the beneficial effects of phytase supplementation were more pronounced with the low dose (1,000 PU/g diet) rather than the high dose (4,000 PU/g diet) of phytase. Whereas the high dose of phytase improved performance to levels which were comparable (P > 0.05) to control levels in terms of all measured parameters, the low dose of phytase supported weight gains and feed consumption which exceeded (P < 0.05) those of the control fish, and yielded values for all other measured parameters which were at least comparable to those obtained by the controls. The low and high phytase, high fibre canola meal diets which were reasonably close to the NFM diet in protein, lipid, ash, and gross energy content (Refer to Table 5.6), as well as phytic acid content (Refer to Table 5.8) differed with respect to glucosinolate content, which may account for the observed difference in performance. Above a dietary glucosinolate content of 158 (xg/g DMB 3-butenyl isothiocyanate (= 1.40 umol/g 3-butenyl isothiocyanate equivalents, where 3-butenyl isothiocyanate equivalents, as measured in grams is the sum of 3-butenyl-, 4 butenyl-, 2-hydroxy-3-butenyl and 2-hydroxy-4 pentyl- glucosinolates multiplied by a factor of 113; Hardy and Sullivan, 1983; Hilton and Slinger, 1986), glucosinolates and products of glucosinolate hydrolysis are known to impair thyroidal function, and reduce growth response in rainbow trout (Higgs et al., 1995). Additionally, a high dietary glucosinolate content is also characteristically associated with reduced feed intake, and feed efficiency of salmonids (Higgs et al., 1979; 1990). Both diets (UHFCMLP and UHFCMHP) had glucosinolate contents that exceeded 1.40 umol/g. However, the undephytinized, high fibre, high phytase canola meal diet had a slightly higher alkenyl glucosinolate content (2.30 vs 2.15 umol/g alkenyl glucosinolates) than was present in the undephytinized, high fibre, low phytase canola meal diet did (2.15 umol/g alkenyl glucosinolates), which might explain the slightly lower 185 weight gain, growth rate, feed intake, and feed efficiency of fish fed the former diet. As high dietary glucosinolate content is associated with hepatic and thyroidal impairment in salmonids (Higgs et al., 1979; 1990), one might expect to find evidence for altered hepatic and thyroidal status, (e.g. T 3 , T 4 , 5'D levels or histopathological changes), of fish in the UHFCMLP and UHFCMHP treatment groups, possibly being more severe in the UHFCMHP treatment group which had a higher concentration of alkenyl glucosinolates. Yet, given that the thyroidal and hepatic responses of fish fed the undephytinized, unsupplemented, high fibre canola meal diet did not differ significantly (P > 0.05) from the responses given by control fish, and both the low and high phytase high fibre canola meal diet treatment groups (UHFCMLP and UHFCMHP) out-performed the unsupplemented canola meal treatment group, it is entirely possible that the thyroidal and hepatic responses of fish in the low and high phytase RPC treatment groups would also have closely approximated control values, showing little evidence of impairment. This would suggest that a factor other than glucosinolate content might be responsible for the observed antinutritive effects, and the observed differences in performance of the low and high phytase RPC treatment groups. Phytic acid removal by commercial dephytinization was also effective in significantly improving the feeding value of commercial (high fibre) canola meal, wherein all performance and digestibility measures were not statistically different (P > 0.05) from those of the low phytase high fibre canola meal diet (UHFCMLP), with weight gains, growth rates and feed intakes being second only to values for the UHFCMLP treatment group. The apparent digestible energy of the diet was, however, slightly lower than that of the control diet. Consistent with the findings of McCurdy and March (1992), who determined that fibre reduction of canola meal did not alter protein quality, and did not significantly improve growth or feed efficiency of juvenile rainbow trout (6 g) fed diets with fibre-reduced canola meals at a level of 40% of the dietary protein, in this investigation fibre reduction was ineffectual in improving the feeding value of canola meal, as measured by trout weight gain, growth rate, feed intake, feed efficiency, hepatic or thyroidal status, protein efficiency, percentage of protein deposited as flesh, carcass proximate and mineral composition and feed nutrient (e.g. protein) digestibility and/or utilization. In contrast to the results obtained upon phytase supplementation of high fibre canola meal, a high dose (4,000 PU/g) of microbial phytase improved the 186 feeding value of low fibre canola meal for rainbow trout, but did not improve it to such an extent that the ULFCMHP diet was able to support equal fish performance as that of the control treatment (NFM). Weight gain, feed efficiency, protein efficiency, and the conversion of dietary protein to flesh (PPD), as well as the digestible energy content of the diet were significantly reduced when the entire fish meal (animal protein) component of the control diet was replaced by undephytinized rapeseed protein concentrate (95 % dietary protein). Plasma titres of T 3 of the URPC treatment fish were depressed and average liver weight was reduced, indicating impairment of thyroidal function, as has been previously described for salmonids fed diets high in glucosinolate content (Higgs et al., 1979; Yurkowski et al., 1978; Hardy and Sullivan, 1983). As was the case with microbial phytase supplementation of the high fibre canola meal diets, the low dose (1,000 PU/g) treatment group outperformed the high dose (4,000 PU/g) treatment group. However, neither the low (1,000 PU/g diet) nor high (4,000 PU/g diet) dose of phytase effectively improved the feeding value of rapeseed protein concentrate for rainbow trout, suggesting incomplete or absent in vivo dephytinization. In fact poorer performances were given by the microbial phytase supplemented- RPC treatment groups than by the unsupplemented, undephytinized-RPC (URPC) treatment group, and the poorest performance was given by the the high phytase (4,000 PU/g diet), undephytinized-RPC treatment group (URPCHP). Most measured parameters of the URPCHP treatment group were significantly inferior (P < 0.05) to those of the control group (NFM). The reason for the poor performance of this group is unclear. However, because fish from this treatment group (URPCHP) as well as fish in the (URPC) and (URPCLP) treatment groups initially showed a strong aversion to their diets, (expelling pellets and avoiding feed for a period of ten days), there is a strong suggestion that the palatability of the RPC diets may have been substantially less than that of those diets containing animal protein. Indeed feed intakes of the URPC, the URPCLP and the URPCHP treatment groups were significantly lower than those noted for the control or canola meal treatment groups. Observations of fish feeding behaviour and fish growth responses also indicated that microbial phytase, added as a "coating" to the pellets, may have further reduced the palatability of the diets. Higher in phytase, the URPCHP diet may have had the most objectional taste, 187 countering the chemoattractant properties of dietary FINNSTIM™. Hunger likely overcame the poor taste of the pellets, causing the fish to try, then to become accustomed to the diet. During the period of time that feed pellets were repeatedly mouthed before ingestion, it is highly possible that the phytase coating of the pellets may have been reduced or removed before final ingestion of the pellet, thus reducing or precluding in vivo dephytinization of the digesta. However, what may yet explain the lack of any evidence for in vivo dephytinization of RPC over the remaining investigation period is a mystery, for after the first ten days of rejecting the pellets, fish fed quickly and agressively, leaving little time for possible leaching losses of phytase. The results suggest that FINNSTIM™ at 1.5 % diet DMB may have been ineffective in improving the palatability of the RPC diets at least during the initial part of the study. Perhaps phytase incorporation into the canola and RPC diets before pelleting might prove beneficial in reducing the observed palatability problems. However, careful attention would have to be paid to the temperature of pelleting, as the activity of phytase derived from Aspergillus sp. has been shown to retain 46% of its activy at 87° C by Simons et al. (1990), and to be completely destroyed at an internal pellet temperature of 80 0 C by Jongbloed and Kemme (1990). To avoid the loss of phytase activity through heat processing, cold pelleting might have been used. In this regard, cold-pelleting of the diets was attempted prior to the commencement of this investigation, but this resulted in pellets that were soft, powdery and of poor consistency, and durabilty. Therefore it was decided to externally coat the pellets with phytate in the manner described. Contrary to the findings of Teskeredzic et al., (1995), dephytinization of rapeseed protein concentrate by commercial pretreatment with microbial phytase was effective in significantly improving the nutritive value of RPC for trout. This suggests that the dephytinization procedure employed by Aiko Biotechonology Ltd., for this investigation was a superior method to that used by the same company for the previous study by Teskeredzic et al. (1995). However, although commercial dephytinization of the RPC improved the feeding value of RPC for rainbow trout, yielding better fish performance and diet digestibility than any of the other rapeseed protein concentrate diets, the performance of the trout was generally inferior to that obtained for the control (NFM) treatment. 188 The DRPC diet, had a glucosinolate content that was lower than the levels determined for the canola meal or other RPC diets, and accordingly the fish fed diet DRPC had comparable plasma titres of T 3 and T 4 and liver weights to respective values noted in the control (NFM) fish. However, hepatic 5'D activity levels of the dephytinized-RPC treatment group (DRPC) were almost three times the levels found in the control (NFM) group (6.72 ± 1.068 vs 1.54 ± 0.135 ng/ml, respectively). Thyroid function and particularly 5'D activity in trout are known to be stimulated by caloric intake, protein intake, and increased somatic growth (Eales and MacLatchy, 1989; Riley et al., 1993). However, 5'D activity is not always positively correlated with somatic growth and may be elevated under conditions of compromised growth (Riley et al., 1993). Additionally, it has previously been demonstrated with salmonids that below maximum tolerable levels for thyrotoxic compounds (e.g. glucosinolates and/or their products of hydrolysis), the thyroid may compensate and counter suppression of function in order to maintain normal growth and normal levels of thyroid hormones (Yurkowski et al., 1978; Higgs et al., 1979; 1982; Hardy and Sullivan, 1983). Compensatory thyroidal response may be manifested as elevated thyroid hormones and/or 5'D activity, indicating increased conversion of T 4 to the highly active T 3 , although increased hepatic generation of T 3 is not always associated with an increase in plasma T 3 as elevated uptake of T 3 in tissues may lead to increased plasma T 3 clearance (Eales et al., 1992). Consequently, the control-level plasma titres of T 3 and T 4 , as well as the significantly elevated 5'D activity levels of the DRPC treatment group, appear to be consistent with compensatory thyroidal response to dietary glucosinolate content and/or possibly other unknown factors in the DRPC diet. Although it has been demonstrated previously that dietary inclusion of rapeseed protein concentrate at a level of 59% of the dietary protein does not significantly alter whole carcass composition of rainbow trout (see experiment 1), carcass composition, particularly mineral retention, was significantly affected when rapeseed protein concentrate was included at a level of 95% of the dietary protein. Relative to the levels found in the control fish (NFM), Mg, Ca and P were significantly reduced in fish given the high phytase supplemented RPC treatment, but not in those given the unsupplemented RPC treatment (URPC) or other RPC diets. The poorer retention of these minerals, and consequently poorer bioavailability appears to reflect the overall poor performance of fish in the URPCHP treatment group. However, one 189 might have expected the undephytinized, unsupplemented RPC treatment group, in the absence of phytase, to show poorer mineral retention, as previous research has found that dephytinization or supplementation of plant-protein diets with microbial phytase may improve the bioavailability of minerals (Cromwell et al., 1993; Pallauf etal., 1992; Riche, 1993). The poorer retention of minerals in the fish fed high phytase RPC diet (URPCHP) relative to those fed the phytase-free, URPC diet is a surprising finding. This observation appears to be even more unusual given that a high dose of phytase to the fish meal diet (NFMHP) significantly increased the carcass retention of Mg above that of the control fish, suggesting successful in vivo dephytinization of the phytic acid contributed by the wheat in the diet, and consequently a reduced complexing of cationic minerals and proteins with phytic acid. Another surprising and somewhat anomalous finding was that carcasses from the fish meal free rapeseed protein concentrate treatment groups generally had significantly higher zinc contents than those of the control fish. This is quite unexpected as it is well established that the dietary presence of phytic acid is strongly associated with induced deficiency or reduced bioavailability of zinc in trout (Spinelli et al., 1983) and salmon (Richardson et al., 1985; NRC, 1993). Given that the rapeseed protein concentrate diets contained substantially more magnesium than other experimental diets and magnesium is recognized to accentuate phytate's adverse effects on zinc bioavailability (Forbes et al., 1984; Reddy et al., 1989), this observation appears even more unusual. No explanation for this anomaly may be offered at this time. Whole carcass composition was unaffected by the inclusion of canola meal protein at a 20% dietary protein inclusion level. 5.5.3 Influence of diet treatment on general fish health and behaviour No variation in fish health or behaviour was found between treatments except where noted above for differences among groups in initial feeding behaviour. 190 5.6 Conclusion The findings from this investigation indicate that the feeding value of canola meal for rainbow trout may be significantly enhanced by commercial dephytinization of canola meal by pretreatment with microbial phytase, and by microbial phytase application directly to the pelleted diet, at a level of 1,000 PU/g dry diet or 4,000 PU/g dry diet. However, the results indicate that the greatest nutritional improvement was given by supplementation of the diet with a low dose of phytase (1,000 PU/ g diet), which supported better fish performance than the control high quality fish meal diet. Commercial dephytinization of canola meal prior to dietary incorporation reduced the digestible energy level of the diet but otherwise supported fish performance and diet digestibility values equal to those of the best performing low phytase, canola meal treatment group. At an inclusion level of 20% of the dietary protein, fibre reduction was ineffectual in improving the feeding value of canola meal for rainbow trout, although microbial phytase supplementation did slightly enhance the nutritional value of the low fibre canola meal diet. The supplementation of rapeseed protein concentrate-based diets with microbial phytase proved ineffectual in improving the feeding value of the diets. In fact, microbial phytase addition directly to the diet adversely affected the feeding value of the undephytinized rapeseed protein concentrate for rainbow trout. Reduced palatability of the RPC diets relative to all other diets was apparent, and suggests that FINNSTIM at 1.5% diet DMB was not effective overcoming the poor palatability of these diets. It is unclear whether the top dressing of the phytase itself was responsible for the reduced diet palatability or whether some unknown factor or interaction was the cause. Commercial dephytinization of rapeseed protein concentrate by pretreatment with microbial phytase was effective in improving the nutritional value of rapeseed protein concentrate for rainbow trout. Although it was shown in experiment that at an inclusion level of 59% of the dietary protein commercially dephytinized rapeseed protein concentrate can successfully replace the entire fish meal component of a rainbow trout diet without compromising fish performance, several fish performance parameters and diet energy digestibility were compromised with the dietary inclusion of dephytinized rapeseed protein concentrate at a level of 95% of the dietary protein in animal protein-free diets. It is possible that the initial 191 poor feeding responses of the fish fed diet DRPC contributed to this fmding and if the experiment had been of longer duration, then less difference in the performance of the fish fed diets NFM and DRPC would have been noted. The results of this investigation indicate that phytic acid reduction is a viable means of enhancing the nutritional value of canola or rapeseed protein concentrate for rainbow trout. Also, they indicate that the application of microbial phytase (1,000 PU/g DMB) directly to the finished diet is a more effective, less costly, more practical means of improving the nutritional value of commercial canola meal through dephytinization than is commercial dephytinization by phytase pretreatment of commercial canola meal. However, reduced palatability of diets suggests that this mode of phytase application, although practical, is not optimal. Although diet digestibility was reduced when unsupplemented commercial canola meal was used in place of some the fish meal in the diet (20 % of the dietary protein), the results indicate that fish weight gain, growth rate, feed intake, appetite, feed efficiency, protein efficiency, protein deposition and hepatic and thyroidal responses were equal to those obtained for the fish meal control treatment. The finding has important cost implications for trout culture. For example, given that the cost of commercial canola meal represents 56 % of the cost of premium quality fish meal, on a per kilogram protein basis. The addition of the Finase™ (Aiko Biotechnology) phytase preparation to the diet at a supplemental dose of lOOOPU/g dry diet however, does not appear to be cost effective at this time since this strategy would add 40.7 0/kg dry diet (Finase™ currently costs approximately $203.42 (Cdn)/kg; activity is 500,000 PU/g). • The results also suggest that whereas phytic acid reduction clearly enhances the feeding value of rapeseed protein concentrate for rainbow trout, direct application of microbial phytase to the rapeseed protein concentrate finished diets may not be a practical means of dephytinization. Furthermore, it may be said that while commercial phytase pretreatment of RPC improves its nutritional value, the extent of nutritional enhancement conferred by the methods used in this investigation were not sufficient to enable complete replacement of all animal protein (95 % dietary protein), in a trout diet with dephytinized RPC, without compromising diet digestibility or growth performance. 192 5.7 Acknowledgements This study was supported, administered and funded by NRC/IRAP, Aiko Ltd., Biotechnology, Finland, Moore Clark Co. (Canada) Inc., in collaboration with the Department of Fisheries and Oceans of Canada, West Vancouver Laboratory. Thanks are extended to representatives of NRC/IRAP (Ms. Yvonne Jones, Dr. Warren Ngata and Dr. John Howard), representatives of Moore Clark (Mr. Greg Deacon and Ms. Jill Mills), and representatives of Aiko Ltd., Biotechnology, Finland (Mr.Timo Vaara, Ms. Marja Turuner and Mr. Dan Haglund). For their analyses of samples, their interest and their proferred advice thanks are also extended to Dr. R. Hardy and Dr. K. Shearer of Northwest Fisheries Centre, Seattle Wa., U.S.A., Dr. G. Eales of the Department of Zoology, University of Winnipeg, Manitoba, and Dr. P. Reaney of Agriculture Canada Research Station, Saskatoon, Saskatchewan. Additionally, Dr. J. Bell of the Department of Animal and Poultry Science, University of Saskatchewan, is thanked for generously supplying the fibre-reduced canola meal and Dr. S. Satoh of the Department of Aquatic Biosciences, Tokyo University of Fisheries, Tokyo, Japan is thanked for his assistance in sampling. 193 CHAPTER SIX 6.0 Summary, conclusions and recommendations Investigations in this thesis were used to assess the effects of phytic acid removal on the nutritional value and acceptability of canola meal and rapeseed protein concentrate (RPC) as partial and total replacements for fish meal in rainbow trout diets. The first investigation examined whether the nutritional value of fish meal-free diets containing RPC at a dietary inclusion level of 59% of the dietary protein could be enhanced by commercial phytase pretreatment, with or without the supplementation of essential amino acids and/or cation-anion balancing of these diets to mimic the essential amino acid profile of a control diet based on high quality B.C. whole herring meal which provided 59% of the dietary protein. The second investigation was used to examine whether commercial dephytinization by microbial phytase pretreatment or in vivo dephytinization through the ingestion of diets supplemented with microbial phytase (1,000 PU/g dry diet and 4,000 PU/g dry diet) could improve the nutritional value of canola meal (20 % of the dietary protein) or RPC (95 % of the dietary protein) for rainbow trout. Additionally, the effect of fibre reduction on the nutritional value of canola meal was assessed. In the absence of phytase pretreatment or supplementation, fish performance and/or diet digestibility were reduced when canola meal replaced some (20 % of the total dietary protein) of the fish meal, and rapeseed protein concentrate replaced all (59 % of the total dietary protein in the first investigation and 95 % of the total dietary protein in the second investigation) of the fish meal in rainbow trout diets. Fish performance and diet digestibility were significantly improved with phytic acid reduction. However, dephytinization procedures were not equally effective. Diets containing canola meal (20 % of the dietary protein) which had been commercially pretreated with microbial phytase had lower digestible energy concentrations than the control diets, but values for protein digestibility, protein efficiency, feed efficiency, protein deposition and growth rate were comparable to control values. Weight gain was ~ 13 % greater and feed consumption was « 16 % greater 194 than control values. Similarly, fish performance and diet digestibility were significantly better when the diets contained canola meal to which microbial phytase was supplemented (1,000 PU/g dry diet and 4,000 PU/g dry diet), than they were when unsupplemented, phytase-free canola meal was included. With the supplementation of 4,000 PU/g dry diet, diet digestibility, fish performance and feed utilization did not significantly differ from values obtained with the control diet. The best overall performance was given by the canola meal treatment for which a low dose of phytase (1,000 PU/g dry diet) was supplemented, wherein all fish performance, feed utilization and diet digestibility parameters were found to be equivalent to those of fish on the control diet, except weight gain and feed intake were - 2 1 % greater than control values. Fibre reduction was of little nutritional consequence when canola meal was included at a level of 20 % of the dietary protein in trout diets. At an inclusion level of 59% of the dietary protein, commercially dephytinized RPC successfully replaced all the fish meal in a trout diet without compromising fish performance. However, commercially dephytinized RPC could not entirely replace the fish meal component in diets in which fish meal furnished 95% of the dietary protein and was the only source of animal protein in the diet. In this situation, significant reductions in the digestible energy content of the diet, and fish performance were noted. The latter was manifested by significantly poorer weight gains, growth rates, feed intakes, feed efficiency, protein efficiency, protein deposition and digestible energy values than noted for fish ingesting the fish meal control diet, whether or not the diets were supplemented with microbial phytase (1,000 PU/ g dry diet or 4,000 PU/ g dry diet). The ingestion of phytase-coated pellets did not appear to facilitate sufficient in vivo dephytinization of the diet to support fish performance or diet digestibility equal to that of control groups. The findings from both investigations suggest that in the presence of supplemental inorganic phosphorus and FINNSTIM™ (1.5 % dry diet) as a palatability enhancer, commercial canola meal may be used as a less expensive, partial replacement (20% of the dietary protein) for expensive fish meal in a rainbow trout diet without compromising growth, feed efficiency, protein efficiency, protein deposition, carcass mineral composition or thyroidal and hepatic responses. For an additional cost, weight gain may 195 be significantly improved with the addition of a low dose of phytase (1,000 PU/g dry diet), but the additional cost associated with the use of a higher dose (4,000 PU/g dry diet) of phytase, whether aclministered directly or commercially applied as a pretreatment to the feedstuff, is unwarranted. Fibre reduction also appears to be an unneccessary procedural expense when canola meal is included at a level of 20% of the dietary protein in a rainbow trout diet. The findings from both investigations also suggest that rapeseed protein concentrate has an amino acid profile which is optimal for rainbow trout growth, and if the concentrate is commercially dephytinized by pretreatment with microbial phytase, it may be used in the presence of supplemental inorganic P and FINNSTIM™ (1.5 % dry diet), to replace all of the fish meal (59% of the dietary protein), without compromising fish performance or carcass mineral composition. This is believed to be the first occassion in which a plant source of protein has successfully replaced the entire fish meal component of a salmonid diet. Based on the results of both investigations of this thesis it may be concluded that phytic acid removal enhances the nutritive value or rapeseed and canola feedstuffs, and this strategy improves their acceptability as replacements for fish meal protein in rainbow trout diets. However, before the replacement of fish meal with rapeseed and canola protein feedstuffs can become an accepted practice, additional research concerning the means and effects of phytic acid removal from rapeseed and canola feedstuffs deserve further consideration. It has been shown that nutritional improvements in the feeding value of phytic-acid-reduced soybean meal for pigs persists even in the absence of inorganic, supplemental phosphorous (Khan and Cole, 1995). However, whether the beneficial effects of phytic acid removal on the nutritive value of canola and rapeseed protein feedstuffs for salmonids might persist in the absence of inorganic, supplemental phosphorus is not known, as this was not evaluated in the investigations of this thesis. Consequently, future research in this area is needed. In the interest of identifying practical means of phytic acid removal, future research should examine the possible reasons for the failure of direct microbial phytase addition to effectively promote in vivo dephytinization of RPC (perhaps different application methods, phytase doses and/or levels of dietary FINNSTIM™ might prove beneficial). 196 Future research should also be directed at determining to what extent dephytinization will affect the storability of canola and rapeseed protein products, as phytic acid is well recognized to have preservative qualities, involving strong chelating and/or antioxidant properties which discourage oxidative damage and lipid peroxidation. Consequently, the antioxidant properties of phytic acid are believed to be primarily responsible for the longevity and retained nutritional value of seeds (Ohlrogge and Kernan, 1982) and plant-derived feedstuffs. The preservative properties of phytic acid are thought to be largely responsible for the observed stability and nutritive integrity of the rapeseed protein concentrate used in the investigations of this thesis, wherein the concentrate was already 19 years old at the time it was employed in these studies. Originally prepared in January of 1974, proximate analyses of the concentrate conducted at that time determined that dry matter content was 96.4 %, crude protein was 65.4% (DMB), lipid was 8.3 % (DMB), and ash was 6.1% (DMB). After 19 years of unrefrigerated storage under conditions of low humidity, proximate determinations conducted immediately before the RPC was incorporated into the diets that were prepared for this thesis found dry matter to be 91.2 %, crude protein to be 65.4 % (DMB), lipid to be 10.0 % (DMB) and ash content to be 6.7 % (DMB). Considering that the proximate analyses were conducted 19 years apart, by different labs, the apparent stability of the nutritive integrity of the protein product is remarkable. The replacement of fish meal with rapeseed and canola protein products provides an opportunity to extend the use of domestically produced feedstuffs, financially benefitting the Canola industry by creating a larger value-added market for rapeseed and canola seeds, for which oil content currently accounts for approximately 80% of agronomic value. Additionally, the successful replacement of fish meal in salmonid diets may also provide a means of addressing the economic as well as the aesthetic or environmental issues often associated with the intensive culture of trout and salmon. Widespread incorporation of substantial proportions of canola meal and rapeseed protein concentrate in salmonid feeds and potentially in the feeds of other aquatic or carnivorous species, has the potential to greatly increase the income of the rapeseed and canola crushing industry, for which the non-oil portion of the seeds have an agronomic value of approximately 20%. Further economic benefits might also be realized if it were possible to create a value-added market with some of the non-oil and non-protein 197 constituents or rapeseed and canola seeds, namely phenolics and phytic acid. The economic benefits to the trout and salmon farming industry which may result from the replacement of fish meal with canola and rapeseed protein products is centered around reduced feed costs as a consequence of reduced supplementation of diets with minerals, (e.g. phosphorous), and/or reduced use of expensive fish meal. Phosphorus is an expensive component in pig diets (Pontillart 1991), and its supplementation to poultry and salmonid diets is also recognized to substantially add to the cost of diets. In 1989 it was estimated that if in the the poultry industry, the phosphorous that is complexed with phytic acid in soybean meal (which is less than that potentially available in canola meal), which is fed to chicks could be made biologically available to the animal, there would be a cost saving of $200 x 10 6 (U.S.), associated with the 600-700 billion chicks raised in the U.S.A. per year (Han, 1989). Should the phosphorus content of canola and rapeseed protein products be rendered biologically available through dephytinization, similar cost benefits would undoubtedly be associated with the use of dephytinized canola meal and RPC for poultry, salmonids or other animals, potentially reducing the cost of dietary phosphorus supplementation. Improved bioavailability of other minerals as a result of dephytinization might also serve to decrease the associated cost of mineral supplementation. The high cost of salmonid production is largely associated with the high cost of feed for which fish meal is the primary protein source. On a per kilogram protein basis, fish meal is expensive relative to plant protein sources such as canola meal or rapeseed protein concentrate. Consequently, the replacement of fish meal with a less expensive protein source would reduce the high cost of feed, and increase the margin of profitability in salmon farming. Currently the cost of canola meal is approximately 46-50% of the cost of high quality fish meal, and although RPC or canola protein concentrate (CPC) are not currently produced commercially, the projected cost including shipping and FINNSTIM™ supplementation has been estimated to be 80 % of the cost of premium fish meal. However, fish meal cost is expected to rise as fish meal production is in a decline and is predicted to decrease in quantity and quality. Nonetheless, regardless of possible positive economical benefits to the salmonid culture industry, the replacement of fish meal with rapeseed and canola protein feedstuffs is viewed by many as a worthwhile pursuit simply as a means of 198 addressing what may be deemed as a negative aesthetic or moral practice of feeding high quality protein to fish rather than to humans. Based on the fact that approximately 5 kg of raw fish is required for the production of 1 kg of fish meal, it has been estimated that 3 kg of raw fish is required to produce 1 kg of weight gain of intensively farmed carnivorous fish or shrimp (New and Wijkstrom, 1990). Consequently, because salmon are traditionally harvested at 2-4 kg (Libey and Bosworth, 1990), 6-12 kg of raw fish may ultimately be required for the production of one farmed salmon. It is not difficult to understand why some may view the practice of feeding good quality protein to fish as objectionable. The replacement of fish meal with canola and rapeseed protein products may also be beneficial for the state of the environment, wherein the reduced supplementation of dietary minerals such as phosphorus to feed, and the increased utilization of intrinsic minerals as a result of dephytinization may lead to reduced pollution and eutrophication of rearing water. That is, improved phosphorus retention and absorption may serve to reduce the phosphorus discharged into the environment from poorly utilized feed. Given that it has been estimated that fish such as Atlantic salmon retain only about 14-22% of the dietary phosphorous found in typical commercial feeds, and thus discharge approximately 80% into the the sun-ounding water (Ketola and Harland, 1993), any improvement in phosphorus retention would prove beneficial. There are many potential benefits associated with the replacement offish meal with plant protein feedstuffs. Prior to this study, the complete replacement of fish meal in a salmonid diet had not yet been achieved without compromising growth performance. Through the reduction of phytic acid content and its suppressive effects on the protein and general nutritional quality of rapeseed and canola protein products, it has now been demonstrated that fish meal may be partially replaced at a level of 20% of the dietary protein with canola meal and completely replaced at a level of 59% of the dietary protein with dephytinized rapeseed protein concentrate without compromising growth performance of rainbow trout. 199 BIBLIOGRAPHY Abdellatif, A .M.M. and Vies, R.O., (1970) Physiopathological effects of rapeseed oil and canbra oil in rats. In: Proceedings international conference on the science, technology and marketing of rapeseed and rapeseed products. Rapeseed Association of Canada, Ste-Adele, p. 423-434 Adron, J.W. and Mackie, A.M. , (1978) Studies on the chemical nature of feeding stimulants for rainbow trout, Salmo gairdnen'Richardson. J. Fish Biol. 12: 303-310 Alisone, R.M. (1971). In: N.W. Pirie (ed). Leaf protein. Blackwell Scientific Publications, Oxford, p. 78 Alii , I. and Houde, R., (1986) Characterization of phytate in canola. In: Research on canola seed, oil, meal and meals fractions. Canola Council of Canada, Winnipeg, p. 159-165 Aman, P. (1970) Polysaccharides and lignin in rapeseed. In: Proceedings international conference on the science, technology and marketing of rapeseed and rapeseed products. Ste- Adele , Quebec. Sept 20-23. Rapeseed Association of Canada in co-operation with the Department of Industry, Trade and Commerce, Ottawa p. 38-44 Anjou, K., Krook, C.G., Fecski, A.J., Ohlson, R. (1976) Can. Patent 999,186 AOAC, (1990) Vol.1. Official methods of analysis of the association of official analytical chemists. 15th ed. Association of Official Analytical Chemist Inc. p 96 Appelqvist, L.A. (1971) Composition of crucifereous oil crops. J. Am. Oil Chem. Soc. 48: 851 Appelqvist, L.A. (1972) Chemical constituents of rapeseed. In: L.A. Appelqvist and Ohlson, R. (eds.) Rapeseed. Elsevier Publ., Amsterdam p 168-180 Aref, M.M. , (1970) Refined protein products from rapeseed. In: Proceedings: International conference on the science, technology and marketing of rapeseed and rapeseed products. Rapeseed Association of Canada, Winnipeg, p. 525-538 Asgard,T. (1988). Nutritional value of animal protein sources for salmonids. In: Proceedings of Aquaculture International Congress, British Columbia Pavilion Corp., Vancouver. P. 411-418 Atkinson, M.R. and Morton, R.K. (1960) Free energy and the biosynthesis of phosphates. In: M . Florkin, and H.S. Mason, (eds.). Comparative Biochemistry . Academic Press, New York, vol. 2:1 Barton, B.A., Peter, R.E., Paulencu, C. (1980) Plasma-cortisol levels of fmgerling rainbow trout (Salmo-gairdneri) at rest, and subjected to handling, confinement, transport and stocking. Can. J. Fish. Aquat. Sci. 37: 805-811 Bayley, H.S. and Thompson, R.G., (1969) Phosphorus requirements of growing pigs and effect of steam pelleting on phosphorus availability. J. Anim. Sci. 28: 484 Beach, D.H.C., (1983) Rapeseed crushing and extraction. In: J.K. Kramer, F.D. Sauer and W.J. Pigden (eds.). High and low erucic rapeseed oils: production, usage, chemistry, and toxicological evaluation. Academic Press, Toronto, p. 181-195 200 Beare-Rogers, J.L., (1970) Nutritional aspects of long-chain fatty acids. In: Proceedings: International conference on the science, technology and marketing of rapeseed and rapeseed products. Ste. Adele, Quebec Sept. 20-23, (1970) Rapeseed Association of Canada, Winnipeg, p. 450-465 Bechyne, M . , Kondra, Z.P. (1970) Effect of seed pod location on fatty acid composition of seed oil from rapeseed (Brassica napus and Brassica campestris). Can. J. Plant Sci. 50: 151 Bell, J.M. (1984) Rapeseed meal and rapeseed by-product evaluation and utilization. Can. J. Anim. Sci. 64: 194-195 Bell, J.M., (1990) Canola meal and animal nutrition. Proceedings: International canola conference (2-6 April): 67-96 Bell, J.M., (1993) Factors affecting the nutritional value of canola meal: A review. Can. J. Anim. Sci., 73: 679-697 Bell, J.M. and Keith, M.O. (1989). Factors affecting the digestibility by pigs of energy and protein in wheat, barley and sorghum diets supplemented with canola meal. Anim. Feed Sci. Technol. 24: 253-265 Bell, J.M. and Shires, A., (1982) Composition and digestibility by pigs of hull fractions from rapeseed cultivars with yellow or brown seed coats. Can. J. Anim. Sci., 62: 557-565 Bengtsson, L., Hofstein, A.V. and Loof, B., (1972) Botany of rapeseed. In: L A . Appelqvist and R. Ohlson (eds.). Rapeseed cultivation, composition, processing and utilization. Elsevier, Amsterdam, p. 36-44 Bergot, F. 1979. Carbohydrates in rainbow trout diets: effects of the level and source of carbohydrates and the number of meals on growth and body composition. Aquaculture 18: 157-167 Bergot, F. and Breque, J. (1983) Digestibility of starch by rainbow trout: effects of the physical state of starch and of the intake level. Aquaculture 22:81-96 Besecker, R.J.Jr., Plumlee, M.P., Pickett, R.A., Conrad, J.H., 1967. Phosphorus from barley grain for growing swine. J. Anim. Sci., 26: 1477 Beudeker, R.F., Geerse, C. and Verschoor, G.J., (1990) Biotechnological products for the compound feed industry. In: Biotechnology International. Century Press Ltd., London, p. 310-313. Bhatty, R.S. and Sosulski, F.W., (1972) Diffusion extraction of rapeseed glucosinolates with ethanolic sodium hydroxide. J. Am. Oil Chem. Soc, 49: 346 Bimbo, A.P., (1990) Fish Meal and Oil. In: R E . Martin and G.J. Flick (eds.). The seafood industry. Van Nostrand Reinhold, New York, p. 325-350 Blair, R., Reichert, R.D. (1984) Carbohydrates and phenolic constituents in a comprehensive range of rapeseed and canola fraction: nutritional significance for animals. J. Sci. Food Agic. 35: 29-35 Blair, R., Robblee, A.R., Dewar, W.A., Bolton, W. and Overfield, N.D., (1975) Influence of dietary rapeseed meals and selenium on egg production and egg tainting in laying hens. J. Sci. Food Agric, 26: 311-318 Bligh, E.G. and Dyer, W.A., (1959) A rapid method of total lipid extraction and purification. Can. J. Biochem. Physiol., 37: 911-917 201 Block, E., (1992) Dietary cation-anion balance and its effect on the performance of ruminants. 121-137 Boulter, G.S., (1983) The history and marketing of rapeseed oil in Canada. In: . J.K. Kramer, F.D. Sauer and W.J. Pigden (eds.). High and low erucic rapeseed oils: production, usage, chemistry, and toxicological evaluation. Academic Press, Toronto, p. 61-83 Bres, M. , (1989) The effects of prey relative abundance and chemical cues on prey selection in rainbow trout. J. Fish Biol., 35: 1803-1811 Brooks, J.R., and Morr, C V . (1982) Phytate removal from soybean protein isolates using ion-exchange processing treatments. J. Food Sci. 47: 1280-1282 Brown, E.C., Heit, M.L and Ryan, D.E. (1961) Phytic acid: an analytical investigation. Can. J. Chem. 39: 1290 Bruce, J.A.M. and Sundstol, F. (1994) The effect of microbial phytase in diets for pigs on apparent ileal and faecal digestibility, pH and flow of digesta measurements in growing pigs fed a high-fibre diet. Can. J. Anim. Sci. 75: 121-127 Byrd, C A . and Matrone, G. (1965). Investigations of chemical basis of zinc-calcium-phytate interaction in biological systems. Pro. Soc. Exp. Biol. Med. 119: 347 Cain, K. D. and Garling, D.L. (1995) Pretreatment of soybean meal with phytase for salmonid diets to reduce phosphorus concentrations in hatchery effluents. The Prog. Fish Cult. (57): 114-119 Campbell, G.L. and Bedford, M.R., (1992) Enzyme applications for monogastric feeds: A review. Can. J. Anim. Sci., 72: 449-466 Campbell, L.D. and Slominski, B.A., (1991) Causes of liver hemorrhage in laying hens emphasizing indole glucosinolates. In: 9th project report. Canola Council of Canada, Winnipeg, p. 33-56 Canola Council of Canada, (1994) Canada's canola. Canola council of Canada, 19 p. Carr, E.S., (1982) Chemical stimulation of feeding behavior. In: T.J. Hara (ed.). Chemoreception in fishes: developments in aquaculture and fisheries science, Vol. 8. Elsevier Scientific Publishing Company, Amsterdam, Oxford, New York, p. 259-273 Chamberlain, G.W., (1993) Aquaculture trends and feed projections. World Aquaculture, 24: 19-34 Cheeke, P.R. and Shull, L.R., (1985) Natural toxicants in feeds and poisonous plants. The AVI Publishing company, Inc., 492 p. Chen, B.H.Y. and Morr, C V . (1985) Solubility and foaming properties of phytate-reduced soy protein isolate. J. Food. Sci. 50: 1139 Cheryan,M. (1980) Phytic acid interactions in food systems. CRC Critical Review of Food Science and Nutrition. 13: 297-335 Chiu, Y.N. , Austic, R.E. and Rumsey, G.L., (1984) Effect of dietary electrolyte and histidine on histidine metabolism and acid-base balance in rainbow trout (Salmo gairdneri). Comp. Biochem. Physiol., 78: 777-783 202 Chiu, Y.N. , Austic, R.E. and Rumsey, G.L., (1987) Interactions among dietary minerals, arginine and lysine in rainbow trout (Salmo gairdneri). Fish Physiol. Biochem. 4: 45-55 Cho, C.Y., (1990) Fish nutrition, feeds, and feeding: with special emphasis on salmonid aquaculture. Food Reviews International, 6: 333-357 Cho, C.Y. and Slinger, S.J., (1979) Aparent digestibility measurement in feedstuffs for rainbow trout. In: J.E. Halver and K. Tiews (eds.), Finfish nutrition and fishfeed technology, Vol. II. Heenemann Verlagsgesellschaft mbH„ Berlin, p. 239-267 Church, D.C., Pond, W.G. (1974) Basic animal nutrition and feeding . (Third ed.) John Wiley & Sons, New York. p . 123-187 Clarke, W.C, Virtenen, E., Blackburn, J. and Higgs, D.A., (1994) Effects of a dietary betaine/amino acid additive on growth and seawater adaptation in yearling chinook salmon. Aquaculture, 121: 137-145 Classen , H.L., and Bedford, M.R. (1991) The use of enzymes to improve the nutritive value of poultry feeds In: W.Haresign and D.J.A. Cole (eds.) Recent advances in animal nutrition.. Butterworth, Heinemann Ltd., Berlin, p. 95-116 Cooper-Driver, G.A., (1983) Chemical substances in plants toxic to animals. In: M . Rechcigl Jr. (ed), Handbook of naturally occurring food toxicants. CRC Press, Inc., Boca Raton, Florida, p. 213-240 Cosgrove, D.J., (1980) Inositol phosphates: their chemistry, biochemistry and physiology. Elsevier, 85 p. Costello, A.J.R., Glonek, T., and Myers, T.C. (1976) 31P-nuclear magnetic resonance-pH titrations of myo-inositol hexaphosphate. Carbohydr. Res. 46: 159 Crimson, R.E., Graham, W.C. and Leshures, D.K., (1991) Thyroid hormone and organoleptic factor supplementation of canola meal diets for rainbow trout. In: 9th project report Canola: research on canola seed, oil and meal. Canola Council of Canada, Winnipeg, p. 289-298 Cromwell, G.L., Stanly, T.S., and Randolph, J.H., (1992) Effectts of phytase on the utilization of phosphorus in com-soybean meal diets by growing finishing pigs. J. Anim. Sci. 69 (suppl.) p. 358 Cromwell, G.L., Stahly, T.S., Coffey, R.D., Monegue, H.J. and Randolph, J.H., (1993) Efficacy of phytase in improving the bioavailability of phosphorus in soybean meal and com-soybean meal diets for pigs., J. Anim. Sci.: 1831 Cyr, D.G. and Eales, J.G., (1988) The influence of thyroidal status on ovarian function in rainbow trout Salmo gairdneri. J. Exp. Zool., 248: 81-87 Dabrowski, K. and Sosulski, F., (1983) Extraction of phenolic compounds from canola during protein concentration and isolation. In: Proceedings of the sixth international rapeseed conference, Vol. II., Paris, p. 338-3432 Dabrowski, K., Poczycznski, P., Kock, G. and Berger, B., (1989) Effect of partially or totally replacing fish meal protein by soybean meal protein on growth, food utilization and proteolytic enzyme activities in rainbow trout (Salmo gairdneri) New in vivo test for exocrine pancreatic secretion. Aquaculture, 77: 29-49 Darroch, C S . and Bell, J.M. (1991) Goitrogenicity of indoylglucosinolates. In: 9 t h project report: research on canola seed, oil and meal. Canola Council of Canada, Winnipeg, p. 356-361 203 Daun, J.K., (1983) The introduction of low erucic acid rapeseed varieties into C a n a d i a n production. In: J.K. Kramer, F.D. Sauer and W J . Pigden (eds.). High and low erucic rpaeseed oils: production, usage, chemistry, and toxicological evaluation. Academic Press, Toronto, p. 161-180 Daun, J.K., Buhr, N. , Diosady, L.L., Mills, J.T. and Mag, T. (1993)a. Oilseeds-processing. In: Grains and oilseeds: handling, marketing, processing, 4 t h ed., Vol II Canadian International Grains Institute, Winnipeg. P. 883-935 Daun, J.K., Kenaschuk, E.D., Deido, W., Gubbels, G., Downey, R.K., Scarth, R., Woods, D., Buzzell, R.I., Adolphe, D. and Hume, D., (1993)b. Oilseeds-production. In: Grains and Oilseeds: Handling, marketing, processing, 4th ed., Vol. II. Canadian International Grains Institute, Winnipeg, p. 831-881 Daun, J.K. and McGregor, DI (1981) Glucosinolate analysis of rapeseed (canola). Method of the Canadian Grain Commission. Revised edition. Grain Research Laboratory, Canadian Grain Commission. Winnipeg, Davies, N.T., Carswell, A. J. P., and Mills, C F . (1985) The effect of variation in dietary C intake on the phytate-Zn interaction in rats. In: C F . Mills, I. Bremmer, and J.K. Cheters (eds.). Trace elements in man and animals. Commonwealth Agricultural Bureaux, Slough,, p 456-457 Davies, N.T. and Nightingale, R. (1975) The effects of phytate on intestinal absorption and secretion of zinc, and whole-body retention of zinc, copper, iron and manganese in rats. Br. J. Nutr. 34: 243-258 Davies, N.T. and Olpin, S.E. (1979) Studies on the phytate: Zn molar contents in diets as a determinant on Zn bioavailability to young rats. Br. J.Nutr. 41: 590-603. De, B.P. and Biswas, B.B., (1979) Evidence for the existence of a novel enzyme system: myoinositol-1-phosphate dehydrogenase in Phasolusaurus. J. Biol. Chem., 254: 8717 de Rham, O. and Jost, T., (1979) Phytate-protein interactions in soybean extracts and low-phytate soy protein products. J. Food Sci., 44: 596-600 Deshpande, S.S., Sathe, S.K. and Salunkhe, D.K., (1984) Chemistry and safety of plant polyphenols. In: M . Friedman (ed.). Nutritional and toxicological aspects of food safety. Plenum Press, New York, p. 457-495 Diosady, L.L., Rubin, L.J. and Tzeng, Y .M. , (1991) Canola meal upgrading by membrane processes. In: 9th project report, Canola: research on canola seed, oil and meal. Canola Council of Canada, Winnipeg, p. 23-31 Dishington, I.W., (1975) Prevention of milk fever by dietary salt supplements. Acta Veterinaria Scandinavica, 16: 503-512 Donaldson, E.M., Fagerlund, U.H.M., Higgs, D.A. and McBride, J.R., (1979) Hormonal enhancement of growth. In: W.S. Hoar and D.J. Randall (eds.). Fish Physiology, Vol. VIII. Academic Press, London, p. 456-497. Downey, R.K., (1990) Breeding canola for yield and quality. Proceedings:International canola conference (2-6 April): 41-50. Downey, R.K. and Bell, J.M. (1990) New developments in canola research. In: Canola and Rapeseed: production, chemistry, nutrition and processing technology. Van Nostrand Reinhold, New York, p 37-46 204 Downey, R.K. and Robbelen, G., (1989) Brassica species. In: Oil crops of the world. McGraw-Hill Publishing company, New York, p. 339-362 Durkee, A.B. and Thivierge, P.A., (1975) Bound phenolic acids in Brassica and Sinalpis oilseeds J. Food Sci., 40: 820 Dy Penaflorida, V. (1989) An evaluation of indiginous protein sources as potential component in the diet formulation for Tiger Prawn, Penaeus monodon, using essential amino acid indes (EAAI). Aquaculture 83:319-330 Eales, J.G. (1985) The peripheral metabolism of thyroid hormones and regulation of thyroidal status in poikilotherms. Can. J. Zool. 63: 1217-1231 Eales, J.G. and MacLatchy, D.L., (1989) The relationship between T 3 production and energy balance in salmonids and other teleosts. Fish Physiol. Biochem, 7: 289-293 Eales, J.G., MacLatchy, D.L., Higgs, D.A. and Dosanjh, B.S., (1992) The influence of dietary protein and caloric content on thyroid function and hepatic thyroxine 5'-monodeiodinase activity in rainbow trout, Oncorhynchus mykiss. Can. J. Zool., 70: 1526-1535 Eapen, K.E., Tape, N.W., Sims, R.P.A., (1969) New process for the production of better quality rapeseed oil and meal. II. Detoxification and dehulling of rapeseed-feasibility study. J.Am. Oil Chem. Soc. 46:52-69 El Nochrashy, A.S., (1976) New varieties of rapeseed as a source of proteins. Fette Seifen Anstrichmittel, 77: 451-452 El Nochrashy, A.S., Mukherjee,K.D, Mangold,H. K.. (1977) Rapeseed protein isolates by countercurrent extraction and isoelectric precipitation. J. Agric. Food Chem. 25:193-197 Erdman, J.W., (1979) Oilseed phytates: nutritional implications. J. Am. Oil Chem. Soc, 56: 736-741 Fagerland, U.H.M., Higgs, D.A., McBride, J.R, Archedekin, C , Dosanjh, B.S. and Eales, J.G. (1987) Nutritional value of canola meal protein for juvenile coho salmon (Oncorhynchus kisutch). In: 8th Progress report, research on canola seed, oil, meal and feal fractions, Canola Caouncil of Canada, Publ. p. 5-13 Fenton, T.W. and Fenton, M , (1979) Procedure for the determination of chromic oxide in feed and feces. Can. J. Anim. Sci, 59: 631-634 Flanagan, P.R. (1984) A model to produce pure zinc deficiency in rats and its use to demonstrate that dietary phytate increases the excretion of endogenous zinc. J. Nutr. 114: 493-502 Forbes, R . M , Parker, H.M. and Erdman, J.W. (1984) Effects of dietary phytate, calcium and magnesium levels on zinc bioavailability to rats. J. Nutr. 114: 1421-1425 Fordyce, E.J , Forbes, R . M , Robbins, K.R. , Erdman, Jr., J.W. (1987) Phytate X calcium/zinc molar ratios: Are they predictive of zinc bioavailability? J.Food Sci. 52: 440-444 Fowler, L . G , (1980) Substitution of soybean and cottonseed products for fish meal in diets fed to chinook and coho salmon. Prog. Fish. Cult, 42(2): 87-91 2 0 5 Gifford-Steffen, A.R. and Clydesdale, F.M. (1993) Effect of varying concentrations of phytate, calcium, and zinc of the solubility of protein, calcium, zinc, and phytate in soy protein concentrate. J. Food Protection. 56 (1): 42-46 Gillberg, L. and Tornell, B., (1976) Preparation of rapeseed protein isolates. J. Food Sci., 41: 1063-1069 Goh, Y.K. , Clandinin, D.R., Robblee, A.R. and Darlington, K. (1979) The effect of level of sinapine in laying ration on the incidence of fishy odor in eggs from brown-shelled layers. Can J. Anim. Sci. 53: 1006-1010 Graf, E., (1986) Chemistry and applications of phytic acid: An overview. In: E. Graf (ed.). Phytic acid: Chemistry & applications. Pilatus Press, Minneapolis, p. 1-22 Graf, E. and Eaton, J.W., (1984) Effects of phytate on mineral bioavailability in mice. J. Nutr., 114: 1192-1198 Graf, E. and Eaton, J.W., (1990) Antioxidant functions of phytic acid. Free Radical Biology & Medicine, 8: 61-69 Graf, E., Empson, K. and Eaton, J.W. (1987) PA: a natural antioxidant. J. Biol. Chem. 262: 11,647-11,650 Grice, H C . and Heggtveit, H.A., (1983) The relevance to humans of myocardial lesions induced in rats by marine and rapeseed oils. In: J.K. Kramer, F.D. Sauer and W.J. Pigden (eds.). High and low erucic rapeseed oils: production, usage, chemistry, and toxicological evaluation., Academic Press, Toronto, p. 551-561 Haard, N.F., (1995) Digestibility and in vitro evaluation of plant protein for salmonid feed. In: C. Lim and D.J. Sessa (eds.), Nutrition and utilization technology in aquaculture. AOCS Press, Champaign, II., p. 199-219 Hajen, W.E., (1990) Evaluation of feedstuff digestibility in post-juvenile chinook salmon Oncorhynchus tshawytscha in seawater. Master of Science Thesis, University of British Columbia, Vancouver, p. 196 Hajen, W.E., Beames, R.M., Higgs, D.A. and Dosanjh, B.S., (1993)a. Digestibility of various feedstuffs by post-juvenile chinook salmon (Oncorhynchus tshawytscha) in sea water. 1. Validation of technique. Aquaculture, 112: 321-332. Hajen, W.E., Higgs, D.A., Beames, R.M. and Dosanjh, B. (1993)b. Digestibility of various feedstuffs by post juvenile chinook salmon (Oncorhynchus tshawytscha ) in sea water. 2. Measurement of digestibility. Aquaculture 112: 333-348 Halver, J. E. (1972) Fish nutrition. J.E. Halver (ed.). Academic Press, New York p. Han, Y.W., (1989) Use of microbial phytase in improving the feed quality of soya bean meal. Anim. Feed Sci. Tech., 24: 345-350. Han, Y.W. and Wilfred, G., (1988) Phytate hydrolysis in soybean and cottonseed meals by Aspergillus ficuum phytase. J. Agric. Food Chem., 36: 259-262. Hara, T.J., (1973) Olfactory responses to amino acids in rainbow trout, Salmo gairdneri. Comp. Biochem. Physiol., 302 (2): 316-328 2 0 6 Hara, T.J, (1977) Further studies on the structure-activity relationships of amino acids in fish olfaction. Comp. Biochem. Physiol, 56A: 559-565 Hara, T.J, (1982) Structure-activity relationships of amino acids as olfactory stimuli. In: T.J. Hara (ed.), Chemoreception in fishes: developments in aquaculture and fisheries science, Vol. 8. Elsevier scientific publishing company, Amsterdam, p. 135-157 Hardy, R.W. and Shearer, K.D. (1993) The use of zinc amino acid chelates in high calcium and phosphorus diets of rainbow trout. In: H.D. Asmead (ed.) Role of amino caid chelates in animal nutrition. Noyes Publication, p. 424-439 Hardy, R.W. and Sullivan, C V , (1983) Canola meal in rainbow trout (Salmo gairdneri) production diets. Can. J. Fish. Aquat. Sci, 40: 281-286 Hatje, G , (1989) World importance of oil crops and their products. In: Oil crops of the world. McGraw-Hill Publishing company, New York, p. 1-21 Higgs, D A , Markert, J.R, Macquarrie, D.W, McBride, J.R. and Dosanjh, B.S , (1979) Development of practical dry diets for coho salmon, Oncohynchus kisutch, using poultry-by-product meal, feather meal, soybean meal and rapeseed meal as major protein sources. In: J.E. Halver and K. Tiews (eds.), Finfish nutrition and fishfeed technology, Vol. II. Heenemann Verlagsgesellschaft mbH„ Berlin, p. 191-218. Higgs, D A , McBride, J.R, Markert, J.R, Dosanjh, B.S, Plotnikoff, M.D. and Clarke, W.C , (1982) Evaluation of Tower and Candle rapeseed (canola) meal and Bronowski rapeseed protein concentrate as protein supplements in practical dry diets for juvenile chinook salmon (Oncorhynchus tshawytscha). Aquaculture, 29: 1-31 Higgs, D A , Fagerlund, H . M , McBride, J.R, Plotnikoff, M . D , Dosanjh, B.S, Markert, J.R. and Davidson, J , (1983) Protein quality of altex canola meal for juvenile chnook salmon (Oncorhynchus tshawytscha) considering dietary protein and 3,5,3'-Triiodo-L-thyronine content. Aquaculture, 34: 213-238 Higgs, D . A , McBride, J.R, Dosanjh, B.S, Clarke, W.C, Archedekin, C. and Hammons, A . - M , (1988) Nutritive value of plant protein sources for fish with specieal emphasis on canola products. In: Proceedings of Aquaculture International Congress and Exposition. Aquaculture International Congress (ed.), British Columbia Pavillion Corp, Vancouver, p. 427-435 Higgs, D . A , Dosanjh, B.S, Little, M , Roy, R.J.J, and McBride, J.R, (1989) Potential for including canola products (meal and oil) in diets for Oreochromis mossambicus x O. aureus hybrids. In: Proc. Third Int. Symp. on Feeding and Nutr. in Fish, p. 301-314 Higgs, D . A , McBride, J.R, Dosanjh, B.S. and Fagerlund, U .H.M, (1990) Potential for using canola meal and oil in fish diets. In: Fish physiology, fish toxicology, and fisheries management: proceedings of an international symposium. R.C. Ryans (ed.), U.S. Environmental Protection Agency, Athens, p. 88-107 Higgs, D , Hardy, R , Teskeredzic, Z , Dosanjh, B , Forster, I, McBride, J , Jones, J. and Beames, R , (1991) nutritive value of rapeseed protein concentrate for rainbow trout (Oncorhynchus mykiss). In: Proceedings: GCIRC Eight International Rapeseed Congress, Vol. 5. D.I. McGregor (ed.), The Organizing Committee of The Eighth International Rapeseed Congress, p. 1612-1617 Higgs, D . A , Macdonald, J.S, Levings, C D . and Dosanjh, B.S, (1994)a. Nutritional and feeding habits of Pacific salmon (Oncorhynchus sp) in relation to life history stage. In: Physiological ecology of Pacific salmo. C. Groot, L. Margolis and W.C. Clarke (eds.), UBC press, Vancouver, p. 159-315 207 Higgs, D.A., Prendergast, A.F., Dosanjh, B.S., Beames, R.M., Deacon, G. and Riley, G., (1994)b. Canola protein offers hope for efficient salmon production in finfish diets. In: High performance fish, fish physiology association., : D.D. MacKinlay (ed.), Vancouver, p. 337-382. 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: Nutrition and Utilization Technology in Aquaculture. C. Lim and D.J. Sessa (eds.), AOCS Press, Champaign, p. 130-156 Hilton, J.W. (1989) The interactions of vitamins, minerals and diet composition in the diet of fish. Aquaculture 79: 223-244 Hilton, J.W., (1991) Thyroid hormone and organoleptic factor supplementation of canola meal diets for rainbow trout. In: 9th project report, Canola: research on canola seed, oil and meal. Canola Council of Canada, Winnipeg, p. 289-298 Hilton, J.W. and Atkinson, J.L. (1982) Response of rainbow trout (Salmo gairdneri) to increased levels of available carbohydrate in practical trout diets. Br. J. Nutr. 47: 597-608 Hilton, J. and Slinger, S., (1986) Digestibility and utilization of canola meal in practical-type diets for rainbow trout (Salmo gairdneri). Can. J. Fish. Aquat. Sci., 43: 1149-1155 Hofer, R. and Sturmbauer, C. (1985) Inhibition of trout and carp a-amylase by wheat. Aquaculture 48: 277-283 Hofsten, A.V. (1970) In: Proceedings international conference on the science, technology and marketing of rapeseed and rapeseed products. Ste- Adele , Quebec. Sept 20-23. Rapeseed Association of Canada in co-operation with the Department of Industry, Trade and Commerce, Ottawa, p 70-85 Holmes, M.R.J., (1980) Nutrition of the oilseed rape crop. Applied Science Publishers Ltd. p. 158 Hossain, M.A. and Jauncey, K. (1989) Nutritional evaluation of some Bangladeshi oilseed meals as partial substitutes for fish meal in the diet of common carp. Cyprinus carpio L. Aquacult. Fish Man. 20: 255-268 Hossain, M.A. and Jauncey, K., (1993) The effects of varying dietary phytic acid, calcium and magnesium levels on the nutrition of common carp, Cyprinus carpio. Fish Nutrition in Practice: p.705-715. House, W.A., Welch, R.M. and van Campen, D.R. (1982) Effect of phytic acid on the absorption, distribution and endogenous excretion of zinc in rats. J. Nutr. 112: 941-953 Huisman, J. and Tolman, G.H., (1992) Antinutritional factors In: the plant proteins of diets for non-ruminants. In: P.C. Garnsworthy, W. Havesign and D.J.A. Cole (eds.), Recent Advances In Animal Nutrition. Butterworth-Heinemann Ltd., Berlin, p. 4-19 Hyatt, K.D., (1979) Feeding strategy. In: W.S. Hoar, D.J. Randall and J.R. Brett (eds.), Fish physiology: bioenergetics and growth, Vol. 8. Academic Press, New York, p. 71-119 LAFMM (International Association of Fish Meal Manufacturers), (1990) World fish meal production down. Feedstuffs, 62 (25): 7 Irving, G.C.J., (1980) Nutritional aspects of inositol phosphates. In: Studies in organic chemistry, 4: Inositol phosphates: their chemistry, biochemistry and physiology. Elsevier Scientific Publishing Company, Amsterdam, p. 157-172 208 Irving, G.C.J., and Cosgrove, D.J. (1971). Inositol phosphate phosphatases of microbiological origin: Some properties of a partially purified bacterial (Pseudomonas sp.) phytase. Aust. J. Biol. Sci. 24:-547-557 Jackson, A . J . , Caper, B.S. and Matty, A.J. (1982) Evaluation of some plant proteins in complete diets for the tilapia, Sarotherodon mossambicus. Aquaculture 27: 97-109 Jaffe, W.G., (1983) Nutritional significance of lectins. In: M. Rechicgl Jr. (ed.), Handbook of naturally occurring food toxicants. CRC Press, Inc, Boca Raton, Florida, p. 31-38 Jensen, S.K., Olsen, H.S. and Sorensen, H., (1990) Aqueous enzymatic processing of rapeseed for production of high quality products. In: F. Shahidi (ed.). Canola and rapeseed, production, chemistry, nutrition and processing technology. Von Nostrand, New York, p. 331-343 Jones, J.D. (1970) Rapeseed protein concentrates-toxicology and nutrition. In: Proceedings international conference on the science, technology and marketing of rapeseed and rapeseed products. Ste- Adele , Quebec. Sept 20-23. Rapeseed Association of Canada in co-operation with the Department of Industry, Trade and Commerce, Ottawa., p. 128-141 Jones, J.D., (1979) Rapeseed protein concentrate preparation and evaluation. J. Am. Oil Chem. Soc, 56: 716-721 Jones, J.D. and Holme, J. (1979) United States. Patent 4,158,656 Jongbloed, A.W. and Kemme, P.A., (1990) Effects of pelleting mixed feeds on phytase activity and the apparent absorbability of phosphorus and calcium in pigs. Anim. Feed Sci. Technol., 28: 232-242 Jongbloed, A.W., Mroz, Z. and Kemme, P.A., (1992) The effect of supplementary Aspergillus niger phytase in diets for pigs on concentration and apparent digestibility of dry matter, in different sections of the alimentary tract. J. Anim. Sci., 70: 1159-1168 Kaufman, H.W. (1986) Interaction of inositol phosphates with mineralized tissues. In: Graf, E. (ed). Phytic acid: chemistry, and applications. Pilatus Press, Minneapolis, p 303 Kaushik, S.J., Cravedi, J.P., Lalles, .P., Sumpter, J., Fauconneau, B. and Laroche, M . (1995) Partial or total replacement of fishmeal by soybean protein on growth, protein-utilization, potential estrogenic or antigenic effects, cholesterolemia, and flesh quality in rainbow trout, Oncorhynchus mykiss. Aquaculture 133: 257-274 Ketaren, P.P., Batterham, E.S., Dettmann, E.B. and Farrell, D.J., (1993) Phosphorus studies in pigs: Effect of phytase supplementation on the digestibility and availability of phosphorus in soyabean meal for grower pigs. Br. J. Nutr., 70: 289-312 Ketelsen, S.M., Stuart, M.A., Weaver, C M . , Forbes, R.M., Erdman, J.W. (1984). Bioavailability of zinc to rats from defatted soy flour, acid-precipitated soy concentrate and neutralized soyconcentrate as determined by intrinsic and extrinsic labelling tecniques. J. Nutr. 114: 536-542 Ketola, H.G. (1979) Influence of dietary zinc on cataracts in rainbow trout (Salmo gairdneri), J. Nutr. 109: 965-969 Ketola, H.G. and Harland, B.F., (1993) influence of phosphorus in rainbow trout diets on phosphorus discharges in effluent water. Trans. Amer. Fish. Soc, 122: 1120-1126 209 Khan, N . and Cole, D.J.A, (1995) The impact of phytase on pig performance. In: . Lyons and K.A. Jacques (eds.). Proceedings of Alltech's eleventh annual symposium : Biotechnology in the feed Industry. Knottingham University Press, Nottingham, p. 355-361 Kilpatrick, (1994) Protein sources in aquaculture feeds-present and future. New Developments in Seafood Science and Technology: 1-21 Kitamikado, M , Morishita, T. and Tachino, S. (1964) Digestibility of dietary components in fishes II. Bull. Jpn. Fish. Soc. 30:50-54 Koh, L .S , Erickson, L.R. and Beresdorf, W.D, (1990). In : F. Shahidi (ed.), Canola and rapeseed: production, chemistry, nutrition and processing technology. , Van Nostrand Reinhold, New York, p. 47-78 Kozlowska, H , Naczk, M , Shahidi, F. and Zademowski, R , (1990) Phenolic acids and tannins in rapeseed and canola. In: Shahidi, F. (ed). Canola and rapeseed: production, chemistry, nutrition and processing technology. Van Nostrand Reinhold, New York, p. 193-209 Kramer, J.K.G. and Sauer, F .D, 1983a. Cardiac lipid changes in rats, pigs, and monkeys fed high fat diets. In : J.K. Kramer, F.D. Sauer and W.J. Pigden (eds.). High and low erucic rapaseed oils: production, usage, chemistry, and toxicological evaluation.. Academic Press, Toronto, p. 475-513 Kramer, J.K.G. and Sauer, F.D, 1983b. Results obtained with feeding low erucic acid rapeseed oils and other vegetable oils to rats and other species. In: J.K. Kramer, F.D. Sauer and W.J. Pigden (eds.). High and low erucic rapaseed oils: production, usage, chemistry, and toxicological evaluation. Academic Press, Toronto, p. 413-474 Kramer, J .K.G, Sauer, F.D. Pigden, W.J. (eds.) (1983) High and Low Erucic acid rapeseed oils. Academic Press, Toronto, p. 582 Krieberg, H , (1992) Metomidate sedation minimizes handling stress in chinook salmon. Bull. Aquacul. Assoc. Canada, 92: 52-54 Krogdahl, A , (1989) Alternative protein sources form plants contain antinutrients affecting digestion in salmonids. In: M . Takeda and T. Watanabe (eds.), The current status of fish nutrition in aquaculture. Proceedings of the third international symposium on feeding and nutrition in fish, Toba, p. 253-261 Krygier, K , Sosulski, F. and Hogge, L , (1982)a. Composition of phenolic acids in rapeseed flour and hulls. J. Agric. Food Chem, 30: 334-336 Krygier, L , Sosulski, F.W. and Hogge, L , (1982)b. Free, esterified, and insoluble phenolic acids. 2. Composition of phenolic acids in rapeseed flour and hulls. J. Agric. Food Chem, 30: 334-336. Kuvayeva, E.B. , Kretovic, V.L. (1978) Phytase of germinating pea seeds. Sov. Plant P. 25(2): 290-295 Lacroix, M , Amiot, J , Cheour, F , De La Noiie, J , Goulet, G. and Brisson, G.J, (1988) Effect of methanol/acetone/water extraction and enzymatic hydrolysis on the nutritional value of unheated rapeseed proteins. Plant Foods for Human Nutrition, 38: 343-353 Lall, S.P. (1989) The minerals. In: Fish nutrition. 2 n d ed. J.E. Halver (ed.) Academic Press, New York, p 219-257 Langer, P , (1983) Naturally occurring food toxicants: goitrogens. In: Handbook of naturally occurring food toxicants. M . Rechcigl Jr. (ed.), CRC Press, Inc, Boca Raton, p. 101-129. 210 Leatherland, J., Hilton, J. and Slinger, S. (1987) Effects of thyroid hormone supplementation of canola meal based diets on growth, and interrenal and thyroid gland physiology of rainbow trour (Salmo gairdneri). Fish Physiol. Biochem. 3:73-82 Lei, X.G., Ku, P.K, Miller, E.R., Ullrey, D.E., Yokoyama, M.T. (1992) Supplemental dietary microbial phytase improves bioavailability of zinc as well as phytate phosphorus in a corn-soybean meal diet for weanling pigs. J. Anim. Sci. 70 (sup. 1): abstract 365 Lei, A., Ku, P.K., Miller, E.R., Ullrey, D.E. and Yokoyama, M.T., (1993) Supplemental microbial phytase improves bioavailability of dietary zinc to weanling pigs. J. Nutr., 123: 1117-1123 Leung, J., Fenton, T.W., Mueller, M.M. and Clandinin, D.R., (1979) Condensed tannins of rapeseed meal. J. Food Sci., 44: 1313-1316 Libey, G.S. and Bosworth, B.G., (1990) Aquaculture. In: The seafood industry. Van Nostrand Reinhold, New York, p. 291-301 Lim, C. and Akiyama, D.M., (1992) Full-fat soybean meal utilization by fish. Asian Fisheries Science, 5: 181-197 Lim, P.E. and Tate, M.E., (1973) The phytases II. properties of phytase fractions F l and F2 form wheat bran and the myo-inositol phosphates produced by fraction F2. Biochim. Biophys. Acta., 302: 316 Lips, H.J. and Grace, N.H., and Hamilton, E.M. (1948). Canadian erucic acid oil: edible use of rape and mustard seed oils. Can. J. Res., 26: 360-364 Liu, R., Thompson, L.U. and Jones, J.D., (1982) Yield and nutritive value of rapeseed protein concentrate. J. Food Sci., 47: 977-981 Lott, J.NA- (1985) Accumulation of seed reserves of P and other minerals. In: Seed Physiology Vol. 1. D.R. Murray (ed). Academic Press. Australia p. 139-166 Lott. J.N.A., Ockenden, I. (1986) Energy dispersive x-ray analysis and chemical analysis in the study of mineral storage deposits in seeds. Cereal. F.W. 31: 606 Love, H.K., Rakow, G., Raney, J.P. and Downey, R.K., (1990) Can. J. Plant Sci., 70: 419-424 Love, R.M., (1970) The chemical biology of fishes. Academic Press, 266 p. Liihs, W. and Friedt, W., (1994) The major oil crops. In: D.J. Murphy (ed.). Designer oil crops: breeding, processing and biotechnology. VCH Verlagsgesellschaft MbH, Weinheim, p. 5-71 Mackie, A.M. , (1982) Identification of the gustatory feeding stimulants. In: Chemoreception in fishes: developments in aquaculture and fisheries science, Vol. 8. T.J. Hara (ed.), Elsevier scientific publishing company, Amsterdam, Oxford, New York, p. 275-291 Mackie, A .M. and Mitchell, A.I., (1985) Identification of gustatory feeding stimulants for fish-applications in aquaculture. In: C B . Cowey, A.M. Mackie and J.G. Bell (eds.). Nutrition and Feeding in Fish. Academic Press, Toronto, p. 177-189 Maddaih, V.T., Kurnick, A.A., and Reid, B.L. (1964) Phytic acid studies. Proc. Soc. Exp. Biol. Med. 115:391 211 Matthews, G . M , Park, D . L , Achord, S, Ruehle, T.E. (1986) Static seawater challenge test to measure relative stress levels in spring chinook salmon smolts. Trans. Am. Fish Soc. 115: 236-244 Mattson, F . H , (1973) Potential toxicity of food lipids. In: Toxicants naturally occurring in foods. National Academy of Sciences, Washington, p. 189-209 Maynard, L.A. and Loosli, J .K, (1969) Animal nutrition. 6th ed. McGraw-Hill, 613 p. McCallum, I.M. and Higgs, D .A , (1989) An assessment of processing effects on the nutritive value of marine protein sources for juvenile chinook salmon (Oncorhynchus tshawytscha). Aquaculture, 77: 181-200 McCurdy, S.M. and March, B .E , (1992) Processing of canola meal for incorporation in trout and salmon diets. J. Am. Oil Chem. Soc, 69: 213-220 McDonald, B . E , (1983) Studies with high and low erucic acid rapeseed oil in man. In: J.K.G. Kramer, F.D. Sauer and W.J. Pigden (eds.). High and low erucic acid rapeseed oils: production, usage, chemistry and toxicological evaluation. Academic Press, New York, p. 135 Meams, K . J , (1986) Sensitivity of brown trout (Salmo trutta L.) and Atlantic salmon (Salmo salar L.) fry to amino acids at the start of exogenous feeding. Aquaculture, 48: 253 Messerich, J , (1994) A review of technical advancements in fish meal production and their use in aquaculture feeds, International Proteins Corporation, St. Paul, p. 1-24 Miller, E.L. and DeBoer, F , (1988) Feedstuffs. 6. By-products of animal origin. Livst. Prod. Sci, 19: 189-195 Miller, W.J , (1988) Canada's Canola. Canola Council of Canada, p 24 Mitaru,B.M, Blair, R , Bell, J . M , Reichert, R.D. (1982). Tannin and fibre contents of rapeseed and canola hulls. Can. J. Anim. Sci. 62: 661-663 Mongin, P , (1981) Recent advances in dietary anion-cation balance in poultry. In: W. Haresign (ed.). Recent advances in Animal Nutrition. : Butterworths, Boston, p. 109-119 Moyano, F.-J, Cardenete, G. and De La Higuera, M , (1992)a. Nutritive value of diets containing a high percentage of vegetable proteins for lroul,Oncorhynchus mykiss. Aquat. Living Resour, 5: 23-29 Moyano, F.J, Cardenete, G. and de la Hiuera, M , (1992)b. Use of two vegetable by-products as protein sources in rainbow trout feeding. Anim. Prod, 55: 277-284 Mundheim, H. and Opstvedt, J , (1989) Effect of dietary level of protein and fiber on apparent protein digestibility in the rainbow trout (Oncorhynchus mykiss) and salmon (Salmo salar) and comparison of protein digestibility in mink (Mustela vison), rainbow trout and salmon. Proc. Third Int. Symp. on Feeding and Nutr. in Fish (28 August-1 September): 195-200 Murai, T , (1992) Protein nutrition of rainbow trout. Aquaculture, 100: 191-207 Naczk, M , Diosady, L . L , Rubin, L.J. (1986) The phytate and complex phenol content of meals produced by alkanol-ammonia/hexane extraction of canola. Lebensm. Wiss, u. Technol. 19:13-16 212 Naczk, M . and Shahidi, F., (1990) Carbohydrates of canola and rapeseed. In: : F. Shahidi (ed.). Canola and rapeseed, production, chemistry, nutrition and processing technology. Von Nostrand Reinhold, New York, p. 211-220 Nair, V.C., Laflamme, J. and Duvnjak, Z., (1991) Production of phytase by Aspergillus ficuum and reduction of phytic acid content in canola meal. J. Sci. Food Agric, 54: 355-365 Nasi, M. (1990) Microbial phytase supplementation for improving availability of plant phosphorus in the diet of the growing pig. J. Agri. Sci. in Finland. 62: 435-443 NRC (National Research Council), (1981) Nutrient requirements of cold water fishes. National Academy Press, p.63 NRC (National Research Council), (1993) Nutrient requirements of coldwater fishes. National research council,, p. 114 Nayani, N.R. and Markakis, P., (1986) Phytases. In: phytic acid: chemistry & applications. E. Graf (ed.), Pilatus Press, Minneapolis, p. 101-118 Nelson, T.S., Daniels, L.B., Hall, J.R., Shields, L.G. (1976) Hydrolysis of natural phytate phosphorus in digestive tract of calves. J. Anim. Sci. 42: 1509-1512 Nelson, T.S., Shieh, T.R., Wodzinski, R.J. and Ware, J.H., (1971) Effect of supplemental phytase on the utilization of phytate phosphorus by chicks. J. Nutr., 101: 1289-1294 Neshem, H.E., (1990) Overview of the developing U.S. canola industry. Proceedings: International canola conference (2-6 April): 22-28 New, M.B., (1991) Turn of the millennium in aqaculture: Navigating troubled waters or riding the crest of the wave. World Aquaculture, 22: 22-49 New, M.B. and Wijkstrom, U.N., (1990) Feed for thought. World Aquaculture, 21: 17-23 Niewiadomski, H., (1990) Vol. 23.Developments in food science 23: Rapeseed Chemistry And Technology. Elsevier, PWN-Polish Scientific Publishers, p 429 Nolan, K.B., Duffin, P.A. and McWeeny, D.J., (1987) Effects of phytate on mineral bioavailability : in vitro studies on magnesium, calcium, iron, coper, and zinc (also cadmium) solubilities in the presence of phytate. J. Sci. Food Agric, 40: 79-85 Noland, P.R., Funderburg.,M., Johnson,Z. (1968) Phosphorus availability in a practical diet for swine. J. Anim. Sci. 26:1155 Norton, G., (1989) Nature and biosynthesis of storage proteins. In: Oil crop of the world: their breeding and utilization. G. Robbelen, R.K. Downey and A. Ashri (eds.), McGraw-Hill Publishing Company, New York, p. 165 Nwokolo, E.N. and Bragg, D.B., (1977) Influence of PA and crude fibre on the availability of minerals from four protein supplements in growing chicks. Can. J. Anim. Sci., 57: 475-477 Nwokolo, E.N. and Bragg, D.B., (1980) Biological availability of minerals in rapeseed meal. Poultry Sci., 59:155-158 213 O'Dell, B.L. (1969) Effect of dietary components upon Zn availability: A review of original data. Am.J. Clin. Nutr. 22: 1315-1322 O'Dell, B . L , deBoland, A.D. (1976) Complexation of phytate with proteins and cations in com germ and oilseed meals. J. Agric. Food. 24: 804-808 O'Dell, B . L , deBoland, A. and Koidrtoyahann, R , (1972) Distribution of phytate and nutritionally important elements among the morphological components of cereal grains. J. Agric. Food Chem, 20: 718 O'Dell, B . L , Yohe, J.M. and Savage, J.E. (1964) Zinc availability in the chick as affected by phytate, calcium and ethylenediaminetetraacetate. Poult. Sci. 43: 415-419 Ogino,C. and Yang, G. (1978). Requirement of rainbow trout for dietary zinc. Bull. Jpn. Soc. Sci. Fish 44:1015-1018 Ohlrogge, J.B. and Kernan, T.P, (1982) Oxygen-dependent aging of seeds. Plant physiol, 70: 791-794 Ohlson, R. (1972) Non-nutritional uses of rapeseed oil and rapeseed fatty acids. In: L.A. Appleqvist, and R. Ohlson. (eds). Rapeseed: Cultivation, composition, processing and utilization, p 274 Okubo, K. Myers, D.V. and Iacocobucci, G.A. (1976) Binding of phytic acid to glycinin. Cereal Chem. 53:513 Okubo, K. and Waldrop, A . B , (1975) Preparation of low phytate soybean protein isolates and concentrate by ultrafiltration. Cereal chemistry, 52: 263 Olivier, S.L, McDonald, B.E. and Opuszynska, T , (1971) Weight gain, protein utilization and liver histochemistry of rats fed low ans high thioglucoside content rapeseed meals. J. Physiol. Pharmacol, 49: 448-456 Olli, J , Krogdahl, A. and Berg-Lea, T , (1989) Effects of soybean trypsin inhibitor activity on nutrient digestibility in salmonids fed practical diets containing various soybean meals. In: M . Takeda and T. Watanabe (eds.). Proceedings of the Third International Symposium on Feeding and Nutrition in Fish . Tokyo University of Fisheries, Tokyo, p. 263-271 Omasaiye, O. and Cheryan, M . (1979) Low-phytate, full-fat soy protein products by ultrafiltration of aqueous extracts of whole soybeans. Cereal Chem. 56: 68-70 Omeljaniuk, R.J , Cook, R.F. and Eales, J.G, (1984) Simultaneous measurement of thyroxine and triiodothyronine in trout plasma using a solid-phase radioimmunoassay. Can. J. Zool, 62: 1451-1453 Oser, B . L , (1959) An integrated essential lamino acid index for predicting the biological value of proteins. In: . A.A. Albanese (ed.). Protein and amino acid nutrition. Academic Press, New York, p. 282-295 Paik, I.K. and Robblee, A .R , (1980) Products of the hydrolysis of rapeseed glucosinolates. Can. J. Anim. Sci, 60: 481-493 Paullauf, J. von Hohler, D , Rimbach,G. andNeusser, H. (1992) Einfluss einar zulage au mikrobieller phytase zu einar mais-soja-diat auf die sceinbare absorption von phosphor und calcium beim ferkel. J. Anim. Nutr. 52:310-324 Palmer, T.N. and Ryman, B . E , (1972) Studies on oral glucose intolerance in fish. J. Fish Bio l , 4: 311-319 214 Patience, J.F, Austic, R .E , Forsberg, N.E. and Boyd, R.D, (1984) Nutritional and physiological implications of dietary electrolyte balance in swine diets. Abstracts Cornell Nutrition Conference for Feed Manufacturers ed, p. 55-62 Pen, J , Verwoerd, T.C, van Paridon, P A , Beudeker, R.F, van den Eizen, P.J .M, Geerse, K , van der Klis, J.D, Versteegh, H.A.J , van Ooyen, A.J.J, and Hoekema, A , 1993. Phytase-containing transgenic seeds as a novel feed additive for improved phosphorus utilization. Biotechnology, 11: 811-814. Pfeffer, E. (1994) Carbohydrate utilization and its determination. In: Report of the EIFAC workshop on methodology for determination of nurient requirements in fish. J.M. Grop. And A.G.J. Tacon (eds). Occasional paper No. 29. Rome. p. 56 Pike, I H , Andorsdottir, G. and Mundheim, H , (1990) The role of fish meal in diets for salmonids. IAFMM, 24: 1-35. Pointillart,A. (1988) Phytate phosphorus utilization in growing pigs. Proc. 4 t h Intern. Sem. digestive physiology in pigs. Polish Academy of Sciences, Jablonna. P 319-326 Pointillart, A , (1991) Enhancement of phosphorus utilization in growing pigs fed phytate-rich diets by using rye bran. J. Anim. Sci, 69: 1109-1115 Raboy, V. and Dickinson, D.B, (1987) The timing and rate of phytic acid accumulation in developing soybean seeds. Plant Physiol, 85: 841 Ramachandra, G , Virupaksha, T. and Shadaksharas, W , (1977) Relationship between tannin levels and in vitro protein digestibility in finger millet. J. Agric. Food Chem, 25: 1101-1104 Rao, P.V. and Clandinin, D.R. (1972) Chemical determination of available carbohydrates in rapeseed meals. Poultry Science 51: 1474-1475 Rapoport, S, Leva, E. and Guest, G . M , (1941) Phytase in plasma and erythrocytes of various species of vertebrates. J. Biol. Chem, 139: 621-632 Reddy, N.R. and Pierson, M . D , (1989) Nutritional consequences of phytates. In: Phytates in cereals and legumes. CRC Press Inc, Boca Raton,Florida, p. 81-110 Reddy, N .R , Sathe, S.K. and Salunkhe, D.K, (1982) Phytate in legumes and cereals. Adv. Food Res, 28: 1-9 Reddy, N .R , Pierson, M . D , Sathe, S.K. and Salunkhe, D.K, (1989) Phytates in cereals and legumes. CRC Press, Inc., 152 p. Rehnberg, B . G , JonassOn, B. and Schreck, C B , (1985) Olfactory sensitivity during parr and smolt developmental stages of coho salmon. Trans. Amer. Fish. Soc. 114: 732-736 Ressurrenccion, A .P , Juliano, B.O. and Tanaka, Y. (1979) Nutrient content and distribution in milling fractions of rice grain. J. Sci. Food Agric. 30: 475 Richardson, N . L , Higgs, D .A , Beames, R.M. and McBride, J.R, (1985) Influence of dietary calcium, phosphorus, zinc and sodium phytate level on cataract incidence, growth and histopathology in juvenile chinook salmon (Oncorhynchus tshawytscha). J. Nutr, 115: 553-567 215 Riche, M. , (1993) Phosphorus absorption coefficients for rainbow trout (Oncorhynchus mykiss) fed commercial sources of protein. MSc. Thesis, West Lafayette, 145 p. Riley, W.W., Higgs, D.A., Dosanjh, B.S. and Eales, J.G., (1993) Influence of dietary amino acid composition on thryroid function of juvenile rainbow trout, Oncorhynchus mykiss. Aquaculture, 112: 253-269 Robblee, A.R., Clandinin, D.R., Summers, J.S. and Slinger, S.J., (1986) Canola meal for poultry. In: D.R. Clandinin (ed.), Canola meal for livestock and poultry. Canola Council of Canada, Winnipeg, p. 8-14. Rubin,L. J., Diosady, L.L., Phillips,C.R. (1984) United States Patent 4,460,504 Rubin, L.J., Diosady, L.L. and Tzeng, Y. -M. , (1990) Ultrafiltration in rapeseed processing. In: F. Shahidi (ed.), Canola and rapeseed: production, chemistry, nutrition and processing technology. Von Nostrand, New York, p. 307-329 Rumsey, G.L. (1994) History of early diet development in fish culture, 1000BC to A.D. 1955- Review. Prog. Fish Cult. 56(1): 1-6 Rumsey, G.L., (1988) Chemical control of feed intake in fishes. In: Proceedings of the VI world conference of animal production. Finnish animal breeding association, Helsinki, p. 100-110 Rumsey, G.L., (1993) Fish meal and alternative sources of protein in fish feeds update (1993) Fisheries 18: 14-19 Rumsey, G.L. and Ketola, H.G., (1975) Amino acid supplementation of casein in diets of Atlantic salmon (Salmo salar) fry and of soybean meal for rainbow trout (Salmo gairdneri) Finger lings. J. Fish. Res. Board Can., 32: 422-426 Rumsey, G., Endres,J.G., Bowser, P.R., Earnest-Koons, K.A., Anderson, D.P. and Siwicki, A.K. (1994) Soy protein in diets for rainbow trout: Effects on growth, protein absorption, gastrointestinal histology, and nonspecific serological and immune response. In: Nutrition and utilization technology in aquaculture. C. Lim and D.J. Sessa (eds.) AO AC Press, Champlaign, p. 166-188 Salunkhe, D.K., Chavan, J.K., Adsule, R.N. and Kadan, S.S., 1992. World oilseeds: chemistry, technology and utilization. Van Nostrand Reinhold, p. 554 Sandberg, A.-S. and Svanberg, U., (1991) Phytate hydrolysis by phytase in cereals; effects on in vitro estimation of iron availability. J. Food Sci., 56: 1330-1333 Sandberg, A.-S., Larsen, T. and Sandstrom, B., (1993) High dietary calcium level decreases colonic phytate degradation in pigs fed a rapeseed diet. J. Nutr., 123: 559-566 Satoh, S., W.E. Poe and Wilson, RP. (1989) Effect of supplemental phytate and/or tricalcium phosphate on weight gain, feed efficiency and zinc content in vertebrae of channel catfish. Aquaculture 80: 155-161 Scott, J.J. and Loewus, F.A. (1986) Phytic metabolism in plants. In: Phytic acid: chemistry, and applications. E. Graf, (ed.) Pilatus Press, Minneapolis, p. 43 Serraino, M.R. and Thompson, L.U., (1984) Removal of phytic acid and protein-phytic acid interactions in rapeseed. J. Agric. Food Chem., 32: 38-40 Serraino, M.R., Thompson, L.U., Savoie, L. and Parent, G., (1985) Effect of phytic acid on the in-vitro rate of digestibility of rapeseed protein and amino acids. J. Food Sci., 50: 1689-1692 216 Shah, B.G., Giroux, A., Belonje, B. and Jones, JD. (1979). Optimal level of Zn supplementation for young rats fed rapeseed protein concentrate J. Agric. Food Chem. 27: 387-390 Shah, B.G., Jones, J.D., Mclaughlan, J.M. and Beare-Rogers, J.L. (1976). Beneficial effecct of Zn supplementation in young rats fed protein concentrate from rapeseed or mustard. Nutr. Rep. Int. 13: 1-8 Shahidi, F. (1990) North American production of canola. In: Canola and rapeseed, production, chemistry, nutrition and processing technology. F. Shahidi (ed.), Von Nostrand Reinhold, New York, p. 15-23 Shahidi, F. and Naczk, M. , (1990) Removal of glucosinolates and other antinutrients from canola and rapeseed by methanol/ammonia processing. In: Canola and rapeseed, production, chemistry, nutrition and processing technology. F. Shahidi (ed.), Von Nostrand Reinhold, New York, p. 291-306 Sharma, C.A., Goel, M . and Irshad, M. , (1978) Myoinositol hexaphosphate as a potential inhibitor of a-amylase of different origins. Phytochemistry, 17: 201-204 Shearer, K.D., (1984) Changes in elemental composition of hatchery reared rainbow trout,Salmo gairdneri, associated with growth and reproduction. Can. J. Fish. Aquat. Sci., 41: 1592-1600 Shimeno, S., Hosokawa, H. and Takeda, M. , (1979) The importance of carbohydrate in the diet of a carnivorous fish. In: Finfish nutrition and fishfeed technology, Vol. I. J.E. Halver and K. Tiews (eds.), Heenemann Verlagsgesellschaft mbH„ Berlin, p. 127-143 Shparkovsky, I.A., Pavlov, I.D. and Chinarina, A.D., (1983) Behavior of young hatchery-reared Atlantic salmon, Salmo salar (Salmonidae) influenced by amino acids. J. Ichthyol., 4: 140 Simons, P.C.M., and Versteegh, H.A.J. (1991) Application of microbial phytase in poultry nutrition. Poult. Sci., PSA and SPSS abstracts Vol 70 (supp.): 110 Simons, P .C .M, Versteegh, H.A.J., Jongbloed, A.W., Kemme, P.A., Slump, P , Bos, K.D., Wolters, M.G.E., Beudeker, R.F. and Verschoor, G.J., (1990) Improvement of phosphorus availability by microbial phytase in broilers and pigs. Br. J. Nutr. 64: 525-540 Singh, B. and Sedeh, H.G., (1979) Characteristics of phytase and its relationship to acid phosphatase and certain minerals in triticale. Cereal chemistry, 56: 267 Slominski, B.A. and Campbell, L.D., (1990) Indoleacetonitriles- thermal degradation products of indole glucosinolates in commercial rapeseed (Brassica Napus) meal. J. Sci. Food Agric, 47: 75-84 Slominski, B.A., Simbaya, J., Campbell, L.D. and Guenter, W., (1995) Nutritive profile of yellow-seeded canola/rapeseed. In: Rapeseed today and tomorrow; 9th International rapeseed congress, Vol. I. D. Murphy (ed.), Organizing committee of th 9th International rapeseed congress, Dorchester, p. 148-150 Sobolev, A. M . and Rodionova, M.A. (1966) Phytin synthesis by alerone grain in ripening sunflower seeds. Soviet Plant Phsiol. (Enlish transl.) 13: 958 Solmes, J. (1969) Tastes of amino acids, peptides and proteins. J. Agric. Food 17: 686 Sorensen, H., (1990) Glucosinolates: structure, properties, function. In: Canola and rapeseed: production, chemistry, nutrition and processing technology. F. Shahidi (ed.), Van Nostrand Reinhold, New York, p. 149-172 217 Sosulski, F , (1979) Organoleptic and nutritional effects of phenolic compounds on oilseed protein products: a review. J. Am. Oil Chem. Soc, 56: 711-715 Spannhof, L. and Plantikow, H , (1983) Studies on carbohydrate digestion in rainbow trout. Aquaculture, 30: 95-108 Specker, J.L. and Schreck, C B . (1980) Stress responses to transportation and fitness for marine survival in coho salmon (Oncorhynchus kistutch) smolts. Can. J. Fish. Aquat. Sci. 37: 765-769 Spinelli, J , Mahnken, C. and Steinberg, M , (1979) Alternate sources of proteins for fish meal in salmonid diets. Proceedings of the World Symposium on Finfish Nutrition and Fishfeed Technology . In: J.E. Halver and K. Tiews (eds.). Finfish nutrition and fishfeed technology, Vol. II. Heenemann Verlagsgesellschaft mbH, Berlin (Hamburg, 20-23 June), Vol. II. P. 131-142 Spinelli, J , Houle, CR. and Wekell, J.C, (1983) The effect of phytates on the growth of rainbow trout (Salmo gairdneri) fed purified diets containing varying quantities of calcium and magnesium. Aquaculture, 30: 71-83 Stefansson, B.R, (1983) The development of improved rapeseed cultivars. In: High and low erucic rapeseed oils: production, usage, chemistry, and toxicological evaluation. . In: Finfish nutrition and fishfeed technology, Vol. II. J.E. Halver and K. Tiews (eds.), Heenemann Verlagsgesellschaft mbH, Berlin Academic Press, Toronto, p. 143-159 Stone, F .E , Hardy, R.W. and Spinelli, J , (1984) Autolysis of phytic acid and protein in canola meal (Brassica sp), wheat bran (Triticum sp) and fish silage blends. J. Sci. Food Agric, 35: 513-519 Sutterlin, A .M. and Sutterlin, N , (1970) Taste response in Atlantic salmon (Salmo salar) parr. J. Fish. Res. Brd. Can, 27: 1927-1942 Tape, N W , Sabry, W.I. and Eapen, K . E , (1970) Production of rapeseed flour for human consumption. Can. Inst. Food Technol. J , 3: 78 Taylor, R.H. (1979) Gastric emptying, fibre absorption. Lancet 1: 872 Teskeredzic, Z , Higgs, D .A , Dosanjh, B.S, McBride, J.R, Hardy, R.W, Beames, R . M , Jones, J.D, Simell, M , Vaara, T. and Bridges, R .B , (1995) Assessment of undephytinized and dephytinized rapeseed protein concentrate as sources of dietary protein for juvenile rainbow trout (Oncorhynchus mykiss). Aquaculture, 131: 261-277 Thompson, L . U , (1986) Phytic acid: a factor influencing starch digestibility and blood glucose response. In: E. Graf (ed.), Phytic Acid: Chemistry & Applications. Pilatus Press, Minneapolis, p. 173-194 Thompson, L . U , (1990) Phytates in canola/rapeseed. In: canola and rapeseed, production, chemistry, nutrition and processing technology. F. Shahidi (ed.). Von Nostrand Reinhold, New York, p. 173-191 Thompson, L.U. (1993). Potential health benefits and problems associated with antinutrients in foods. Food Research International: 131 Thompson, L.U. and Cho, Y , (1984) Effect of acylation upon extractability of N , minerals, PA in rapeseed flour, and protein concentrate. J. Food Sci, 49: 771-776 Thompson, L . U , Reyes, E. and Jones, J.D. (1982) Modification of the Na hexametaphosphate extraction-precipitation technique of rapeseed protein concentrate preparation. J. Food Sci. 47: 982-988 218 Thompson, L.U. and Serraino, M.R., (1985). Effect of germination on protein, fat, and PA concentration of rapeseed. J. Food Sci. 50:1200 Thompson, L.U. and Serraino, M R . , (1986) Effect of PA reduction on rapeseed protein digestibility and amino acid absorption. J. Agric. Food Chem., 34: 468-469. Tietz, N.W. (ed). Fundamentals of clinical chemistry. W.B. Saunders Company, Philadelphia.p 916 Tiews, K., Koops, H., Grop, J. and Beck, H., (1979) Compilation of fish meal-free diets obtained in rainbow trout {Salmo gairdneri) feeding experiments at Hamburg (1970-1977/78). In: J.E. Halver and K. Tiews (eds.). Finfish nutrition and fishfeed technology, Vol. II. Heenemann Verlagsgesellschaft mbH„ Berlin, p. 219-228 Tkachuk, R., (1981) Oil and protein analysis of whole rapeseed kernels by near infrared reflectance spectroscopy. J. Am. Oil Chem. Soc, 58: 819-822 Tzeng, Y - M . (1987) Process developmnet for the production of high quality rapeseed (Canola) protein isolates by membrane tchnology. PhD. Thesis. Univ. of Toronto. Tzeng, Y - M . , Diosady, L. and Rubin, L., (1988)a. Preparation of rapeseed protein isolate by Na hexametaphosphate extraction, ultrafiltration, diafittration, and ion exchange. J. Food Sci., 53 (5): 1537-1541 Tzeng, Y - M , . , Diosady, L.L., and Rubin,L.J. (1988)b. Preparation of a rapeseed protein isolate using sodium hexametaphosphate extraction, ultrafiltration, and diafiitration. Can. Inst. Food Sci. Technnol. J. 21(4): 419-424 Tzeng, Y - M . , Diosady, L.L., and Rubin,L.J. (1990) Production of canola protein materials by alkaline extraction, precipitation, and membrane processing. J. Foo Sci. 55:1147-1156 Unger, (1990). In: canola and rapeseed, production, chemistry, nutrition and processing technology. F. Shahidi (ed.), F. Shahidi (ed.), Von Nostrand Reinhold, New York, p. 230 Vaisey, M . , Latta, M. , Bruce, V .M. and McDonald, B.E., (1973) Assessment of the intake of high and low erucic acid rapeseed oils in a mixed Canadian diet. Can Inst. Food Sci. Technol. J., 6: 142-146 Vaisey, M . , Latta, M. , Bruce, V.M. , McDonald, B.E. (1974) Assessment of intake and digestibility of high and low erucic acid rapeseed oils in a mixed Canadian diet. Can. I. Food 6: 142-147 van Etten, C.H. and Tookey, H.L., (1983) Glucosinolates. In: J. Rechcigl M (ed.), CRC Handbook of naturally occurring food toxicants. CRC Press, Boca Raton, p. 15-30 Viola, S., Arieli, Y. and Zohar, G., (1988) Animal protein-free feeds for hybrid tilapia (Oreochromis niloticusx O. aureus) in intensive culture. Aquaculture, 75: 115-125 Virtanen, E., (1992) Betaine supplementation enhances the seawater adaptation of salmonids. In: V. International Union of Nutritional Sciences. Fundacion Chile/International Union of Nutritiona Sciences, Santiago, p. 40 Virtanen, E., Junnila, M. and Soivio, A., (1989) Effects of food containing betaine/amino acid additive on the osmotic adaptation of young Atlantic Salmon, Salmo salar L. Aquaculture, 83: 109-122 219 Vohra, P , Gray, G .A, Kratzer, F.H. (1965) Phytic acid-metal complexes. P. Soc. Exp. M. 120: 447 Vohra, P , (1989) Carbohydrate and fiber content of oilseeds and their nutritional importance. In: Oil crops of the world. McGraw-Hill Publishing company, New York, p. 208-225 Ward, N . E , (1991) Chemoattractants for trout and salmon: amino acids beyond protein synthesis. Feed Management, 42 ( 3,March): 6-10 Ward, A .T , Thacker, P A , Rotter, B. and Campbell, L. (1991) The effect of enzymes on the availability of minerals and on the growth of chicks fed canola based diets. In: 9 t h project report: research on canola seed, oil and meal. Canola Council of Canada. Winnipeg. P. 33-56 Watanabe, T. and Pongmaneerat, J , (1993) Potential of soybean meal as a protein source in extruded pellets for rainbow trout. Nipon Suisan Gakkaishi, 59(8): 1415-1423 Webster, C D . and Tidwell, J.H. (1992). Use of distillers by-products in aquaculture diets. World Aquaculture 23: 55-57 White, T , (1957) Tannins-their ocurrence and significance. J. Sci. Food Agric, 8: 377-385 Williams, P.J. and Taylor, T .G, (1985) A comparative study of phytate hydrolysis in the gastrointestinal tract of the golden hamster (Mesocricetus auratus) and the laboratory rat. Br. J. Nutr, 54: 429-435 Wilson, R.P, (1989) Amino acids and proteins. In: Fish Nutrition, 2nd. ed. J. Halver (ed.), Academic Press, San Diego, p. 111-151 Wilson, R.P, (1994) Utilization of dietary crbohydrate by fsh. Aquaculture, 124: 67-80 Wilson, R.P, Cowey, C B . and Adron, J.W, (1985) effect of dietary electrolyte balance on growth and acid-base balance in rainbow trout (Salmo gairdnerii). Comp. Biochem. Physiol, 82A: 257-260 Wise, A , (1983) dietary factor determining the biological activities of phytate. Nutrition abstracts and reviews, 53: 791-806 Yiu, H , Alstar, I , and Fulcher, R.G. (1983) The effect of processing on the structure and microchemical organization of rapeseed. Food Mocrostrut. 2: 165-173 Youngs, C G , (1991) Technical status assessment of food proteins from canola. In: Canola: research on canola seed oil and meal. Canola Council of Canada, Winnipeg, p. 309-341 Youts, S.E. (1990) Canola, a world class oilseed crop. Proceedings: International canola conferences (2-6 April). Potash and Phosphate Institute of Canada, Saskatoon, Canada. Foundation for Agronomic Research , Atlanta, Georgia, p. 1-8 Yurkowski, M , Bailey, J .K, Evans, R .E , Tabachek, J.-A.L, Ayles, G.B. and Eales, J .G, (1978) Acceptabilty of rapeseed proteins in diets of rainbow trout (Salmo gairdneri). J. Fish. Res. Board Can, 35:951-962 

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:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0089028/manifest

Comment

Related Items