University of New Hampshire University of New Hampshire Scholars' Repository
Master's Theses and Capstones Student Scholarship
Winter 2009
Laver (Porphyra spp) and soy protein concentrate as alternative proteins in diets for juvenile Atlantic cod, Gadus morhua
Abigail B. Walker University of New Hampshire, Durham
Follow this and additional works at: https://scholars.unh.edu/thesis
Recommended Citation Walker, Abigail B., "Laver (Porphyra spp) and soy protein concentrate as alternative proteins in diets for juvenile Atlantic cod, Gadus morhua" (2009). Master's Theses and Capstones. 503. https://scholars.unh.edu/thesis/503
This Thesis is brought to you for free and open access by the Student Scholarship at University of New Hampshire Scholars' Repository. It has been accepted for inclusion in Master's Theses and Capstones by an authorized administrator of University of New Hampshire Scholars' Repository. For more information, please contact [email protected]. LAVER (PORPHYRA SPP.) AND SOY PROTEIN CONCENTRATE AS ALTERNATIVE PROTEINS IN DIETS FOR JUVENILE ATLANTIC COD, GADUS MORHUA
BY
Abigail B. Walker
B.S., Springfield College, 2004
THESIS
Submitted to the University of New Hampshire in Partial Fulfillment of the
Requirements for the Degree of
Master of Science
In
Zoology
December, 2009 UMI Number: 1481705
All rights reserved
INFORMATION TO ALL USERS The quality of this reproduction is dependent upon the quality of the copy submitted.
In the unlikely event that the author did not send a complete manuscript and there are missing pages, these will be noted. Also, if material had to be removed, a note will indicate the deletion. UMT Dissertation Publishing
UMI 1481705 Copyright 2010 by ProQuest LLC. All rights reserved. This edition of the work is protected against unauthorized copying under Title 17, United States Code.
ProQuest LLC 789 East Eisenhower Parkway P.O. Box 1346 Ann Arbor, Ml 48106-1346 This thesis has been examined and approved.
Thesis Director, Dr. David Berlinsky, Associate Professor-of Biological Sciences
UW^^j^ Dr. W. Huntting Howell, ( Professor of Biological Sciences
Dr. Christopher Neefus, Professor of Biological Sciences
\\ J 3o /cPoo^ Date DEDICATION
This thesis is dedicated to my husband, Norbert, who siphoned tanks, sprayed filter pads, lugged buckets and always packed a snack and a pair of dry socks. I am ever grateful for your love and support.
in ACKNOWLEDGEMENTS
I would like to acknowledge Dr. David Berlinsky, my advisor, for his support and guidance at all stages of my research and in the writing of this thesis. Also, I thank my committee members, Dr. W. Hutting Howell and Dr.
Christopher Neefus for their suggestions and encouragement. I would like to thank Dr. Inga Sidor of the New Hampshire Veterinary Diagnostic Laboratory for her generous insight into the proper methods for preparing and analyzing my histological samples and Barbara Brothwell, who prepared the histology slides.
George Nardi and his associates at GreatBay Aquaculture gave helpful guidance throughout my studies. I am indebted to my labmates and to several undergraduates who gave countless hours to help me with animal husbandry and sample collections. In particular, I would like to thank Heidi Colburn, Timothy
Breton, Ryan Brown, Amanda Clapp, Danielle Duquette, and Lauren Wyatt. This research was funded by grants from the UNH Center for Atlantic Marine
Aquaculture and the United Soybean Board. iv TABLE OF CONTENTS
DEDICATION .. . iii
ACKNOWLEDGEMENTS iv
LIST OF TABLES vii
LIST OF FIGURES xi
ABSTRACT...... xiii
CHAPTER PAGE
I. INTRODUCTION 1
RATIONALE 1
GROWTH AND NUTRITION 5
ALTERNATIVE PROTEIN SOURCES 11
MACROALGAE-BASED ALTERNATIVE PROTEINS 15
SOYBEAN-DERIVED ALTERNATIVE PROTEINS 18
HISTOLOGY AND HISTOPATHOLOGY OF SOY-FED ATLANTIC COD
: 24
II. LAVER (PORPHYRA SPP.) AS AN ALTERNATIVE PROTEIN IN DIETS FOR JUVENILE ATLANTIC COD 26
ABSTRACT 26 v INTRODUCTION 26
MATERIALS AND METHODS 29
RESULTS 35
DISCUSSION 41
III. SOY PROTEIN CONCENTRATE AS AN ALTERNATIVE PROTEIN IN DIETS FOR JUVENILE ATLANTIC COD 45
ABSTRACT 45
INTRODUCTION 46
MATERIALS AND METHODS 48
RESULTS 57
DISCUSSION 72
IV. CONCLUSIONS 79
LITERATURE CITED 81
APPENDIX ...92
ANIMAL CARE AND USE APROVAL DOCUMENTATION 92
VI LIST OF TABLES
CHAPTER 1 TABLE 1. Factors Effecting Fishmeal and Fish Oil Use in Aquaculture 2
CHAPTER 2
TABLE 1'. Proximate and Amino Acid Composition of Oven-Dried Porphyra sp.
: 31
TABLE 2. Experimental Diet Formulations 36
TABLE 3. Proximate (% wt. weight) and fatty acid composition (% of fatty
acids) of experimental diets 37
TABLE 4. Amino acid profile of experimental diets (% of sample) 38
TABLE 5. Growth and feed performance in juvenile Atlantic cod fed the experimental diets over 12 weeks (IBW=15.55 ± 0.36 g) 39
vii TABLE 6. Muscle fatty acid composition of juvenile Atlantic cod fed the
experimental diets over 12 weeks. Presented as % of total fatty acids
(IBW=15.55±0.36g)... 40
CHAPTER 3
TABLE 1. Trial 1: Experimental Diet Formulation 49
TABLE 2. Trial 2: Experimental Diet Formulation 50
TABLE 3. Histopathological Scoring 56 TABLE 4. Trial 1: Proximate (% wt. weight) and fatty acid composition (% of
fatty acids) of experimental diets 58
TABLE 5. Trial 2: Proximate (% wt. weight) and fatty acid composition (% of
fatty acids) of experimental diets 59
TABLE 6. Trial 1: Growth and feed performance in juvenile Atlantic cod fed
the experimental diets over 12 weeks (IBW= 87.9 ± 4.9 g) 60
TABLE 7. Trial 2: Growth and feed performance in juvenile Atlantic cod fed
the experimental diets over 12 weeks (IBW=15.92 ± 0.36 g) 60
TABLE 8. Trial 1: Muscle proximate composition of juvenile Atlantic cod fed
the experimental diets over 12 weeks. Presented as g/100 g tissue ...... 62
VIM TABLE 9. Trial 1: Muscle fatty acid composition of juvenile Atlantic cod fed the
experimental diets over 12 weeks. Presented as % of total fatty acids
(IBW= 87.9 ± 4.9 g) 62
TABLE 10. Trial 1: Muscle fatty acid composition of juvenile Atlantic cod fed
the experimental diets over 12 weeks. Presented as g/100 g tissue
(IBW= 87.9 ± 4.9 g) 63
TABLE 11. Trial 1: Hepatic proximate composition of juvenile Atlantic cod fed
the experimental diets over 12 weeks. Presented as g/100 g tissue .....63
TABLE 12. Trial 1: Hepatic fatty acid composition of juvenile Atlantic cod fed
the experimental diets over 12 weeks. Presented as % of total fatty
acids (IBW= 87.9 ± 4.9 g) 64
TABLE 13. Trial 1: Hepatic fatty acid composition of juvenile Atlantic cod fed
the experimental diets over 12 weeks. Presented as g/100 g tissue
(IBW= 87.9 ± 4.9 g)... 65
TABLE 14. Trial 2: Muscle proximate composition of juvenile Atlantic cod fed
the experimental diets over 12 weeks. Presented as g/100 g tissue
(IBW=15.92 ± 0.36 g) 65
TABLE 15. Trial 2: Muscle fatty acid composition of juvenile Atlantic cod fed
the experimental diets over 12 weeks. Presented as % of total fatty
acids (IBW=15.92 ± 0.36 g) 66
IX TABLE 16. Trial 2: Muscle fatty acid composition of juvenile Atlantic cod fed
the experimental diets over 12 weeks. Presented as g/100 g tissue
(IBW=15.92 ± 0.36 g) 66
TABLE 17. Trial 1: Results of Histopathological Scoring 70
TABLE 18. Trial 2: Results of Histopathological Scoring 71
TABLE 19. Number of fish with inflammatory cells present throughout the
intestinal tissues 72
J
X LIST OF FIGURES
CHAPTER 1
FIGURE 1. The dissimilarity diagrams of the A/E ratio profiles among fish: (a)
for dietary essential amino acid requirements; (b) for body amino acid
compositions 10
CHAPTER 3 FIGURE 1. Histopathology sections 54
FIGURE 2. Trial 1: Histomicrographs of juvenile Atlantic cod intestinal
mucosae: (a) fish fed a soy-free control diet with a total inflammation
score of 5, (b) fish fed a soy-free control diet with a total inflammation
score of 12, (c) fish fed the SPC40 diet with a total inflammation score
of 7, and (d) fish fed the SPC40 diet with a total inflammation score of
12 68
FIGURE 3. Trial 2: Histomicrographs of juvenile Atlantic cod intestinal
mucosae: (a) fish fed a commercially available control diet with a total
inflammation score of 5, (b) fish fed a commercially available control diet
with a total inflammation score of 12, (c) fish fed the SPC50 diet with a
total inflammation score of 6, and (d) fish fed the SPC50 diet with a total
inflammation score of 9 69
XI ABSTRACT
LAVER {PORPHYRA SPP.) AND SOY PROTEIN CONCENTRATE AS ALTERNATIVE PROTEINS IN DIETS FOR JUVENILE ATLANTIC COD, GADUS MORHUA
by
Abigail B.Walker
University of New Hampshire, December, 2009
Fishmeal is the main ingredient in formulated fish feeds, and the cost of this commodity is the largest and most volatile recurrent expenditure by the finfish aquaculture industry. Additionally, the reliance on wild-harvested forage species raises concerns about the industry's ecological impacts. Marine macroalgae and soy-derived feed ingredients both have merits as alternative proteins for piscivorous marine fish. Atlantic cod, in particular, have a highly adaptable digestive system and perform well with alternative protein diets.
Atlantic cod, Gadus morhua, juveniles (initial body weights: 15.6 g, 87.9 g, and 15.92 g) were fed iso-nitrogenous, iso-caloric diets to evaluate two alternative proteins. In the first experiment, purple laver {Porphyra spp.) replaced
0%, 15%, or 30% of the fishmeal in a commercially available marine finfish diet.
In the second experiment, diets contained soy protein concentrate (SPC) to provide 0%, 10%, 20%, 30% or 40% dietary soy protein or to replace 0%, 25%, and 50% of the fishmeal in a commercially available diet for marine finfish. No
xii differences in survival, growth, or hepatic-somatic indices were found among any
of the treatment groups. The fish attained an average weight of 41.0 g with a
specific growth rate (SGR) of 1.19% and a feed conversion ratio (FCR) of 1.26 in the first experiment. In the second experiment, the fish attained average weights of 162.7 g in the first trial and 39.1 g in the second trial with SGRs of 0.76% and
1.12% and FCRs of1.28 and 1.29, respectively. These results indicate that soy protein concentrate can entirely replace fish meal in diets for juvenile Atlantic cod. Additionally, SPC or Porphyra can be combined with other common feed ingredients to replace dietary fishmeal by at least 50% or 30%, respectively.
XIII CHAPTER 1
INTRODUCTION
The Food and Agriculture Organization of the United States has predicted that the aquaculture industry will continue to expand globally for some time, while many fisheries will remain static (FAO 2002; FAO 2009). These fisheries include those for the clupeids (Clupea sp., Brevoortia sp., and Ethmidium sp.j, anchovy
(Engraulidae sp.J, and capelin (Mallotus villosus), which are processed into fishmeal and incorporated into formulated fish feeds. The inclusion of fishmeal in aquaculture diets will, therefore, become less economical in the future.
Innovations in feed composition with regard to the quantity and sources of dietary proteins are necessary for the aquaculture and aquatic feed industries to become sustainable, and are essential if these industries are to continue to expand at their current rate (Gatlin et al. 2007). Fishmeal is the main ingredient in formulated fish feeds, and the cost of this commodity is the largest and most volatile recurrent expenditure by the finfish aquaculture industry (De Silva and
Anderson 1994; Naylor et al. 2009). For this reason, the profitability of aquaculture worldwide is inextricably tied to the fishmeal market. The cost of fishmeal is highly variable from season-to-season and year-to-year. In 2006, fishmeal attained its highest price ever. The Norwegian price paid in July of that year was $1600 per metric ton and was 2.6 times more costly than in the
l previous year (Hardy 2006). This unprecedented leap in the market price was
due to the combined effects of an El Nino weather pattern, hurricane Katrina, and
the cyclic collapse of capelin, menhaden, and anchovy fisheries around the world
(Wilson 2002). The price for fishmeal was relatively stable in 2007, but rose again in 2008 due to high demand, most notably from China and other Asian countries (57% of all fishmeal consumed; FAO 2009).
Table 1
Factors Affecting Fishmeal and Fish Oil Use in Aquaculture (adapted from Naylor et al 2009)
Reduces Use Increases Use
• development of substitutes for » consumer awareness of the health fishmeal in feed formulations benefits of n-3 fatty acids • expanded culture of lower-trophic <» reduced consumption of beef and level species other terrestrial meats • consumer and industry awareness <» consumer concern over high of sustainability concerns mercury levels in wild fish • development of sustainability » expansion of marine aquaculture certification for farmed fish • concerns regarding PCBs, dioxins, <• increased use of formulated feeds and other fish feed contaminants to grow omnivorous species • volatile prices of reduction fishery <» slow adoption of alternative and vegetable commodities formulations by feed manufacturers > volatile prices of reduction fishery and vegetable commodities
The reliance on fishmeal also raises concerns about the aquaculture industry's impact on the environment and consumer's acceptance of farmed fish.
Table 1 outlines several factors which affect worldwide consumption of fishmeal.
The large majority of fishmeal comes from a dedicated reduction fishery, focused on small planktivorous forage species, that totals 1/4 to 1/3 of the annual harvest of fish worldwide (20-30 million metric tons; Naylor et al. 2009). When these
2 wild-caught fish are fed to farmed carnivorous species, the energetic efficiency
inherent in the change in trophic level dictates that an larger biomass of wild- caught fish is required to grow a relatively smaller mass of aquaculture product.
This "fish-in to fish-out" ratio (FI/FO) averaged 1.04 in 1995 and, due in large part to the replacement of fishmeal with alternative proteins, decreased by 39% to
0.63 in 2007. These values include omnivorous and herbivorous fish, and fish grown extensively. In intensive aquaculture of carnivores, however, 2 to 5 times more protein is required to grow fish than is harvested from those fish. Currently, the FI/FO ratio for farmed Atlantic salmon (Salmo salar) is 5.0 (Naylor et al.
2009).
The future of the aquaculture industry also depends on the development of culture methods for new and niche species that are well-suited to local growing conditions and consumer markets. Atlantic cod, Gadus morhua, is regionally important in New England and has constituted a major component of the northeast Atlantic fishery for as long as the fishery has been in existence. Due to population declines and fishing restrictions, culture of Atlantic cod has increased significantly in North America and Western Europe. Currently, cod production in
Norway totals 15,000 metric tons and is predicted to exceed 300,000 metric tons in the next few years (Jorstad et al. 2008). Because Atlantic cod can be grown in net pens that were designed for culture of Atlantic salmon, the decreasing market price for salmon has caused some salmon farmers to initiate the culture of
Atlantic cod.
3 Considerable research has recently been conducted on cod nutrition,
including assessment of their ability to digest and assimilate diets containing
alternative proteins (Aksnes et al. 2006; Albrektsen et al. 2006; Forde-Skjaervik et al. 2006; Hansen et al. 2006; Refstie et al. 2006a; Tibbetts et al. 2006; Hansen et ai. 2007a; Karalazos 2007). These studies indicate that cod have high apparent digestibility coefficients (ADCs) for several protein sources, including several derived from plants, and grow well over a wide dietary protein range (36-
66%) (Hemre et al. 2004; Tibbetts et al. 2006; Hansen et al. 2007a).
One main objective in the study offish nutrition is to reduce the aquaculture industry's reliance on fishmeal by identifying fishmeal replacements derived from more sustainable sources of protein. A second objective is to expand the application of alternative diets to new species. The experiments reported in this thesis addressed these industry priorities by meeting the following objectives:
Objective 1: To assess Porphyra as a fishmeal replacement in diets for
juvenile Atlantic cod (Gadus morhua)
Objective 2: To assess soy-protein concentrate as a fishmeal replacement
in diets for juvenile Atlantic cod (Gadus morhua)
4 Growth and Nutrition
While Atlantic cod may tolerate a wide range of dietary compositions, commercially formulated diets for Atlantic cod and other marine carnivores typically contain 50-55% protein in the form of herring or menhaden-derived fishmeals (Refstie et al. 2006a). These diets are, by design, low in dietary starch and typically contain terrestrial ingredients only in the form of binding agents
(glutens) or wheat flour fillers.
The dietary protein requirements for Atlantic cod, as with all fish, can be evaluated from two perspectives. First, the dietary protein requirement is the amount of crude protein in the diet that maximizes growth. Second, it is the balance of essential and non-essential amino acids that provide sufficient and efficient protein synthesis. Progress is being made to improve knowledge offish nutrition and ideal feed formulations from both perspectives (Wilson 2002).
Of the 20 amino acids common to all proteins, ten have been identified as essential for growth in fish species. The remaining ten are considered non essential because they can be synthesized within the animal from other metabolic intermediates. Amino acids are processed by the fish in three ways.
They can be used as energy for physiological functions, as building blocks for protein synthesis, or they can be transformed into another amino acids or compounds such as citric acid cycle intermediates. Most amino acids are degraded for energy (Cowey 1994; De Silva and Anderson 1994).
5 Amino acids are available or sequestered in two separate and ever-
exchanging pools. They exist as free amino acids in the blood and tissues or are
incorporated into the proteins that comprise body tissues. The synthesis of
proteins from amino acids, and the degradation of proteins into amino acids
occurs continuously, as does the movement of amino acids between the tissues
and the blood. Fish obtain less than half of the amino acids required for protein
synthesis from the catabolism of body tissues. The majority of free amino acids
come from dietary sources (Lyndon et al. 1993; De Silva and Anderson 1994).
Individual amino acid requirements have be studied for the most
commonly farmed species including salmon and trout species, channel catfish,
common carp, Nile tilapia, milkfish, Japanese eel, red drum, yellowtail, Japanese
flounder and white sturgeon (Akiyama et al. 1997). Generally, these
requirements are established by evaluating the species' growth response to
graded levels of amino acids in test diets (Wilson 2002). Due to high variability in
the published results, both among and within species, the actual requirements for
many species are questionable (Akiyama et al. 1997). In Atlantic cod, the amino
acid requirements are currently unknown (Lyndon et al. 1993).
The estimation or determination of the amino acid requirements of a given
species is typically carried out with one of two approaches. The first is to
evaluate the requirement of one particular amino acid in exclusion of all others.
Typically, dose-response or graded test diets are formulated such that all amino
acid levels are held constant except the one being tested. The diets are
evaluated by plotting the amino acid concentration against the measured weight
6 gain. The relationship is generally linear and then plateaus at the required concentration. At this point, the excess amino acid is no longer used for protein synthesis and does not contribute to growth (Wilson 2002). In these tests, diet formulation is crucial. Test diets that use free synthetic amino acids can overestimate the required amount. Dabrowski and Guderley (2002) demonstrated that fish grew more slowly when fed a diet that included crystalline amino acids and casin as compared to complete protein diets. Because growth efficiency trials are both expensive and labor intensive, they have been conducted only for the most commercially important species (Wilson 2002).
A second method for estimating amino acid requirements measures the effect of dietary amino acid levels on the amount of free amino acids in the serum, blood, or muscle tissue. The hypothesis is that plasma levels remain low until the requirement of a given amino acid is met, at which point plasma levels rise. This method has been useful in a few cases, but does not hold for all fish species or all amino acids. For example, this design gave credible results for lysine, threonine, histidine, and methionine in channel catfish, but arginine did not fit the pattern. This amino acid rose steadily in the blood and muscle and no specific requirement was indicated. In striped bass, lysine and methionine requirements could be established, but results for other amino acids were variable in this and other species (Wilson 2002).
An interesting, but more difficult test for amino acid requirements evaluates amino acid oxidation. Studies using amino acid oxidation aim to determine the quantity of dietary amino acid that is converted into CO2. The
7 hypothesis is that when dietary amino acids are below the required level, they
are used almost entirely for protein synthesis. When the requirement level is
met, excess amino acids are used for energy and are incorporated into the
respired C02 (Akiyama et al. 1997). Amino acid oxidation studies measure C02
production from 14C labeled amino acids to determine the amount of amino acid oxidized, and a dose-response curve is used to evaluate the requirement.
Results produced using this technique for lysine were very similar to the levels obtained from growth data for that amino acid. A study to evaluate tryptophan requirements, however, gave a lower than expected result. The consensus from these studies is that oxidation studies lack precision (Wilson 2002).
A different approach, the Ideal Protein Concept, is borrowed from swine nutrition and focuses on the profile of all essential amino acids, or on the ratio of essential to non-essential amino acids. This method postulates that there is a correlation between the amino acid profile required for maximum growth and the pattern of amino acids in the whole fish flesh (Akiyama et al. 1997). This theory is appealing because the profile in whole body tissues offish are fairly conservative among species and are not affected by the various environmental factors that might impact growth studies (Wilson 2002).
The Ideal Protein Concept can be applied only if the requirement for one amino acid is known. The ratio (A/E) applied to whole body tissue, is:
A/P _ content of each essential AA total essential AA content (including cystine and tyrosine)
8 The known amino acid is often lysine. The requirements of other amino acids are determined relative to the known amino acid. The accuracy of the amino acid analysis of the whole-fish tissue is of utmost importance in this method
(Akiyama et al. 1997; Wilson 2002).
Akiyama et al. (1997) compiled published data to compare the essential amino acid requirements using the A/E ratio among several species. They plotted their results using dissimilarity diagrams (Figure 1). There was much more variability among species when requirements were determined using dose- response testing than when compared using calculated values derived from A/E ratios. This is as expected since there is limited variability in the tissue amino acid profile among fish species. The dose-response derived requirements also showed results which were more consistent within phylogenetic categories
(Wilson 2002).
9 Milkfish. Chinook (K} me sataon •(a) •Japanese^ ffiapia | Cherry Japanese eei eel *"
Nile tilapia Channel catfish salmon »n»1,B,fc (Jh Common carp Catla Coi» salmon Figure 1. Chum salmon Chinook salmon The dissimilarity diagrams of the A/E ratio profiles among fish: (a) for dietary essential amino acid requirements; (b) for body amino acid compositions. From Akiyama etal. 1997, used with permission. The requirement for essential amino acids is further complicated by the role of non-essential amino acids. For example, cystine is a non-essential amino acid that is made in the degradative pathway of methionine. Likewise, tyrosine is made from phenylalanine. For these amino acid pairs, the requirement for the essential amino acid is greater if the non-essential amino acid is not supplied (De Silva and Anderson 1994). If non-essential amino acids are not provided by the diet, they can be made using the essential amino acids as non-specific nitrogen sources. This mechanism is often inefficient and requires more essential amino acids from the diet. Peres and Oliva-Teles (2006) found that European sea bass attained maximum growth at an essential to non-essential amino acid ratio of 50/50. In this study, maximum retention of nitrogen occurred at a ratio of 60/40 10 and the least efficient nitrogen retention was at a ratio of 40/60. Similar ratios of 0.55 to 0.66 are found in terrestrial animals, however, unlike terrestrial species, voluntary food intake increases in fish as essential to non-essential amino acid ratios decrease (Peres and Oliva-Teles 2006) Alternative Protein Sources Without clear guidelines for the specific amino acid requirements of farmed fish, the aquaculture industry must rely on sources of crude protein that are as balanced as possible. In keeping with the ideal protein theory, fish meal is the most suitable protein for use in marine fish feeds because it is highly digestible and has amino acid and fatty acid profiles that are similar, not only to those of natural prey items, but also to the desired final fillet. Additionally, the cost of fishmeal has yet to become cost prohibitive in commercial aquaculture feeds (Wilson 2002; Hansen et al. 2007a). Current trends in fish nutrition research have focused on alternative protein sources to reduce reliance on the costly and variable fishmeal market. The nutritional value of a protein source is evaluated using several measurements. The protein efficiency ratio (PER), is the ratio of wet weight gained to wet weight of crude protein fed. This ratio does not take into account the protein required for physiological maintenance and repair (De Silva and Anderson 1994). A more accurate measurement is the net protein utilization (NPU). 11 „T„TT final body protein - initial body protein inn NPU = — — x 100 total protein fed Several alternative sources of protein have been suggested and tested for effectiveness as replacements for fishmeal. A desirable foodstuff would be high in protein with a well-balanced amino acid profile. Further, it would be low in fiber and starch, free of anti-nutritional factors, highly digestible, and sufficiently palatable to fish. From a production standpoint, potential fishmeal alternatives would also be readily and continuously available and would be amenable to shipping, storage, and pellet formation, all at a low cost (Gatlin et al. 2007; Hansen et al. 2007a). Some alternatives are more appropriate for herbivores than carnivores, and no single protein offers the ideal amino acid and digestibility profile. Combinations of various protein sources have also shown potential as fishmeal replacements (Naylor et al. 2009). The protein sources which offer the most similar amino acid profile to fishmeal suffer the same problems related to cost and availability. Krill and squid meals are very effective but are not yet harvested in quantities that would be commercially useful. Fisheries by-catch has hardly been considered because it has the same seasonality and market restrictions as fishmeal. By-products and silage from food-fish processing have nutritional profiles that are nearly equivalent to standard reduction fish meals, however, the infrastructure necessary to process them into meals of consistent availability and quality do not yet exist (Trushenski et al. 2006; Naylor et al. 2009). Further, the occurrence of 12 PCBs and dioxins in fish processing by-products represent regulatory and monitoring constraints for these ingredients (Naylor et al. 2009). Slaughterhouse waste from terrestrial livestock, while plentiful, has low palatability and undesirable amino acid profiles. Beef products including meat and bone meal have been voluntarily avoided or banned since the outbreaks of bovine spongiform encephalopathy in the early 1990's (Forster et al. 2003; Friedland et al. 2009). Silk-worm meal and earth worms are effective from a nutritional standpoint; but are also of limited quantity (Trushenski et al. 2006). Plant-based protein sources offer the most potential due to the availability and predictability of the supply, as well as the modest expense (Moyano et al. 1992; Davies et al. 1997b; Kikuchi 1999; Pereira and Oliva-Teles 2003; Hansen et al. 2006). Plant proteins, however, have some undesirable qualities, especially for carnivorous fish (Trushenski et al. 2006). Plant-derived options include the well-studied soybean meal (SBM), corn and wheat gluten meals, barley, cottonseed, rapeseed, and lupine meals. Fishmeal replacement with terrestrial plant proteins has been associated with reduced growth, increased feed conversion ratio (FCR), and reduced protein utilization in marine fish, including Atlantic cod (Von der Decken and Lied 1993; Albrektsen et al. 2006; Refstie et al. 2006a; Hansen et al. 2007a), Japanese flounder (Paralichthys olivaceus; Kikuchi 1999; Choi et al. 2004; Deng et al. 2006), gilthead sea bream (Sparus aurata; Sitja-Bobadilla et al. 2005) and Atlantic salmon (Krogdahl et al. 2003). This reduced feed performance may be explained by low nutrient digestibilities due to the presence of antinutritional factors (Francis et al. 2001) 13 and/or increased dietary fiber. Unbalanced amino acid profiles, low palatability, and the relatively high costs related to undeveloped supply chains can further limit the inclusion of plant-derived ingredients (Sanz et al. 2000; Deng et al. 2006). Depending on the species in question, plant proteins may require processing to remove anti-nutritionai factors, which include phytic acid and protease inhibitors that reduce the digestibility and nutritional value of the protein source (Trushenski et al. 2006). When used, plant proteins are generally combined with supplemental amino acids to counteract nutritional deficiencies (Deng et al. 2006). Amino acid supplementation in any aquaculture diets must be done with care, as the uptake of synthetic free amino acids is not the same as the uptake of amino acids from intact proteins. When the diet is digested, the free amino acids reach the protein synthesis sites first and the amino acids supplied by the dietary proteins are delayed. Lacking the proper amino acid ratios for protein synthesis, the crystalline-supplied amino acids are disproportionately catabolized for energy. Terrestrial plant proteins also fall short of providing the essential fatty acids (EFAs) necessary for fish health and quality fillets. The nutritive value of marine fish as a human food source is due in large part to the highly unsaturated n-3 fatty acids found in marine fish oil. These desirable fatty acids, including 20:5n-3 (eicosapentaenoic acid, EPA) and 22:6n-3 (docosahexaenoic acid, DHA), are considered essential and, therefore, must be obtained from the diet (Sargent et al. 2002). In humans, dietary polyunsaturated fatty acids have been shown to reduce the risk of death from coronary heart disease and are especially 14 important in visual and cognitive gestational development (Mozaffahan and Rimm 2009). Typically, the nature of dietary lipids is reflected in the fatty acid composition of the tissues. Since n-3 fatty acids are rare in terrestrial oils, marine fish fed these oils produce fillets that are higher in n-6 fatty acids, especially 18:2n-6 (Bell et al. 2003). Starving the fish of n-3 fatty acids may not only reduce the quality of the final product, but may also impact fish health, as the EFAs are important contributors to membrane structure and fluidity, egg quality, and disease resistance (Sargent et al. 2002). N-3 fatty acid deficiency can manifest as a swollen, fatty liver with reduced hematopoietic function resulting in anemia (Roberts 2002). Changes related to dietary fatty acid composition are most apparent when the alternative protein contains a significant lipid fraction. For example, Atlantic cod develops increased concentrations of 18:2n-6 and decreased EPA in the fillet when fed diets containing full fat SBM (Karalazos 2007). It is rare to see these changes, however, when the fish are fed protein soybean concentrates or isolates, which have a very low lipid content. Macroalgae-Based Alternative Proteins Macroalgae and single-celled microalgae are particularly attractive options for use as partial fishmeal replacements in diets for marine fish, because they contain lipids with significant quantities of n-3 fatty acids. Unless the lipids are removed for other purposes (e.g. extraction for fuel production), lipid rich algae have the potential for replacing fish oil as well as fishmeal. This is especially important because the manufacture of some aquaculture feeds, including those for Atlantic salmon, rely more heavily on the reduction fishery to supply fish oil 15 than fishmeal. For example, if the inclusion of fish oil in Atlantic salmon diets is reduced by only 4%, the FI/FO for this species is reduced from 5.0 to 3.9, whereas the same reduction in fishmeal would have a negligible effect (5.0 to 4.8; Karalazos et al. 2007; Naylor et al. 2009). Research has shown that the effects of incorporating algae into fish diets are specific to both the type of algae and fish species (Davies et al. 1997a). Several algae species have been evaluated in terms of their amino acid and fatty acid profiles. The compositions of wild-collected Ulva lactuca L. (Chlorophyta, CP: 17.6), Enteromorpha compressa L. Greville (Chlorophyta, CP: 13.6), Padina pavonica L. Thivy (Phaeophyta, CP: 17.4) and Laurencia obtusa Hudson, Lamouroux (Rhodophyta, CP: 24.5) suggest that marine algae from various genera may be used as alternative protein sources (Wahbeh 1997). Amino acid deficiencies and/or high n-6 to n-3 fatty acid ratios found in some macroalgae, however, suggest that these ingredients may need to be used in combination with each other, with other protein sources, or with nutrient supplements. Practical studies indicated that algal inclusion in fish feeds may increase intake and feed efficiency in excess of that for standard ingredients (Mustafa and Nakagawa 1995). Inclusion of algae as a dietary protein was demonstrated with red sea bream (Pagellus bogaraveo; Mustafa and Nakagawa 1995), Japanese flounder (Xu et al. 1993), and rainbow trout (Oncorhynchus mykiss; Soler-Vila et al. 2009). In diets made for fingerling red sea bream, supplemental algae inclusion was shown to improve carbohydrate and protein utilization (Mustafa et al. 1995). 16 These researchers found that replacing 5% of the ingredients in a moist-type diet with Ascophyllum nodosum, Porphyra yezoensis, or Ulva pertusa had favorable results. P. yezoensis preformed most favorably and improved growth and feed efficiency. Mustafa etal. (1995) suggested that diets supplemented with algae may stimulate growth, metabolism, and the absorption of nutrients. Valente et. al. (2006) studied the efficacy of including the macroalgae Gracilaria bursa- pastoris, Ulva rigida, and Gracilaria cornea in diets for European sea bass. They found no negative impact on growth performance or body composition when fish protein hydrolysate was replaced by 10% G. bursa-pastoris or U. rigida or by 5% G. cornea. Because the composition of algae varies notably among species, much more research is required to evaluate marine macroalgae as potential protein replacement. Mustafa and Nakagawa (1995) evaluated macroalgae as a supplement (5% replacement), rather than a large percentage replacement, and indicated that Porphyra had some beneficial properties including stimulation of protein and lipid metabolism. A higher algal incorporation study was performed by Davies et al. (1997a) in diets for the thick-lipped gray mullet (Chelon labrosus). In this study the fishmeal component was replaced by 16 or 33% in iso-energetic, iso- nitrogenous pelleted feeds. They found a reduction in feed performance at the higher inclusion level and suggested that reduced digestible energy may have accounted for the decreased growth and feed utilization. The natural diet of this species is not well known, however. They may be omnivorous, herbivorous or planktivorous, thus making interpretation of the study difficult. 17 With the exception of the aforementioned bream species, there has been little research directed at evaluating the applicability of marine macroalgae in diets for marine carnivores such as Atlantic cod. Aside from amino acid and fatty acid profiles, the digestibility of the carbohydrate fraction is likely to be a limiting factor for macroalgae inclusion. Porphyra contains complex polysaccharides including the sulphated galactan known as porphyran. To further complicate evaluation of potential seaweed ingredients, the standard measurement for the amount of available non-structural carbohydrate, nitrogen-free extract, may not be applicable to macroalgal analysis due to high relative concentrations of gel polysaccharides and water-soluble fiber (Davies et al. 1997a). Heat processing of the algal meal during pellet formation could potentially improve the digestibility of complex carbohydrates, as it does for other plant-derived ingredients, should digestibility present a problem. Soybean-derived Alternative Proteins Soybean protein is perhaps the best example of an alternative protein source that has both beneficial and problematic nutritional properties. Due to their abundance, relative low cost, high protein levels, and sufficient amino acid profiles, soy-derived ingredients have received substantial attention from fish nutrition researchers and feed manufacturers. In 2005, the demand for SBM for aquaculture feeds was estimated at 5 million metric tons (Gatlin et al. 2007). The United Soybean Board has been particularly effective in researching, promoting, marketing, and commercializing soybean-derived feed ingredients for aquaculture markets. Soy-derived feed ingredients include full-fat and defatted 18 SBM (CP: 35-40% and 45-50%, respectively), soy protein concentrates (CP: 6.5- 80%, SPC) and soy protein isolates (CP: 86%, SPI; Storebakken et al. 2000; Tibbetset.al.2006). With the exception of methionine, and to a lesser extent lysine, soy proteins typically have amino acid concentrations that meet or exceed those of fishmeal. The methionine deficiency is partially mitigated by high levels of cysteine and may be further met with the addition of crystalline dl-methoionine (Storebakken et al. 2000a). Defatted SBM (DSBM) is palatable to most fish species. In juvenile Atlantic cod, DSBM has an ADC of 92%, with higher values for soy protein concentrate and soy protein isolate (99% and 97%, respectively). Atlantic cod also have high lipid ADCs for soy-derived proteins with values of 88%, 95%, and 92% for the SBM, SPC, and SPI ingredients, respectively (Tibbetts et al. 2005). Soybean meal inclusion limits were determined by decreased growth and feed conversions at levels above 20% in diets for yellowtail (Serbia quinqueradiata; Viyakarn et al. 1992; Watanabe et al. 1992), 50% of total protein for juvenile red drum (Sciaenops ocellatus; Reigh and Ellis 2002), and 66% of crude protein from dehulled soybean meal for juvenile seabass (Dicentrarchus labrax; Pike et al 1990). These limitations may be due to high fiber in the diet or to the presence heat-sensitive anti-nutritional factors including lectins and protease inhibiting molecules that reduce the fish's ability to efficiently utilize dietary protein (Gatlin et al. 2007). For example, high levels of non-starch polysaccharides (NSPs) in soy-derived ingredients can impact the osmolality of 19 the gut contents, resulting in increased fecal moisture content. This condition can reduce feed performance by decreasing the uptake of nutrients and increasing the rate of gut passage. Arabinogalactan, a particularly viscous fraction of the NSP is thought to induce elevated fecal water in salmonids and may contribute to reduced fat and protein digestibility (Olli et al. 1994a; Refstie et al. 1997; Storebakken etal. 1998). Perhaps the most noteworthy effect from SBM has been the induction of an inflammatory condition in the intestines of Atlantic salmon. This condition was described by Baeverford and Krogdahl (1996) as "soybean-induced non infectious subacute enteritis" The cause of the inflammation was determined to be the alcohol-soluble fraction of SBM (Ingh et al. 1996), of which 45% is nitrogen-free extract. Knudsen et al. (2007) phase-fractioned soybean molasses, the byproduct that results from alcohol washing of SBM, and found saponins to be the most likely compounds for inducing enteritis in salmon. Saponins are compounds produced by wild plants that are known to disrupt the integrity of biological membranes when consumed (Knudsen et al. 2007). The occurrence of soybean-induced enteritis, while common in Atlantic salmon, has only been described in one study in Atlantic cod (Olsen et al. 2007). Atlantic cod are accepting of soy proteins, including fat extracted SBM in combination with other protein ingredients. An early study by Von der Decken and Lied (1993) replaced 10, 20, and 30% of fishmeal with full-fat soybean meal and found that at up to a 20% replacement there were no effects on growth. At 30%, feed intake was reduced by 6% and growth by 67%. In a growth study with 20 larger (220g) fish, Karalazos et al. (2007) replaced 12%, 24%, and 36% of fish meal with full fat SBM and found good growth over all, but reduced specific growth rates (SGR; 0.89-1.08) and increased FCRs (0.68-0.84). These changes occurred at all levels of soy inclusion relative to the control diet, but there were no differences in these parameters among the soy diets. The authors suggested that because the effects of the soy diet were independent of inclusion level, and were seen at lower levels than in Von der Decken and Lied's study, the results may indicate low palatability of the soy diet or other sensitivity of the fish to the change in diet during the first few weeks. Karalazos et al. (2007) also examined the effects of the soy diet on muscle fatty acids and found marked changes relative to controls. The muscles of soy-fed fish were characterized by increased 18:3n-6 and reduced 20:1 n-9, 22:1 n-11, and 22:1n-9 as well as decreased EFAs. Higher levels of EFAs in the muscles relative to the diets, however, did show preferential storage of these fatty acids. Hansen et al. (2006) used a regression-based experimental design to evaluate the effects of a mix of plant materials in 15 extruded diets for juvenile cod (mean weight 139 g). In four of the diets, solvent extracted SBM replaced fishmeal at levels of 4-16%; in four diets soy was combined with corn gluten at increasing levels; and in two diets, SPC and wheat gluten (WG) were used in combination. No differences in growth were found among any of the experimental diets. FCRs ranged from 0.96 to 1.01 for the SBM diets with a general increase in FCR with increasing plant proteins. When coupled with the consistent growth numbers, the results demonstrated that the fish were able to 21 meet their demands for growth by eating more of the diets that were high in plant protein. Protein digestibility levels in this study (77.3-83.0%) were low compared to Tibbetts et al. (2006) but there was no enteritis and the diets containing a mix of SPC and WG had protein and fat digestibilities that were as good as or better than those for the fishmeal diets. Two studies (Hansen et al. 2007b; Olsen et al. 2007) also used a regression design to investigate the effects of replacing fish meal with plant proteins in diets for adult cod in sea cages (mean initial body weight, IBW: 1,652 g). The researchers replaced.0, 25, 50, 75 and 100% of fishmeal with a mixture of plant proteins (PP). The plant protein mixture was 14% bio-processed SBM (BPSBM), 36% SPC, and 50% WG. Growth and protein efficiency were not affected with up to a 50% PP mix, but above that level, reductions in growth, and both feed and protein efficiency were observed. Compensatory feed intake was noted in this and other studies (Refstie et al. 2006a; Refstie et al. 2006b). At 100% replacement, reduced appetite, possibly due to decreased palatability, brought fish close to "food-deprivation status." Olsen et al. (2007), using the same fish as above, evaluated the effects of PP on various health indicators including blood chemistry, fecal moisture content, intestinal histology, and heat shock protein (HSP70) transcription. With elevated PP, they found an increase in fecal moisture content coupled with increased cellularity of the lamina propria and hypertrophy and hyperplasia of the goblet cells. Cellularity refers to the number of nuclei Within the connective tissue of the lamina propria and increased cellularity may indicate increased presence of 22 lymphocytes. This condition peaked in the fish fed the 100% PP diet. A true enteritis-type condition, the only incidence reported in Atlantic cod to date, was observed in two of these fish. Three studies (Forde-Skjasrvik et al. 2006; Refstie et al. 2006a; Refstie et al. 2006b) investigated the effects of 24% (of crude protein) fishmeal replacement by BPSBM, in comparison to the same replacement by standard SBM in diets for 1- and 2- year old cod. In the BPSBM, the a-galactosides were largely removed but some NSPs remain. The researchers found no difference in growth among dietary treatments, but found decreased digestibility of amino acids, protein, and energy in the soy diets. This decreased digestibility was met with compensatory feed intake and gastrointestinal growth which resulted in lowered FERs. The researchers also observed increased moisture in the feces but found no changes in the morphology of the intestinal mucosa. Restie et al. (2006b) described an increase in the size of the stomach, pyloric intestine and mid intestine in fish fed SBM, but only the pyloric intestine increased in those fed BPSBM. Increased fecal moisture was observed in both standard and bio- processed SBMs. Olsen et al. (2007) asserted that the increased fecal moisture content ("diarrhea-like condition") observed in their study may have been linked to the "osmotic influence" of indigestible soybean a-galactosides or NSPs. Although not directly measured (calculated from 100% - % protein - % fat - % starch - % ash - % water), the proximate analysis of the diets in their study suggest a ten fold increase in NSPs between the control and their 100% plant protein diet. 23 Histology and Histopatholoqy of Soy-fed Atlantic cod The gross anatomy of the Atlantic cod digestive system differs from that of Atlantic salmon in several ways. The cod intestine has more pyloric caeca, which increase the overall surface area of the intestine and provide more opportunity for nutrient absorption. The pyloric caeca have not yet been linked to any inflammatory reaction due to dietary plant proteins (Hansen et al. 2006), The intestines of Atlantic cod also lack the regular arrangement of basal nuclei and apical absorptive vacuoles that describes healthy epithelial mucosa in salmonids. Instead, the intestinal vaculae are dominated by secretory goblet cells (Morrison 1987). Soybean-induced enteritis in salmonids disrupts the regular basal arrangement of the enterocyte nuclei and presents as a loss of absorptive vacuoles that limits the absorptive capacity of the intestines (Ingh et al. 1991). In cod in contrast, Olsen et al. (2007) reported hypertrophy and hyperplasia of the goblet cells in cod fed diets composed 75% or 100% dietary plant proteins, with no mention of absorptive vacuoles. Refstie et al. (2006b) found no reduction in the absorptive capacity of the intestines of SBM-fed Atlantic cod. Olsen et al. (2007) identified that cod, like salmon, develop thicker and shorter mucosal folds, with increased width and cellularity of the lamina propria. Though this increased cellularity was not described further, an inflammatory reaction to dietary SBM would likely be marked by an infiltration of white blood 24 cells. Olsen et al. (2007) did report an increase in intra-epithelial lymphocytes (lELs) with high levels of dietary plant proteins. Finally, Hansen et al. (2006) observed colonies of rod-shaped bacteria along the brush boarder in the distal intestine of adult cod. These colonies may play a role in the digestion of plant materials. The gut flora may also have a protective role in reducing the potential for infection by pathogenic bacteria that enter the body via the gastrointestinal tract. 25 CHAPTER 2 LAVER (PORPHYRA SPP.) AS AN ALTERNATIVE PROTEIN IN DIETS FOR JUVENILE ATLANTIC COD Abstract Juvenile Atlantic cod Gadus morhua (initial mean weight: 15.6 g) were fed three iso-nitrogenous, iso-caloric diets (55% protein, 16% fat, calculated gross energy = 20.5 MJ/kg) containing dried Laver (nori) Porphyra spp. (red algae) to replace 0, 15, and 30% of the fishmeal in a commercially available marine finfish diet. After 84 d, no differences in survival, growth, or hepatic-somatic indices were found among the treatment groups. The fish attained an average weight of 41.0 g with a specific growth rate (SGR) of 1.19% and a feed conversion ratio (FCR) of 1.26. The fatty acid composition of the muscle was similar among treatment groups, except arachidonic acid levels were greater in fish given diets with 30% Porphyra replacement. These results indicate that Porphyra, at levels up to 30%, appears to be a suitable fishmeal replacement in juvenile cod diets. Introduction The aquaculture industry currently consumes an estimated 34% of the global fishmeal produced and will consume 48% by the year 2010 (FAO 2006). 26 The high cost and uncertain supply of fishmeal, however, necessitate incorporation of alternative protein sources in finfish diets for sustainable growth of the industry. Some considerations in the selection of alternative protein sources include amino acid and fatty acid profiles, digestibility, palatability, availability, and processing costs (Gatlin et al. 2007). Terrestrial plant proteins, particularly soybean products, have been included in the diets of several finfish species because of their abundance and relatively low cost. Their inclusion may be limited, particularly in carnivorous species, because of amino acid and fatty acid imbalances and presence of complex carbohydrates and antinutritional factors (Vielma et al. 2003; Gatlin et al. 2007). Increasing attention has been focused on the use of marine macroalgae (seaweeds) as possible ingredients in fish diets because of their elevated protein content and favorable fatty acid profiles. In contrast to terrestrial plant proteins, seaweeds are relatively high in essential n-3 polyunsaturated fatty acids such as docosahexaenoic acid (DHA; 22:6n-3, where 22 is the number of carbon atoms, 6 is the number of double bonds, and 3 is the position of the first double bond from the methyl end) and eicosapentanoic acid (EPA; 20:5n-3), which are important both for maintaining fish health and imparting neurological and cardiovascular benefits to humans (Harper and Jacobson 2005; Peet and Stokes 2005). In some studies, fish fed diets including seaweed meal showed improved growth, lipid metabolism, body composition and disease resistance (Yone et al. 1986; Satoh et al. 1987; Mustafa and Nakagawa 1995) but contrasting results or no benefits were observed in other studies (Hashim and Mat Saat 1992; Davies 27 et al. 1997a). These inconsistencies may result from the nutritional value of the algal species incorporated, dietary inclusion level, and metabolic differences among experimental finfish species (Mustafa and Nakagawa 1995; Davies et al. 1997a). The purple laver (nori) Porphyra umbilicalis (Bangiales, Rhodophyta), is a marine macroalga native to the North Atlantic including the coast of New England (Blouin et al. 2007). This species is high in protein, EPA, and arachidonic acid (AA; 20:4n-6; Blouin et al. 2006). Furthermore, the high surface : volume ratio, rapid growth (> 25 % per day), and efficient nutrient uptake of this species make it well suited to culture in integrated systems and bioremediation of fish farm effluents (Carmona et al. 2006). Due to population declines and fishing restrictions, culture of Atlantic cod Gadus morhua has increased significantly in North America and Western Europe and is expected to reach 140-180,000 tons by 2010 (Rosenlund and Skretting 2006). Considerable research has recently been conducted on Atlantic cod nutrition, including their ability to digest and assimilate diets containing plant proteins (Albrektsen et al. 2006; Forde-Skjaervik et al. 2006; Tibbetts et al. 2006). These studies indicate that Atlantic cod have high apparent digestibility coefficients for several protein sources, including several derived from plant products, and that growth does not differ significantly over a wide dietary protein range (36-66%;Tibbetts et al. 2006; Hansen et al. 2007). 28 The purpose of the present study was to investigate the value of incorporating laver Porphyra spp. as a replacement for fishmeal in diets for juvenile Atlantic cod. Materials and Methods Experimental Diets The Porphyra spp. was grown in the University of New Hampshire (UNH) integrated recirculating aquaculture system (IRAS), in a greenhouse, at GreatBay Aquaculture, LLC. (GBA, Portsmouth, New Hampshire). The IRAS consisted of four 4,000 L square fiberglass tanks. The seaweed was grown in three of the tanks under vigorous aeration, and Atlantic cod (from 5 to 1,000 g) were grown in the fourth. Mechanical and UV filtered sea water derived from the Piscataqua River was recirculated between the fish and seaweed tanks and the seaweed derived all of the nutrients necessary for growth from the finfish effluent. The culture system water was UV and mechanically filtered and 5-10% of the water volume was exchanged each day. Porphyra sporelings (young plants) were either produced from cultures at the University of Connecticut (Yarish Laboratory, Stamford, Connecticut) or collected from the New Hampshire coast. Additional Porphyra was grown in the seaweed tanks of the IRAS after the Atlantic cod had been harvested; nutrients were supplied via a 0.5x solution of von Stosch seawater enrichment medium (Guiry and Cunningham 1984). Three native North Atlantic species of Porphyra were grown: Porphyra umbilicalis Kutzing, P. linearis 29 - Greville, and P. leucosticta Thuret in LeJolis. The mixture used in the feed was predominantly (>90%) P. umbilicalis. After harvesting, the Porphyra blades were air dried on racks in the greenhouse for approximately 12-24 h then oven dried overnight at 80°C. The dried Porphyra was chopped in a food processor (to approximately 1-2 mm) and packaged in plastic bags for shipment to the feed manufacturer. The final product was analyzed by New Jersey Feed Laboratories, Inc. (Trenton, New Jersey; see below) to determine its nutritional profile (Table 1) before incorporation into experimental diets. The dried Porphyra, was combined with the other ingredients (Table 2). The mixture was finely ground through a hammer mill, moistened with water, and put through a die at room temperature to form.3 mm pellets. Three iso- nitrogenous, iso-caloric diets (55% protein, 16% fat, calculated gross energy ~ 20.5 MJ/kg) were formulated to replace 0, 15, and 30% of the fishmeal in a commercially available diet for marine finfish with the cultured Porphyra (diets are designated NO, N15, and N30 to reflect nori content). 30 Table 1 Proximate and Amino Acid Composition of Oven- Dried Porphyra spp. % of Parameter Sample Moisture 7.32 Protein 32.1 Fat (AH) 3.1 Fiber 3.91 Ash 12.77 Calcium 0.24 Phosphorus 0.46 Sodium 1.58 Potassium 1.65 Methionine 0.71 Cystine 0.9 Lysine 1.75 Phenylalanine 1.17 Leucine 2.14 Isoleucine 1.28 Threonine 1.78 Valine 1.96 Histidine 0.42 Agrinine 1.98 Glycine 2.01 Aspartic Acid 2.85 Serine 1.63 Glutamic Acid 3.75 Proline 1.31 Hydroxyproline 0.01 Alanine 3.36 Animals and Husbandry For this study, 1,800 juvenile Atlantic cod (mean initial body weight, [IBW]= 15.55 g) were hatchery-reared at GBA and transported by truck (~ 7.2 km) to the Aquaculture Research Center (ARC) at UNH. Upon arrival, fish were 31 anaesthetized with 2-phenoxyethanol (J.T. Baker, Phillipsburg, New Jersey) at a concentration of 200 mg/L, batch weighed gravimetrically (5-10 fish/batch) to the nearest 0.01 g with an Aventurer balance (OHAUS Corporation, Pine Brook, New Jersey), and randomly distributed to nine 1,600 L polyethylene tanks (fish density =1.7 g/L). The tanks were arranged in three-tank groups such that two groups made up a 30,000 L recirculating system and the third group made up a 14,600 L recirculating system; each with an appropriately sized chiller, biological filter, and foam fractioner. Each diet was assigned to one tank in each group. Throughout the experiment, the systems were kept on a photoperiod of 12 h light: 12 h dark. Half-hour crepuscular periods were provided with 100 W incandescent bulbs to simulate dawn and dusk. Light intensity was measured with a light meter (Sper Scientific, Scottsdale, Arizona), and ranged from 5 Ix at dawn and dusk to 10 Ix at the water surface during the light period. The water temperature ranged from 9.4 to 12.9°C over the course of the study (mean= 10.9°C). Average (± SD) values of water quality variables were 27 + 1.9 mg/L for salinity, 8.1 ± 0.8 mg/L for dissolved oxygen, 8.1 ± 0.2 for pH, and 0.04 ± 0.06 mg/L for NH3-N. The fish were hand fed three times daily for the first 8 weeks of the study; after 8 weeks, the number of daily feedings was reduced to two. At a given feeding, a ration equal to 1 % of the tank biomass was weighed and offered to the fish in each tank. The pellets were introduced slowly to ensure that all the feed was consumed. The fish were fed the entire ration unless they attained apparent satiation before the entire ration was distributed. Any remaining feed was 32 weighed for the calculation of voluntary feed intake (Fl). At 14-d intervals, feed was withheld for 24 h, after which five batches of five fish were weighed from each tank by randomly netting them into a tared water bath. The daily ration was then adjusted to reflect growth. Upon completion of the experiment (84 d), all fish were fasted for 24 h, enumerated, and weighed according to the initial procedure. Twelve fish from each tank were sacrificed by a lethal dose (>300 mg/L) of tricaine methanesulphonate (MS-222; Argent Laboratories, Redmond, Washington), their livers were excised for hepato-somatic index (HSI) determination and the left and right dorsolateral muscles (fillets) removed for proximate analysis. The muscle (mean weight 7.39 g/fish) was homogenized and pooled to form a single 100 g sample from each tank. The nine samples were kept frozen at -20°C until analyzed for fatty acid composition. Analytical Methods, Calculations, and Statistical Analysis Analysis of the dried Porphyra, experimental diets, and final muscle tissue were performed by New Jersey Feed Laboratory, Inc. (Trenton, New Jersey) in accordance with official methods (AOAC 2006). Moisture was determined by loss in weight on drying (LOD) at 135°C for 2 hours (AOAC 930.15) and ash was determined burning at 600°C for 2 hours (AOAC 942.05). Crude fat was determined gravimetrically after ether extraction (AOAC 920.39). Total protein was determined by nitrogen combustion and calculated as N x 6.25 (AOAC 33 990.03). Fiber was determined as loss on ignition after treatment with 1.25% (weight/volume [w/v}) H2S04 solution and 1.25% (w/v) NaOH solutions (AOAC 978.10). Amino acid composition (excluding tryptophan) was determined by ion exchange chromatography after oxidation with performic acid and hydrolysis with 6-M HCI (AOAC 985.28 and AOAC 994.12). Tryptophan was determined by ion exchange chromatography (AOAC 988.15). Fatty acid methyl esters were separated and determined by gas chromatography (AOAC 963.22). Feed performance was evaluated in terms of voluntary Fl, specific growth rate (SGR), feed conversion ratio (FCR), and HSI. The Fl was expressed as a percentage of average body weight consumed daily and was calculated as: 100 x (amount of feed consumed) * (average tank biomass)"1 * (number of days)"1. The SGR (%/d) was calculated as 100 * [loge(final body weight, FBW) - (loge(IBW)]/(number of days). The FCR was calculated as (dry weight fed)/(live weight gained), and HSI was calculated as 100 * [(wet weight of liver)/FBW]. The effects of dietary treatment on growth and feed performance were evaluated using a randomized complete block experimental design with tank grouping as the blocking factor to reduce any variability incurred by the two different aquaculture systems. The biweekly observations of Fl, FCR, SGR, and growth were evaluated using a repeated measures design and were analyzed using a split-plot analysis of variance (Zar 1999) with biweekly observation as subplots. As necessary, percentages and indices were transformed prior to 34 analysis. Significant (P< 0.05) differences between means were subjected to a post hoc Tukey's honestly significant difference test to distinguish differences among treatment levels. Results are expressed as means (± SE) unless otherwise noted. Pearson's product-moment correlation was used to investigate relationships between dietary and muscle fatty acid compositions. Gut Evacuation At the end of the feeding trial, a qualitative assessment of gut evacuation rate was performed. The fish were fasted for 48 h and then fed their appropriate diet in excess. Three fish from one tank of each dietary treatment group were sacrificed by a lethal dose of MS-222 at 6, 24, 30, 48, 54, and 60 hours after feeding. A different tank from each treatment was sampled at each time point. The digestive tract was immediately excised and divided by scalpel into four sections (stomach with the pyloric caeca and three lengths of the intestine sectioned at the naturally occurring bends). The sections were sliced longitudinally, and examined for the presence of undigested solids. Results Diets The compositional analyses of the diets are presented in Tables 2, 3, and 4. Most of the fish readily accepted the experimental diets which were similar in 35 buoyancy. Proportionately greater amounts of fines were observed in diets with higher Porphyra content, which necessitated sieving prior to feeding. Dark, slightly reflective specs, presumed to be Porphyra, were observed in the feces of fish in the N15 and N30 treatment groups. Due in part to the wide range of IBWs (SD = 2.46 g), the largest individuals dominated during feeding and received a larger portion of the fixed ration. Throughout the study, a few individuals in each tank could not effectively compete for feed. When apparent satiation was attained, feeding cessation was abrupt and all introduced feed was considered consumed. Table"2 Experimental Diet Formulations Ingredient NO N15 N30 Dried Porphyra 0 5.5 11 Fishmeal 37 31.5 26 Soybean 14 14 14 Cereal 23 21 19 Fish Oil 11 11 11 Mammalian 5 7 9 Poultry 5 5 5 Minors/Vitamins/Minerals 5 5 5 Values are presented in g/100g 36 Proximate (% wt. weight) and fatty acid composition (% of fatty acids) of experimental diets ___ Diet NO IM15 N30 Moisture 5.01 4.93 4.28 Protein 55.70 55.00 54.80 Fat 17.11 16.43 16.09 Fiber 0.48 1.10 0.89 Ash 7.88 7.44 6.95 Total % n-3 25.52 26.11 25.98 Total % n-6 11.20 10.77 10.65 16:0 17.43 17.18 17.19 18:0 3.45 3.32 3.44 18:1n9 9.71 9.53 9.49 18:2n6 9.85 9.14 8.92 18:3n3 1.62 1.58 1.59 20:1 n9 2.91 2.85 2.94 20:1n11 0.42 0.42 0.44 20:4n6 0.76 0.86 0.94 20:5n3 8.90 9.30 9.24 22:1n11 3.65 3.59 3.67 22:6n3 9.74 9.82 9.62 n3:n6 2.28 2.42 2.44 37 Table 4 Amino acid profile of experimental diets (% of sample) AA NO N15 N30 Methionine 1.21 1.20 1.16 Cystine 0.66 0.70 0.74 Lysine 3.39 3.33 3.36 Phenylalanine 2.29 2.34 2.38 Leucine 4.97 5.00 5.11 Isoleucine 1.76 1.84 1.66 Threonine 2.28 2.29 2.33 Valine 2.40 2.68 2.58 Histidine 1.37 1.42 1.43 Agrinine 3.30 3.24 3.13 Glycine 3.24 3.07 2.92 Aspartic Acid 4.92 4.85 4.90 Serine 3.16 3.08 3.20 Glutamic Acid 8.63 8.23 8.13 Proline 3.02 2.84 2.88 Hydroxyproline 0.53 0.47 0.42 Alanine 3.92 3.92 4.02 Tyrosine 1.75 1.72 1.73 Tryptophan 0.50 0.47 0.57 Mortality. Growth, and Feed Performance Growth and feed performances are presented in Table 5. Mortality averaged 14.1 ± 0.05% and was independent of dietary treatment. There were no significant differences among treatment groups in growth or feed performance and the fish grew to a mean weight of 41.0 + 0.01 g. •' 38 Table 5 Growth and feed performance' in Juvenile Atlantic cod fed the experimental diets over 12 weeks. flBW-15.,55 ± 0.36 g > Diet NO N1S N30 FBW (g) 41.2? ± 0.010 41.91 ± 0.O10 39.71 ± 0.010 Growth {% IBW) 166.57 ± 8.335 165.46 ± 8.335 157.94 + 8,335 Fl (%ABWday-1) 1.40 ± 0.001 1.42 ± 0.001 1.42 ± 0.001 SGR (%day-1) 1.22 ± 0,001 1.20 ±- 0.001 1.17 ± 0.001 FCR 1.21 ± 0.067 1,25 ± 0.067 1.32 ± 0.067 HSI 0.06 ± . 0.001 0.05 ± 0.001 0.04 ± 0.001 Values presented are means ± SE. The fatty acid composition of the fillet muscle was similar among the treatment groups (Table 6). There was a significantly higher amount of AA (g/100g fatty acids) in the N30 fish and a strong correlation between the amount in the diet and in the muscle (r = 0.90, P < 0.01). There were no significant differences in the arachidonic acid content between the NO and N15 fish. On % of sample basis (g/100g muscle tissue), there was a significant increase in stearic acid (18:0) in the NO fish compared to the N15 fish. There were no significant differences in 18:0 content between the N15 or N30 fish, nor between NO and N30. A decrease (P < 0.09) in HSI was observed as the level of Porphyra replacement increased. HSI was 5.72% for NO fish, 5.47% for N15 fish, and 4.20% for N30 fish. 39 Table 6 Muscle fatty acid composition of juvenile Atlantic cod fed the experimental diets over 12 weeks. Presented as % of total fatty acids. (IBW=15.55 ± 0.36 g) Diet NGI N15 N30 total % n-3 49.86 + 0.010 48.99 + 0.010 48.99 ± 0.010 total % n-6 9.45 + 0.001 9.41 ± 0.001 9.52 ± 0.001 16:0 17.63 ± 0.004 18.24 + 0.004 18.48 + 0.004 18:0 5.01 + 0.004 4.51 ± 0.004 4.59 ± 0.004 18:1n9 7.47 ± . 0.003 7.79 + 0.003 7.64 ± 0.003 18:2n6 5.90 + 0.003 5.72 ± 0.003 5.36 ± 0.003 18:3n3 0.72 ± 0.001 0.70 + 0.001 0.69 + 0.001 20:1n9 0.76 ± 0.001 0.85 + 0.001 0.82 + 0.001 20:1 n11 0.14 ± 0.001 0.16 ± 0.001 0.15 ± 0.001 20:4n6 2.34 + 0.001a 2.50 ± 0.001a 2.89 + 0.001b 20:5n3 13.50 + 0.002 13.62 ± 0.002 13.93 ± 0.002 22:1n11 0.07 +• 0.010 0.12 + 0.010 0.19 + 0.010 22:6n3 31.11 ± 0.009 30.14 ± 0.009 29.70 ± 0.009 Values presented are means ± SE. Values with common postscripts are not significantly different (P < 0.05, Tukey's HSD). Gut Evacuation The rate of gut evacuation varied notably among individuals. At 6 h after feeding, all but two fish showed evidence of stomach evacuation, and all fish had feed present throughout the intestines. In the 24-h sampling, evacuation had progressed through the first section of the intestines in four of the nine fish sampled. Evacuation of the final two sections of the intestine was most variable. At the 30 h sampling, six fish had evacuated the second section and eight had feed remaining in the final section. Evacuation was completed by 48-h in three fish, 54-h in one fish, and 60-h, in four fish. 40 Discussion The results of this study indicate that replacing 30% fishmeal with Porphyra did not have adverse effects on growth or feed conversion in juvenile Atlantic cod. The mean SGR (1.19) in the present study compares favorably with SGRs reported in other studies (Hansen et al. 2007a, 0.95; Bjornsson et al. 2006, 1.05; Albrektsen, 2006, 1.22; Aksnes et al. 2006, 1.27) and the small differences may reflect different experimental conditions and initial size of the experimental fish. A number of other factors, including feeding practices, diet composition, and husbandry can also significantly affect growth and feed conversion. Studies examining the digestibility of Porphyra in Atlantic cod are warranted since undigested material was observed in the feces and conflicting results were reported in other studies of Porpfryra-su pplemented diets, potentially reflecting species-dependent differences in digestive capacity. For instance, replacing 9 or 18% of the fishmeal with P. purpurea in diets fed to thick-lipped grey mullet Chelon labrosus resulted in reduced growth and higher FCRs, proportional to the level of algal inclusion (Davies et al. 1997a). Those authors suggest that porphyran, a complex carbohydrate found in Porphyra, may have contributed to lower digestible energy in the seaweed diets. In contrast, growth and feed performance (feed efficiency ratio, protein efficiency ratio, muscle protein deposition) were improved in juvenile red sea bream Pagrus majorwith 5% dietary inclusion of P. yezoensis meal, although HSIs were significantly elevated (Mustafa et al. 1995). 41 In addition to diet composition, other factors (e.g., gut residence time) can affect digestibility (De Silva and Anderson 1994). In the present study, feeding frequency was reduced from three to two times daily. A feed frequency study on juvenile Atlantic cod (IBW: 192 g) indicated that growth was not different when fish were fed daiiy or every other day (Roseniund et al. 2004). Similarly, in another study, stomach evacuation in juvenile Atlantic cod fed diets containing various plant proteins, was not initiated until 6-12 h after consuming most of the experimental diets; evacuation then continued steadily for 48-60 h. Evacuation of the entire Gl tract was reported after 72 h and was independent of dietary treatment. The authors suggested that feeding more than once every 24 h would not improve growth and may result in enlarged livers (Hansen et al. 2006). Gut evacuation results in the present study, suggest that feeding frequency was excessive, which may have elevated FCRs and contributed to sub-optimal digestion. The saturated fatty acid profiles in muscle were similar to profiles reported in other studies (Dos Santos et al. 1993; Morais et al. 2001; Albrektsen et al. 2006) except that palmitic acid (16:0) was especially high among all treatment groups in the present study. These elevated levels probably reflect the high levels present in dietary components. While, 16:0 is the dominant fatty acid in Porphyra (Blouin et al. 2006) it also tends to increase with increasing mammalian proteins (Ai et al. 2006). Because the control diet in the present study was chosen to represent a commonly used formulation, both of these sources were present. 42 Inclusion of terrestrial plant proteins that lack many n-3 fatty acids and have high n-6:n-3 ratios can disproportionately displace the beneficial inclusion of essential n-3 fatty acids such as DHA and EPA in fish muscle tissue (Sargent et al. 2002). In the present study, the fatty acid composition of the muscle tissue did not differ among treatment groups except for AA leveis, which increased with greater dietary Porphyra inclusion. These levels were higher than those in other published studies (Lie et al. 1986; Morais et al. 2001; Albrektsen et al. 2006) and probably reflect the high levels found in Porphyra. Arachidonic acid in wild P. umbilicalis collected in Maine (18.5 ± 2.9 g/100g fatty acids) was the third-most prevalent fatty acid after EPA (39.4 ± 2.1%) and palmitic acid (21.3 ± 3.1%, Blouin et al. 2006). Although considered an essential fatty acid for marine fish, AA may be deficient in typical marine fish feeds and is absent in most terrestrial plant oils (Bell and Sargent 2003). On a global basis, seaweed aquaculture is a multi-billion dollar industry (FAO 2002; FAO 2003) and Porphyra is one of the most valuable seaweeds cultured. This seaweed is used primarily as the wrapping around sushi rolls and is also the preferred source of the pigment, R-phycoerythrin, which is used as a fluorescent tag for biotechnological applications. Research is underway to incorporate Porphyra and other seaweeds in recirculating systems with finfish to stimulate algal growth, while eliminating nitrogenous compounds (Neori et al. 2004). If grown for this purpose, Porphyra spp. may be a good candidate for fishmeal replacement in marine finfish diets because of its high protein content and favorable fatty acid profile (Blouin et al. 2006). 43 In summary, growth and feed conversion were not affected when up to 30% of the fishmeal in juvenile Atlantic cod diets was replaced with dried Porphyra. Fatty acid profiles were similar in the muscle tissue in all treatment groups, except in terms of the AA levels, which were higher in fish fed the high Porpfryra-supplemented diet. Additional studies are needed to determine the digestibility of Porphyra and to evaluate the economic benefits of feeding seaweed-supplemented diets over the long term. 44 CHAPTER 3 SOY PROTEIN CONCENTRATE AS AN ALTERNATIVE PROTEIN IN DIETS FOR JUVENILE ATLANTIC COD Abstract Two feeding trials were conducted with Atlantic cod, Gadus morhua, juveniles (initial mean weight: 87.9 g or 15.92 g) in which soy protein concentrate (SPC) was used to replace fishmeal in iso-nitrogenous, iso-caloric diets. In the first trial, SPC was included to provide 0%, 10%, 20%, 30% or 40% dietary soy protein (47% protein, 15% fat, 5,059 kcal/kg calculated gross energy) and replaced fishmeal entirely at the highest inclusion level. In the second trial, three diets were formulated to replace 0%, 25%, and 50% of the fishmeal with SPC in a commercially available diet for marine finfish (50% protein, 15% fat, 4,900 kcal/kg calculated gross energy). Upon conclusion of the studies, no differences in survival, growth, hepatic-somatic or viscero-somatic indices were found among the treatment groups. The fish attained mean weights of 162.7 g and 39.1 g in the first and second trials, respectively, with specific growth rates (SGR) of 0.76% and 1.12% and feed conversion ratios (FCR) of 1.28 and 1.29, respectively. Fatty acid composition of the liver and muscle reflected that of the dietary fish oil in the first study but more closely reflected the soybean ingredients in the second. No intestinal enteritis was observed in the histological sections from either study. These results indicate that soy protein concentrate can 45 entirely replace fishmeal in diets of juvenile Atlantic cod, or it can be combined i with other common feed ingredients to replace dietary fishmeal by 50%. Introduction Minimizing the use of fishmeal in aquaculture diets in favor of more cost effective and sustainable ingredients is essential to the continued growth and sustainability of the aquaculture industry (Naylor et al. 2009). Due to their abundance, relative low cost, high protein levels, and favorable amino acid profiles, soy-derived ingredients represent a viable alternative to fishmeal (Gatlin et al. 2007), especially for omnivorous species. Soybean meal (SBM) can be used as the sole protein source with the addition of supplemental amino acids in diets of carp (Viola et al. 1982). Similarly, SBM can constitute 67% and 50% of diets for tilapia (Oreochromisspp.) and channel catfish (Ictalurus punctatus), respectively (Storebakken et al. 2000b). Early research evaluating the applicability of SBM in diets of piscivorous species, however, suggested limited inclusion due to relatively high levels of antinutritional and antigenic factors and indigestible carbohydrates (Ingh et al. 1991; Baeverfjord and Krogdahl 1996). For instance, at levels as low as 10-30%, dietary SBM negatively impacted growth and digestibility (Gatlin et al. 2007) and caused morphological and cellular changes to the intestine in Atlantic salmon (Salmo salar, Baeverfjord and Krogdahl 1996). The alcohol-soluble fraction of the SBM, which contains 46 oligosaccharides and nitrogen-free extracts including saponins and isoflavones, are largely responsible for these effects (Ingh et al. 1996; Gatlin et al. 2007). Soy protein concentrate (SPC), although more expensive, does not contain the alcohol soluble fraction and can be included at much higher concentrations in diets for piscivorous marine species. Soy protein concentrate has higher essential amino acid (EAA) concentrations and nutrient digestibility compared to SBM (Wilson 2002), which permits inclusion levels as high as 100% and 75% in diets of rainbow trout (Oncorhynchus mykiss; Mambrini et al. 1999) and Atlantic salmon (Storebakken et al. 2000a), respectively. Soy protein concentrate can be further processed so that levels of oligosaccharides, B- conglycinin, and trypsin inhibitors are negligible (Adelizi et al. 1998; USSEC 2008). Atlantic cod, although piscivorous, have demonstrated greater tolerance of both processed and unprocessed soy ingredients than Atlantic salmon (Albrektsen et al. 2006; F0rde-Skjaervik et al. 2006; Hansen et al. 2006; Refstie et al. 2006a; Hansen et al. 2007a; Olsen et al. 2007). Hansen et al. (2007b) determined that as much as 50% of dietary protein could be derived from a plant protein mix (SBM, SPC, and wheat gluten) in the diets of adult cod raised in sea cages. Fed at higher quantities (75% or 100%), however, growth, feed efficiency, and protein efficiency were reduced. Cod are opportunistic feeders and their digestive systems are highly adaptive to diets of varying composition. In growth studies, using diets containing various amounts of plant protein, feed intake increased in fish fed diets with lower digestibility such that growth was similar 47 among treatment groups (Refstie et al. 2006a; Refstie et al. 2006b; Hansen et al. 2007b). The purpose of the present study was to investigate the value of incorporating SPC as a fishmeal replacement in cod diets. The dietary formulations fed in the first trial tested the maximum inclusion levels possible, while those in the second replaced fishmeal in a commercially available finfish diet. Materials and Methods Experimental Diets In trial 1, five iso-nitrogenous, iso-caloric diets (47% protein, 15% fat, 5,059 kcal/kg calculated gross energy) were formulated to include 0%, 10%, 20%, 30% and 40% dietary soy protein from SPC. These diets replaced 28, 60, 80, and 100% fishmeal and are designated Control 1, SPC10, SPC20, SPC30, and SPC40, respectively (Table 1). In trial 2, diets were similarly formulated to replace 0%, 25%, and 50% of the fishmeal in a commercially available diet for marine finfish (50% protein, 15% fat, 4,900 kcal/kg calculated gross energy) with SPC (Table 2). These diets were designated Control 2, SPC25, and SPC50, respectively. In both studies, the feed ingredients were ground through a hammer mill, moistened with water, and forced through a die at room temperature to form 4 mm (trial 1) or 3 mm (trial 2) pellets. 48 Table 1 Trial 1: Experimental Diet Formulation Ingredients Control SPC10 SPC20 SPC30 SPC40 Fish, Herring 50.0 36.0 20.0 10.0 0.0 SPC, Profine VF 0.0 14.5 29.1 43.5 58.0 Wheat Gluten ' 2.6 4.0 5.0 3.0 1.3 Blood Meal, spray dried 2.3 1.9 2.9 1.5 1.0 Fish, hydrolyzed fish protein cone. 5.0 5.0 5.0 5.0 5.0 Wheat Middlings 19.0 7.8 0.0 0.0 0.0 Wheat, SWW, whole, ground 10.0 13.0 16.3 15.2 0.0 Corn Starch, feed grade 0.0 4.0 5.0 3.5 14.7 Fish Oil, Menhaden 8.6 9.9 11.3 12.1 13.0 Soy Lecithin 1.5 1.5 1.5 1.5 1.5 Vitamin Premix 0.5 0.5 0.5 0.5 . 0.5 Choline Chloride, liquid 60% 0.0 0.1 0.2 0.4 0.5 Stay C - 35% 0.1 0.1 0.1 0.1 0.1 Mineral Premix 0.3 0.3 0.3 0.3 0.3 Calcium phos. mono, 21 %P 0.0 0.8 1.7 2.1 2.6 Sodium Selinite Premix 54% 0.0 0.0 0.0 0.0 0.0 L-Lysine HCL 78% 0.0 0.1 0.3 0.4 0.4 MHA (methionine) 84% 0.0 0.2 0.3 0.4 0.5 Taurine 95% 0.0 0.3 0.4 0.5 0.5 T-Bind 0.1 0.1 0.1 0.1 0.1 Mold Inhibitor 0.1 0.1 0.1 0.1 0.1 Fish Protein 39.0 29.1 17.8 10.8 3.8 Soy Protein 0.0 10.0 20.0 30.0 40.0 Values are presented in g/100g 49 Table 2 Trial 2: Experimental Diet Formulation Ingredients Control SPC25 SPC50 Soy Protein Concentrate 0 9.6 19.2 Fishmeal1 38.5 27.4 19.3 Soybean meal and Soy lecithin 11 11 11 Corn gluten and Wheat 16.5 18.5 16 Menhaden Fish Oil 10 9.5 10.5 Blood Meal 10 10 10 Poultry By-product Meal 10 10 10 Minors/Vitamins/Minerals2 4 4 4 Values are presented in g/100g 1 Menhaden (min. 62% CP) and herring(min. 75% CP) meals 2 Manganese Proteinate, Zinc Proteinate, Copper Proteinate, Calcium lodate, Iron Proteinate, Cobalt Proteinate, Calcium Carbonate, Sodium Selenite, Vitamin A Acetate, Vitamin D3 Supplement, dl-Alpha Tocopheryl Acetate, Vitamin B12 Supplement, Riboflavin, Niacin, Calcium Pantothenate, Menadione Sodium Bisulfite Complex, Folic Acid, Thiamine Mononitrate, Pyridoxine Hydrochloride, Biotin, Vitamin A Acetate, Vitamin D3 Supplement, L-Ascorbyl-2-Polyphosphate (in no particular order) Animals and Husbandry In trial 1, 1005 juvenile Atlantic cod (mean IBW: 87.9 ± 4.9 g) were hatchery-reared at GreatBay Aquaculture, LLC. (Portsmouth, New Hampshire) and transported by truck to the Aquaculture Research Center at the University of New Hampshire (UNH, Durham, New Hampshire), where they were maintained in recirculating aquaculture systems. At the start of the feed trial, feed was withheld for 24 hours and a subset (n=30) of fish were anesthetized (200 mg/L Finquel MS-222, Tricaine Methanesulfonate, Argent Chemical Laboratories, 50 Redmond, Washington) and weighed gravimetrically to the nearest 0.01 g (OHAUS Adventurer, Pine Brook, New Jersey). All fish were randomly distributed among fifteen 1,500 L polyethylene tanks (approximately 4 g fish/L). Six of the tanks comprised a 9,600 L recirculating system and the remaining tanks comprised a 14,400 L recirculating system, each with appropriately sized chillers, mechanical and biological filters, and foam fractioners. Throughout trial 1, the systems were kept on a 12:12 hour photoperiod. Light intensity (Sper Scientific, Scottsdale, Arizona), ranged from 50 to 100 Ix at the water surface during the light period. The temperature, salinity, dissolved oxygen, pH, and NH3-N averaged ( ± S.D.) 12.2 ± 0.5 °C, 28 ± 2.1 g/L, 7.8 ± 0.6 mg/L, 8.0 ± 0.2, and 0.06 ± 0.04 mg/L respectively. For the first 7 d of the study, the fish were hand fed the experimental diets once daily to apparent satiation to acclimate them to the experimental diets. The remaining 77 days of the study were divided into four 14 d periods and a final 21 d period. For the first 2 days of each period, the fish were fed once daily, at mid day, to apparent satiation to establish the daily voluntary feed intake (Fl). For the remainder of each period, the fish were hand fed a fixed ration, three times daily, equal to one-third of the above-referenced Fl. If the fish attained apparent satiation before the entire ration was fed, the remaining (unfed) portion was weighed and recorded. " At the start of the 4th feeding period (approximate mid-point of the study), the fish were fed for 2 days to calibrate Fl for the period, after which, feed was 51 withheld for 24 h, and 12 fish were weighed from each tank. Upon completion of the experiment (84 d), all fish were fasted for 24 h, enumerated, and weighed according to the initial procedure. Six fish from each tank were sacrificed by a lethal dose of tricaine methanesulphonate (MS-222; Argent Laboratories, Redmond, Washington; >300 mg/L), their livers and viscera, including the digestive system from the stomach to the anus, the liver, spleen, and gallbladder, were excised for determination of hepato-somatic index (HSI) and viscero somatic index (VSI), respectively. The livers and the right dorsolateral muscles (fillets) were retained for proximate and fatty acid analysis. The muscle and liver samples, respectively, were homogenized and pooled to form a single, 100 g sample from each tank. The nine samples were kept frozen at -70°C until analyzed. In trial 2, 1,8.63 juvenile Atlantic cod (mean IBW: 15.92 g) were hatchery- reared at GBA and similarly transported to the ARC, where they were anaesthetized (200 mg/L MS-222), weighed gravimetrically to the nearest 0.01 g, and randomly distributed among nine 1,600 L polyethylene tanks (approximately 2.06 g fish/L). Six of the tanks comprised a 30,000 L recirculating system and the remaining tanks comprised a 14,600 L recirculating system, each with an appropriately sized chiller, biological filter, and foam fractioner. The experimental diets were randomly assigned to the tanks such that one replicate for each diet was reared on the smaller system and two replicates were on the larger system. The fish were hand fed twice daily to apparent satiation and the pellets were introduced slowly to ensure that all the feed was consumed. 52 Throughout trial 2, the systems were kept on a 12:12 hour photoperiod. Half-hour crepuscular periods were provided with 100 W incandescent bulbs to simulate dawn and dusk. Light intensity, ranged from 5 (dawn and dusk) to 10 Ix at the water surface during the light period. The temperature, salinity, dissolved oxygen, pH, and NH3-N averaged ( ± S.D.) 12.4 ±0.8 °C, 26 ± 1.5 g/L, 8.2 ± 0.4 mg/L, 7.8 ± 0.4, and 0.04 ± 0.02 mg/L respectively. Upon completion of trial 2.(84 d), all fish were fasted for 24 h, enumerated, and 25 fish from each tank were weighed according to the initial procedure. Fifteen fish from each tank were sacrificed by a lethal dose of tricaine methane sulphonate (>300 mg/L MS-222) and fillets were removed for determination of muscle proximate analysis and fatty acid composition. The muscle tissues was homogenized in a food processor and pooled to form one, 100 g sample from each tank. The nine samples were kept frozen at -20 °C until they were analyzed. Additionally, 15 fish were euthanized and their livers and digestive system were excised for determination of HSI and VSI. Histopathology In both trials, samples of the digestive system were taken from 3 fish from each tank for histological examination. Approximately 2 mm sections were excised 1-2 cm posterior to the pyloric caeca and 1-2 cm anterior to the sphincter that delineates the distal chamber (Figure 1, approximately the centers of the first and fourth middle intestine, respectively, as described by Refstie et al. 2006). 53 The sections were preserved in 10% formalin for 24 hours, processed for routine histology, sectioned at 5 urn, and stained with hematoxylin and eosin. The slides were randomly numbered for blind analysis. Figure 1 Histopathology sections Analytical Methods, Calculations, and Statistical Analysis For both trials, analyses of the experimental diets, and final tissue samples were performed by New Jersey Feed Laboratory, Inc. (Trenton, New Jersey) in accordance with official methods of AOAC ITERNATIONAL (AOAC 2006). Moisture and ash were determined by loss in weight on drying (LOD) at 135°C for 2 hours (AOAC 930.15) and burning at 600°C for 2 hours (AOAC 942.05), respectively. Crude fat was determined gravimetrically following ether extraction (AOAC 920.39). Total protein was determined by nitrogen combustion and calculated as N * 6.25 (AOAC 990.03). Fiber was determined as loss on ignition after treatment with 1.25% (w/v) H2S04 and 1.25% (w/v) NaOH solutions (AOAC 978.10). 54 Amino acid composition (excluding tryptophan) was determined by ion exchange chromatography following oxidation with performic acid and hydrolysis with 6M HCL (AOAC 985.28 and AOAC 994.12). Tryptophan was determined by ion exchange chromatography according to AOAC 988.15. Fatty acid methyl esters were separated and determined by gas chromatography according to AOAC 963.22. Feed performance was evaluated in terms of voluntary feed intake (Fl, %/day, (feed intake) * (avg. tank biomass)"1 * (no. of days)"1), specific growth 1 rate (SGR, %/day, (loge(FBW)-loge(IBW)) * (no. of days)" * 100), feed conversion ratio (FCR, amount fed (g dry weight) * (live weight gain (g))"1), hepatosomatic index (HSI, (wet weight of liver) * (FBW)"1 * 100) and viscero somatic index (VSI, (wet weight of viscera) * (FBW)"1 * 100). An ordinal scoring system based on Knudsen et al. (2006) and modified for Atlantic cod (Table 3) was used to evaluate changes in the gut morphology and inflammatory responses. Low scores (1-2) and high scores (4-5) indicate normal morphology and marked inflammation/morphometric changes, respectively. The scores for lamina propria (LP) size, LP cellularity, and the presence of lymphocytes were added to provide a total inflammation score ranging from 3 to 15. Goblet cells within multiple (3-5) longitudinally sectioned villi (crypt to apex) were enumerated and the area (urn2) of twenty intact goblet cells per slide was determined using the program ImageJ (Abramoff et al. 2004). Cell area was measured in binary images of slides at 200x magnification. 55 Table 3 Histopathological Scoring (based on Knudsen et al. 2006) Score Histology Lamina Propria (LP): Size 1 LP is a very thin and delicate core of connective tissue 2 LP appears slightly more distinct in some folds 3 LP has markedly increased in most of the folds 4 LP is thick in many folds 5 LP is very thick in many folds Lamina Propria (LP): Cellularity 1 LP is mostly non-inflammatory tissue with some extracellular space 2 LP is mostly non-inflammatory tissue with decreased extracellular space 3 LP contains increased numbers of nuclei in some folds 4 LP contains increased nuclei and inflammatory cells close to the enterocytes 5 LP contains increased nuclei and inflammatory cells throughout Connective Tissue (CT): 1 Almost no CT between glands and stratum compactum 2 Slightly increased amount of CT beneath some of the mucosal folds 3 Clear increase of CT beneath most the mucosal folds 4 Thick layer of CT beneath many folds 5 Extremely thick layer of CT beneath some folds Lymphocytes: 1 Almost no intraepithelial lymphocytes (lELs) 2 Lymphocytes and/or granulocytes present, but at the basal lamina «, Lymphocytes and/or granulocytes present throughout the intestinal mucosa OR the J lamina propria . Lymphocytes and/or granulocytes present throughout the intestinal mucosa AND the lamina propria ,. Lymphocytes, macrophages, and apoptotic cells present throughout intestinal mucosa 56 The effects of dietary treatment on growth and feed performance were evaluated using a randomized complete block experimental design with tank system as the blocking factor to reduce any variability incurred by the different aquaculture systems. The intermediate and final observations of Fl, FCR, SGR, and growth were evaluated using a repeated-measures design and were analyzed using a split-plot analysis of variance with the repeated observations as subplots (ANOVA, Zar 1999). Histological data were analyzed using a Kruskal- Wallis ANOVA for nonparametric ranked sums. As necessary, percentages and indices were arcsine transformed before analysis. Significant (P < 0.05> differences between means were subjected to a post hoc Tukey's honestly significant differences test to distinguish pairwise differences among treatment groups. Results are expressed as means (± SE) unless otherwise noted. A Pearson's product-moment correlation was used to investigate relationships between dietary and tissue fatty acid compositions. Results Experimental Diets The compositional analyses of the diets are presented in Tables 4 and 5. In both trials, most of the fish readily accepted the experimental diets. When apparent satiation was attained, feeding cessation was abrupt and all introduced feed was consumed. In the first trial, with small rations fed throughout the day, feed was consumed entirely in almost every meal. 57 Table 4 Trial 1: Proximate (% wt. weight) and fatty acid composition (% of fatty acids) of experimental diets Diet Control 1 SPC10 SPC20 SPC30 SPC40 Moisture 6.23 6.31 5.78 6.21 5.73 Protein 47.0 48 48.1 47.6 48.2 Fat (AH) 15.00 15.42 15.31 15.38 15.15 Fiber 1.08 1.87 1.18 1.31 1.76 Ash 6.25 7.65 6.62 6.36 7.27 total % n-3 22.76 26.83 28.91 30.06 32.10 total % n-6 4.36 4.55 4.54 4.57 4.57 16:0 17.37 16.98 17.48 17.91 17.86 18:0 3.15 3.10 3.24 3.37 3.49 18:1n9 11.57 9.94 9.14 8.77 8.28 18:2n6 4.36 9.19 8.58 8.36 7.14 18:3n3 2.17 1.75 1.82 1.89 1.92 20:1n9 0.39 0.46 0.32 0.25 0.11 20:1n11 3.31 3.63 2.57 1.89 1.17 20:4n6 0.71 0.72 0.73 0.75 0.81 20:5n3 8.60 8.87 9.88 10.46 11.44 22:1n11 5.87 5.56 3.46 2.11 0.65 22:6n3 11.02 11.40 11.75 11.82 12.23 n3:n6 5.22 5.90 6.36 6.58 7.02 58 Table 5 Trial 2: Proximate (% wt. weight) and fatty acid composition (% of fatty acids) of experimental diets Diet Control 2 SPC25 SPC50 Moisture 7.69 7.61 5.73 Protein 53.2 54.5 55.2 Fat (AH) 14.76 15.06 15.21 Fiber 0.76 1.09 1.16 Ash 10.56 8.09 7.04 total % n-3 25.72 20.43 20.85 total % n-6 9.95 8.96 8.74 16:0 19.46 16.80 16.80 18:0 4.16 3.22 3.17 18:1n9 14.14 10.66 10.47 18:2n6 8.96 8.30 8.04 18:3n3 1.26 1.54 1.53 20:1n11 0.00 0.55 0.54 20:1n9 6.09 5.64 5.55 20:4n6 0.99 0.66 0.70 20:5n3 13.89 7.87 8.05 22:6n3 6.30 6.95 6.98 n3:n6 2.58 2.28 2.39 Mortality, Growth, and Feed Performance Growth and feed performances are presented in Tables 6 and 7. Mortality averaged 6.2 ± 0.7% and 3.7% ± 0.1% in the first and second trial, respectively, and was independent of dietary treatment in both trials. There were no statistically significant differences in growth or feed performance among treatment groups in either trial. The fish grew to a mean weight of 162.7 ± 2.9 g and 39.1 ± 1.3 g, with an SGR of 0.76 ± 0.03% and 1.12 ± 0.04% in the first and second trials, respectively. 59 CO LD LD a LO CM CM ID o a r— K. 00 10 r— J-- d> CO CM o ID q O o o CM CM O CM CD o O o o o-**— O O O T- <3 o -H -H -H +1 -H 41 41 -H 41 41 O CO CO o> to *~t h~ CM LO a> CM 1^, en CO o 60 Proximate and Fatty Acid Composition The proximate and fatty acid compositions of the tissues are presented in Tables 8 -16. In trial 1, the fatty acid composition of the livers, and to a lesser extent of the muscle, followed a similar profile to the feed consumed. This is particularly true with regard to the 18-C fatty acids, which were reduced, and AA and EPA (as % of total fatty acids) in the liver, which increased with increased soy inclusion. The overall fatty acid composition in both the muscles and the diets reflected profiles more typical of the elevated fish oils than terrestrial plant oils. In trial two, the relationship between muscle and dietary fatty acids is similar to that in trial 1 for the 18-C fatty acids. In contrast to trial 1, levels of EPA (as % of total fatty acids) were reduced with increasing dietary SPC in the diet. 61 XJ X3 o t'O CN CO t^- o CD 56 7 96 7 2 07 7 03 4 03 6 CM a o o o O G3 o o o C2> o o +i +1 -H 41 +i +1 +i -Hi +l +i +l +1 CM LC< O © o CM CO CM LO c--\ o CD CN CO co CO •*%• rs. o O o c*l LO o •* CM LO CM co o o o^~ o^— o*~ u> CNi o o o o o o o o o o o o o CN V CM •* CO o 41 CO CO CO 41 -H 4i 4) cd 170 b o > o LO 96 7 07 7 cs o CN 1^. O fe. o 01 to w a. O aooooooooooo CXI CM 3 <*> O CO CO o > Xi o d Ol +1 +! +1 +1 +1 +[ +1 +1 +1 +1 +1 CM ^ 0 2 "c. CM *Q T- Q c CM to •s— r~~ Ill •5 CO 00- CO LO o> o O o^— o o £ E o O o o o ~n >* O O O CM o O o p o d c a> to J^s « —^ o 41 4) 44 41 41 Ol X 0. E » 4> o 1*- CM 0) O CO © CO to co p S3 00 CO O 18 7 CO X +1 56 7 96 7 03 4 CM 74 6 o 03 6 CM 1!— o O T T» CM d ~ o o o o o o o o o o O o 0) 3 o "- (fl ^3 O •£» •D « CN CO t™ r- •o 4L U +i +i +i +1 +1 +1 +1 +i +1 +1 +i O *•* o O o o O^ _ +1 O |S u en o o o o co to o o o o © o> co LO 1ft 10 O CM « o o g co h- CO CM o T o •K ° (0 CO «v co ^r cc T o O o CM LO U> o +) 4) 4) 41 -H K5 -~ a. o v* < v> m o> o2 .. 10 r- •zf ifj *3 » co o CM CO J?0 •9 2> o d o CM (O Q CO S CO CO o s< a ° p r-j CM 3 21.33 . o o o o O a. ar e me a 4> m T5 CO fv- O CO LO r-- 41 +| 4| 4j 41 +| 4| 4| 4| +1 4| CD -*-* to CO o CM zz =5 CO ^~. O 0) o CM d ^ (ft co CO CM 05 s LD 0D CN CO e» 2 f". CO •*^f l~. cc o O O CN "3" Ci5 *— x: 0) 0 Q> £- o U- •S 'o < CO 5E a. ^> CO 62 o O O CM CM CM CD CO cx> O o o O O O O Q o o 001 b 01 2 O o O O O O o o o o o o o o o O o o o o o o o o a. -H +l 41 41 -H 41 41 41 41 41 41 44 w CO CO CM CO CM o O O co CM o o O O O a o O O o cfl "^f T~ f— O o o O O o o o o o O o O CO O o o o o a> X2 CM CM CM CO CD O O o o O o o O CM 00 6 O O o 00 2 001 b o o O o o o o 41 4) 41 41 o o o o o o o o o o o O CO > a 41 4) +1 41 +i 41 4! 41 41 41 41 4) o co co CO 0} CM co CM O o o co CO CM O O O o O o o o O O o o o O o o o o o O o o I a co cb o o o X! 0) X! m co CM ''vf CM co CO £ O o o o o O o o o o ~n ~rl ~H ~H 01 2 O o 00 2 o o o o o o © o CD o o o o o o o © o o o Q. CM o X CD o1L co +i -H •H 41 41 41 41 41 4) 4) 41 41 CM CO to CM O O co JC CM o o O a O O a o o o rr T- T- O •o o O o o O o o o o o oO CoO oO .Si « (8 o o o CM CM CO CO o O o O a o o o 8 o» 001 a 01 2 o 00 2 001 b O o o o o o o •5 ° o o o o o o o o o o o o o to O o ft, -H -H -H 41 +i 41 4) 41 41 4! 41. 41 to 05 a; -a CD CM CO CM CO CM O O O CD co CM o O O O c| o o o O a 5 o co o O o © o o o O o O o O o $g o o o -*•. 0> G> O O XI (tf 03 "5 «L CM CM CO co 41 41 41 41 41 O a O o o C o a o O 01 2 002 a 00 4 o 001 a O o o o o o o to o o o o o o O o o o o o a o to "35 CO CO to O o to CO o o -H +i +1 -H +l 41 41 41 41 41 41 4) a. CM O a. E CM 00 CM CO CO o O o CO o l.f>^ •«tf T— T— T— CO CM O O O o o o o CM o CJ o o o o o o u CJ X o CM to 63 JD -C! X3 XJ P3 FTJ TO O G) O "*• -* f- S3! h~ m o o co , „1 *«- co 00 co co T- CM o co CM h- CUM CM CM O O O Q *~ CM 00 o 000000 Q +1 +1 +1 +1 +| +1 +! +1 +1 +! +1 +1 f- o «a- CM « to O SB i~ CM 00 O CO CO QJ O - O o q it; N « !N d i- n ir « N 1^ d d d ^ T- CO '^w o o co 00 m T- CM O CM h- CM o ea o OOQOOQOOOOO +1 +1 +1 +1 +! +1 +1 +! +1 +! +1 +1 *3" h~ ^ (p ID'U) t- « » U) CO tt- U"j U^ h- CO O t™ ^}- 0 fl) f^ •O xs U JO *~t o !T, n ra .n IB n 1:,-, C::J *rt* *T" t—<- en 05 ^r CO CM CM CM O 0 0 Q sM 17 0 12 8 01 02 0 0 0 Q 0 O 0 O 0 0 C3 0 O >X o ns +1 +! +1 +1 +1 +1 +! +1 +1 +1 +1 +1 w in •r-r, *r K "tt CD Ul kCi ^, Oo &-> 10 cj ra JD ra to n O ra u o , o o> o T "* 1- o> h~ CO CM CO O !cMCNOOOOi-CM ooodddddd.ddd +1 +1 +! +1 +1 +1 +1 +! +! +1 h- K- K LO N 1ft !•) N (S S O CD t CD U5 U"' O O CO c» CO ^ CM *t ffl re re it) _D u •O •o o o co 1-1 C 3 ••=r •f v- O r- OS CM N- CM "1- CTS en a> C-0 f» CM rl n q T- T- CM CM 0 tj 0 O T™ CM odd odddo' 0000 +1 +1 +1 +1 +1 +1 +1 +1 +! +! COK * w * w o a SO cc? CN «-"< O dO CD CS| CD CD CO o> to ^t 0 <*. c e 0> 64 o « ID N. -it CO O CO » »K TO CO •<*• CO ~ ^ ^ CQ o„ . ,j10. «&- - o CN *Cf CN ' - o +! +1 -H +1 +1 +1 +1 +1 +1 +1 +1 44 CM CO CO co co !•». CO •s- CN 00 O § O -* h~ CN CO co Q T- O O <£> U~J -Q ,0 .0 Ca IB 0 CO f~ TT ^^ '_W co t.J CN •et- <(•• ^r O 0 CO CN OQOQOOQOOOO*«* O +1 +! +1 +| +1 +! +1 +1 +1 +1 +i Q. 4-1 44 44 41 44 (cl«^ u 1 0 CO (^ CO !_ < . to CO CO i»* •> vt w "3* 00 CO CN O O ,?r 0 •sf- CN o> " 0 O 36 7 02 3 3 0 45 3 0 CO a. +1 z t diff e < 0 0 0 O 0 0 0 0 O 0 O O 1- en 0 0 05 CD U) M 0 LO O 0 en 0 0 t™u: +1 +1 +1 +1 +1 +1 44 +1 44 +1 +! +1 0 0 0 0 O 0 c , J O) 0 CN co CO co ri 0 CN CN CO OJ CD H-l 44 44 4-i 4-1 CO £ m 0 0 7 Gu .* 1> «& N O Oi N J 11) KK fi N u _ «t n "> 00 U> c's +1 c cm o CO OO „K Q O K CS CO CO CN w CN ™ -* CN £ 2 O 05 ^ O CTJ (X; CO CO M O i- O « O OOOOOOO OO cu a n» 0 d d d d mean s ni w O *§ 4-1 4-1 +1 4-1 4-1 +1 +1 +1 +! +1 +1 +1 +1 +1 +! PJ 0 <= 2 c*> a. ^- u") sa m rt •« 05 CM CN r>- CN CC CO *^ M CO 1^- CO CO .2 CS -et; CN t~ O CO CO O kft CO T3 C o 05 0)0 0 1- CD T3 LTS Lfj CQ' IN « * 6 di 6 d cn 9J S 41 0 a. 5f cn CU 3 J!! CS ra Cti ffi j|£|l CO a CO 3 no » to » « * co CO > CO 04 6 36 7 "*; CN O O O O CO CI O 0 0 0 OOOOOOOO +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 CO co r- •* 0 CN CD CD O O O U"> tfi to r o> <0 0J to 0 0 c C «c c c «c «c •e- p» T- •* 10 to .;- to «o « T~ •$— 80 to 0 0 0 «N T «r» 0 65 Table 15 Trial 2: Muscle fatty acid composition of juvenile Atlantic cod fed the experimental diets over 12 weeks. Presented as % of total fatty acids (IBW=15.92 ± 0.36 g). Diet Control 2 (SPC2 5 SPC50 total % n-3 48.34 + 0.896 48.77 + 0.896 47.05 + 0.896 total % n-6 10.05 + 0.141a 9.39 + 0. 141ab 9.03 + 0. 141b 16:00 17.7 + 0.301 17.5 + 0.301 17.55 + 0.301 18:00 3.69 + 0.061 3.88 + 0.061 3.82 + 0.061 18:1n9 8.84 + 0.170a 8.30 + 0. 170ab 7.90 ± 0.170b 18:2n6 6.95 + 0.169a 6.21 + 0. 169ab 5.59 + 0.169b 18:3n3 0.78 + 0.029 0.80 + 0.029 0.75 + 0.029 20:1 n9 1.18 + 0.051 1.03 + 0.051 1.03 + 0.051 20:1n11 0.31 + 0.059 0.08 ± 0.059 0.22 + 0.059 20:4n6 2.33 + 0.029 2.53 + 0.029 2.57 + 0.029 20:5n3 16.39 + 0.253a 15.91 + 0. 253ab 15.01 + 0.253b 22:6n3 27.04 + 0.837 27.6 + 0.837 26.94 + 0.837 Values presented are means ± SE. Values with common postscripts are not significantly different (P < 0.05, Tukey's HSD). Table 16 Trial 2: Muscle fatty acid composition of juvenile Atlantic cod fed the experimental diets over 12 weeks. Presented as g/100 g tissue (IBW=15.92 ± 0.36 g). Diet Control 2 SPC25 SPC50 total % n-3 0.23 + 0.004 0.23 + 0.004 0.23 ± 0.004 total % n-6 0.05 + 0.001 0.04 + 0.001 0.04 + 0.001 16:00 0.08 + 0.001 0.08 + 0.001 0.08 + 0.001 18:00 0.02 + 0.001 0.02 + 0.001 0.02 + 0.001 18:1n9 0.04 + 0.002 0.04 + 0.002 0.04 + 0.002 18:2n6 0.03 + 0.002 0.03 + 0.002 0.03 + 0.002 18:3n3 0.00 + 0.001 0.00 + 0.001 0.00 + 0.001 20:1n9 0.01 + 0.001 0.00 + 0.001 0.00 + 0.001 20:1n11 0.00 + 0.001 0.00 + 0.001 0.00 + 0.001 20:4n6 0.01 + 0.001a 0.01 + 0.001ab 0.01 + 0.001b 20:5n3 0.08 + 0.002 0.07 + 0.002 0.07 + 0.002 22:6n3 0.13 + 0.002 0.13 + 0.002 0.13 + 0.002 Values presented are means ± SE. Values with common postscripts are not significantly different (P < 0.05, Tukey's HSD). 66 Gut Histology The results of histopathological scoring are presented in Tables 17 and 18. No significant differences among treatment groups were observed in the size or cellularity of the LP or the infiltration of inflammatory cells through the intestinal mucosa or associated connective tissues. No gross enteritis was observed nor were there any increases in size or abundance of goblet cells in either the distal or proximal intestines (Figures 2a-d and 3a-d). In both trials, several samples had.high total inflammation scores, greater than 12, indicating that inflammation was present in some fish. The number offish with lymphocyte scores of 5, indicating the presence of inflammatory cells throughout the intestinal mucosa (Figures 2b, 2d, 3b and 3d) are presented in Table 20. In both trials, variability was high among individuals and no treatment-related trends were apparent. 67 Figure 2. Trial 1: Histomicrographs of juvenile Atlantic cod intestinal mucosae: (a) fish fed a soy-free control diet with a total inflammation score of 5, (b) fish fed a soy-free control diet with a total inflammation score of 12, (c) fish fed the SPC40 diet with a total inflammation score of 7, and (d) fish fed the SPC40 diet with a total inflammation score of 12. em: epithelial mucosae; Ip: lamina propria; gc: goblet cell; ct: connective tissue; iel: intraepithelial lymphocytes A 3 •* . •. i » C D At 68 Figure 3. Trial 2: Histomicrographs of juvenile Atlantic cod intestinal mucosae: (a) fish fed a commercially available control diet with a total inflammation score of 5, (b) fish fed a commercially available control diet with a total inflammation score of 12, (c) fish fed the SPC50 diet with a total inflammation score of 6, and (d) fish fed the SPC50 diet with a total inflammation score of 9. em: epithelial mucosae; Ip: lamina propria; gc: goblet cell; ct: connective tissue; iel: intraepithelial lymphocytes; art: histological artifact 69 O S3 T~ C3 v- O *3- CO CO •* <£> it V- T- Co o o r- CN co co •<$• so f CN CO is. o *- ••• CO rf CO CN co CM o co *- o i™ o •* f- CN CO CO t CO » ^ ^ -i-r *sj~ CN WO r- (ij S O T- r (i) » « CT ?; w ^ r- e CO CN CN CO „J " i- S- d o o d d CD d d d o d CN d d +1 +1 +! +! +1 +| +1 +1 +1 +1 +! +1 +1 +1 CN T- o CD CO' CO uO u> -^~ o T~ ci CD *~ CO CN csi d 00 CM CO CN cx> c\i ^ Is. d 166 . CO a CO CO CN d co CO CO CO CM CN l> CO .15 0 .61 4 u .27 2 .31 8 1.1 4 d d d d o o • o o. d o CM d d +1 +1 +1 +! +1 +1 +1 •H +1 +1 +1 +1 +1 +1 o N n o s o q '<£> <- K i- CS W CO fsi CN d CM d d 00 CN CM d CN h~ CN d CN CO ex:- CN •* CM o .31 0 .31 8 d .24 0 .15 0 .81 4 .33 7 co .13 6 d CM o o d © o o o d d GS CM o d +[ +i +1 +i +i +i -H ^- +i +l '+l +i +l +l +i CM *3* •«- •^j- a CN as <-• CO CO Is. CO 03 CC' d d v" CO d a> oi CM *™ TO .«,.. l-~ ^'"' d o "O w O O C £ T3 o o U o « 3 •3 3 f5 ra W a J2 !2 V) ri E £L a. w o« J5 o "*«^ jn **- 6 _ *i *: £ > >. c CL OL «j *a* *i> "•^ 0 o % o C « 1 V* c £ £ c E Jj c o o ^o # ff3 o »o a j « « O J> . 0 O <> o _» O O < uL 70 O) M « « ^ N CO 23 0 28 8 23 3 CM £L> 3.0 2 O T^ o u o o d o ed o T-' o o o o t£^ ?"=- +i +i +l +i +i +i +i '!> O •JZZ ~2L LCJ (N a) h~ h~ h~ CM *r- uo cq CO e> co CN CN r- W5 r-' CO i; < tN CM *-' CO •'-• T-^ 05 03 4r a «*** . , co "±:. O Z £ Lii 3, n. CO o *~* o ^-^ +1 o m o m CCJ © zzz~ o .~~^ c ^" « T™ w ra >* » %= >1 Sf o tfj ra a> re o m u. 3 £ c»> ^^ c e M ra * (3 o _o eliu l "35 ~E N s T3 01 tf} c ilJ JJ 55 o 3 c m u 3 C^ 3' 3 E S°5 ^ rt « -3Q E i« a JQ >O 0) IS M m es 1 c •j^* m 01 15 b ja tf) •** 01 f2S R IS ra a. a. _0 j«^ a, tft Q. o » ™ o » 71 Table 19. Number of fish with inflammatory cells present throughout the intestinal tissues. Trial 1 Control 1 SPC10 SPC20 SPC30 SPC40 Proximal Intestine 1 0 1 0 0 Distal Intestine 0 0 3 0 2 Trial 2 Control 2 SPC25 SPC50 Proximal Intestine 0 0 0 Distal Intestine 1 ° ° Discussion Atlantic cod perform remarkably well when fed alternative protein diets, including those that contain terrestrial plant proteins (Aksnes et al. 2006; Albrektsen et al. 2006; Forde-Skjaervik et al. 2006; Hansen et al. 2006; Rosenlund and Skretting 2006; Toppe et al. 2006; Hansen et al. 2007a; Hansen et al. 2007b). This species has a robust and dynamic digestive system which, given time, seems capable of adapting to new feed formulations (Refstie et al. 2006a; Ringo et al. 2006). Further, cod have repeatedly compensated for diets with lower digestibility by increasing feed intake (Albrektsen et al. 2006; F0rde- Skjaarvik et al. 2006; Refstie et al. 2006a; Hansen et al. 2007b). In contrast to other studies that included SBM or other less refined plant proteins in the diet (Hansen et al. 2006; Refstie et al. 2006a), the feed intake and FCRs in the present trials do not suggest a need for compensatory feed intake. Although nutrient uptake was not directly measured in the present study, consistent growth and feed intake suggest that it was not hindered by the inclusion of SPC. This is to be expected, as Tibbets et al. (2006) reported high protein and lipid apparent 72 digestibility coefficients for SPC (99% and 95%) when fed to Atlantic cod, and the low-antigen SPC used in the present study was functionally devoid of the antinutritional factors present in other soy-derived proteins. The upper inclusion levels of alternative protein ingredients in fish diets is typically limited by several factors including amino acid and fatty acid profiles, presence of antinutritional factors, digestibility, palatability, availability, and processing costs (Gatlin et al. 2007). The dietary formulation fed in trial 1 was designed to entirely replace fishmeal in Atlantic cod diets without inducing intestinal inflammation or negatively impacting growth. A low antigen SPC was successfully used to replace dietary fishmeal (50% d.w. of the control) such that the experimental diets were composed of 10, 20, 30, or 40% soy protein and the dry weight of fishmeal in the diet was reduced by 28, 60, 80, or 100%. The high cost of SPC, however, is a significant detriment to this formulation. In contrast, trial two tested the applicability of substituting SPC in a commercially available fish diet. This diet contained modest amounts of SBM as well as blood meal and poultry by-product meal, all well-established terrestrial alternative proteins. At the highest SPC inclusion level in this trial, fishmeal comprised only 19.3% of a diet that should be reasonably cost effective. Growth was consistent in both of the present trials and comparable to a previous trial using similarly processed diets (Walker et al. 2009, see Chapter 2). A limitation to both of the diets tested was the research-scale processing method used to produce the compressed pellets. It is likely that growth would have been 73 improved if the diets tested were processed by extrusion technology. Several studies have shown that extruded diets have substantially higher overall digestible energy, particularly for dietary carbohydrates (Barrows et al. 2007). Using extruded diets, fingerling through sub-adult Atlantic cod can attain growth rates and FCRs near 1.0 (Aksnes et ai. 2006; Albrektsen et al. 2006; Hansen et al. 2007a; Karalazos 2007), During the first week of trial 1, feed intake was negatively correlated with SPC inclusion, suggesting decreased palatability of the diets. By the second week, however, there were no apparent differences in feed acceptance. In trial 2, the diets were well accepted throughout the feeding trial. These observations are in contrast with that reported by Hansen et al. (2007b) and Olsen et al. (2007) who reported reduced feed acceptance in sub-adult cod fed a diet comprised of 100% plant protein mix (14% SBM, 36% SPC, 50% wheat gluten) resulting in fish that the authors described as near a "food-deprivation status" at the end of the study. Differences in palatability between the trial 1 diet and the SBM, SPG, and wheat gluten mix used by Hansen could be related to an approximate 5% inclusion of hydrolyzed fish protein concentrate in the present study, or to the MHA-methionine, both of which may have acted as attractants (Huai et al. 2009). Also, Hansen et al. used larger fish (IBW: 1.7 kg) raised in colder water (5-9 °C), which likely affected feed intake. One negative effect of using terrestrial plant proteins in many fish diets is the subsequent incorporation of shorter chain terrestrial plant oils (e.g., 18:2n-6) in the fillet (Sargent et al. 2002; Karalazos 2007; Naylor et al. 2009) at the 74 expense of n-3 fatty acids. Medical research has shown that consuming only 6 ounces of oily marine fish per week can reduce the occurrence of sudden death from coronary heart disease in humans by more than 25% when compared to no fish intake (Mozaffarian and Rimm 2009). In cod, the effects of terrestrial plant oils are most likely to be seen in the liver, which functions as the primary lipid storage site. Atlantic cod have also been shown to preferentially metabolize shorter-chain fatty acids, thereby maintaining the fillet composition even when fed diets rich in plant oils (Karalazos 2007). In trial 1, liver fatty acid profiles were more similar to the diets than were the fillet profiles. The SPC ingredient used is very low in lipid and because fish oil was used to mitigate these lower levels of fat compared to the fishmeal that it replaced, the final liver tissues had increased levels of AA and EPA, and decreased levels of terrestrial 18-C fatty acids. In trial 2, reduced EPA in the diet was reflected in the muscle, but decreased dietary AA levels did not affect muscle composition. The antinutritional factors in soy-derived proteins have long been known to limit the inclusion levels of SBM in fish diets, particularly in salmonids. Atlantic salmon develop "soybean-induced non-infectious subacute enteritis" when fed diets containing soy-derived antigenic compounds (Ingh et al. 1991; Baeverfjord and Krogdahl 1996). Additionally, indigestible carbohydrates in SBM induce increased fecal moisture content and may increase the rate of gut evacuation, thereby lowering nutrient uptake (Storebakken et al. 2000b). These two conditions differ both in their etiology and in their occurrence in Atlantic cod. 75 In cod, replacement of 25% fishmeal with either SBM or BPSBM has increased fecal moisture content and decreased FCRs. Non-starch polysaccharides and a-galactoside oligosaccharides, which make up 20% and 10% of soybean dry weight, respectively, are the most likely causative agents for soybean-induced diarrhea in fish (Storebakken et al. 2000b). In particular, Refstie et al. (2006a) reported diarrhea in BPSBM-fed cod. BPSBM is largely free of NSPs, which further suggests an effect from a-galactoside sugars. Soy oligosaccharides, which are present in the alcohol-soluable soybean molasses, have been implicated for inducing enteritis in salmonids (Ingh et al. 1996). Knudsen et al. (2007), however, phase-separated this fraction and found that soya-saponins, or other compounds with similar n-butanol soluability, are the most likely agents, possibly in combination with other unknown compounds (Knudsen et al. 2008). The processing of SBM to SPC includes alcohol extraction of the soybean molasses, so the lack of enteritis in the present study and others using SPC (Olli et al. 1994b; Ingh et al. 1996) is not unexpected. The same is not true for all refined soy ingredients. Soya-saponins are concentrated in the extraction process used to produce SPI, resulting in a higher saponin concentration than that in SBM (Knudsen et al. 2007). Soybean-induced enteritis is dominated by the infiltration of inflammatory cells in the epithelial mucosae and LP as well as a shortening of the microvilli. The regular arrangement of basal nuclei and supranuclear absorptive vacuoles is lost and, in extreme cases, the enterocyte nuclei and brush border become necrotic. Only one study (Olsen et al. 2006) described necrosis related to dietary 76 plant proteins in cod, and then, only in fish with limited feed acceptance of a diet that replaced 100% of fishmeal with a plant protein mixture (14% bio-processed SBM, 36% SPC, and 50% WG). At lower inclusion levels, histopathological r' changes were limited to an increase in the size and number of the goblet cells and an infiltration of inter-epithelial lymphocytes, relegated primarily to the distal intestine (Olsen et al. 2007). The authors related this phenomenon to the increase in the concentration of indigestible antigenic components as the chyme traveled through the intestinal tract. The cellular composition of the intestines in Atlantic cod differs from that of salmonids in that they lack the regular arrangement of basal nuclei and apical absorptive vacuoles (Inami et al. 2009). The intestinal vuculae are dominated by secretory goblet cells and absorptive, vacuoles are fewer. The villi are not clearly simple or compound as they are described in salmonids (Baeverfjord and Krogdahl 1996; Escaffre et al. 2007) but, at least in the present study, are arranged with distinct groupings of several villi and associated glands separated from the musculature by submucosal connective and lymphoid tissues. Finally, Atlantic cod lack distinct differences in the enterocytes; such as the presence of secretory vacuoles, that delineate the proximate from the distal intestine (Inami et al. 2009). These structural differences, as well as a greater potential for nutrient absorption through the pyloric caeca (Hansen et al. 2006), may contribute to the consistent digestibility, nutrient uptake, growth, and gut tolerance demonstrated by Atlantic cod fed soy- derived proteins. 77 In summary, growth and feed conversion were not affected when fishmeal was entirely replaced by SPC in juvenile cod diets, nor was growth or feed performance reduced by the inclusion of SPC in a commercial diet formulation. To maximize the inclusion of soy-derived proteins in cod diets, further research is necessary to derive efficient SBM/SPC combinations that improve cost effectiveness. 78 CHAPTER 4 CONCLUSIONS The two sets of experiments conducted demonstrate that Atlantic cod are equipped to take advantage of plant-based dietary formulations that may become commonplace for juvenile grow-out in the future. The species does not require the expensive and resource-intensive fi.shmeal-only diets that are associated with growing marine carnivores. Juvenile Atlantic cod have consistently demonstrated the ability to increase feed intake to meet energy requirements and maintain growth. They have highly adaptable digestive systems, which are not prone to acute inflammation in response to soy proteins. In the first study, a novel protein source, dried Porphyra, adequately replaced fishmeal in diets of fingerling cod. Growth and feed conversion were not affected when up to 30% of the fishmeal in juvenile cod diets was replaced with the dried Porphyra. Fatty acid profiles were similar in the muscle tissue in all treatment groups, except AA levels, which were higher in fish fed the high Porpfryra-supplemented diet. In the second study, soy-derived proteins, which are perhaps the best candidates for alternative protein development, were evaluated. Growth and feed conversion was not affected when fishmeal was entirely replaced by SPC, nor was growth or feed performance reduced by the inclusion of SPC in a commercial diet formulation. 79 Juvenile fish, such as those used in the present studies, are especially good candidates for alternative protein diets because they are in a growth period that requires very high feed inputs. Additionally, as long as essential nutrient requirements are met, it is of less concern if juveniles incorporate terrestrial-type fatty acids into their muscle tissues. Nevertheless, research into the long-term effects of soy and other alternative proteins, over the entire grow-out period, is warranted. Currently, there are no perfect alternative proteins on the market. The best suited proteins, probably the protein concentrates, are too expensive for regular use in aquaculture feeds at the levels presented in Chapter 3. At present, soy protein concentrate is more expensive than fishmeal. This could change if fish proteins become less available, or if the cost of soy protein concentrates decrease. New options constantly become available for feed manipulation and improved fish nutrition. Technologies such as the development of transgenic n-3-rich oilseeds, enzymatic processing techniques to remove anti- nutritional factors, and the breeding offish to improve antigen tolerance should increase the options for fishmeal reduction going forward. In addition to these approaches, future research should be aimed at evaluating finishing diets which would allow for lower inclusion levels of fish meal and fish oil for the majority of the growth period, followed by a short-term feeding of diets high in n-3 fatty acids for tissue incorporations prior to slaughter (Naylor et al. 2009). 80 LITERATURE CITED Abramoff, M. D., P. J. Magalhaes and S. J. Ram (2004). Image processing with ImageJ. Biophotonics International 11: 36-42, Adelizi, P. D., R. R. Rosati, K. Warner, Y. V. Wu, T. R. Muench, M. R. White and P. B. Brown (1998). Evaluation offish-meal free diets for rainbow trout, Oncorhynchus mykiss. Aquaculture Nutrition 4(4): 255-262.refs Ai, Q., K. Mai, B. Tan, W. Xu, Q. Duan, H. Ma and L. Zhang (2006). Replacement of fish meal by meat and bone meal in diets for large yellow croaker, Pseudosciaena crocea. Aquaculture 260(1-4): 255-263. Akiyama, T., I. Oohara and T. Yamamoto (1997). Comparison of essential amino acid requirements with A/E ratio among fish species (Review paper). Fisheries Science 63(6): 963-970. Aksnes, A., B. Hope, O. Hostmark and S. Albrektsen (2006). Inclusion of size fractionated fish hydrolysate in high plant protein diets for Atlantic cod, Gadus morhua. Aquaculture 261 (3): 1102-1110. Albrektsen, S., H. Mundheim and A. Aksnes (2006). Growth, feed efficiency, digestibility and nutrient distribution in Atlantic cod (Gadus morhua) fed two different fish meal qualities at three dietary levels of vegetable protein sources. Aquaculture 261(2): 626-640. Baeverfjord, G. and A. Krogdahl (1996). Development and regression of soybean meal induced enteritis in Atlantic salmon, Salmo salar L, distal intestine: A comparison with the intestines of fasted fish. Journal of Fish Diseases 19(5): 375-387. Barrows, F. T., D. A. J. Stone and R. W. Hardy (2007). The effects of extrusion conditions on the nutritional value of soybean meal for rainbow trout (Oncorhynchus mykiss). Aquaculture 265(1-4): 244-252. 81 Bell, J. G., F. McGhee, P. J. Campbell and J. R. Sargent (2003). Rapeseed oil as an alternative to marine fish oil in diets of post-smolt Atlantic salmon (Salmo salar): Changes in flesh fatty acid composition and effectiveness of subsequent fish oil "wash out". Aquaculture 218(1-4): 515-528. Bell, J. G. and J. R. Sargent (2003). Arachidonic acid in aquaculture feeds: current status and future opportunities. Aquaculture 218(1-4): 491-499. Bjornsson, B. and S. R. Olafsdottir (2006). Effects of water quality and stocking density on growth performance of juvenile cod (Gadus morhua L). ICES Journal of Marine Science 63(2): 326-334. Blouin, N., B. L. Calder, B. Perkins and S. H. Brawley (2006). Sensory and fatty acid analyses of two Atlantic species of Porphyra (Rhodophyta). Journal of Applied Phycology 18(1): 79-85. Blouin, N., F. Xiugeng, J. Peng, C. Yarish and S. H. Brawley (2007). Seeding nets with neutral spores of the red alga Porphyra umbilicalis (L.) Kutzing for use in integrated multi-trophic aquaculture (IMTA). Aquaculture 270(1- 4): 77-91. Carmona, R„ G. P. Kraemer and C. Yarish (2006). Exploring Northeast American and Asian species of Porphyra for use in an integrated finfish-algal aquaculture system. Aquaculture 252(1): 54-65. Choi, S.-M., Z. Wan, S.-R. Lim, K.-W. Kim, S. C. Bai and l.-S. Sin (2004). Dietary dehulled soybean meal as a replacement for fish meal in fingerling and growing olive flounder Paralichthys olivaceus (Temminck et Schlegel). Aquaculture Research 35: 410-418. Cowey, C. B. (1994). Amino acid requirements offish: A critical appraisal of present values. Aquaculture 124: 1-11. Davies, S. J., M. T. Brown and M. Camilleri (1997a). Preliminary assessment of the seaweed Porphyra purpurea in artificial diets for thick-lipped grey mullet (Chelon labrosus). Aquaculture 152(1-4): 249-258. Davies, S. J., P. C. Morris and R. T. M. Baker (1997b). Partial substitution offish meal and full-fat soya bean meal with wheat gluten and influence of lysine 82 supplementation in diets for rainbow trout, Oncorhynchus mykiss (Walbaum). Aquaculture Research 28(5): 317-328. De Silva, S. S. and T. A. Anderson (1994). Fish Nutrition in Aquaculture, Chapman and Hall, New York, NY (USA). Deng, J., K. Mai, Q. Ai, W. Zhang, X. Wang, W. Xu and Z. Liufu (2006). Effects of replacing fish meal with soy protein concentrate on feed intake and growth of juvenile Japanese flounder, Paralichthys olivaceus. Aquaculture 258: 503-513. Dos Santos, J., V. C. Burkow and M. Jobling (1993). Patterns of growth and lipid deposition in cod (Gadus morhua L.) fed natural prey and fish-based feeds. Aquaculture 110(2): 173-189. Escaffre, A.-M., S. Kaushik and M. Mambrini (2007). Morphometric evaluation of changes in the digestive tract of rainbow trout (Oncorhynchus mykiss) due to fish meal replacement with soy protein concentrate. Aquaculture 273(1): 127-138. FAO. (2002). The State of World Fisheries and Aquaculture 2002. Electronic edition from http://www.fao.org/docrep/005/y7300e/y7300e00.htm. FAO. (2003). Review of the state of world aquaculture. FAO Fisheries Circular No. 886, Rev. 2. FAO (2006). State of World Aquaculture 2006: Advance Copy. Rome, Inland Water Resources and Aquaculture Service Fishery Resource Division: 145. FAO (2009). State of World Aquaculture 2008. Rome, Food and Agriculture Organization of the United Nations. Forde-Skjasrvik, O., S. Refstie, M. A. Aslaksen and A. Skrede (2006). Digestibility of diets containing different soybean meals in Atlantic cod (Gadus morhua); comparison of collection methods and mapping of digestibility in different sections of the gastrointestinal tract. Aquaculture 261(1): 241- 258. 83 Forster, I. P., W. Dominy, L. Obaldo and A. G. J. Tacon (2003). Rendered meat and bone meals as ingredients of diets for shrimp Litopenaeus vannamei (Boone, 1931). Aquaculture 219(1-4): 655-670. Francis, G., H. P. S. Makkar and K. Becker (2001). Antinutritional factors present in plant-derived alternate fish feed ingredients and their effects in fish. Aquaculture 199(3-4): 197-227. Friedland, R. P., R. B. Peterson and R. Rubenstein (2009). Bovine Spongiform Encephalopathy and Aquaculture. Journal of Alzheimer's Disease 17(2): 277-279. Gatlin, D. M., F. T. Barrows, P. Brown, K. Dabrowski, T. G. Gaylord, R. W. Hardy, E. Herman, G. Hu, A. Krogdahl, R. Nelson, K. Overturf, M. Rust, W. Sealy, D. Skonberg, E. J. Souza, D. Stone, R. Wilson and E. Wurtele (2007). Expanding the utilization of sustainable plant products in aquafeeds: a review. Aquaculture Research 38: 551-579. Hansen, A.-C, O. Karlsen, G. Rosenlund, M. Rim bach and G. I. Hemre (2007a). Dietary plant protein utilization in Atlantic cod, Gadus morhua L. Aquaculture Nutrition 13(1): 200-215. Hansen, A.-C, G. Rosenlund, O. Karlsen, W. Koppe and G.-l. Hemre (2007b). Total replacement offish meal with plant proteins in diets for Atlantic cod (Gadus morhua L.) I - Effects on growth and protein retention. Aquaculture 272(1-4): 599-611. Hansen, A.-C, G. Rosenlund, O. Karlsen, P. A. Olsvik and G.-l. Hemre (2006). The inclusion of plant protein in cod diets, its effects on macronutrient digestibility, gut and liver histology and heat shock protein transcription. Aquaculture Research 37(8): 773-784. Hardy, R. W. (2006). "Fish Meal Prices Drive Changes in Fish Feed Formulations". Aquaculture Magazine: 29-31. Harper, C. R. and T. A. Jacobson (2005). Usefulness of omega-3 fatty acids and the prevention of coronary heart disease. American Journal of Cardiology 96(11): 1521-1529. 84 Hashim, R. and N. A. Mat Saat (1992). The utilization of seaweed meals as binding agents in pelleted feeds for snakehead (Channa striatus) fry and their effects on growth. Aquaculture 108(3-4): 299-308. Hemre, G., O. Karlsen, K. Eckhoff, K. Tveit, A. Mangor-Jensen and G. Rosenlund (2004). Effect of season, light regime and diet on muscle composition and selected quality parameters in farmed Atlantic cod, Gadus morhua L Aquaculture Research 35(7): 683-697. Huai, M.-Y., Y.-J. Liu, L.-X. Tian, S.-X. Deng, A.-L. Xu, W. Gao and H.-J. Yang (2009). Effect of dietary protein reduction with synthetic amino acids supplementation on growth performance, digestibility, and body composition of juvenile Pacific white shrimp, Litopenaeus vannamei. Aquaculture International. Online publication: 15 February 2009. Inami, M., A. J. Taveme-Thiele, M. B. Schroder, V. Kiron and J. H. W. M. Rombout (2009). Immunological differences in intestine and rectum of Atlantic cod (Gadus morhua L). Fish & Shellfish Immunology 26(5): 751- 759. Ingh, T. S. G. A. M. v. d., A. Krogdahl, J. J. Olli, H. G. C. J. M. Hendriks and J. G. J. F. Koninkx (1991). Effects of soybean-containing diets on the proximal and distal intestine in Atlantic salmon {Salmo salar): a morphological study. Aquaculture 94(4): 297-305. Ingh, T. S. G. A. M. v. d., J. J. Olli and A. Krogdahl (1996). Alcohol-soluble components in soybeans cause morphological changes in the distal intestine of Atlantic salmon, Salmo salar L. Journal of Fish Diseases 19(1): 47-53. Jorstad, K., T. van der Meeren, O. I. Paulsen, T. Thomsen, A. Thorsen and T. Svasand (2008). Escapes of Eggs from Farmed Cod Spawning in Net Pens: Recruitment to Wild Stocks. Reviews in Fisheries Science 16: 285- 295. Karalazos, V. T., J; Cutts, CJ; Alderson, R; Galloway, TF; Albrektsen, S; Arnason, J; MacDonald, N; Pike, I; Bell, JG (2007). Effects of Fish Meal Replacement with Full-Fat Soy Meal on Growth and Tissue Fatty Acid Composition in Atlantic Cod (Gadus morhua). Journal of Agricultural and Food Chemistry 55(14): 5788-5795. 85 Kikuchi, K. (1999). Partial replacement offish meal with corn gluten meal in diets for Japanese flounder Paralichthys olivaceus. Journal of the World Aquaculture Society 30(3): 357-363. Knudsen, D., F. Jutfelt, H. Sundh, K. Sundell, W. Koppe and H. Frokaer (2008). Dietary soya saponins increase gut permeability and play a key role in the onset of soyabean-induced enteritis in Atlantic salmon (Salmo salarL.). British Journal of Nutrition 100: 120-129. Knudsen, D., P. Uran, A. Arnous, W. Koppe and H. Frokiaer (2007). Saponin- containing subtractions of soybean molasses induce enteritis in the distal intestine of Atlantic salmon. Journal of Agricultural and Food Chemistry 55(6): 2261-2267. Krogdahl, A., A. M. Bakke-McKellep and G. Baeverfjord (2003). Effects of graded levels of standard soybean meal on intestinal structure, mucosal enzyme activities, and pancreatic response in Atlantic salmon (Salmo salar L). Aquaculture Nutrition 9(6): 361-371. Lie, O., E. Lied and G. Lambertsen (1986). Liver Retention of Fat and of Fatty Acids in Cod Gadus-Morhua Fed Different Oils. Aquaculture 59(3-4): 187- 196. Lyndon, A. R., I. Davidson and D. F. Houlihan (1993), Changes in tissue and plasma free amino acid concentrations after feeding in Atlantic cod Fish Physiology and Biochemistry 10(5): 365-375. Mambrini, M., A. J. Roem, J. P. Cravedi, J. P. Lalles and S. J. Kaushik (1999). Effects of replacing fish meal with soy protein concentrate and of DL- methionine supplementation in high-energy, extruded diets on the growth and nutrient utilization of rainbow trout, Oncorhynchus mykiss. Journal of Animal Science 77(11): 2990-2999. Morais, S., J. G. Bell, D. A. Robertson, W. J. Roy and P. C. Morris (2001). Protein/lipid ratios in extruded diets for Atlantic cod (Gadus morhua L.): effects on growth, feed utilisation, muscle composition and liver histology. Aquaculture 203(1-2): 101-119. Morrison, C. M. (1987). Histology of Atlantic Cod, Gadus morhua: An Atlas. Part One. Digestive tract and asociated organs. Ottowa, National Research Council of Canada. 86 Moyano, F. J., G. Cardenete and M. de la Higuera (1992). Use of two vegetable by-products as protein sources in rainbow trout feeding. Animal Production 55(2): 277-284. Mozaffarian, D. and E. Rimm (2009). Fish Intake, Contaminants, and Human Health: Evaluating the Risks and the Benefits. Journal of the American Medical Assocaition 296(15): 1885-1889. Mustafa, M. G. and H. Nakagawa (1995). A review: Dietary benefits of algae as an additive in fish feed. Israeli Journal of Aquaculture 47(3-4): 155-162. Mustafa, M. G., S. Wakamatsu, T.-A. Takeda, T. Umino and H. Nakagawa (1995). Effects of algae meal as feed additive on growth, feed efficiency, and body composition in Red Sea Bream. Fisheries Science (Tokyo) 61(1): 25-28. Naylor, R. L, R. W. Hardy, D. P. Bureau, A. Chiu, M. Elliott, A. P. Farrell, I. Forster, D. M. Gatlin, R. J. Goldburg, K. Hua and P. D. Nichols (2009). Feeding aquaculture in an era of finite resources. Proceedings of the National Academy of Sciences. 106: 15103-15110. Neori, A., T. Chopin, M. Troell, A. H. Buschmann, G. P. Kraemer, C. Hailing, M. Shpigel and C. Yarish (2004). Integrated aquaculture: Rationale, evolution and state of the art emphasizing seaweed biofiltration in modern mariculture. Aquaculture 231 (1 -4): 361 -391. Olli, J. J., K. Hjelmeland and A. Krogdahl (1994a). Soybean trypsin inhibitors in diets for Atlantic salmon (Salmo salar, L): effects on nutrient digestibilities and trypsin in pyloric ceaca homogenate and intestinal content. Comparative biochemistry and physiology 109A: 923-928. Olli, J. J., A. Krogdahl, T, S. G. A. M. van den Ingh and L. E. Brattaas (1994b). Nutritive value of force soybean products in diets for Atlantic salmon (Salmo salar L). Acta Agriculturae Scandinavica Section A 44: 50-60. Olsen, R. E., A.-C. Hansen, G. Rosenlund, G.-l. Hemre, T. M. Mayhew, D. L. Knudsen, O. Tufan Eroldogan, R. Myklebust and O. Karlsen (2007). Total replacement of fish meal with plant proteins in diets for Atlantic cod (Gadus morhua L) II - Health aspects. Aquaculture 272(1-4): 612-624. 87 Peet, M. and C. Stokes (2005). Omega-3 fatty acids in the treatment of psychiatric disorders. Drugs 65(8): 1051-1059. Pereira, T. G. and A. Oliva-Teles (2003). Evaluation of corn gluten meal as a protein source in diets forgilthead sea bream (Sparus aurata L.) juveniles. Aquaculture Research 34(13): 1111-1117. Peres, H. and A. Oliva-Teles (2006). Effect of the dietary essential to non essential amino acid ratio on growth, feed utilization and nitrogen metabolism of European sea bass (Dicentrarchus labrax). Aquaculture 256(1-4): 395-402. Pike, I. H., G. Andorsdottir and H. Mudheim (1990). The role of fish meal in diets for salmonids. International Association For Fishmeal Manufacturing Technical Bulletin: No. 24, 35 pp. Refstie, S., O. Forde-Skjaervik, G. Rosenlund and K. Rorvik (2006a). Feed intake, growth, and utilization of macronutrients and amino acids by 1-and 2-year old Atlantic cod (Gadus morhua) fed standard or bioprocessed soybean meal. Aquaculture 255(1-4): 279-291. Refstie, S., T. Landsverk, A. M. Bakke-McKellep, E. Ringo, A. Sundby, K. D. Shearer and A. Krogdahl (2006b). Digestive capacity, intestinal morphology, and microflora of 1-year and 2-year old Atlantic cod (Gadus morhua) fed standard or bioprocessed soybean meal. Aquaculture 261(1): 269-284. Refstie, S., T. Storebakken and A. J. Roem (1997). Adaptation to soybean meal in diets for rainbow trout, Oncorhynchus mykiss. Aquaculture 153: 263- 272. Reigh, R. C. and S. C. Ellis (1992). Effects of dietary soybean and fish-protein ratios on growth and body composition of red drum (Sciaenops ocellatus) fed iso-energetic diets. Aquaculture 104: 279-292. Ringo, E., S. Sperstad, R. Myklebust, S. Refstie and A. Krogdahl (2006). Characterisation of the microbiota associated with intestine of Atlantic cod (Gadus morhua L.) - The effect of fish meal, standard soybean meal and a bioprocessed soybean meal. Aquaculture 261(3): 829-841. 88 Roberts, R. J. (2002). Nutritional Pathology. Fish Nutrition. In: J. E. Halverand R. W. Hardy (Eds.). San Diego, Academic Press: 453-504. Rosenlund, G., O. Karlsen, K. Tveit, A. Mangor-Jensen and G. I. Hemre (2004). Effect of feed composition and feeding frequency on growth, feed utilization and nutrient retention in juvenile Atlantic cod, Gadus morhua L. Aquaculture Nutrition 10(6): 371-378. Rosenlund, G. and M. Skretting (2006). Worldwide status and perspective on gadoid culture. ICES Journal of Marine Science 63(2): 194-197. Sanz, A., M. Garcia Gallego and M. De la Higuera (2000). Protein nutrition in fish: Protein/energy ratio and alternative protein sources to fish meal. Journal of Physiology and Biochemistry 56(3): 275-282. Sargent, J. R., D. R. Tocher and G. J. Bell (2002). The Lipids. Fish Nutrition. In: J. E. Halver and R. W. Hardy (Eds.). San Diego, USA, Academic Press: 181-257. Satoh, K., H. Nakagawa and S. Kasahara (1987). Effect of Ulva meal supplementation on disease resistance of red sea bream. Nippon Suisan Gakkaishi 53(7): 1115-1120. Sitja-Bobadilla, A., S. Pena-Llopis, P. Gomez-Requeni, S. K. Medale, F. and P.- S. J. (2005). Effect offish meal replacement by plant protein sources on non-specific defence mechanisms and oxidative stress in gildhead sea bream (Spams aurata). Aquaculture 249: 387-400. Soler-Vila, A., S. Coughlan, M. Guiry and S. Kraan (2009). The red alga Porphyra dioica as a fish-feed ingredient for rainbow trout (Oncorhynchus mykiss): effects on growth, feed efficiency, and carcass composition. Journal of Applied Phycology 21(5): 617-624. Storebakken, Shearer, Roem and Storebakken (2000a). Growth, uptake and retention of nitrogen and phosphorus, and absorption of other minerals in Atlantic salmon Salmo salar fed diets with fish meal and soya-protein concentrate as the main sources of protein. Aquaculture Nutrition 6(2): 103-108. 89 Storebakken, T., I. S. Kvien, K.D. Shearer, B. Grisdale-Helland, S. J. Helland and G. M. Berge (1998). The apparent digestibility of diets containing fish meal, soybean meal or bacterial meal fed to Atlantic salmon (Salmo salar): evaluation of different fecal collection methods. Aquaculture 169: 195-210. Storebakken, T., S. Refstie and B. Ruyter (2000b). Soy Prodcuts as Fat and Protein Sources in Fish Feeds for Intensive Aquaculture. Soy in animal nutrition. In: J. K. Drackley (Eds.). Savoy, IL, Federation of Animal Science Societies: 170. Tibbetts, S. M., S. P. Lall and J. E. Milley (2005). Effects of dietary protein and lipid levels and DP DE super(-1) ratio on growth, feed utilization and hepatosomatic index of juvenile haddock, Melanogrammus aeglefinus L. Aquaculture Nutrition 11: 67-75. Tibbetts, S. M., J. E. Milley and S. P. Lall (2006). Apparent protein and energy digestibility of common and alternative feed ingredients by Atlantic cod, Gadus morhua (Linnaeus, 1758). Aquaculture 261(4): 1314-1327. Toppe, J., A. Aksnes, B. Hope and S. Albrektsen (2006). Inclusion offish bone and crab by-products in diets for Atlantic cod, Gadus morhua. Aquaculture 253: 636-645. Trushenski, J. T., C. S. Kasper and C. C. Kohler (2006). Challenges and Opportunities in Finfish Nutrition. North American Journal of Aquaculture 68(2): 122-140. USSEC (2008). Soy Protein Concentrate Aquaculture Feeds Technical Bulletin. St. Louis, MO US Soybean Export Council. Valente, L. M. P., A. Gouveia, P. Rema, J. Matos, E. F. Gomes and I. S. Pinto (2006). Evaluation of three seaweeds Gracilaria bursa-pastoris, Ulva rigida and Gracilaria cornea as dietary ingredients in European sea bass (Dicentrarchus labrax) juveniles. Aquaculture 252(1): 85-91. Vielma, J., J. Koskela, K. Ruohonen, I. Jokinen and J. Kettunen (2003). Optimal diet composition for European whitefish (Coregonus lavaretus): carbohydrate stress and immune parameter responses. Aquaculture 225(1-4): 3-16. 90 Viola, S., S. Mokady, U. Rappaport and Y. Arieli (1982). Partial and complete replacement offish meal with soybean meal in feeds for intensive culture of carp. Aquaculture 26: 223-236. Viyakarn, V., T. Watanabe, H. Aoki, H. Tsuda, H. Sakamoto, N. Okamoto, N. Iso, S. Satho and T. Takeuchi (1992). Use of soybean meal as a substitute for fish meal in a newly developed soft-dry pellet for yellowtail. Nippon Suisan Gakkaishi 58: 1991-2000. Von der Decken, A. and E. Lied (1993). Metabolic Effects on Growth and Muscle of Soybean Protein Feeding in Cod (Gadus-Morhua). British Journal of Nutrition 69(3): 689-697. Wahbeh, M. I. (1997). Amino acid and fatty acid profiles of four species of macroalgae from Aqaba and their suitability for use in fish diets. Aquaculture 159(1-2): 101-109. Walker, A. B., H. R. Fournier, C. D. Neefus, G. C. Nardi and D. L. Berlinsky (2009). Partial Replacement of Fish Meal with Laver Porphyra spp. in Diets for Atlantic Cod. North American Journal of Aquaculture 71 (1): 39- 45. Watanabe, T., V. Viyakarn, H. Kimura, K. Ogawa, N. Okamoto and N. Iso (1992). Utilization of soybean meal as a protein source in a newly developed soft- dry pellet for yellowtail. Nippon Suisan Gakkaishi 58: 1761-1773. Wilson, R. P. (2002). Amino acids and proteins. Fish Nutrition. In: J. E. Halver and R. W. Hardy (Eds.). San Diego, Academic Press: 143-179. Xu, B. T., S. Yamasaki and H. Hirata (1993). Supplementary Ulva sp. var. meal level in diet of Japanese flounder, Paralihthys olivaceus. Suisanzoshoku 41:461-468. Yone, Y., M. Furuichi and K. Urano (1986). Effects of dietary wakame Undaria pinnatifida and Ascophyllum nodosum on absorption of dietary nutrients, and blood sugar and plasma free amino-N levels of red sea bream. Bulletin of the Japanese Society of Scientific Fisheries 52: 1465-1468. Zar, J. H. (1999). Biostatistical Analysis. Upper Saddle River, NJ, Prentice Hall. 91 APPENDIX J live**-si *» i*/ Nit 'A I L-pii|-* S!M»% 2i-Ayf^ffll. •In , ' : I r ••-! MI JOT--1 Preset:"8 I "-\->;,.n» • • *• I s« .'. . I-I-I». •. . ii '. pr.^ui.U'. i .' Mid*'. . .•.- Category i" • ?• e«t Rev^w Date: "" ' * ~ > !n "^ ;:"£-'. u-s. 4nr*ii _.'?». i. u J."." 'in • 1?\" X1 r»s re "OWIL =I*L 4K.:.. <_'_ •, -r>" tJ > j.iniirn, - ft,. '" *>T "rt' s-nn- ><* g -rf ' I'll II"* 'N* -< "-v*-/, , |1 -' * »I i rj* — •,-X „, ,;,. ft i '„•. ci- *.',' . :.">" T»-- = .,ih frrrr-- HE r /;»i,x ,UO'J ,.c'- >. i~ •MK, t_r* .JM. i'^c- " > y r 1, K^I i-hi ,i -•» nftl. I.JI--' * 1-hr _f) .fit -,'••riL.-'il jifvn^'1 i i 4 i - n_ j»^".'|/T!it= t: ' . if . i«" nin»'-»! *,> •( -fir ir *-^-ri'i 'i p"; -v *i ,i ' /> "K"lc ~rn-?ry coarcbil*' --nar^^ r< yo„- J-OL, I'I.M " h in-d 93 trie *A>.U*_ -r-.-.-u.uu •-f ™r * I f ",i; ' Please Note: , 1 4 -17-' p" • IrercriTc, i'.t t'"i-( i. ~^LU -L v i"?! r fn.i ^.LLL » u-.ti 11.1 _ 2 U" u' a n i >• n*rcc-cr i" d if -r .1 ,= '.-in fur 11 1 *-ir s^rt r\ >~ n rht .1.- V* jr |-vsa Lrrl* fn',J"-'"i "of -'* -"- Hi . I (_ ) JT\)S. '-*.< f. ,JjtP"! leil- '.Uiy U jj pi-Ji't: r«._ j „,-].] air.-ut c ' vr -if U~ /Alps'- Ir vn_ ia c a<-« .u^i. ., 1 _-dL-, LL '.. L . f* .*^J fV*- "i'i" ir?ft or h'h Sin^n H f/i?- "c V-i. *i" I**" 16551T A * I f i'i K. i.nai! Ffl« 92 r> c> r-- r- r*- r*~ o> o « tt r O *T^~ •«"* u} o o o CM CM O 00 o o o o a 'Zi d CM O O O o o o CO *-- O O t~ o o o o o o o LO «+» CM CM 41 41 4t 41 41 41 41 +1 41 41 •« +1 -M •H 41 M 41 41 S5 «M o <» T» CO O LO CO C7> o u ta.o -=f CO 00 a? 00 CO CO ^_ ,_ O CO *"> CO LO Tf CO r-~ 00 T- f— •r- ^™- ^— a. V— 00 oo r»- O CD o E II o in •^r o id o o o -t— ne— LO XT 1— LO -st **— ^-~ *— CO ?5 T~ = EQ CD lO LO o lO CM CM LO o O .£ en h- l-~ CO w 1**- O CO IN o r~- CO T~^ io o a o a *L~ CM CM o CO TO CM CM •^T o <•') LO o o f") o CO o O O o •>- C- > o o o o T~ o o T- O o o CO > CM "* CM o o o o Q o o 41 41 41 -H 41 4! 41 4) 41 41 C t- O o 41 41 41 41 41 41 41 41 a. «. 4, ^_ <=t CO CM' LO CO OO t-- LO h- F" CD 0) q TO CM CM LO o> LO * S O CO CM ^, LO CM m o r~ **» Q. O O CO •^r •>* "-T CCi O o O ^J -*r d LO' a> o r*- K ^ *-~ -** •u £ co •s— ID O T- -' £ o -H •» 41 41 41 41 s 41 +1 * 41 O ft— 5 to o CI m cs o LO h- o r-~ to O CM i*-~ CO o 0 4> i» UL O CO r~ T— a> LO o> h- co lo **"- ***** CN Q 5 r T|" o r- o T— ^— ***** ~*$o co t-- T3 O CO «) ^~~u> x— : ' m ^ (0 < ... J2 w a £K! t- 5 S t;0 k. o «) m CO o 1 o u. 0 U. 0} 51. i > o 5 >. O m CO to >. 11L vP ^o CO Q^ .22 5 *D E I of ^„^, aa O O) £**** * > < s^ 2 = 3 5 Q £ o c m -1— 0 o S « o o u. 0 E 0> 1- u. X > o S