@prefix vivo: . @prefix edm: . @prefix ns0: . @prefix dcterms: . @prefix dc: . @prefix skos: . vivo:departmentOrSchool "Land and Food Systems, Faculty of"@en ; edm:dataProvider "DSpace"@en ; ns0:degreeCampus "UBCV"@en ; dcterms:creator "Azana, Cynthia Dy Prieto"@en ; dcterms:issued "2009-08-17T19:55:39Z"@en, "2002"@en ; vivo:relatedDegree "Master of Science - MSc"@en ; ns0:degreeGrantor "University of British Columbia"@en ; dcterms:description """Steller sea lions (Eumetopias jubatus) in the Gulf of Alaska declined since the late 1970s. Their population decline might be related to a shift in their diet from fatty, high-calorie fishes such as herring (Clupea pallasi) to low energy density fish such as walleye pollock (Theragara chalcogrammd). I compared the nutritional value of herring with pollock and explored seasonal changes in the nutrient content of pollock. I also compared the nutritional status of three captive Steller sea lions fed pollock and herring. Herring was a more concentrated in dietary lipid (p<0.001) and energy source (p<0.001) than pollock. The protein of herring was also higher in digestibility (p=0.015) than pollock protein, which could indicate that even if ingested energy was equal in both diets, absorbed energy for body functions may be reduced when pollock is eaten. There was little difference in the protein quality of pollock and herring with the exception that valine was more abundant in herring (p=0.004). The energy content of pollock changed seasonally, with the peak in energy concentration occurring in the summer and fall (July to November) and then declining over the winter prior to spawning. Captive Steller sea lions lost mass or increased mass at a slower rate on a pollock diet than when they consumed herring, at which time, they all increased in mass. The sea lions had lower levels of plasma cholesterol when fed pollock. Their red blood cells were also more susceptible to oxidation, which corresponded with lower plasma vitamin E levels. These findings suggest that consumption of predominantly pollock has nutritional consequences for the Steller sea lion. Even if they are able to increase their caloric intake to maintain their body mass, Steller sea lions may still be more susceptible to disease originating from oxidative stress."""@en ; edm:aggregatedCHO "https://circle.library.ubc.ca/rest/handle/2429/12322?expand=metadata"@en ; dcterms:extent "4505468 bytes"@en ; dc:format "application/pdf"@en ; skos:note "Seasonal variation in nutrient composition of Alaskan walleye pollock (Theragra chalcogramma) and its effect on the nutritional status of Steller sea lions (Eumetopias jubatus) B y Cynthia D y Prieto Azana B.Sc. University of British Columbia, 1997 A THESIS S U B M I T T E D I N P A R T I A L F U L F I L L M E N T O F T H E R E Q U I R E M E N T S F O R T H E D E G R E E O F M A S T E R O F S C I E N C E In T H E F A C U L T Y OF G R A D U A T E S T U D I E S Department of Food Science We accept this thesis as conforming to the required standard T H E U N I V E R S I T Y OF B R I T I S H C O L U M B I A October 2002 © Cynthia D . Azana, 2002 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. 11 Abstract Steller sea lions (Eumetopias jubatus) in the G u l f of Alaska declined since the late 1970s. Their population decline might be related to a shift in their diet from fatty, high-calorie fishes such as herring (Clupea pallasi) to low energy density fish such as walleye pollock (Theragara chalcogrammd). I compared the nutritional value of herring with pollock and explored seasonal changes in the nutrient content o f pollock. I also compared the nutritional status of three captive Steller sea lions fed pollock and herring. Herring was a more concentrated in dietary lipid (p<0.001) and energy source (p<0.001) than pollock. The protein o f herring was also higher in digestibility (p=0.015) than pollock protein, which could indicate that even i f ingested energy was equal in both diets, absorbed energy for body functions may be reduced when pollock is eaten. There was little difference in the protein quality of pollock and herring with the exception that valine was more abundant in herring (p=0.004). The energy content o f pollock changed seasonally, with the peak in energy concentration occurring in the summer and fall (July to November) and then declining over the winter prior to spawning. Captive Steller sea lions lost mass or increased mass at a slower rate on a pollock diet than when they consumed herring, at which time, they all increased in mass. The sea lions had lower levels o f plasma cholesterol when fed pollock. Their red blood cells were also more susceptible to oxidation, which corresponded with lower plasma vitamin E levels. These findings suggest that consumption of predominantly pollock has nutritional consequences for the Steller sea lion. Even i f they are able to increase their caloric intake to maintain their body mass, Steller sea lions may still be more susceptible to disease originating from oxidative stress. I l l Table of Contents Abstract i i Table of Contents i i i List o f Tables v List o f Figures v i Acknowledgements v i i Chapter I Literature Review 1 1. Marine Mammals in Alaska 1 2. Commercial Fishing Methods 3 3. Nutrition - Review of macronutrients 4 4. Nutritional Stress in the Steller sea lion 8 5. Nutrition Evaluation o f Food Stuffs Proximate Analysis 11 6. Oxidative status and its health implications 14 7. Antioxidants 17 8. Factors in the variation o f fish composition within species 19 9. Study objectives and Hypotheses 21 Chapter II Proximate Analysis o f Walleye Pollock - Seasonal Variation and Comparison with Herring 23 Introduction 23 Materials and Methods 28 Sample preparation .. .26 Proximate analysis 29 , Moisture analysis 29 Crude lipid analysis 29 Nitrogen determination 30 Crude ash analysis 30 Fatty acid profiling 31 Energy density 32 In vitro digestibility assay 32 Amino acid analysis 33 Statistics 33 Results 34 Seasonal differences in pollock 34 Seasonal changes in fatty acid composition of pollock 38 Seasonal changes in amino acid composition and protein digestibility in pollock 42 Species differences between pollock and herring caught in the fall season 46 Differences in protein quality and digestibility between pollock and herring caught in the fall season 51 Discussion 54 Seasonal changes in proximate analyses and energy density of walleye pollock 54 Seasonal changes in fatty acids ofpollock 57 Seasonal changes in protein digestibility of walleye pollock 57 Implications for Steller sea lions 58 Limitations 59 Macronutrient content and energy density: walleye pollock versus herring 60 Protein digestibility : pollock versus herring 61 Implications for the differences between pollock and herring for Steller sea lions 62 Fatty acid profiles 64 Protein Quality 65 Summary 67 Chapter III Effect o f a Pollock diet on the Nutritional Status of the Steller sea lion 68 Introduction 68 Materials and Methods 70 Feeding trials on captive Steller sea lions 70 Sea lion plasma lipid analyses 70 In vitro forced peroxidation assay 71 Vitamin E Assay of Sea Lion Plasma 70 Results 73 Steller sea lion changes in feeding behaviour 73 Changes in mass of the Steller sea lion 75 Blood and Serum analyses 77 Forced red blood cell peroxidation assay 81 Discussion 85 Changes in body mass 85 Change in feeding behaviour 87 Plasma cholesterol 89 Oxidative stress and the role of Vitamin E 90 Future directions 93 Summary 94 References 95 Appendix I Monthly variation in proximate values of pollock 102 Appendix II Sex variation in proximate values of pollock based on months 103 Appendix III Summary o f morphological data collected from ground pollock and herring 104 List of Tables Table 2.1 Summary o f walleye pollock used for analysis 28 Table 2.2 Seasonal changes in macronutrient content of walleye pollock 35 Table 2.3 Sex differences in macronutrient content of walleye pollock 36 Table 2.4 Seasonal changes in saturated and monounsaturated fatty acid composition of walleye pollock 39 Table 2.5 Seasonal changes in polyunsaturated fatty acid composition o f walleye pollock 40 Table 2.6 Seasonal changes in non.essential amino acid composition of walleye pollock 43 Table 2.7 Seasonal changes in essential amino acid composition of walleye pollock 44 Table 2.8 Species differences in macronutrient content - Comparison of Walleye pollock and herring caught in the Fal l season 47 Table 3.1 Changes in eating behaviour in Steller sea lions : Pollock vs. Herring diets 74 Table 3.2 Changes in mass of Steller sea lions : pollock diet vs. herring diet 76 Table 3.3 Blood and Serum analyses of Steller sea lions when diet consists of Herring vs. Pollock 78 Table 3.4 Plasma lipid profiles o f Steller sea lions when diet consists of Herring vs. Pollock 79 vi List of Figures Figure 2.1 Seasonal changes in energy density of pollock 37 Figure 2.2 Seasonal changes in percentage of polyunsaturated fatty acid content in pollock 41 Figure 2.3 Seasonal changes in protein digestibility of pollock 45 Figure 2.4 Differences in proximate analyses parameters between herring and pollock on a wet basis 48 Figure 2.5 Differences in energy density between herring and pollock on a wet basis 49 Figure 2.6 Differences in fatty acid composition between pollock and herring 50 Figure 2.7 Differences in protein digestibility between pollock and herring 52 Figure 2.8 Differences in specific amino acid contents between pollock and herring...53 Figure 3.1 Changes in plasma vitamin E concentration in Steller sea lions when switched from herring to pollock diets 80 Figure 3.2 In vitro forced peroxidation assay on red blood cells of Male 1 82 Figure 3.3 In vitro forced peroxidation assay on red blood cells o f Female 1 83 Figure 3.4 In vitro forced peroxidation assay on red blood cells o f Female 2 84 Acknowledgements vii I would like to thank my supervisors, Dr. David Kitts and Dr. Andrew Trites, and committee member, Dr. Timothy Durance for their guidance, patience and support throughout this process. Thank you to Pamela Rosenbaum for keeping it all together. I would also like to thank David Rosen and the staff at the Vancouver Aquarium for their work with the Steller sea lions. I would like to thank Andrea Hunter and my brother Robert Azana Jr. for doing the messy job of fish grinder. Thank you to Sherman Yee for showing me the ropes with all the equipment. A big thank you goes to Ruth Joy for teaching me about statistics, since I did not recall anything from my second year stats course. I would also like to send many thanks to Siva K . and Gilles Galzi o f Animal Science, Charles Hu , Katie Du , and Rebecca Shields for their parts in the data collection for this project. I also would like to thank At-Sea Processors Association for providing the fish for this study. Most of all, thank you to my husband, whose support has been essential in getting through this program and through life. 1 Chapter I: Literature Review 1. Marine Mammals in the Gulf of Alaska The G u l f o f Alaska is home to several different marine mammals, which either live there for their entire lives or migrate through the area. For example, many whales and dolphins move into the G u l f o f Alaska during the summer to feed and then migrate south during the winter months (Calkins 1986). The Steller sea l ion (Eumetopias jubatus), the harbour seal (Phoca vitulina richardsi), the sea otter (Enhydra lutris), and some species of Cetacea make their home in the G u l f o f Alaska for their entire lifetime (Calkins 1986). Other pinnipeds, such as the northern fur seal (Callorhinus ursinus) and the northern elephant seal (Mirounga angvstirostris) make seasonal appearances. Significant population declines have occurred among harbour seals, northern fur seals, and Steller sea lions (Calkins 1986; Trites 1992; Trites et al. 1996; Hobson et al. 1997). The focus of my study is the Steller sea lion, which is the largest of the eared seals and range in the northern regions o f the Pacific Ocean (Calkins 1986). The species is named after a German naturalist, who described the sea lions in 1751. Males and females o f the species have very obvious physical differences. The adult male can reach, on average, a length o f 3 m and a body weight of 681 kg (Winship et al. 2001). Females can reach a length of 2.3 m and 283 kg in body weight (Winship et al. 2001). Breeding season begins in mid-May, when adult sea lions begin to gather on rookeries. Males defend their claimed territory from other males but allow females to move freely in the areas. Pups are born between mid-May and mid-July, but the adult females breed again only ten days after giving birth (Pitcher et al. 2 1998). Although conception occurs at this time, the embryo does not implant into the uterine wall until the autumn months. Males are generally capable o f breeding between the ages o f three to seven years old, however they are physically unable to defend a territory until they are more mature, generally at about ten years o f age. Females can start to breed at ages three to six and continue to bear a pup each year afterwards. Steller sea lions feed upon a variety o f different prey species, the bulk of which is pollock, followed by squid, pacific cod, herring and capelin (Merrick et al. 1997). The food requirement o f the Steller sea lion was predicted at 5% o f body mass for 14 year old males and 6% for 22 year old females, with higher food requirements for younger individuals (Winship et al. 2002). Other pinniped species in the region of the Steller sea lions seem to feed on many o f the same prey species and thus, may compete for food. Northern fur seals, Steller sea lions, and harbour seals each have a preference for certain species, but are also opportunistic predators that wi l l take advantage of prey that is abundant (Hobson et al. 1997). Pollock in particular, seems to be a large part of all the diets of these species, but there are preferences with regards to size o f the pollock. Steller sea lions appear to prefer juvenile-sized pollock, but are capable of consuming larger fish (Hobson et al. 1997). Northern fur seals and harbour seals consume pollock with a mean fork length that is smaller than what is consumed by the sea lions (Hobson et al. 1997). This common prey species may be the shared link that has led to a decline in the populations of all three pinniped species. 3 2. Commercial Fishing Methods It is important to know about commercial fishing methods when doing a study that involves analysis of fish caught by commercial fisheries. While research fisheries may use the same equipment, the focus of a research fishery is to obtain a sample o f fish that would be most representative o f the population to be studied. In contrast, commercial fisheries are involved in catching fish in a way that is economical and marketable rather than representative. When using commercially caught fish, one should understand this difference. Fisheries generally classify species o f commercial catch into three categories: pelagic species, demersal species, and shellfish. Pelagic fish mainly reside near the ocean surface but may be found anywhere between the ocean bottom and surface (Sainsbury 1971). These would include herring, mackerel, and salmon, among others. They are normally caught with gear that has no contact with the ocean floor. Some species such as herring may gather in shallow waters during spawning and may be caught with gear that has contact with the ocean floor (Sainsbury 1971). Demersal fish live in deeper ocean waters and are closer to the ocean floor. These include pollock, cod, and whiting. Flounder live on or close to the bottom o f the ocean while shellfish such as crab, mussels, lobsters, and shrimp, live on the bottom of the ocean. The type o f fishing gear used to make the catch is based on several factors. One is the depth of the waters where the fish w i l l be found. If the net is to touch the ocean floor, the quality o f the sea bed becomes a factor. Different gear is used for rough and even sea beds than is used for soft and sandy sea beds. Lastly, different methods of fishing are used for fish that are to be used in bulk for fish meal, animal feeds, etc than for fish that have high individual value. Fish that have high individual value must be caught using methods that 4 ensure the condition of the fish is o f high quality compared to those caught in bulk that may be physically damaged in the nets. 3. Nutrition M y study involved the nutrition of Steller sea lions, so the function o f various nutrients must first be explored. The categories o f macronutrients are as follows: carbohydrate, protein, and lipid. For the purposes o f my study, I mainly focused on protein and lipid since the relative amount of carbohydrate in the Steller sea lion diet is minimal in comparison. 3.1 Protein and Amino Acids Proteins have multiple functions in the body. First, proteins are necessary for the growth and maintenance of tissues. This includes growth o f embryos and developing young animals, as well as replacement of worn out cells such as in the blood, gastrointestinal tract, and skin (Williams 1995). Proteins are required to form the structures o f enzymes and some hormones in the body. Enzymes are proteins that act as catalysts for various biochemical reactions in the body. Amino-acid-based hormones, such as thyroid, insulin, and glucagons, are chemical messengers in the body that are released in response to changes in the internal environment o f the body. Their release enables the body to respond to the changes and restore the normal conditions (Sizer et al. 1994). Another role o f protein is in building immunity against foreign proteins that may enter the body through the production o f antibodies. Proteins are also used in fluid and electrolyte balance as well as p H regulation. Proteins can act in this way by either passively attracting water from in or out of the cell or 5 by the use of transport proteins embedded in the cell membranes, which actively transport ions in and out of the cell that w i l l draw water with them (Zeman 1991; Sizer et al. 1994). The p H balance o f the body is maintained by proteins because they can act as buffers to acid or alkaline conditions (Zeman 1991). Lastly, protein can be utilized by the organism for energy in the event that other sources are unavailable (Sizer et al. 1994). Proteins have a complex, three-dimensional structure. The primary structure of the protein consists o f the sequence o f amino acids contained within it. This sequence is closely linked to the function o f the protein (Lehninger et al. 1993). The secondary structure consists o f regularly occurring formations in the chain o f amino acids, called a polypeptide, such as the alpha-helix and the beta-conformation. These formations are determined by the interactions o f the amino acids in the primary structure. The tertiary structure of the protein is the three-dimensional structure of a polypeptide, which contains within it the secondary structures of the protein. Finally, the three-dimensional polypeptides, which are subunits of a protein, come together to form the quaternary structure o f the protein. In my study I w i l l look at both crude protein content o f the fish as well as the individual amino acid contents. Amino acids are important because animals require them in their diets for maintenance, growth, reproduction, and lactation. The metabolism o f amino acids by the animal in their gastrointestinal tract can change the composition o f the dietary amino acids. Enrichment, impoverishment, or changes o f the proportional quantities o f the amino acids absorbed into the circulation can occur (Williams 1995). These occurrences can be influenced by the microbial flora in the gut and also by the amount of protein ingested. Amino acids can also be classified as essential and non-essential. The conventional definition o f an essential amino acid is one that \"cannot be synthesized by the animal 6 organism out of materials ordinarily available to the cells at the speed commensurate with the demands for normal growth\" (Reeds 2000). These amino acids include histidine, isoleucine, leucine, lysine, methionine, phenylalanine, threonine, tryptophan, and valine (Sizer et al. 1994). Non-essential amino acids are ones that can be synthesized by the body at a rate that w i l l meet demands and these include alanine, arginine, aspartate, cystine, glutamate, glycine, proline, serine, and tyrosine (Sizer et al. 1994). The dietary requirements o f essential amino acids are difficult to assess due to adaptations that can be made by the animal in times o f need such as decreasing urinary nitrogen and increasing digestive efficiency o f proteins (Mil lward 1998). The requirements are generally a function o f the animal's metabolic demands and the digestive efficiency o f the animal. Little is known about the specific dietary requirements of the Steller sea lion, let alone the amino acid requirements, but there are a few general facts about protein requirements in mammals. One study found a common amino acid pattern in mammalian milk, which could suggest common amino acid requirements o f mammalian babies (Davis et al. 1995). Pinniped milk was found to contain large amounts of glutamate, leucine, and proline - approximately summing to 40% of total milk amino acids (Davis et al. 1995). The stage of lactation did not affect the amino acid composition in that study. Another study found that dietary protein deficiency caused delayed maturation and a slowed rate of growth in lab rodents (McAdam et al. 1999). The nestling deer mice of mothers who were fed a low quality plant diet prior to weaning grew at only half the rate of the control group whose mothers were fed high protein cat food. The absolute levels o f protein did not affect the nestlings growth as the protein quality did, and they found arginine and valine may have particular importance (McAdam et al. 1999). Protein deficiency during pregnancy was also shown to have detrimental results on the young. Protein malnutrition can 7 cause decreased placental and fetal growth and cause postnatal damage such as permanently retarded growth and permanent alterations in the structure and function o f some organs (Wu et al. 1998). For these reasons, I looked at the protein and amino acid content o f the Steller sea lion diet as the protein content and quality may have an effect on their ability to reproduce and also maintain their health as adults. 3.2 Lipids Fats are nutritionally important because they are the primary form of stored energy in the body. Because they are able to provide more than twice the energy of proteins and carbohydrates, fats provide a compact source of energy, which is valuable in periods o f low food availability. Fats are also important as shock absorbers and insulation around vital organs, as structural components of cell membranes, and as a medium for obtaining fat-soluble vitamins A , D , E , and K and essential fatty acids from the diet (Sizer et al. 1994). There are three main categories o f lipids in the body. Triglycerides, also known as triacylglycerols ( T A G ) , represent stored energy and consist of a glycerol molecule with fatty acids esterified to it. Sterols, such as cholesterol are present in most eukaryotic cells and can have various functions in the body such as forming some hormones. Lastly there are structural lipids that form things such as cell membranes. These include phospholipids, which form the phospholipid bilayers of cells and also sphingolipids, which form some structural lipids in the nervous system and are present in small amounts in other membranes (Lehninger et al. 1993). Aside from their structural function and their function as energy storage, some lipids are biologically active. A s mentioned previously, sterols are components of some hormones 8 such as sex hormones, Cortisol, and aldosterone. Eicosanoids are compounds that act in the tissue where they are produced, unlike hormones, and are involved in reproduction, inflammation, fever, pain from injury and disease, blood clotting, blood pressure, gastric acid secretion as well as other functions (Lehninger et al. 1993). Eicosanoids are derived from long-chain polyunsaturated fatty acids, such as arachidonic acid and eicosapentanoic acid. Fat-soluble vitamins also have important functions. Vitamin A is important for night vision and also acts as an antioxidant. Vitamin D has a role in calcium metabolism. Vitamin E is another antioxidant and vitamin K is involved in blood clotting. Essential fatty acids are ones that the body cannot synthesize and are needed for basic functions. These include alpha-linolenic acid, which can be found in foods such as fatty fish, and linoleic acid, which can be found in seeds, nuts, and whole grains (Sizer et al. 1994). Deficiency in these fatty acids is rare but can result in growth retardation, reproductive failure, skin abnormalities, and problems with kidney and liver function (Sizer et al. 1994). Linolenic acid in particular is the precursor for very long chain omega 3 fatty acids, which are important in brain and visual development, reproduction, skin integrity and inflammatory response (Sizer et al. 1994). I looked at the fatty acid composition of the Steller sea lion diet to assess how well their diet meets the needs for the important polyunsaturated fatty acids mentioned here. 4. Nutritional stress in the Steller sea lion Steller sea lions have been experiencing a population decline in Alaska between Prince Wil l iam Sound and the far Aleutian Islands since the late 1970s (Trites et al. 1996). One theory is that changes in prey base in these regions have caused nutritional stress on these animals (Trites et al. 2002). The diet of the Steller sea lion in regions of population 9 decline is composed o f primarily walleye pollock (Theragra chalcogramma), whereas prior to the decline it consisted o f mainly small, fatty, schooling fish such as herring (Clupea pallasi) (Merrick et al. 1997). Nutritional stress for the purposes o f this study was defined as a decline in health as indicated by decreases in weight or changes in any o f the hematological measures taken. Nutritional stress can be caused by the absence o f one or more nutrients that are required for optimal health. In the Steller sea lions, small schooling fish may have a greater ability to meet their nutritional needs than pollock. O f special interest in my study is the energy density o f pollock compared to other major components o f the sea lion diet, such as herring. L o w energy density and poor protein quality can have detrimental effects on the health of sea lions. The poor nutritional quality o f their diets may also play a role in the body size reduction o f the animals since the 1970s (Calkins et al. 1998). In order to test this hypothesis, I measured and compared the nutrient content o f pollock to the nutrient content o f herring caught in the same season and year. In addition, captive Steller sea lions were fed diets of exclusively herring or pollock so that their nutritional status on both diets could be assessed. 4.1 Evidence for nutritional stress in the Steller sea lion Research on the decline o f the Steller sea lion has turned its focus to the possibility o f nutritional stress in the animals due to a number of physiological changes that have been observed in the population during the years of the decline (Trites et al. 2002). Sea lions in the years o f the decline were smaller in mass, length, and girth than sea lions prior to the decline (Calkins et al. 1998). In addition to the size reduction, the body fat was reduced (Castellini et al. 1993) and there was a greater reduction in girth than in length, which is 10 indicative o f nutritional stress in animals (Calkins et al. 1998). Animals may change their reproductive performance in relation to their nutritional status as well. In the Steller sea lion, the rate of pregnancy decreased in lactating females (Pitcher et al. 1998), which implies that lactating females did not obtain enough energy to sustain a pregnancy and nurse a pup at the same time. 4.2 Seasonal variation in energy requirements of Steller sea lions The energy requirements of the Steller sea lion has been shown to have seasonal variation coinciding with the breeding season, which is from May to July (Winship et al. 2002). During this time, food consumption of male sea lions decreases, as they hold terrestrial territory (Kastelein et al. 1990). The mass o f males fluctuate seasonally when they reach sexual maturity and fat stores built outside o f the breeding season become especially important. While the female sea lion does not have the same degree o f variation in its dietary intake or body mass, the female does have changing nutrient needs due to pregnancy. Typically from January to July, the female carries a growing fetus and is in the later stages of pregnancy. From August to December, most females would be carrying a fetus and suckling a pup as well (Kastelein et al. 1990). Female sea lions with pups have 70% greater food requirements than females o f the same age without pups (Winship et al. 2002). Aside from seasonal changes related to the mating season, variation may originate from changes in environmental temperature as well. When the environmental temperatures were higher, captive sea lions consumed less because the animals required less fat stores for insulation (Kastelein et al. 1990). In low temperatures, food consumption increased in order to build up fat stores. It has also been shown that the more time the sea lion spends in the 11 water, the more insulation it w i l l need because o f the increased body heat losses (Kastelein et al. 1990). Sea lions could very well be affected by seasonal fluctuations in the nutrient content o f pollock. For example, an increase in pollock nutrient density during the winter months, when the sea lions are building up the fat stores and have increased energetic needs, could be beneficial to the sea lions because this is the period when their energetic needs are highest. I f the opposite were true, the result could be detrimental to the sea lions because the diet would not be adequate to meet the increased energy requirements. For this reason, I explored the seasonal variation in the nutrient composition o f pollock. 5. Nutrition Evaluation of Food Stuffs : Proximate Analysis Proximate analysis consists of determining the amount of moisture, ash, protein, lipid, and carbohydrate in a food item or feed stuff. There are many official methods to accomplish the evaluation of moisture, protein, lipid, and ash, but I w i l l describe only the ones used in my study. Carbohydrate was not analyzed as its content in fish is minimal in comparison with the other components that were measured. 5.1 Moisture Moisture analysis of foods is important because it is the main constituent of food and allows food processors to have information that could affect the storage and processing o f the food item (Bradley 1998). Knowing the moisture content means that other analytical results, such as lipid content, can be expressed as a value based on the dry food. Doing so allows results to be more consistent as many factors can influence the moisture content o f samples. 12 Moisture is also important for pinniped species that live in marine environments because they obtain most of their fresh water from their food (Geraci 1975). I used the vacuum oven drying method o f moisture analysis by placing samples in an oven under reduced pressure. This allowed for more complete removal of water without the decomposition associated with the higher temperatures needed when using a conventional oven (Bradley 1998). This was particularly important with my fish samples because o f their high l ipid content, which make the fish susceptible to lipid oxidation. Use o f a conventional or forced air oven would have altered the chemical composition o f the fish and my moisture analyses due to the degradation associated with higher temperatures and exposure of samples to oxygen. 5.2 Protein I used the Dumas method of protein analysis for the fish samples, which measures total nitrogen in a sample rather than protein. The protein content must be calculated from the nitrogen content using predetermined estimates o f the nitrogen content o f protein. The factor to determine protein content from nitrogen content varies with different foods. For fish, I used the factor o f 6.25, which is the estimate for eggs and meats (Chang 1998) and originates from the fact that most of these proteins contain 16% nitrogen (100/16=6.25) (Chang 1998). Protein content determined by the Dumas method was achieved by combusting the fish samples. A n y nitrogen gas that was subsequently released was then quantified by gas chromatography (Chang 1998). The disadvantage to this method is that protein content can be overestimated due to the presence o f nitrogen that is not associated with protein in the food. 13 5.3 Lipid Crude lipid is often measured using a solvent extraction, such as the Folch's method that I used in my study. Neutral lipids such as triglyceride, wax, and pigment can be extracted from tissues using ethyl ether, chloroform or benzene as solvents (Lehninger et al. 1993). Membrane lipids such as phospholipid are more effectively removed from the sample using polar organic solvents such as ethanol and methanol (Lehninger et al. 1993). Many methods, such as Folch's, include using both types o f solvents to remove all lipids from the sample. A common mixture is chloroform and methanol, where the lipids remain in the chloroform layer and the polar molecules, such as protein and carbohydrate, remain in the methanol layer (Lehninger et al. 1993). Since the Folch's method of solvent extraction o f lipid is able to remove both polar and neutral lipids, it was used in my study to gain an accurate determination of the total crude lipid o f our fish samples. 5.4 A s h Ash is defined as the total mineral content of a food. This includes trace minerals such as zinc and minerals present in larger amounts such as calcium and iron. I chose the dry ashing method, which involves combusting the samples at high temperatures so that any organic material is burned off, leaving only inorganic minerals behind. This method is advantageous because many samples can be analyzed at once and it is safe in that no reagents are needed and no blanks are required for the analysis (Harbers 1998). There may be loss o f some volatile elements, but it is less o f a concern because I focused on crude ash rather than trace mineral analysis. 14 6. Oxidative stress and its health implications 6.1 Oxidative Stress Oxidative stress is basically a change in the pro-oxidant to antioxidant equilibrium in favor o f the pro-oxidants that leads to damage in a biological system (Kehrer et al. 1994). It could also be that the rate of free radical formation is greater than the ability o f the cell to transform them into less toxic species (Salem et al. 1997). Oxidative stress can lead to cellular dysfunction or death and can also produce chemical changes in lipids, protein, and D N A , which can lead to changes in their functionality (Salem et al. 1997). D N A damage can result in mutations, cancer formation, aging, and cellular death and the hydroxyl radical is often the species associated with this type o f damage (Acworth et al. 1997). When oxidation affects amino acids in a protein, the protein can lose its functionality especially when the damage happens in a critical part of the amino acid sequence (e.g. the active site o f an enzyme) (Acworth et al. 1997). This damage can be a result o f oxidation that occurs with aging and can result in a loss o f biochemical and physiological function. Oxidized proteins are also more susceptible to proteolysis (Kehrer et al. 1994). 6.2 Free Radicals Free radicals are species o f atoms or molecules that contain one or more unpaired electrons (Ternay et al. 1997). In contrast, a stable molecule has an even number o f electrons 15 in complete orbitals (Pryor 1994). Free radicals can have either a net charge of zero or may carry a charge, usually have very short lifetimes, and decompose very quickly (Ternay et al. 1997). One way that decomposition can occur is through dimerization where two free radicals join their unpaired electrons to form a stable compound. Another way is through disproportionation, which involves the simultaneous oxidation o f one radical and the reduction of another to form two stable compounds (Ternay et al. 1997). 6.3 Reactive Oxygen Species There are a few different species o f oxygen radicals that may be present in biological systems. The superoxide anion, O2\", is produced by the reduction o f O2. It has been implicated in disease states and its toxicity seems to be related to the Fenton reaction shown below (Ternay et al. 1997). The hydroperoxyl radical, OFT, is more reactive than the superoxide anion and has a larger role in biological damage (Ternay et al. 1997). It is produced during the Fenton reaction (Ternay et al. 1997). F e 2 + + H2O2 -> F e 3 + + O H - + OH* Hydrogen peroxide is not a radical, but acts as an oxidant. Though it may react slowly on its own, it can be toxic and result in the formation o f hydroxyl radicals when coupled with some metals (Ternay et al. 1997). It also has the ability to diffuse into membranes and lipid deposits (Ternay et al. 1997). 6.4 Lipid peroxidation Polyunsaturated fatty aeids(PUFA) are very susceptible to oxidative damage, particularly those present in structural lipids (Acworth et al. 1997). Polyunsaturated fatty 16 acids are susceptible to oxidation more so than other fatty acids because of the double bonds in their structure. L ip id peroxidation follows three stages: initiation, propagation, and termination. Oxidation o f P U F A , usually by the hydroxyl radical, can result in the formation of a peroxyl radical, which then oxidizes more fatty acids, creating a chain reaction. This is only terminated when either the lipid to protein ratio decreases so the radicals begin to attack protein, or when the lipid radicals encounter an antioxidant (Acworth et al. 1997). Initiation o f lipid peroxidation starts when a carbon attached to a double bond becomes the subject o f attack by a free radical (Banks 1997). In biological systems, free radicals can originate from various enzymatic and non-enzymatic reactions (Banks 1997). L ip id peroxidation can also be initiated by iron and other transition metals through a reaction similar to the Fenton reaction in which the metal itself (e.g. F e 2 + ) reacts with the l ipid molecule or the metal can form a complex with hydrogen peroxide that subsequently results in the formation o f the hydroxyl radical (Banks 1997). The attack o f P U F A by oxidants or free radicals results in the formation o f a l ipid radical. After the initiation phase, propagation o f the l ipid peroxidation process begins and this phase is followed by termination. The initial l ipid radical formed by the initiation phase combines with molecular oxygen to form the l ipid peroxyl radical (Banks 1997). The peroxyl radical then attacks another polyunsaturated fatty acid, forming lipid hydroperoxides and more l ipid radicals (Banks 1997). Iron can also promote the rate o f peroxidation by converting lipid hydroperoxides to reactive alkoxyl or peroxyl radicals, which in turn attack another polyunsaturated fatty acid (Banks 1997). These reactions continue in a cycle until the termination phase o f the process, in which two radical species combine to form 17 conjugated lipid dienes, aldehydes, polymers, and hydrocarbons (Banks 1997). The products of lipid peroxidation can be measured to determine the extent of oxidation in a sample. 7. Antioxidants Antioxidants are essential in controlling the oxidative stress that can cause cellular damage in organisms. The following are some o f the antioxidants that aid in fulfilling this function in the body. 7.1 Ascorbic acid Ascorbic acid is a water-soluble compound that has strong reducing power. This reducing power enables it to be very effective as a free radical scavenger and ascorbic acid works to protect lipids and membranes by scavenging free radicals that may initiate l ipid peroxidation (Briviba et al. 1994). It also regenerates lipid-soluble antioxidants such as vitamin E , which get transformed during lipid peroxidation (Briviba et al. 1994). A deficiency in ascorbic acid in the human diet can lead to scurvy, but vitamin C is readily found in many fruits and vegetables so deficiency is rare (Briviba et al. 1994). 7.2 Glutathione Glutathione is a reducing agent that is made o f glutamate, cysteine, and glycine, and is not required in the diet (Briviba et al. 1994). There are two ways in which glutathione works to fight oxidation. First, glutathione can be a substrate for antioxidant enzymes that can reduce hydroperoxides, preventing the accumulation of lipid hydroperoxides (Briviba et al. 1994). Second, it can also react directly with free radicals and in so doing, protect cells 18 from reactive oxygen species. This causes the formation of G S S G that can be reduced back to G S H with an N A D P H dependent enzyme. The G S H to G S S G ratio should be kept high in tissues for it to be effective (Briviba et al. 1994). 7.3 Vitamin E Vitamin E consists o f eight different compounds, which exhibit similar biological activity with the most active o f these being alpha-tocopherol (Kij ima 1993). The other compounds are alpha, beta, gamma, and delta tocotrienols and beta, gamma, and delta tocopherols (Landvik 1997). It is widely distributed in animals and plants and is especially rich in plant oils (Kij ima 1993). Vitamin E in its purified form is either colorless or pale yellow. It is insoluble in water, but soluble in lipids and organic solvents (Kij ima 1993). Vitamin E must be absorbed with dietary fat through the gastrointestinal tract and lymphatic system, and absorption is dependent on the individual animal's ability to do so (Landvik 1997). It is transported in the plasma mainly through low-density lipoproteins and primarily in the form o f alpha-tocopherol (Landvik 1997). The exact vitamin E requirements of Steller sea lions are unknown, although it has been shown that animals with high levels o f polyunsaturated fats in their diet have great variability in their requirements (Landvik 1997). Functionally, vitamin E is essential for cellular growth and maintenance o f membrane permeability, and is an effective free radical quencher (Lee et al. 2000). Vitamin E has the ability to reduce peroxyl radicals, hydroxyl radicals, superoxide radicals and singlet oxygen in biological membranes (Banks 1997). Vitamin E protects the polyunsaturated fatty acids present in membranes and has also been shown to have a role in mitochondrial function, nucleic acid metabolism, protein metabolism, hormone production, vitamin A protection, and 19 selenium sparing (Landvik 1997). Vitamin E also appears to be able to modulate heart attack risk by inhibiting smooth muscle cell proliferation (involved in blood vessel wall thickening) and by inhibiting platelet aggregation, adhesion, and platelet release reactions (Traber 2001). Another important function o f vitamin E is in the function o f the immune system, which deficiency has been shown to depress and supplementation has been shown to stimulate (Meydani 1995). It also has a roles in growth, reproduction, prevention o f some diseases, and integrity of tissues (McDowel l et al. 1996). 7.4 Others Other antioxidants include ubiquinone, carotenoids, retinoids, and flavonoids. Ubiquinone is found in soybean oi l , meat, fish, nuts, wheat germ, and some vegetables. In humans there are high levels o f ubiquinone in the vital organs. In the heart, kidney, and liver, for example, 70-100% of it is present in the reduced form, whereas in the brain and lungs 80% of it is in the oxidized state (Briviba et al. 1994). Retinoids are forms o f vitamin A that can be found mainly in vegetables and fruits Flavonoids are the red, blue, and yellow pigments o f plants. They are radical scavengers that work also to chelate iron and inhibit radical-producing enzymes (Briviba et al. 1994). 8. Factors in the variation of fish composition within species When analyzing the composition of a fish species, it must be understood that results from one study done in a certain region may not apply to fish from other regions and that there is high variability even within a given area. For this reason, in my study, I used a large sample size for the proximate analysis and energy determination to keep the variation in the 20 results relatively low. Fish caught in different months were also assessed separately because there could be variation between fish caught from one season to the next. There are several factors that can account for the variation in fish composition. Fish muscle can be classified into two types: white and dark. White muscle is fast muscle used for sudden bursts of speed such as in predation or escape (Love 1988). Dark muscle is slow muscle that is used for continuous swimming (Love 1988). White muscle is less metabolically active than dark muscle and is present in different amounts in different fish. For example, cod, haddock, and whiting are relatively inactive and have more white muscle than active species such as herring, tuna, and mackerel (Love 1988). Dark muscle is different in color than white muscle due to the presence of haem pigments that facilitate the transport o f oxygen to the tissues and dark muscle is also higher in the amounts o f mitochondria, vitamins, and trace elements than white muscle. Rancidity probably begins in the dark muscle o f fish due to the haem pigments and also the greater amount of lipid contained in dark than white muscle (Love 1988). Dark muscle is stronger tasting and probably more nutritious than white muscle due to these inherent differences. The quantities o f these muscle types can also vary within species depending on environmental factors. Water movement is one such factor. Fish that live in an area of high water flow would theoretically have a higher level o f swimming effort than fish in still waters and therefore have more dark muscle. This was the case for brown trout that were subjected to an environment that caused them to swim a distance of 1.5 body lengths per second and resulted in an increased proportion o f l ipid and glycogen content (Davison et al. 1977). Fish in moving waters may in fact be in better condition than fish from still waters. 21 The p H levels o f the water may also contribute to the composition and condition of the fish. Fish are known to die when the p H o f their environment falls outside of their range o f tolerance. Such changes can cause interference with their metabolic processes especially in the case o f freshwater fish and fish in the early stages o f development (Love 1988). The depth o f water that fish reside in can affect the composition o f the fish and in particular, their fatty acid composition. Medium chain saturated fatty acids and long chain polyunsaturated fatty acids were seen to decrease in the flesh of fish from deep water, but CI8:1 increased (Lewis 1967). Another study found that cod caught at 90 to 100m were soft and watery compared to cod caught at closer to the surface at 45m (Love 1988). These changes could improve the buoyancy o f fish at greater depths or it could be a reflection o f the differences in food supply from shallow to deep water. Temperature is another factor that affects the fatty acid composition o f the fish. A s the temperature o f the water drops, the proportion of unsaturated fatty acids in the fish increases and saturated fatty acids decrease. This is because the unsaturated fatty acids allow structural lipids to remain flexible in the low temperatures due to the low melting point (Love 1988). This effect is not only seen in fish, but also in zooplankton and algae. 9. Study Objectives and Hypotheses The overall objective for my study was to determine i f there were any differences in the nutritional quality o f pollock and herring species and i f these differences were sufficient to cause a nutritional stress on Steller sea lions. O f course, knowing the nutrient composition o f the fish alone cannot tell the whole story of the possible nutritional stress the sea lions may be under. Variations in nutrient digestion and absorption have direct effects on the 22 nutritional status of the sea lions, which could be different than would otherwise be predicted from the dietary analysis. For this reason, I studied the effects o f an exclusively pollock diet on captive Steller sea lions and compared them to the effects of an exclusively herring diet. I hypothesized that pollock is of less nutritional value than herring, particularly in energy density and that this difference wi l l reflect itself by causing a decline in the health o f the sea lions while they are on a pollock diet. 23 Chapter It: Proximate Analysis of Walleye Pollock - Seasonal Variation and Comparison with Herring Introduction Changes in the species composition and the abundance o f forage fish in Alaska happened rapidly in the late 1970s (Van Pelt et al. 1997). In particular, walleye pollock appear to have proliferated while herring stocks declined. There also appears to have been a corresponding shift in the diets o f marine mammals and birds towards consuming more walleye pollock and fewer fattier fishes such as herring (Merrick et al. 1997). Some have specified that the decline o f Steller sea lions is nutritionally based and tied to the apparent dietary shift (Alverson 1992; Rosen et al. 2000a; Trites et al. 2002). Walleye pollock (Theragra chalcogramma) are gadids and occur in the Pacific Ocean from Canada to the G u l f o f Alaska, Bering Sea, Aleutian Islands, the Sea o f Okhotsk, and Sea o f Japan. They live predominantly near the ocean bottom around the continental shelf, but are also found near the surface and in mesopelagic areas of deep waters ( O C S E A P 1986). Walleye pollock is an important commercial fish species and is rated one of the highest in total world catches (Janusz et al. 1997). In addition to humans, pollock have numerous predators in the G u l f o f Alaska including several species of fish, seabirds, and marine mammals (Brodeur et al. 1996). The three main predators of pollock in the G u l f o f Alaska are Pacific halibut (Hippoglossus stenolepis), arrowtooth flounder (Atheresthes stomias) and the Steller sea l ion (Eumetopias 24 jubatus) (Hollowed et al. 2000). Cannibalism o f juvenile pollock by adults is significant, but is less prevalent in the G u l f o f Alaska than in the Bering Sea (Hollowed et al. 2000). Pollock eggs and larvae can be found in regions o f the G u l f o f Alaska throughout the year. However, large groups of pollock spawn in certain areas of the Gulf, such as the Sheilikof Strait, in the spring between March and Apr i l ( O C S E A P 1986; Schabetsberger et al. 1999). A t this time, pollock may inadvertently ingest eggs during respiration. Spawning males have been found to have the highest numbers o f eggs in their stomachs, mainly due to the amount of time spent in areas o f high egg densities (Schabetsberger et al. 1999). Seasonal changes associated with spawning or feeding may result in seasonal changes to the nutritional value of the pollock. Since Steller sea lion nutritional requirements fluctuate during various seasons o f the year due to breeding behaviours or water temperatures (Winship et al. 2002), it is of interest to see i f there are any seasonal changes in pollock nutrient content. Pacific herring are a subspecies o f Atlantic herring and range from southern California to Korea ( O C S E A P 1986). Herring also feed on copepods and euphausiids as do juvenile pollock, and spawn in the G u l f o f Alaska only in the spring. They are used commercially for oil , fertilizer, fish meal, bait, and roe and their eggs are harvested. Herring may also be salted and pickled and are the prey o f many marine birds, marine mammals, and other fish ( O C S E A P 1986). The shift in the Steller sea lion diet from small schooling fishes such as herring to mainly walleye pollock prompted me to analyze the nutrient content and compare the differences between o f these two species. I also examined the digestibility o f the protein. In 25 this way I sought to determine i f there were any differences in the ability o f each species o f fish to meet the nutritional requirements o f their predators. M y objective was to identify seasonal and sex differences in the protein, lipid, ash, moisture and energy content of whole walleye pollock. In addition, I wanted to determine how walleye pollock and herring differed. Finally, I wanted to compared the fatty acid profiles, amino acid profiles and protein digestibility to obtain a more complete picture o f the nutritional value of pollock and herring. M y main hypothesis o f this experiment was that pollock would be o f less nutritional value than herring and, in particular, have a lower energy density than herring. Moreover, both species o f prey fish would satisfy the requirements for essential amino acids, as animal proteins are generally o f good quality. A further hypothesis was that both fish would have a high polyunsaturated fat content, and high content of omega-3 fats in particular would be similar in content, as fish oils are generally rich in these types o f fatty acids. I also expected that pollock would have increased l ipid content during the spawning season due to the presence of roe in the females and the ingestion o f roe by the males. The increase in l ipid would correspond with an increase in energy density of pollock from the spawning season. 26 Materials and Methods a. Sample preparation Whole frozen pollock caught in the Bering Sea were obtained monthly in 1998 and 1999 from commercial fisheries (At-Sea Processors Association). A single sample of whole frozen herring was also obtained in 1998 for the purpose of comparison. Pollock and herring were stored either at the Food Science Department at U B C or at the Vancouver Aquarium in a freezer of at least -18°C. Fish were frozen in boxes immediately after being caught and were thawed overnight for grinding. Only fish that were free o f major deformities such as missing fins, were selected for grinding and dissection. The morphology o f each fish was evaluated using body weight, girth, length, and weight of gonads. Subsequently the fish were ground in a Hobart Silentcutter, bagged, vacuum-sealed, and frozen at -18°C. Between individual fish, the grinder was washed using a high pressure water spray and thoroughly dried before the next sample. For proximate analysis, small amounts of ground fish were thawed overnight at 4°C and any sample that was not used was freeze dried for subsequent assays. For the purposes of the study, the pollock were divided into seasons by the month in which they were caught (Table 2.1). Attempts were made at obtaining samples from all months o f the year, but as these attempts were unsuccessful, I analyzed samples according to season. For this reason, samples sizes for the seasons were different as I evaluated an equal number of samples from each month that was available. The results from the various months were then pooled according to season. These included winter (January and February), spring (March), summer (July and August), and fall (September, October, and November). Some months were missing due to periods when no pollock fishing occurred. Herring were caught 27 in November 1998 and were compared only to pollock that were caught in the fall season. Pollock from both 1998 and 1999 were pooled as no significant differences were found between years. Table 2.1 Summary of walleye pollock used for analysis Month Year Season January 1998 Winter February 1999 Winter March 1999 Spring July 1999 Summer August 1999 Summer September 1998 Fal l October 1998 Fal l 1999 Fal l November 1999 Fal l All fish provided by At-Sea Processors Association and were caught in the Bering Sea 29 b. Moisture analysis Moisture determination was done by first pre-drying aluminum pans in a forced air oven at 130 °C for four hours and cooling pans in a dessicator containing silica gel. Analysis of the ground pollock continued by using an analytical balance to weigh 3 to 5 grams o f wet sample in dried aluminum pans. The pans with the sample were placed in a vacuum oven at 80°C, 25mmHg, and dried overnight. After removal from the vacuum oven, the samples were cooled in a dessicator. Once completely cooled, the samples were weighed once again on the analytical balance to evaluate moisture loss and total solids. Moisture content (as a percent) was determined by dividing the difference between the initial and final weight o f the sample by the initial weight o f the sample. c. Crude Lipid Analysis Total crude lipid was evaluated by weighing two grams o f ground fish in a dried, desiccated Erlenmeyer flask and using the Folch's Double Phase Method (Folch et al., 1957). 50mL o f Folch I solution (2: 1 chloroform: methanol) was added to the flask and the mixture was blended. Flasks were then sealed with Parafilm and left overnight. The next day the solution was filtered through fluted Whatman filter paper 4 into a lipid-free glass graduated cylinder. The flasks were rinsed with lOmL o f Folch I solution and the rinse solution was then added to the filter. The solid sample left in the filter paper was recovered, stored, and used for protein analysis. Ten milliliters o f 0.88% N a C l solution was added to the cylinder, which was sealed, tilted twice, and left overnight. During this time the solution separated into 2 layers. The next day the top layer was suctioned off and lOmL o f Folch II solution 30 (3:48:47 chloroform: methanol: water) was added to the bottom layer in the cylinder that was again sealed, tilted twice and left overnight. The final volume o f the bottom layer was recorded the next day. The solvent was evaporated from a known volume o f the bottom layer (CHCI3) and the lipid left after evaporation was weighed to determine total crude lipid content. d. Nitrogen Determination The total organic nitrogen present was determined using the Leco Method as described by A O A C ( A O A C , 992.15, 1995). Air-dried samples (0.25 g) were combusted in an automated Leco FP-328 Nitrogen Analyzer (Leco Corp. Joseph Michigan, U S A ) . The calibration standard was E D T A (9.58% nitrogen). Nitrogen values were multiplied by 6.25 to obtain crude protein values. e. Crude Ash Analysis Total ash determination started with preparation of the porcelain crucibles. Crucibles and lids were soaked in detergent overnight, then rinsed and soaked in 3 N hydrochloric acid overnight. The crucibles were then rinsed with distilled deionized water and placed overnight in the muffle furnace (lab-heat box type with solid state, vari-watt power level control; Blue M Electric Co. , Blue Island IL) at 550°C to remove any contaminants or leftover organic material. Crucibles were removed the next day and cooled in a dessicator before being weighed with the analytical balance. Three grams o f wet ground fish was placed in each crucible and heated at 550°C for 32 hours. The crucibles and ash were weighed after being cooled in a dessicator. The percent crude ash was measured in triplicate 31 and calculated according to weight of crucible and weight of sample before and after ashing, multiplied by 100 percent. f. Fatty acid profiling When measuring the fatty acid content of the fish lipid, lOmL o f the CHCI3 layer left from the total crude l ipid analysis was placed in a fat-free test tube in a 30°C to 40°C water bath to evaporate to dryness. The sample was then cooled to room temperature and 5 m L o f 0.5N CH3OH-KOH was added to the test tube, which was then shaken vigorously. The sample was left overnight at room temperature and 2.5mL o f petroleum ether was then added. After the sample separated into two layers, the top layer (consisting o f ether and non-saponifiables) was suctioned off and discarded. A drop of 0.4 M HC1 and then 5mL o f BF3 were added to the samples and caps were placed loosely on the test tubes to allow gas to escape. The test tube containing the sample was then placed in boiling water and then gradually cooled over 15 to 20 minutes. After the samples reached room temperature, 2 drops of dHaO were added, followed by 2.5mL of hexane. After the phases separated, the top hexane layer was transferred to an eppendorf tube. Methyl esters were analyzed using a Varian 3700 gas chromatograph (GC-17A; Shimadzu, Scientific Instruments Inc. Columbia, M . D . ) equipped with a flame ionization detector and an A O C 1400 auto injector (Shimadzu, Scientific Instruments, Columbus, M D ) . Samples were injected onto a silicone fused Omegawas T M 320, 30 m x 0.32 mm ID capillary column (Supelco Inc, Bellefonte, P A ) with a 0.25 mm film thickness. Helium was the carrier gas. The injector temperature was set at 200°C, with the detector temperature set at 220°C and the column temperature was set at 220°C. The column flow rate was set at 1.9 mL/min. Each sample was analyzed in triplicate 32 to. A mixture of short chain, medium chain and long chain saturated, monounsaturated and polyunsaturated fatty acids was used as a standard (Sigma Chemical Co. , St. Louis MO). g. Energy Density Gross energy determinations of pollock and herring were conducted using an adiabatic bomb calorimeter on vacuum dried samples (Department o f Animal Science, U B C ) according to the Parr method (Parr, 2001). Samples (l.Og) were weighed and individually placed in a sealed bomb calorimeter container, which was sealed with excess oxygen and ignited electrically inside the bomb container. Sample heats o f combustion were calculated from the rise in temperature o f the water jacket inside the bomb container. h. In Vitro Digestibility Assay In vitro digestibility o f pollock and herring protein was determined according to the method o f Yuan et al. (1991). Ground fish samples were suspended in distilled water and the p H adjusted to 1.9 using hydrochloric acid. Pepsin (porcine stomach mucosal :10,000-Sigma chemicals ) was added to the suspended fish solution, and samples were placed in a shaking 37°C water bath for 30 minutes. Samples were then adjusted to a p H o f 8.0 and incubated with pancreatin (porcine pancrease - Sigma, chemicals) and incubated again at 37°C in a shaking water bath. At defined 1 minute intervals, aliquots were removed over a 30 minute period and deproteinized with 20% T C A . Trinitrobenzenesulfonic acid ( T N B S ) was added to each sample to measure for protein digestion products Protein digestibility was determined from the initial slope (e.g. 0-10 minutes) using linear regression analysis o f 33 T N B S absorption-time data. Relative digestibility o f the different fish protein sources was made in comparison to a casein standard. i. Amino Acid Analysis Fish samples (2.0 g) were weighed and refluxed with 6.0 M H C L for 24 hours at 110+1°C under vacuum to obtain complete hydrolysis. Additional samples were hydrolyzed with performic acid to specifically recover cysteine. Moreover, tryptophan was analyzed from samples hydrolyzed with 4.2 M sodium hydroxide for 16 hours at 121 + 1°C under vacuum. Individual amino acids were quantified using an amino acid analyzer (Pharmacia BiaCore 20) equipped with a cation-exchange column and ninhydrin detection. j . Statistics Results were displayed as mean + S E M . Statistical analysis was done using A N O V A with Tukey post-hoc test for multiple comparisons for results that displayed homogeneity o f variances. Kruskal-Wallis and Mann-Whitney tests were done for results that required a non-parametric analysis. Significance was defined as p < 0.05. 34 Results Seasonal differences in pollock Specific differences in proximate analysis parameters were apparent in the walleye pollock between seasons as indicated in Table 2.2. O f particular interest were the seasonal changes in energy density (Fig 2.1), which was highest in the fall at 5.41 ± 0.029 kcal/g and decreased during the winter to 5.08 ± 0.025 kcal/g (p<0.001). Energy density increased in parallel with increases in the l ipid content. L i p i d content was 17.04 + 0.677% in spring and increased significantly to 21.96 + 0.479% in the summer months (p<0.001). A drop in moisture accompanied this rise from 76.67 + 0.201% in spring to 75.88 + 0.168% in summer (p=0.041). L i p i d content was significantly higher in the fall at 21.25 ± 0.358% than in winter value at 15.44 ± 0.341% (pO.OOl) . Protein content in winter was 66.28 + 0.513 % and decreased in the spring months to 62.68 ± 1.077% (p=0.001). It reached its lowest levels in the fall at 60.79 ± 0.442 %, which was significantly, lower than during winter (pO.OOl) . Ash content was lowest in the summer at 10.59 + 0.310 %, which was significantly lower than the value for spring (p=O.038). The highest amount o f ash was found in the winter, which showed a significant difference from fall (p=0.003). Differences between male and female pollock in their proximate analyses values are indicated in Table 2.3. None of the parameters, which include moisture, protein, lipid, and ash, displayed any distinction between males and females caught in the same season. Energy also did not show significant differences between males and females within seasons. Since no difference was found between sexes, values for males and females could be pooled. o o NO o N O §3^ DO » i s IW o o -h-l ^ \"o m o o m o f<\"> o O o o o + 1 +1 +1 +1 00 o NO i n /-> ctj O o o © + 1 +1 +1 +1 Tt; o o iri r-' CN •c a, CO s e > .3 O ^ > • ex § « > 36 »-r GO IW o o co (L> -t-» co e o co VO r -m VO co co 00 VO 00 m Os d +1 o + 1 + 1 + 1 d + 1 d + 1 d + 1 d + 1 CS 00 1 .—1 co 00 o CO o vb vo vo\" VO VO co VO VO CS VO VO d vo i n IT) VO OO i n vo O + 1 o +1 +1 d +1 d + 1 d + 1 d + 1 d +1 i o o m m s -a s CO CD <: > males s c2 B 1 £ o 60 c •c CO O O d v o o d V o o o o o o d V CD > a o CO c3 CD CZ3 CS m d d o d cs o Os VO (3H CO Os CO oo o ir> Os CD CO m co I T ) o o d V OS oo o , o t- d ^ V CS o CO 14 CS CS 00 o s <* 1 - 1 oo I CO CS VO d cs vo 00 . £ O m\" CD > I e o CO cd C/5 37 6.00 5.00 L 4. 00 Winter Spring Summer Fall Figure 2.1 Seasonal changes in the energy density of pollock where * indicates a significant difference between fall and winter, p<0.001 38 Seasonal changes in fatty acid composition of pollock Fatty acid profiles o f pollock were affected during different seasons as shown in Tables 2.4 and 2.5. The concentration o f omega-3 fatty acid in pollock (Fig 2.2) was highest during the summer, and significantly reduced in the fall (p<0.001). Although levels o f 18:3n3 and 22:6n3 dropped from spring to summer, the 20:5n3 significantly increase from 14.53% (spring, p=0.016) to 17.00% of total fatty acids (summer) making the overall omega-3 fatty acid concentration high in the summer. This corresponded with the lowest concentrations of omega-6 fatty acids in walleye pollock occurring during the summer months, which were significantly lower from the fall (p<0.001) and winter (p<0.001) omega-6 fatty acid concentrations. 39 C O <—i 00 m r -oo 00 ON en N O O + 1 00 C O CN CS o 00 d + 1 00 NO NO C O d + 1 NO i n £ 2* C O d +1 ON ON CN CN ON 0 o NO Q 1 - 1 V o o 00 d + 1 I T ) NO d + 1 i n NO d +1 r -co i n i n d + 1 N O o r - o 00 o N O o V ' I g i i U O ' 1 NO C N t--d + 1 ON 00 ON ON O C N + 1 C O C N O N C N C O d + 1 C O C O 00 C N d +1 O N i n i n o O •c O o N O o o © c o O CD > O h o s ' I U E I *3 1 O ' 1 2 w A U A. U S3 A > T3 o § 00 o 1 VO CS CS i n d CS 1 d CS 00 1 00 c o to on ,—1 ,—1 00 CO i n 00 r - CS r~-cs m t- VO m 00 co CO d d d d +1 + 1 +1 + 1 o CO CS 1—1 CS CO CO CO CS , ; i n r-~ rt CS o i n CS VO vo CO CS d d d d + 1 +1 + 1 + 1 OS CO o Os CO i n o m i—i CS Os 00 cd IT) vo CS o o o o d d d d +1 + 1 + 1 +1 o oo r -CO CS o d d d d VO o OS CS Os 00 CO O o o o d d d d + 1 +1 + 1 + 1 r - 00 CO i n CS CO d d d d ir> Os CO i n CO i n O o d d d d +1 + 1 + 1 +1 OS CO CO Os CS ; Os CO d 60 -t-» e .s •e cx s is in 3 00 00 o i n co o o co d OS .-> r - o r - o 2 ?' 2 ? VO CS o o o OS . o V 00 41 Figure 2.2 Seasonal changes in percentage of polyunsaturated fatty acid content of pollock (•significantly different from summer, rxO.OOl) 42 Seasonal changes in amino acid composition and protein digestibility in pollock There were no significant differences in non-essential vs. essential amino acid content of pollock between seasons or in the amounts o f individual amino acids as indicated in Tables 2.6 and 2.7. Moreover, no statistically significant changes in the digestibility o f the pollock protein (Fig. 2.3) were detected between seasons, although this may be due to the small sample size. 43 Table 2.6 Seasonal changes in non-essential amino acid composition in walleye pollock expressed as % of total amino acids Amino Winter Spring Summer Fal l F 3 J 3 p-value A c i d Alanine 7.08 ± 0 . 2 1 2 7.46 ± 0 . 0 2 2 7.00 ± 0 . 1 8 9 7.28 ± 0 . 1 7 2 0.74 0.542 Arginine 6.40 ± 0 . 1 8 9 6.48 ± 0 . 0 0 3 6.82 ± 0 . 6 2 1 6.23 ± 0 . 1 2 8 0.68 0.820 Aspartate 9.87 ± 0 . 1 9 3 10.28 ± 0 . 1 3 6 9.69 ± 0 . 2 9 6 9.98 ± 0 . 0 7 7 1.24 0.335 Cystine 1.01 ± 0 . 0 3 9 0.98 ± 0 . 0 0 2 0.91 ± 0 . 0 4 3 0.96 ± 0 . 0 9 6 0.23 0.871 Glutamate 14.60 ± 0 . 4 0 1 14.55 ± 0 . 3 5 7 14.41 ± 0 . 2 4 4 14.84 ± 0 . 0 9 4 0.68 0.578 Glycine 9.56 ± 0 . 8 5 6 8.18 ± 0 . 5 7 5 8.66 ± 0 . 3 7 9 8.62 ± 0.483 0.70 0.568 Proline 6.33 ± 0.588 4.90 ± 0.062 5.58 ± 0 . 5 5 6 5.71 ± 0 . 3 1 8 1.05 0.402 Serine 5.34 ± 0 . 1 7 2 5.59 ± 0 . 2 4 2 5.46 ± 0 . 0 8 2 5.45 ± 0 . 0 6 7 0.48 0.700 Tyrosine 2.54 ± 0 . 1 4 5 2.91 ± 0 . 4 3 9 3.22 ± 0 . 4 6 2 2.49 ± 0 . 1 7 4 1.53 0.255 1. Values are presented as mean +/- SEM. No significant differences found between seasons. 44 Table 2.7 Seasonal changes in essential amino acid composition in walleye pollock expressed as % of total amino acids 1 Amino Winter Spring Summer Fal l F 3,i3 A c i d Histidine 2.25 ± 0.078 2.24 + 0.058 2.18 ± 0 . 0 4 8 2.43 ± 0 . 1 6 4 0.669 0.586 Isoleucine 3.80 ± 0 . 1 8 0 3.62 ± 0 . 0 1 7 4.21 ± 0 . 1 8 2 3.88 + 0.193 1.074 0.394 Leucine 7.78 ± 0 . 3 0 5 8.24 ± 0 . 2 5 7 8.19 ± 0 . 2 3 5 8.09 ± 0 . 2 8 9 0.354 0.787 Lysine 7.29 ± 0 . 3 4 3 7.86 ± 0 . 1 2 2 7.48 ± 0 . 2 1 5 7.60 ± 0 . 1 3 1 0.780 0.526 Methionine 2.16 ± 0 . 1 2 8 2.15 ± 0 . 1 2 4 2.21 ± 0 . 0 8 3 2.33 ± 0 . 1 3 3 0.423 0.740 Phenylalanine 3.51 ± 0 . 1 6 4 3.75 ± 0 . 1 9 7 3.47 ± 0 . 1 9 0 3.56 ± 0 . 1 3 3 0.293 0.830 Threonine 4.97 ± 0 . 1 6 3 5.26 ± 0 . 1 6 8 4.95 ± 0 . 1 1 4 4.96 ± 0.088 0.835 0.498 Tryptophan 1 .07±0 .064 1.18 ± 0 . 0 3 5 1.01 ± 0 . 0 4 8 1 . 1 0 ± 0 . 1 4 2 0.201 0.894 Valine 4.43 ± 0 . 1 8 8 4.38 ± 0 . 0 3 9 4.56 ± 0 . 1 8 3 4.50 ± 0 . 1 1 1 0.185 0.905 n 1. Values are presented as mean +/- SEM. No significant differences found between seasons. 45 Figure 2.3 Seasonal changes in protein digestibility of pollock (KW3.28 = 8.356, p - 0.039) 46 Species differences between pollock and herring caught in the fall season There were definite differences in proximate analyses parameters between herring and pollock (Table 2.8). Herring had lower moisture content than pollock (p<0.001) and therefore had an inherently higher concentration of energy when comparing whole wet fish. In addition, herring had a higher lipid concentration, when expressed as a proportion o f the dry matter of the fish than pollock (p<0.001), resulting in an even higher energy content (Fig 2.5). Figure 2.4 shows the differences between species on a wet basis, enabling one to see how this disparity in nutrient and energy content could potentially affect sea lions that were eating whole wet fish. Herring had a lower concentration o f protein (p<0.001) and ash (p<0.001) than the pollock samples due to the bulk of the solid matter consisting of l ipid material. O f these lipids, herring has a higher proportion o f saturated fatty acids (p<0.001) than pollock (Fig. 2.6). In addition comparison o f the polyunsaturated fatty acid portion o f both species^ showed that pollock had a higher proportion o f omega-3 fatty acids (p<0.001) as opposed to omega-6 than did herring (Fig. 2.6). 47 n H \\o f l N 1 r - H CS m cs m co co vo IW o PH ^ '3 cu a* ICO i n o o d d + 1 +1 o o i n iri r~ Os d d + 1 +1 00 VO d 00 vo i n m vo d d + 1 +1 cs CS CS co o d + 1 i n d + 1 00 d vo d + 1 co cs VO m o o co co CU « eg o o PH CT5 O O + 1 o d + 1 o + 1 i n o vd vd vb 1—1 cs 00 ^ —I d + 1 +1 +1 .—1 OS 00 Os OS OS co co co m r-~ <—1 + 1 +1 +1 CO CS vo CS ^ ' t t CO CO cs CO CO CS d d d d d d + 1 + 1 + 1 +1 +1 + 1 i n 00 0 r - Os d d vb vb cs cs VO m d d d d d d +1 + 1 +1 + 1 + 1 + m CO CO 00 >—< -+ i n ^ ' i n r - VO VO vo CO co CU 1 S, 13 rt § 6 2 6 3 > 3 0 > 0 > 5 5 a. X cu CO CO VO CS o o o ' d m V vo i n CO PH CO o d <-• . r -« °°. «. 0 PH\" CS o o o g • Os 2 d PH 00 Os i n co o --H O 1 d \" V > H3 O ^ > • a. x CU CO I co < > Q ta • PH .a ! I CO > U 0, 1 03 JO 5 -s 03 1 d 9 a 48 Figure 2.4 Differences in proximate analyses parameters between pollock and herring species on a wet basis (*significant differences, p<0.001) 49 Figure 2.5 Differences in energy density between pollock and herring on a wet basis, *p<0.001 50 Figure 2.6 Differences in fatty acid composition between pollock and herring (* significant difference, pO.OOl) 51 Differences in protein quality and digestibility between pollock and herring caught in the fall season Serine was higher in pollock than herring (p<0.001) and is a non-essential amino acid (Fig 2.8). Valine was higher in herring than in the pollock (p=0.004) and is essential (Fig 2.8). There was a slightly higher percentage of total essential amino acids in the herring at 47.8% of total amino acids than in the pollock at 44.9%, but these differences were not considered significant (p=0.062) which could be due to a lack of statistical power. Herring protein was found to have an initial rate of digestion at 100.4% of the rate of digestion o f casein after five minutes. Pollock protein, on the other hand, had a lower initial rate of digestion compared to casein at 94.8%. These digestibility results were statistically significant (p=0.015) based on a non-parametric analysis o f variance (Fig 2.7). 52 Figure 2.7 Di {Terences in protein digestibility between pollock and herring <*sigmfkant difference, M W U ; - 44.00, p~(K01.5) 53 6.0 * 5.5 o c a 5.o 2 o 4.5 0s-Herring Pollock 4.0 • • 1 n= 2 1 7 x= 5.4 4.5 - i ** Herring Pollock Figure 2.8 Differences in specific amino acid contents between pollock and herring (*p<0.001, **p=0.004) 54 Discussion Seasonal changes in proximate analyses and energy density of walleye pollock The nutrient and energy content of walleye pollock changed seasonally and energy content was lowest in winter and highest in the summer and fall. This loss o f energy may reflect changes in eating habits due to fluctuations in prey availability or seasonal behavioral patterns. Winter conditions in the area o f the Bering Sea and Alaska are harsh and prey populations may decline to low levels, although there has been no data collected on the specific seasonal changes of food availability in this region (Sogard et al. 2000). A study on hake (Merluccius hubbsi) found that the total nitrogen content as well as the lipid content decreased during the spawning months, which in this case were March and September (Mendez et al. 1997). Hake nitrogen content also remained constant during the winter, whereas l ipid content decreased as it was used for energy during times o f starvation. This could be comparable to what happens to the pollock during the winter and spring. The energy in pollock may have decreased during the winter due to food shortage. Recovery during spring, when food becomes more available, may have been slow due to the use o f muscle lipids and proteins as energy reserves in times o f spawning. Slow recovery in spring may also be related to a decrease in feeding during spawning, when starvation appears to occur ( O C S E A P 1986; Pedersen et al. 2001) and fish rely on energy stores for survival. Supporting this idea is that pollock are known to be seasonal feeders (Ciannelli et al. 1998). Steller sea lions, and especially juvenile individuals may target smaller pollock, which would be easier to swallow, and possibly easier to capture than older, larger fish. Juvenile pollock feed at a high rate and grow rapidly during the spring and summer months 55 in order to survive the winter (Ciannelli et al. 1998). For this reason there may be the same seasonal pattern of body condition for age-0 and juvenile pollock as in adult pollock (Ciannelli et al. 1998). Another study demonstrated that the size and condition of juvenile pollock had a significant correlation with its ability to survive low temperatures and starvation conditions (Sogard et al. 2000). Contradicting Cianelli 's results is a study by Paul (1998) who found juvenile pollock increased in energy content in the fall and winter, to a peak level in the spring. Paul's result also contrasts with my data on adult pollock, which had lowest values for winter and spring and the increased energy density occurred in the summer and fall. Juvenile fish may have lower energy density than adult fish because the ingested energy of a juvenile fish is allocated towards growth rather than storage of lipid (Sogard et al. 2000). This is supported by data in which a previous study done on juvenile age-0 pollock found fish caught in the spring contained 0.955 kcal/g wet mass (Paul et al. 1998) compared to our spring pollock that contained 1.20 kcal/g wet mass. The diet o f pollock changes with age. Juvenile pollock consume copepods and euphausiids, but as they grow their diet shifts to fish, such as juvenile pollock, as well as shrimp and crab ( O C S E A P 1986). These dietary changes can affect the nutritional value o f the pollock. Age-0 pollock exhibit diel migration patterns that enable them to feed and grow efficiently. Cold water reduces the metabolic rate of pollock. Age-0 pollock w i l l swim to warmer waters where prey is more abundant and then retreat into colder waters to use the ingested energy for growth rather than basic metabolic needs (Ciannelli et al. 1998). For this reason, the temperature at which pollock grew optimally decreased as the availability o f food decreased (Ciannelli et al. 1998). However, the opposite is true as well . In times when the waters were warmer and the prey 56 was abundant, optimal growth temperatures would be higher. Strong year classes came from years that were warmer than years where the waters were colder, in addition to other environmental factors (Wespestad et al. 2000). There are no data to state that spawning and seasonal feeding behaviours associated with it occur or do not occur until adulthood in pollock. From the contradictory results of two studies on juveniles, the seasonal variation in energy o f juveniles can probably be more closely associated with prey abundance and the diel feeding pattern used as an energy-conserving strategy in juveniles that is non-existent in adults Other parameters also changed seasonally such as the protein and ash, but it was probably the changing proportion o f lipid in the fish that accounts for the changes in energy content, as lipid stores are used up during times o f reduced feeding. A l l o f the proximate values in Table 2.1 are expressed on a dry basis, so these changes in proportions are not due to fluctuation in water content. Moisture content also experienced seasonal changes. It was highest in the winter, when lipid and therefore energy content were low, and was lowest in the fall when energy and lipid contents were high. Energy density in my study was expressed as dry weight values, so changes in moisture content would affect how much energy would actually be available to the sea lions when a whole wet fish is ingested. For example, the higher moisture content of the fish in the winter, coupled with the lower energy content in the dry matter of the fish leads to an even greater energy depression than as it may initially appear based oh dry matter values. Likewise, the lower moisture content of the fish in the fall months when the energy content is higher on a dry basis leads to a greater amount o f energy available to the sea lions compared to winter fish. 57 Seasonal changes in fatty acids of pollock Total omega-3 fatty acids were highest in pollock during the summer months. This contrasted with the expectation that long-chain polyunsaturated fatty acid content would be higher in winter to help keep the membranes o f the fish more fluid in the colder temperatures. The result may instead be related to the food available to them during these months. In some Kodiak Island bays, pollock consumed primarily chaetognaths and copepods in May, and primarily euphausiids and shrimp in July ( O C S E A P 1986). The fatty acid composition of these prey may affect the fatty acids in pollock. However, different species o f fish incorporate dietary fatty acids into their cell membranes at different levels (Roy et al. 1999). Dietary fats of carp have been shown to have a limited effect on liver membrane composition due to temperature (Roy et al. 1999) but this was not found in pollock. The high polyunsaturated fatty acid content of the summer pollock could potentially make their predators, who incorporate these fatty acids into their own tissues, more susceptible to oxidative stress. This is most likely not the case in territorial male Steller sea lions, which are largely fasting during these months, but it may affect the milk o f nursing females, and therefore the oxidative status o f suckling pups at this time o f year. Seasonal Changes in Protein Digestibility of Walleye pollock The in vitro method o f determining protein digestibility measures the initial rate of protein digestion, which is compared to a standard protein (casein). The products of protein digestion were not removed during the experiment, so the accumulation o f digestive products 58 can inhibit the enzymatic digestion over time. Therefore, only the initial rate of digestion should be considered (Yuan et al. 1991). Protein digestibility in pollock appeared to be higher in the summer than in the other seasons but the difference was not statistically significant due possible to a lack of statistical power. A larger sample should be used to confirm any differences that were seen in my study. Variability in protein digestibility can occur from variations in amino acid composition, although I did not find any seasonal changes in protein quality in my study. Experiments with larger samples o f fish for amino acid and protein digestibility should be used to find the true affects o f season on protein digestibility in relationship to protein quality. Implications for Steller sea lions The seasonal changes in energy density o f pollock are disadvantageous to the sea lion when considering the seasonal feeding habits o f the species. During the winter, male sea lions build up energy stores that must sustain them throughout the breeding season in the spring. However, I found the lowest levels o f energy in pollock occurred during the winter and spring months. The highest energy density as well as the highest level o f protein digestibility occurred in the summer and fall. This coincides with behavioral changes of territorial male sea lions, which are fasting during the summer breeding months As well during the winter months, there are increased requirements for fat stores to provide insulation in the colder environmental temperatures for both sexes. The low lipid content o f pollock during these months might not allow for the sea lions to ingest the amount o f energy needed to meet their needs. It is unlikely that the sea lions would lower their basal metabolic rate in response to the decrease in ingested energy. A 31% drop in basal metabolism was found to 59 occur in Steller sea lions during periods o f fasting, but the same response did not occur during periods of food restriction (Rosen et al. 2002). The extent to which the Steller sea lion diet, in terms of prey composition, changes seasonally is unknown, but generally pinnipeds are believed to be opportunistic predators and take advantage of locally and seasonally abundant food sources (Hobson et al. 1997). Variations in prey consumption would be more likely due to the migratory patterns o f fish than selectivity by the animal (Brown et al. 1998). Harbour seals in the United Kingdom have been shown to have significant changes in their diet seasonally. In the spring the seals prey mainly on sand eels, whereas gadids are very dominant in the diet for the rest of the year, being most important during the winter months (Brown et al. 1998). This pattern o f feeding generally follows the seasonal availability o f the prey species as it most likely would for Steller sea lions as well . Limitations Caution should be taken when using the results of my study to support the theory that a pollock diet can be detrimental to the health of the Steller sea lion. While I found significant differences in nutrient content between pollock and herring, it must be noted that my study used samples provided by commercial fisheries. This means that the fish were selected for size by the fishing nets used. A comparison o f research and commercial fisheries, however, did find significant correlation between data collected from both groups (Fox et al. 1996). Both groups did use the same gears and methods, which themselves produce bias, such as the mesh size of the net selecting for fish size. In addition, since the nutrient content of fish varies seasonally, I could not apply the pollock versus herring data to other seasons of the year. 60 Most of the pollock used in my study were large adults, whereas in the wi ld , sea lions may be more apt to catch juvenile pollock, which would be more similar in size to the adult herring. Pollock being a large and bony fish may be difficult for sea lions, especially juveniles, to swallow when they are full-grown. It is unknown what the differences between juvenile and adult pollock are nutritionally and further study is needed to find any differences. Furthermore, the results could also be affected by the location in which the pollock were caught. Different geographical locations could cause differences in nutrient content due to water temperatures and prey availability. Since my samples came from more than one fishing company, it is possible that various samples were caught at different locations, meaning the differences between seasons could be due to a location difference rather than a seasonal difference. While I did have at least one month's worth of samples from each season, ideally I would have liked to have had samples from every month o f the year to assess the pattern o f change in nutrient content. There were no pollock caught in the months o f Apr i l , May, or June and I had no samples from December as well because fisheries do not harvest pollock year round. Macronutrient content and energy density : walleye pollock versus herring It is generally known that large demersal and pelagic fish such as pollock are lower in lipid content and therefore energy, than small schooling species, such as herring (Van Pelt et al. 1997). In my study there were significant differences between herring and pollock for all the measured parameters in the proximate analyses. Most importantly is the fact that herring was much more energy dense at 6.1 ± 0.04 kcal/g dry matter than pollock which had an energy density of 5.4 + 0.03 kcal/g on a dry basis. This would be due to the fact that the lipid 61 content o f herring was much greater than pollock, and lipids provide the greatest contribution to energy content (9.2 kcal/g) when compared to protein and carbohydrate (4.2 kcal/g) (Trayhurn 1992). This observation would be further magnified when the data are expressed on a wet basis as herring is considerably lower in water content that pollock (Figs 2.4 and 2.5). It was shown previously that there exists a negative correlation between water content and lipid content in fish (Van Pelt et al. 1997), but it is important to use dry matter to perform bomb calorimetry in order to compare samples where loss of water in samples may occur prior to analysis. Bomb calorimetry is also known to be an overestimate o f available or net utilization energy as it measures all combustible material in the sample and does not account for materials that may be indigestible to animals and not available for body maintenance or growth. Pollock is a bonier fish with 10.9 + 0.23 % of its dry matter as ash, compared to herring, which was found to have 6.9 + 0.19% ash. This fact probably further contributed to the lower energy density of pollock, but could also make the pollock less palatable than the herring when eaten whole, as with the sea lions. Rosen and Trites (2000b) found that pollock did have a lower dry-matter digestibility compared to herring due to the higher amount of bony material in pollock. Protein digestibility: Pollock vs. Herring Tests on pollock for protein digestibility showed that it was less digestible than herring, whose digestibility was on par with casein, the standard used. Coupled with the lower energy density o f pollock, the lower digestibility suggests that there would be even less energy available to the sea lion compared to herring. 62 There are limitations in assuming that in vitro digestibility is equivalent to the gut o f an animal, which has interactions o f microflora with some o f the dietary components. The method of in vitro digestibility used for my study was not specific to the sea lion, which limits its ability to show what would happen to protein in the sea lion's digestive system. However, studies done in vivo with pinnipeds support my in vitro results. For example, Rosen and Trites (2000b) found that the digestive efficiency o f Steller sea lions fed herring was greater than when the same sea lions were fed pollock. The dry matter digestibility, which is also known as assimilation efficiency, of herring was also found to be greater than for pollock (Rosen et al. 2000b). This supports my finding that herring protein was more digestible than pollock. Another study done in ringed seals found that the animal's efficiency at assimilating ingested energy also decreased as food quality (defined as energy and lipid) decreased (Lawson et al. 1996). The ringed seals had the highest assimilation efficiency for herring and the lowest efficiencies were found for low energy fish (Lawson et al. 1996). The studies with ringed seals and Steller sea lions did not measure only protein digestibility, but the digestibility o f the entire fish. Protein was independently evaluated in another in vivo study done on rats, which found that herring containing diets had a higher protein efficiency ratio than pollock-containing diets (Donnelly et al. 2002). Implications of the differences between pollock and herring for Steller sea lions Differences in nutrient content of pollock and herring are fundamental to understanding the plight of the Steller sea lion. Herring stocks make up a lower proportion o f the sea l ion diet than they once did in the areas of the population decline, while pollock have increased. One theory on why this change has occurred is that long-term climate 63 changes have affected the abundance o f various fish species (Klyashtorin 1998; Benson et al. 2002). Pollock seem to prefer warmer temperatures, with strong year classes generally seen in warmer years (Wespestad et al. 2000). In contrast, herring seem to prefer lower temperatures and strong year classes have been associated with low sea surface temperatures in a study done in Japan (Nagasawa 2001). Regardless o f the reason for the prey population shift, the abundance of pollock in the sea lion diet may cause nutritional stress due to the energetic needs o f the sea lion not being met. The lower digestibility o f pollock protein may have compounded the lower energy density by having less ingested energy available after digestion and absorption by the sea lion. Animals that consume pollock instead of herring would have to increase their food intake to compensate for the difference in energy density. They also have to compensate for the energy that is not absorbed and for the increased energy they require to digest larger meal sizes (Rosen et al. 1997). Strictly using the values I found in my study, without considering digestibility o f energy, sea lions would have to consume 35.5% more pollock than herring to get the same caloric content. The disparity increases i f consideration is given to the amount o f energy actually absorbed from the fish and the energy used for digestion (heat increment of feeding). Using previously determined digestive efficiency values o f 93.9% for pollock and 95.4% for herring (Rosen et al. 2000b) and the energy lost from the heat increment of feeding (15.7% for pollock and 11.9% for herring) (Rosen et al. 1997; Rosen et al. 2000a), I found that the Steller sea lion would need to consume 60.4% more pollock than herring to obtain the same amount of usable energy. This is close to the value found by Rosen and Trites (2000a), 64 which found that an average o f 56% more pollock than herring would be needed for the sea lions to obtain the same utilizable energy as herring. Fatty acid profiles Not only does the amount of l ipid in the fish influence nutritional quality, but the specific types of fatty acids present in the lipid are also important. For example, i f more saturated fat is ingested in the diet than polyunsaturated fat, then plasma cholesterol may become elevated (Bruckner 1992). Polyunsaturated fatty acids, and in particular, omega-3 fatty acids are widely known to be beneficial to health because of the effect they have on modifying blood lipid characteristics. Long chain omega-3 polyunsaturated fatty acids ( P U F A ) are instrumental in preventing thrombosis and atherosclerosis and can also increase the proportion of omega-3 P U F A in cardiac phospholipids, which affect the physiological functioning o f the heart (Sergiel et al. 1998). Omega-3 fatty acids include the essential linolenic acid, as well as docosahexanoic acid ( D H A ) and eicosapentanoic acid (EPA) . These fatty acids are precursors to compounds that can cause vasodilation and can reduce platelet synthesis, thereby reducing the risk o f a heart attack (Bruckner 1992). Omega-6 fatty acids, such as linoleic acid and arachidonic acid are precursors to compounds that can encourage platelet formation and vasoconstriction. Since both omega-3 and omega-6 fatty acids compete for the same biochemical pathway to form respective vasoactive compounds, the ratio o f omega-3 to omega-6 in the diet is also important to cardiovascular health. The herring in my study had a higher ratio of omega-6 to omega-3 fatty acids than pollock. Pollock also had a higher ratio o f polyunsaturated fatty acids to saturated fatty acids when compared to herring. 65 In terms of human health, the lower fat content of pollock as well as the low saturated fat content and high ratio o f omega-3 fatty acids to omega-6 fatty acids would be beneficial for cardiovascular health. It is not known however, what the effects o f these parameters would be on the Steller sea lions. Sea lions need the high fat content of herring to have sufficient energy stores for breeding and also for protection from the cold water temperatures in Alaska. It is also unknown how prevalent any sort o f cardiovascular disease is in the Steller sea l ion population. The genetics o f each mammal population can make them more or less susceptible to cardiovascular disease when fed cholesterol or saturated fat because of the varying lipoprotein profiles in each species (Kwiterovich 1997). A more important consideration would be the affect on oxidative status of the sea lion. The greater amount of P U F A in pollock than in herring could also make the sea l ion more susceptible to oxidative stress when eating pollock. I f the sea lion incorporates these fatty acids into its tissues, those tissues could undergo oxidation whereas saturated and monounsaturated fats would be more resistant to oxidation. Protein quality Amino acids are the building blocks of proteins and each amino acid has an important physiological function. Not only is the appropriate amount of dietary protein important, but the protein should also be complete and include all essential amino acids. Deficiency in any essential amino acid could lead to any number of health problems. The needs o f Steller sea lions in particular are unknown, so I can only hypothesize about the effects o f differences in the amino acid content of their prey. Protein deficiency in laboratory rodents has been shown to delay maturation and slow growth rates (McAdam et al. 1999). Arginine and valine 66 deficiencies in particular have been implicated in the slowed growth and delayed maturation (McAdam et al. 1999). The lower relative amount o f valine I found in pollock compared to herring could explain the growth depression seen in sea lion pups, i f they encounter the same problems as the lab rodents. However, the higher serine content of pollock than herring probably has little or no effect on the health of sea lions because serine is a dispensable amino acid that can be synthesized endogenously i f no dietary source is available (Reeds 2000). It is difficult to come to a conclusion about the effects o f these differences in amino acids in sea lions, not only because of the lack of data on their amino acid needs, but also because metabolism o f proteins in the gut can change the amount of any particular amino acid being absorbed as compared to being ingested Amino acids can be enriched by an increased absorption rate of a particular amino acid and anabolism. Impoverishment of amino acids from decreased absorption or catabolism, and modification of amino acid composition with increased protein intakes can also occur. Microbes can also alter amino acid absorption. It is unlikely that the small differences in amino acid content between pollock and herring would be o f great detriment to the sea lions because my analyses shows that pollock protein does not appear to lack any essential amino acids. More likely, the three dimensional conformation o f the protein structure may be what is affecting the differences in digestibility o f the pollock and herring proteins. 67 Summary The steady decline o f Steller sea lions from Prince Wi l l i am Sound through the Aleutian Islands may be related to the shift in their diet from small schooling fish such as herring (Clupea pallasi) to primarily walleye pollock (Theragra chalcogramma). I compared the nutritional value of herring with pollock and explored seasonal changes in the nutrient content o f pollock. Pollock caught in the winter had the lowest energy density compared to pollock caught in the summer and fall that had the highest energy density (p<0.001). Herring had significantly greater energy content than pollock (p<0.001) when fish caught in the same season were compared. Herring was lower in ash, moisture, and protein than pollock because a large proportion of the herring mass was made up o f lipid, hence the higher energy density. In herring, the proportion o f fat that was made up o f saturated fatty acids and omega-6 fatty acids was significantly higher than in pollock (p<0.001). A significantly higher proportion o f the fat of pollock was omega-3 fatty acids compared to herring which had a higher proportion o f fat as omega-6 fatty acids (p<0.001). There were no major differences in protein quality between the fish, although herring was significantly lower in serine (p<0.001) and higher in valine (p=0.004) than pollock. Herring protein did exhibit significantly higher digestibility than pollock protein (p=0.015). These differences could mean that Steller sea lions that consume primarily pollock are at risk for not meeting their energetic requirements and possibly being under greater oxidative stress due to the higher levels o f P U F A in pollock. 68 Chapter III: Effect of a Pollock Diet on the Nutritional Status of the Steller Sea Lion Introduction A shift in diet from fatty fishes to low-fat fishes is thought by some to underlie the decline o f Steller sea lions in the G u l f o f Alaska and Aleutian Islands since the late 1970s (Alverson 1992; Rosen et al. 2000a; Trites et al. 2002). Diets dominated by any single species o f fish can result in intoxication, increased effects o f antimetabolites, and malnutrition in pinnipeds, regardless o f diet quality Furthermore, some species o f fish such as hake and pollock and other gadids can induce anemia in mammals (Geraci 1975) and may cause oxidative stress (Stout et al. 1960; Thompson et al. 1997). Oxidative stress can cause a number of health problems in animals such as organ failure, carcinogenesis, immune deficiencies and cardiovascular disease. These health problems are caused by damage derived from the products of oxidation reactions in the body. Antioxidants such as vitamin E are therefore very important in regulating the oxidative status of animals. Deficiency o f vitamin E in pinnipeds has been shown to cause steatites, muscular degeneration, liver necrosis, and anemia. It can result from dietary inadequacy or destruction of the vitamin E through oxidation (Geraci 1981). The goal o f my study was to evaluate the nutritional status o f captive Steller sea lions fed walleye pollock and Pacific herring. Field observations o f Steller sea lions in Alaska indicate that body size was reduced during the decline (Calkins et al. 1998)and plasma haptoglobin was elevated (Zenteno-Savin et al. 1997). I therefore measured body size and 69 hematology and tested the resistance of red blood cells (drawn from the animals while on pollock and herring diets) to oxidation induced in vitro. The main hypothesis o f my study was that the captive sea lions should experience a decline in health (as measured by anemia and weight loss) while fed the pollock diet as compared with the herring diet. The underlying cause for these changes was expected to be due to the greater relative susceptibility to oxidative damage when the sea lions were switched to the pollock diet. 70 Materials and Methods a. Feeding Trials on captive Steller sea lions Three juvenile Steller sea lions, one male (Male 1 - M 9 7 K O ) and two females (Female 1 - F97HA, Female 2 - F97SI) participated in this study at the Vancouver Aquarium Marine Science Centre. A control diet o f herring was administered prior to the experimental feeding trials. The sea lions were allowed to consume the diets ad libitum. The trial was done as a crossover with each of the three sea lions eating pollock for six-week treatment periods, followed by six weeks o f a herring diet for a control/recovery period. The trial was done twice on each sea lion. The ages o f the sea lions ranged between 2.3 to 2.4 years at the beginning o f the first pollock trial. Weight, length and girth o f the sea lions were recorded during the feeding trials and blood samples were drawn after each feeding period was completed. Blood samples were sent to a veterinary lab for analyses (Central Lab for Veterinarians, Langley, .B .C. ) , which included hematocrit, hemoglobin concentration, glucose concentration, blood urea nitrogen(BUN), creatinine, serum iron, iron saturation, and total iron binding capacity (TEBC). b. Sea Lion Plasma Lipid Analyses Plasma was collected from centrifuged ice-chilled whole blood and analyzed for total cholesterol (Siedel et al., 1983), triacylglycerols (Ziegenhorn, J. 1975:Yuan et al., 1998), and phospholipids (Takayama et al., 1977) using biochemical assay kits (Boehringer Mannheim, Laval, Quebec). 71 c. In Vitro Forced peroxidation assay for R B C The method of determining the susceptibility o f red blood cells to oxidation was taken from the studies o f (Yuan et al. 1996) and (Yuan et al. 2002). Briefly, aliquots (50uL) of packed R B C were diluted into a 10% suspension with 0.9% NaCl-2 m M NaN3 and incubated at 37°C for 5 min. The peroxidizing solution (500uL hydrogen peroxide in varying concentrations, freshly prepared in saline azide) was added to the R B C and the mixture incubated at 37°C for 30 min. The reaction was stopped by the addition of 0.5 m L cold 28% TCA-0 .1 M Na-arsenite ( B D H chemicals, Poole, England), followed by centrifugation at 12,000 x g for 5 min, at 4°C. A 1.0 ml aliquot of supernatant was assayed with 0.5 ml., 0.5%-2-thiobarbituric acid ( T B A , Eastman Organic Chemicals, Rochester, N Y ) , which was freshly prepared in 0.025 M N a O H for the formation o f malondialdehyde. Samples were boiled for 15 min and cooled to room temperature. Absorption readings were made at 532 nm to determine the amount o f thiobarbituric acid reactive substances ( T B A R S ) . d. Vitamin E Assay of Sea Lion Plasma Extraction o f a-tocopherol was performed according to the method of Desai (1984) with modification. Briefly, 0.25 ml of sea lion plasma was mixed with 0.25 ml o f water and 0.5 ml o f ethanol. 1 ml o f hexane was added and mixture was mechanically mixed for 10 minutes, followed by centrifugation at 4000 R P M for 5 minutes at 4°C (Eppendorf 4050). 0.4ml o f hexane layer was then removed to a new test tube and evaporated to dry under nitrogen stream. Residue was then dissolved with 0.1ml methanol. 25 ml o f the residue was used for injection. Reverse phase column C18 Luna column (5m, 250'4.6mm, Phenomnex, 72 Torrane, C A ) and isocratic flow with methanol: water = 98/2 (v/v) at 1.5 ml/min was used. Tocopherol was monitored at 292nm (Hoglen and Liebler, 1998). The amount o f a-tocopherol quantified according to a standard curve obtained from standard a-tocopherol. 73 Results Steller sea lion changes in feeding behaviour The three sea lions that were a part o f the feeding trial increased their food intakes when their diets were switched from herring to pollock (Table 3.1). For example, Male 1 consumed, on average, 7.92 ± 0.882 kg of herring per day, which was significantly lower than the 11.00 + 1.278 kg of pollock he consumed per day (p=0.05). Female 1 also increased its food intake from 5.40 ± 0.261kg of herring to 9.11 ± 0.887 kg of pollock (p<0.001). Female 2 did not significantly increase the mass of pollock she ingested. The daily caloric intake, which was calculated by multiplying the mass o f fish eaten per day by the energy density values found in Chapter II, did not differ significantly for any o f the animals. I also examined the dietary intake in terms o f the percentage o f body mass consumed and found Male 1 (p=0.030) and Female 1 (p=0.001) consumed significantly more pollock as a percentage of their body mass (Table 3.1). 74 Table 3.1 Changes in eating behaviour in Steller sea lions : Pollock diet vs Herring diet 1 Animal Diet mass o f kcal eaten % of body mass fish eaten (kg) per day eaten Male 1 Herring 7.92 ± 0.882 3 Pollock 11.00+ 1.278 16954 ± 1887 15191+ 1765 4.11 ± 0 . 4 9 7 a 5.98 + 0.631 Female 1 Herring Pollock 5 . 4 0 ± 0 . 2 6 1 a 9.11+0.887 11561 ± 5 5 8 12575 + 1225 5.20 ± 0.282\" 8.57 + 0.874 Female 2 Herring Pollock 7.29 ± 0.792 8.58 + 0.796 15603 ± 1 6 9 7 11851+1099 5.23 ± 0 . 7 1 9 6.62 + 0.658 A N O V A F i , 9 4 (diet) p-value 13.723 <0.001 1.507 0.223 17.550 <0.001 1 .Values are presented as mean +/- SEM. significant difference is defined as p<=0.05. The last week of each trial was used to calculate results, a - values were significantly different between diets 75 Changes in mass of the Steller sea lions The masses o f the sea lions changed when their diets were switched between pollock and herring as indicated in Table 3.2. For Male 1, the first six weeks on pollock had him gaining mass at a rate of 0.0914 kg/day, but he gained mass at a higher rate o f 0.3914 kg/day (p<0.001) when he was switched to the herring diet. On the second trial with pollock, Male 1 lost weight at a rate o f 0.1683 kg/day, and gained back the mass lost at a rate o f 0.8646 kg/day (pO.OOl) when put back on a herring diet for six weeks. Female 1 lost mass with the first feeding period o f pollock at a rate of 0.0394 kg/day and gained mass when switched to herring at a rate o f 0.1464kg/day (pO.OOl) . This trend in Female 1 continued into the second pollock trial in which she gained weight at a rate of 0.1629 kg/day. When put back on the herring diet, she continued to gain mass, but at a slightly lower rate of 0.1078 kg/day (pO.01) . Female 2 lost mass on both o f her pollock trials at a rate o f 0.1953 kg/day for trial 1 and 0.0989 kg/day during the second trial. She started to gain mass when switched to herring at a rate of 0.0827 kg/day for trial 1 and 0.1113 kg/day for trial 2. The differences in her rates o f mass change were also statistically significant for both feeding trials (pO.OOl) . 76 Table 3.2 Changes in mass of Steller sea lions : pollock diet versus herring diet 1 Animal Diet Rate of Change in mass (kg/day) Trial 1 Trial 2 M a l e l Herring 0.391 a 0.673 a Pollock • 0.091 -0.168 ta.61 = 10.76 t2.es = 8.01 p<0.001 p<0.001 Female 1 Herring 0.146 3 0.103 a Pollock -0.039 0.163 t 2, 7 2= 6.92 t2,79= 2.784 p<0.001 p<0.01 Female 2 Herring 0.083 a 0.060 a Pollock -0.195 -0.099 t2,7,= 13.571 t2,79 = 8.011 p<0.001 p<0.001 1. Rates were taken from plot of mass over time of trial, a - values were significantly different between diets A t - test for comparison of two slopes from linear regressions was used to find differences. 77 Blood and Serum analyses There was a decrease in cholesterol levels (Table 3.4) of the sea lions while on the pollock diet. There was also a decrease in plasma vitamin E content (Fig. 3.1) for all three sea lions when they consumed pollock. Female 1 had a plasma vitamin E content of 19.5 + 1.78 ug/mL while on the herring diet, which decreased to 12.9 ± 0.30 ug/mL while eating pollock. Male 1 also experienced a drop in vitamin E , from 14.71 + 1.10 ug/mL on herring to 10.5 + 1.21 ug/mL on pollock. The results for the individual sea lions could not be statistically compared because the sample size was only two for each o f the measures. When the results for all three animals were pooled I did find there to be a significant decrease in plasma vitamin E levels when the sea lions were consuming pollock (p=0.032, n=6). o o \"o CO > > \"3 s 5 CU CO T3 § O O .S3 co © ^H t~-d d o — i CS OS O 00 d d OS r- oo o r-d d >-H oo oo CS CM <-! d cs ^H r-d d o so d d so oo CO CO CS SO d d o Os CS SO d d cu 3 > PH 79 Table 3.4 Plasma profiles o f Steller sea lions when diet consists of herring vs. pollock 1 Diet Triglycerides Phospholipids Cholesterol Vitamin E mmol/L mmol/L mg/dL ug/mL Male 1 Herring Pollock 0.8 ± 0 . 1 5 0.6 + 0.02 6.0 ± 0 . 8 3 5.5 + 0.26 480.8 ± 3 1 . 4 0 327.2 + 5.14 14.7 ± 0 . 7 8 10.5 + 1.21 Female 1 Herring Pollock 0.6 ± 0 . 1 0 0.5 + 0.04 6.0 ± 0 . 3 8 5.3 + 0.62 438.2 ± 2 6 . 7 0 347.4 + 84.65 19.5 ± 1.78a 12.9 + 0.30 Female 2 Herring Pollock 0.5 ± 0 . 0 7 0.8 + 0.03 7.2 ± 0 . 1 8 7.0 + 0.30 514.8 ± 2 1 6 . 1 4 509.5 + 5.50 17.2 ± 0 . 2 7 16.5 + 0.50 A N O V A Fi,io 0.072 (diets) p-value 0.794 0.851 0.378 3.237 0.102 6.231 0.032 1 .Values are presented as mean +/- SEM. significant difference is defined as p<=0.05. a - values were significantly different between diets The last week of each feeding trial was used to calculate results. n=2 for each animal during each feeding trial 80 Figure 3.1 Changes in plasma vitmain E concentration in Steller sea lions when switched from herring to pollock diets (n=2 for all measurements) 81 Forced red blood cell peroxidation assay A l l o f the Steller sea lion 'pollock' blood samples that were exposed to hydrogen peroxide at varying concentrations experienced an increase in absorbance values at 532nm, an indication of thiobarbituric acid reactive substances, compared to 'herring' blood samples (Figs. 3.2, 3.3, and 3.4). A l l three sea lions experienced an increase in oxidative stress when fed a diet consisting of pollock. 82 Figure 3.2 In vitro forced peroxidation assay on red blood cells o f Male 1 83 Figure 3.3 In vitro forced peroxidation assay on red blood cells of Female 1 84 Figure 3.4 In vitro forced peroxidation assay on red blood cells o f Female 2 85 Discussion Changes in body mass The size o f a Steller sea l ion can be an indication o f health status (Trites et al. 2002). Generally, animal populations, which have a reduction in body size over a period o f time, are thought to suffer from food limitation and malnutrition. This is especially true of animals that undergo a larger reduction in mass than in length, as was found in female Steller sea lions in the G u l f o f Alaska (Calkins et al. 1998). A shift in diet from fishes such as herring and capelin, to gadoid fish such as walleye pollock may be the cause o f the nutritional stress the sea lions may be under. The weights of the sea lions in my study indicate that Male 1 and Female 2 clearly had an easier time gaining body mass when fed the herring diet than when fed the pollock diet. Steller sea lions should still be growing until about age 8 for males and age 5 for females (Kastelein et al. 1990; Winship et al. 2001). The weight losses or reduced rates of weight gain that occurred in the animals during the pollock feedings therefore suggest that their nutritional needs were not being met by the pollock as they were on the herring. Although these trials occurred over a short period of time (four six-week periods) these changes in body mass are probably not a result of innate seasonal feeding patterns. Male Steller sea lions do not appear to develop their strong cyclical pattern o f feeding until they reach 7 years o f age, and females do not appear to have a strong monthly pattern of food intake until they reach 10 years o f age (Kastelein et al. 1990). However, other studies have shown some seasonality in juveniles and in adult male sea lions that starting at age 6 between 86 November and March with little growth from M a y to September (Winship et al. 2001). This study found that sea lions under the age o f 6 grew minimally or decreased in size during the breeding season from June to July (Winship et al. 2001). M y study was done between the end of January to the end o f August and crosses between the pre-breeding season period and the breeding season. The first trial in both females began in late winter and into the spring, and the first trial o f the male sea l ion occurred in the spring months. Both o f these periods are thought to be times o f increased feeding and growth. The loss o f body weight that occurred when the sea lions were fed the pollock diet followed by the weight gain due to the herring diet is probably not an effect that can be attributed to a seasonal change, as I would expect growth to occur throughout this time interval. The second feeding trial of the females occurred from late Apr i l into July, and the second trial o f the male occurred from early June into late August. During this time, I expected to see a decreased rate of feeding and growth, rather, in Female 2 and Male 1 a similar pattern to the one that occurred in their first feeding trial was observed. In Female 1, however, I saw an increased rate o f growth on pollock and a slightly lower rate o f growth on herring. From these results, seasonal patterns o f growth probably did not have an impact on the rate of growth o f the sea lions as much as the changes in diet did because I would have seen a decrease in growth rates across the trial had it been a seasonal change. M y results agree with studies on rats that were fed pollock and herring, where rats fed herring or pollock supplemented with herring oil were heavier than rats fed pollock or pollock supplemented with its own oil (Donnelly et al. 2002). 87 Changes in feeding behaviour Instead o f looking only at the changes in mass or the amount of food ingested, the influenced o f food intake as a relationship with the mass o f the sea l ion was also examined to correct for increases in food intake related to growth of the animals. Doing so showed Female 1 had a high rate of weight gain on her second trial o f pollock, after losing mass during her first trial. This same individual also ingested the greatest amount of pollock for its size (8.57 + 0.874 % of body mass), which was a 65% increase over the amount of herring consumed. These results indicate that Female 1 was able to increase her caloric intake on pollock as shown in the previous energy analyses o f pollock and herring (Chapter II). I estimated that a sea lion would have to consume an average o f 60.4% more pollock than herring to get the same utilizable energy. Thus, Female 1 was able to consume enough pollock to meet her energy demand. Both Male 1 and Female 2 also increased the mass o f food they consumed when on the pollock diet, but not to the extent that they were able to consume as many calories as they did on the herring diets. Male 1 increased his intake by 45% and Female 2 increased her consumption by 27%, which is clearly not enough to compensate for the lower energy density and digestibility of pollock. The increased amount o f food ingested is in contrast to the study by Rosen and Trites (1999) in which other young Steller sea lions fed a low energy diet of squid did not eat more to compensate decrease in dietary energy density. Another study done with pollock and herring also found that the sea lions did not increase their food intake in relation to the lowered energy intake (Rosen et al. 2000a). However the sea lions were only fed pollock for periods between eleven and twenty-four days (Rosen et al. 2000a), which may not have been enough time to see a significant increase in food intake or may not have been enough 88 time for the sea lions to adapt to the new food source. The same may be true for the study using squid, where the sea lions were on an ad libitum diet o f squid for fourteen days (Rosen et al. 1999). Time may be needed for the sea lions to physically stretch their stomach capacity to accommodate the increase in food ingested, although palatability o f the food given may also be an issue. Rats fed pollock or herring have been shown to increase their consumption o f low energy diets to acquire the same amount of energy as rats on the high energy density diets (Donnelly et al. 2002). However, the rats only had to consume 10% more due to the dilution of the fish with other feed ingredients. The larger amount of pollock eaten by the sea lions in my study does come at an energetic cost. The heat increment o f feeding, which is a measure o f the cost o f processing food, increases as meal size increases (Rosen et al. 1997). In addition protein has the highest energetic cost o f digestion and l ipid has the lowest (Rosen et al. 1997). Pollock would therefore increase the heat increment of feeding in sea lions due to the larger amount of food ingested to meet metabolic demands, and the higher proportion of protein to lipid than in herring. The Steller sea lions fed pollock in previous studies lost mass and had depressed resting metabolic rates due to the lower amount of energy in the pollock diet (Rosen et al. 1999; Rosen et al. 2000a). These sea lions had a higher rate of weight loss than the animals in my study because they did not increase their dietary intake of the low-energy diet. A more recent study found that Steller sea lions that were fasted had depressed resting metabolic rates, but that sea lions whose food (herring) was restricted did not experience this depression (Rosen et al. 2002). 89 A bioenergetic model for the Steller sea lion showed that the energy requirements of males and females increased during the winter and spring, even between 2 and 3 years o f age (Winship et al. 2002). The sea lions in my study were not tested for changes to their metabolic rates, so it is not known what their metabolic needs were. However, it is clear that they were not adequately met by the pollock diet. Steller sea lions in the wi ld may have increased needs in comparison to captive animals due to increased activity or time spent in the water, so the effect o f eating pollock may be magnified in wi ld specimens that have to expend more energy in the foraging effort. Winship et al. (2002) estimated that 3-year-old male sea lions required 21 + 5.0 kg of food and females required 17 + 3.8kg of food per day in February. This is calculated as 11% of body mass for both males and non-pregnant females using values for predicted mass (189 kg for males, 156 kg for females without a fetus) (Winship et al. 2001). However, the captive animals in my study only ate in the range of 5.40 kg to 11.00kg for both herring and pollock. These values are calculated as in the range of 4.1% to 8.6% of their body masses. Plasma cholesterol The sea lions in my study did not exhibit changes in many o f the blood analyses that were examined. There was a decrease in plasma total cholesterol for all three animals (32% decrease in Male 1 and 21% decrease in Female 1), as well as a decrease in plasma vitamin E levels when fed pollock. The decrease in plasma total cholesterol was probably due to the higher level o f polyunsaturated fat in pollock than in herring (Chapter n , Fig. 2.6). Diets supplemented with menhaden oi l , another marine oil which is high in omega 3 fatty acids, caused plasma total cholesterol to decrease in hypercholesteremic chicks and rats (Castillo et 90 al. 1999; Yuan et al. 2002). In humans high levels o f plasma cholesterol and specifically low-density lipoprotein cholesterol, is a major factor in cardiovascular disease. The effect o f cholesterol and incidence of cardiovascular disease is unknown in sea lions. However, cholesterol can have a stabilizing affect on cell membranes by protecting against lipid oxidation. A study done in Japanese quail showed that diets high in cholesterol and saturated fat increased the cholesterol levels o f tissue membranes as well as plasma cholesterol (Yuan et al. 1999). While the increase in plasma cholesterol caused the deposition o f plaques in the arteries, the increased cholesterol in tissue membranes was associated with greater resistance to peroxidation of liver tissues. If this is correct, then an increased susceptibility to oxidation of the sea lions would occur when they are fed pollock diets due to decreased cholesterol in cell membranes. Oxidative stress and the role of Vitamin E I found increased susceptibility to oxidation in the forced peroxidation analyses for all three sea lions that were fed pollock. This may have been due in part to lower levels o f cholesterol, but lowered vitamin E levels were probably the primary cause o f oxidative stress. Vitamin E is an antioxidant, and the decrease in serum vitamin E for all three o f the sea lions may account for the lower resistance to oxidation o f their red blood cells. Oxidative stress can result from the formation o f free radicals in biological systems when an animal suffers environmental stress. Damage from oxidation can result in a reduced growth rate, fertility, carcinogenesis, immunodeficiencies, and neurological damage in mammals (Vichnevetskaia et al. 1999). Animals require antioxidants, such as vitamin E and vitamin C in their diet to combat oxidative damage. Vitamin E protects cell membranes 91 from oxidative damage, but can also react with chemical toxins such as carbon tetrachloride and benzene (Vichnevetskaia et al. 1999). If an increase in pollock intake results in a loss of vitamin E , and an increased chance o f oxidative damage as a result, then the decline o f the Steller sea lions could be related to an increase in illnesses related to oxidative damage and a higher mortality rate as a result. Studies on rats have shown that a deficiency in vitamin E can cause neuropathy as evidenced by increased levels of oxidation in all neural tissues (MacEvi l ly et al. 1996). The loss in serum vitamin E caused by the pollock diet can be due to a lowered dietary vitamin E intake from pollock than herring. It could also be due to pollock having a higher polyunsaturated fat content than herring, which is higher in saturated fat than pollock. P U F A is very susceptible to lipid oxidation due to the double bonds in the chemical structure of the fatty acids. The dietary and body stores o f vitamin E in the captive sea lions may have been used up to combat this oxidation. Subsequently, dietary vitamin E may not have been enough to replenish the vitamin E pool in the body. A high level o f dietary vitamin E has been shown to suppress both enzymatic and non-enzymatic lipid oxidation in rats (Sodergren et al. 2001), so it is an important dietary component. Chronic dietary vitamin E deficiency in rats resulted in undetectable levels o f vitamin E in serum and liver after one year and minimal amounts in the muscle and nervous system (MacEvi l ly et al. 1996). The hypothesis that a lowered serum vitamin E concentration may lead to oxidative stress was supported by the in vitro forced peroxidation experiment with the red blood cells o f the three sea lions. A drop in plasma levels o f vitamin E coincided with a reduced ability to protect against oxidative challenges in all three sea lions (Figs. 3.2, 3.3, and 3.4). Another study found that rats fed a diet supplemented with menhaden oi l had increased levels o f 92 omega-3 fatty acids in heart tissue, which in turn, caused the heart tissue to be more susceptible to oxidation than in rats fed corn oil or lard (O'Farrell et al. 1997). This effect occurred regardless o f what oil supplement the rats were given (menhaden oi l , corn oi l , or lard), although the greatest levels o f oxidation occurred from a combination of low dietary vitamin E and increased dietary intake o f menhaden oil . Deficiencies in vitamin E can also have effects on the fertility and reproduction o f animals (Azz i et al. 2000). One study involving bovine embryos created in vitro found that vitamin E allowed more embryos to develop into blastocysts and that embryos cultured with vitamin E were larger in surface area than controls (Olson et al. 2000). Cows supplemented with vitamin E were also found to need fewer numbers o f days and inseminations for conception to occur (Baldi et al. 2000). Such parameters are difficult to assess in wi ld populations o f the Steller sea lion, but there is data that suggests lactating females had reduced pregnancy rates during the population decline (Pitcher et al. 1998). There is some evidence of stress among Steller sea lions from declining populations. Plasma haptoglobin levels, which are an indicator o f disease or sub-lethal damage, were elevated in sea lions from declining populations when compared to sea lions from stable populations (Zenteno-Savin et al. 1997). Haptoglobin can increase in response to inflammation, infections, trauma, myocardial infarction, rheumatoid arthritis, leukemia, and tuberculosis and other conditions (Zenteno-Savin et al. 1997). A number o f these conditions can originate from oxidative stress but also the lower levels o f vitamin E in sea lions fed pollock. Vitamin E also has a role in the prevention o f disease and is found in higher concentration in immune cells than in other cells of the body, suggesting that immune cell membranes are at greater risk o f oxidation (Wang et al. 2000). Studies have shown that 93 vitamin E supplementation improves cell-mediated immunity and oxidative stress in humans (Azz i et al. 2000; Lee et al. 2000). The lowered plasma vitamin E levels in sea lions fed pollock may result in an increased incidence of disease and infection. From my results with the captive sea lions, it is apparent that the red blood cells of the animals did indeed become more susceptible to oxidative damage, both from increased dietary P U F A and low serum vitamin E . Future Directions Although I obtained some promising results in my study in linking nutritional status with the decline o f the Steller sea l ion population, more work needs to be done to find the specifics of why pollock is detrimental to the sea lions' health. O f particular interest was the lowered vitamin E status in the animals fed the pollock diet. However, a larger sample size would increase the reliability o f the results. A s well , it would be helpful to analyze the amount of vitamin E in the pollock itself so that we may know i f the problem lies in the content of vitamin E , or its bioavailability from the pollock. Perhaps post-mortem analysis on sea lions carcasses found in the areas of sea lion decline would also give clues as to whether the sea lion in declining populations are under oxidative stress. I f cancers, neuropathy, cardiovascular disease or other types of diseases caused by oxidation are apparent, then we could better show a causative link between the pollock diet, oxidative stress, and the decrease in Steller sea lions numbers in Alaska. 94 Summary Declining populations o f Steller sea lions (Eumetopias jubatus) in Alaska have been consuming higher proportions of low-energy prey than they did in the past. M y study compared the nutritional status of three captive Steller sea lions fed pollock (the current primary prey in the wild) and herring (the historic dominant prey). Compared to herring, all three animals increased the amount o f pollock they ate to compensate for its lower energy density, and either lost weight or failed to gain weight at the same rate. N o significant changes were detected in the nutritional status o f the sea lions in terms of common hematology parameters such as plasma glucose, blood-urea nitrogen, and hematocrit. However, there was a decrease in plasma vitamin E concentration as well as a decrease in plasma total cholesterol for all three animals. When the red blood cells from sea lions fed pollock or herring were challenged with hydrogen peroxide, they exhibited increased susceptibility to oxidation when the sea lions were on the pollock diet. Thus the captive Steller sea lions were nutritionally stressed while on the pollock diet. 95 References Acworth, I. N . , D . R. McCabe and T. J. Maher (1997). The analysis o f free radicals, their reaction products, and antioxidants. Oxidants, Antioxidants, and Free Radicals. S. Baskin and H . Salem. Washington, D . C . , Taylor and Francis: 23-78. Alverson, D . L . (1992). A review of commercial fisheries and the Steller sea l ion (Eumetopias jubatus): the conflict arena. Reviews in Aquatic Sciences 6: 203-256. Arts, M . T., R. G . Ackman and B . J. Holub (2001). \"Essential fatty acids\" in aquatic ecosystems: a crucial link between diet and human health and evolution. Canadian Journal of Fisheries and Aquatic Sciences 58: 122-137. Banks, M . (1997). Vitamin E and polyunsaturated fats: Antioxidant and pro-oxidant relationship. Oxidants. Antioxidants, and Free Radicals. S. Baskin and H . Salem. Washington, D .C . , Taylor and Francis: 95-111. Benson, A . J. and A . W. Trites (2002). Ecological effects of regime shifts in the Bering Sea and eastern North Pacific Ocean. Fish and Fisheries 3: 95-113. Bradley, R. L . (1998). Moisture and Total Solids Analysis. Food Analysis. S. S. Nielsen. Gaithersburg, M D , Aspen Publishers: 119-139. Briviba, K . and H . Sies (1994). Nonenzymatic antioxidant defense systems. Natural Antioxidants in Human Health and Disease. B . Frei. San Diego, C A , Academic Press Inc: 107-128. Brodeur, R. D . and M . T. Wilson (1996). A review o f the distribution, ecology and population dynamics o f age-0 walleye pollock in the G u l f o f Alaska. Fisheries Oceanography 5(suppl. 1): 148-166. Brown, E . G. and G . J. Pierce (1998). Monthly variation in the diet of harbour seals in inshore waters along the southeast Shetland ( U K ) coastline. Marine Ecology Progress Series 167: 275-289. Bruckner, G . (1992). Fatty Acids and Cardiovascular Disease. Fatty Acids in Foods and Their Health Implications. C. K . Chow. N e w York, N Y , Marcel Dekker Inc. 53: 735-752. Calkins, D . G . (1986). Marine Mammals. The G u l f o f Alaska : Physical Environment and Biological Resources. D . W . Hood and S. T. Zimmerman. Alaska, N O A A : 527-550. Calkins, D . G . , E . F. Becker and K . W . Pitcher (1998). Reduced body size o f female Steller sea lions from a declining population in the G u l f of Alaska. Marine Mammal Science 14(2): 232-244. 96 Castellini, M . A . and D . G . Calkins (1993). Mass estimates using body morphology in Steller sea lions. Marine Mammal Science 9(1): 48-54. Castillo, M . , F. Amalik, A . Linares and E . Garcia-Peregrin (1999). Dietary fish oi l reduces cholesterol and arachidonic acid levels in chick plasma and very low density lipoprotein. Molecular and Cellular Biochemistry 200: 59-67. Chang, S. K . C. (1998). Protein Analysis. Food analysis. S. S. Nielsen. Gaithersburg, M D , Aspen Publishers: 237-249. Ciannelli, L . , R. D . Brodeur and T. W. Buckley (1998). Development and application o f a bioenergetics model for juvenile walleye pollock. Journal of Fish Biology 52: 879-898. Davis, T. A . , H . V . Nguyen, D . P. Costa and P. J. Reeds (1995). Amino acid composition o f pinniped milk. Comparative Biochemistry and Physiology 110B(3): 633-639. Davison, W. and G . Goldspink (1977). The effect o f prolonged exercise on the lateral musculature o f the brown trout (Salmon trutta). Journal o f Experimental Biology 70: 1-12. Donnelly, C , A . W . Trites and D . D . Kitts (2002). Possible effects o f pollock and herring on the growth and reproductive success of Steller sea lions (Eumetopias jubatus): Insights from feeding experiments using an alternative animal model, Rattus norvegicus. British Journal of Nutrition in review. Fox, D . S. and R. M . Starr (1996). Comparison of commercial fishery and research catch data. Canadian Journal o f Fisheries and Aquatic Sciences 53: 2681-2694. Geraci, J. R. (1975). Pinniped nutrition. Rapports et Proces-verbaux des Reunion Conseil International pour l'Exploration de la Mer 165: 312-323. Geraci, J. R. (1981). Dietary disorders in marine mammals: synthesis and new findings. Journal of the American Veterinary Association 179(11): 1183-1191. Harbers, L . H . (1998). Ash Analysis. Food analysis. S. S. Nielsen. Gaithersburg, M D , Aspen Publishers: 141-149. Hobson, K . A . , J. L . Sease, R. L . Merrick and J. F. Piatt (1997). Investigating trophic relationships o f pinnipeds in Alaska and Washington using stable isotope ratios o f nitrogen and carbon. Marine Mammal Science 13(1): 114-132. Hollowed, A . B . , J. N . Ianelli and P. A . Livingston (2000). Including predation mortality in stock assessments: a case study for G u l f of Alaska walleye pollock. Journal o f Marine Science 57: 279-293. ^ 97 Janusz, J. and J. Horbowy (1997). The state of the walleye pollock in the northern part of the Okhotsk Sea, North Pacific. Fisheries Research 30: 87-102. Kastelein, R. A . , N . Vaughan and P. R. Wiepkema (1990). The food consumption of Steller sea lions (Eumetopias jubatus). Aquatic Mammals 15(4): 137-144. Kehrer, J. and C. Smith (1994). Free radicals in biology: sources reactivities and roles in the etiology o f human disease. Natural Antioxidants in Human Health and Disease. B . Frei. San Diego, C A , Academic Press Inc: 25-62. Ki j ima, S. (1993). Chemistry of Vitamin E . Vitamin E : its usefulness in health and in curing diseases. M . Mino , H . Nakamura, A . T. Diplock and H . J. Kayden. Tokyo, Japan, Japan Scientific Societies Press: 3-12. Klyashtorin, L . B . (1998). Long-term climate change and main commercial fish production in the Atlantic and Pacific. Fisheries Research 37: 115-125. Kwiterovich, P. O. (1997). The effect of dietary fat, antioxidants, and pro-oxidants on blood lipids, lipoproteins, and atherosclerosis. Journal o f the American Dietetic Association 97:s31-s41. Landvik, S. (1997). Vitamin E . Oxidants. Antioxidants, and Free Radicals. S. Baskin and H . Salem. Washington, D C , Taylor and Francis: 79-94. Lawson, J. W. , J. A . Hare, E . Noseworthy and J. K . Friel (1996). Assimilation efficiency o f captive ringed seals (Phoca hispida) fed different diets. Polar Biology 18: 107-111. Lee, C . - Y . J. and J. M . - F . Wan (2000). Vitamin E supplementation improves cell-mediated immunity and oxidative stress in Asian men and women. Journal o f Nutrition 130: 2932-2937. Lehninger, A . L . , D . L . Nelson and M . M . Cox (1993). Principles o f biochemistry. New York, Worth Publishers. Lewis, R. W. (1967). Fatty acid composition o f some marine animals from various depths. J Fish Res B d Can 24: 1101-1115. Love, R. M . (1988). The food fishes : their intrinsic variation and practical implications. London , N e w York, Farrand Press ;Van Nostrand Reinhold. MacEvi l ly , C. J. and D . P. R. Mul le r (1996). L ip id peroxidation in neural tissues and fractions from vitamin E deficient rats. Free Radical Biology and Medicine 20(5): 639-648. 98 McAdam, A . G . and J. S. Mi l la r (1999). The effects o f dietary protein content on growth and maturation in deer mice. Canadian Journal o f Zoology 77: 1822-1828. Mendez, E . and R. M . Gonzalez (1997). Seasonal changes in the chemical and l ipid composition o f fillets of the Southwest Atlantic hake (Merluccius hubbsi). Food Chemistry 59(2): 213-217. Merrick, R. L . , M . K . Chumbley and G . V . Byrd (1997). Diet diversity o f Steller sea lions (Eumetopias jubatus) and their population decline in Alaska: a potential relationship. Canadian Journal o f Fisheries and Aquatic Sciences 54: 1342-1348. Meydani, M . (1995). Vitamin E . Lancet 345(8943): 170-176. Mil lward , D . J. (1998). Metabolic Demands for Amino Acids and the Human Dietary Requirement: Mi l lwa rd and Rivers (1988) Revisited. Journal o f Nutrition 128: 2563s-2576s. Nagasawa, K . (2001). Long term variations in abundance o f Pacific herring (Clupea pallasi) in Hokkaido and Sakhalin related to changes in environmental conditions. Progress in Oceanography 49: 551-564. Ney, D . M . (1991). The Cardiovascular System. Clinical Nutrition and Dietetics. F. Zeman. N e w York, N Y , Macmillan Publishing Company: 339-397. O C S E A P (1986). Marine Fisheries: Resources and Environments. The G u l f o f Alaska : Physical Environment and Biological Resources. D . W . Hood and S. T. Zimmerman. Alaska, N O A A : 417-450. O'Farrell, S. and M . J. Jackson (1997). Dietary polyunsaturated fatty acids, vitamin E . and hypoxia/reoxygenation-induced damage to cardiac tissue. Clinica Chimica Acta 267: 197-211. Paul, A . J., J. M . Paul and R. L . Smith (1998). Seasonal changes in whole-body energy content and estimated consumption rates of age 0 walleye pollock from Prince Wi l l i am Sound, Alaska. Estuarine. Coastal, and Shelf Science 47: 251-259. Pedersen, J. and J. R. G . Hislop (2001). Seasonal variations in the energy density o f fishes in the North Sea. Journal o f Fish Biology 59. 380-389: Pitcher, K . W. , D . G . Calkins and G . W . Pendleton (1998). Reproductive performance o f female Steller sea lions: an energetic-based reproductive strategy? Canadian Journal ofZoologv 76: 2075-2083. Pryor, W. (1994). Free Radicals and L i p i d Peroxidation: What they are and how they got that way. Natural Antioxidants in Human Health and Disease. B . Frei. San Diego, C A , Academic Press Inc: 1-24. 99 Reeds, P. J. (2000). Dispensable and Indispensable Amino Acids for Humans. Journal o f Nutrition 130: 183 5s-1840s. Rosen, D . A . S. and A . W. Trites (1997). Heat Increment of Feeding in Steller sea lions, Eumetopias jubatus. Comparative Biochemistry and Physiology 118A(3): 877-881. Rosen, D . A . S. and A . W . Trites (1999). Metabolic effects of low-energy diet of Steller sea lions, Eumetopias jubatus. Physiological and Biochemical Zoology 72(6): 723-731. Rosen, D . A . S. and A . W. Trites (2000a). Pollock and the decline o f Steller sea lions: testing the junk-food hypothesis. Canadian Journal o f Zoology 78: 1243-1250. Rosen, D . A . S. and A . W . Trites (2000b). Digestive efficiency and dry-matter digestibility in Steller sea lions fed herring, pollock, squid, and salmon. Canadian Journal o f Zoology 78: 234-239. Rosen, D . A . S. and A . W. Trites (2002). Changes in metabolism in response to fasting and food restriction in the Steller sea l ion (Eumetopias jubatus). Comparative Biochemistry arid Physiology 132B: 389-399. Roy, R., E . Fodor, K . Kitajka and T. Farkas (1999). Fatty acid composition o f the ingested food only slightly affects physicochemical properties o f liver total phospholipids and plasma membranes in cold-adapted freshwater fish. Fish Physiology and Biochemistry 28: 1-11. Sainsbury, J. C. (1971). Commercial fishing methods : an introduction to vessels and gears [by] John C. Sainsbury. London, Fishing News. Salem, H . and S. Baskin (1997). Toxicity of oxygen and ozone. Oxidants. Antioxidants, and Free Radicals. S. Baskin and H . Salem. Washington, D C , Taylor and Francis: 227-236. Schabetsberger, R. , R . D . Brodeur, T. Honkalehto and K . L . Mie r (1999). Sex-biased egg cannibalism in spawning walleye pollock: the role of reproductive behaviour. Environmental Biology o f Fishes 54: 175-190. Sergiel, J. P., L . Martine, D . Raederstorff, A . Grynberg and L . Demaison (1998). Individual effects of dietary E P A and D H A on the functioning of the isolated working rat heart. Canadian Journal o f Physiology and Pharmacology 76: 728-736. Sizer, F. S. and E . N . Whitney (1994). Nutrition : Concepts and Controversies. St. Paul, M N , West Publishing Company. 100 Sodergren, E . , J. Cederberg, B . Vessby and S. Basu (2001). Vitamin E reduces l ipid peroxidation in experimental hepatotoxicity in rats. European Journal o f Nutrition 40: 10-16. Sogard, S. M . and B . L . O l l a (2000). Endurance of simulated winter conditions by age-0 walleye pollock: effects of body size, water temperature and energy stores. Journal o f Fish Biology 56: 1-21. Stout, F. M . , J. E . Oldfield and J. Adair (1960). Nature and cause of the \"cotton fur\" abnormality in mink. Journal o f Nutrition 70: 421-426. Ternay, A . and V . Sorokin (1997). Redox, Radicals, and Antioxidants. Oxidants. Antioxidants, and Free Radicals. S. Baskin and H . Salem. Washington, D C , Taylor and Francis: 1-22. Thompson, P. M . , D . J. Tollit, H . M . Corpe, R. J. Reid and H . M . Ross (1997). Changes in haematological parameters in relation to prey switching in a wi ld population o f harbour seals. Functional Ecology. Traber, M . G. (2001). Does vitamin E decrease heart attack risk? Summary and implications with respect to dietary recommendations. Journal o f Nutrition 131: 395s-397s. Trayhurn, P. (1992). Dietary Fatty Acids and Thermogenesis: Implications for Energy Balance. Fatty Acids in Foods and Their Health Implications. C. K . Chow. New York, N Y , Marcel Dekker Inc. 53: 517-530. Trites, A . W. (1992). Northern fur seals : why have they declined? Aquatic Mammals 18: 3-18. Trites, A . W. and C. Donnelly (2002). The decline o f Steller sea lions in Alaska: A review o f the nutritional stress hypothesis. Mammal Review in review. Trites, A . W . and P. A . Larkin (1996). Changes in the abundance o f Steller sea lions (Eumetopias jubatus) in Alaska from 1956 to 1992: how many were there? Aquatic Mammals 22(3): 153-166. Van Pelt, T. I., J. F. Piatt, B . K . Lance and D . D . Roby (1997). Proximate composition and energy density o f some North Pacific forage fishes. Comparative Biochemistry and Physiology 118A: 1393-1398. Vichnevetskaia, K . D . and D . N . Roy (1999). Oxidative stress and antioxidative defense with an emphasis on plant antioxidants. Environmental Reviews 7: 31-51. Wespestad, V . G . , L . W . Fritz, W . J. Ingraham and B . A . Megrey (2000). On relationships between cannibalism, climate variability, physical transport, and recruitment success 101 of Bering Sea walleye pollock (Theragra chalcogramma). Journal of Marine Science 57: 272-278. Williams, P. E . V . (1995). Digestible amino acids for non-ruminant animals: theory and recent challenges. Animal Feed Science and Technology 53: 173-187. Winship, A . J., A . W. Trites and D . G . Calkins (2001). Growth in body size o f the Steller sea lion (Eumetopias jubatus). Journal o f Mammology 82(2): 500-519. Winship, A . J., A . W. Trites and D . A . S. Rosen (2002). A bioenergetic model for estimating the food requirements o f Steller sea lions Eumetopias jubatus in Alaska, U S A . Marine Ecology Progress Series 229: 291-312. Wu, G . , W. G . Pond, T. Ott and F. W. Bazer (1998). Maternal dietary protein deficiency decreases amino acid concentrations in fetal plasma and allantoic fluid o f pigs. Journal o f Nutrition 128: 894-902. Yuan, Y . , D . Kitts and D . Godin (1996). Heart and red blood cell antioxidant status and plasma lipid levels in the spontaneously hypertensive rat and normotensive Wistar-Kyoto rat. Canadian Journal o f Physiology and Pharmacology 74: 290-297. Yuan, Y . V . and D . D . Kitts (2002). Dietary fat source and cholesterol interactions after plasma lipids and tissue susceptiblity to oxidation in spontaneously hypertensive and normotensive Wistar-Kyoto rats. Molecular and Celluar Biochemistry 232: 33-47. Yuan, Y . V . , D . D . Kitts and D . V . Godin (1999). Influence o f increased saturated fatty acid intake from beef tallow on antioxidant status and plasma lipids in atherosclerosis-susceptible Japanese quail. Nutrition Research 19(3): 461-481. Yuan, Y . V . , D . D . Kitts, T. Nagasawa and S. Nakai (1991). Paracellular calcium absorption, femur mineralization, and biomechanics in rats fed selected dietary proteins. Food Chemistry 39: 125-137. Zenteno-Savin, T., M . A . Castellini, L . D . Rea and B S. Fadely (1997). Plasma haptoglobin levels in threatened Alaskan pinniped populations. Journal of Wildlife Diseases 33(1): 64-71. 102 Appendix I: Summary of Proximate Analysis on Pollock1 Month Moisture % Ash % Lipid % Protein % Energy kcal/g n January 77.44 ±0.180 12.23 + 0.341 14.02 ± 0.450 68.54 ±0.481 4.98 ± 0.033 83 February 73.28 + 0.225 11.42 + 0.450 17.44 + 0.400 63.10± 0.883 5.23 + 0.028 59 March 76.67 ± 0.201 11.68 + 0.405 17.04 ±0.677 62.48 ± 1.077 5.16±0.037 61 July 76.33 + 0.242 11.56 + 0.424 21.81 ±0.735 61.03 ±0.657 5.34 ±0.093 60 August 75.45 + 0.221 9.65 ±0.421 22.10 ±0.624 63.31 + 0.665 5.28 ±0.051 61 September 75.49 ±0.258 12.25 ±0.523 21.43 + 0.761 60.33 + 0.881 5.23 + 0.050 63 Oct-98 73.66 ±0.261 10.11+ 0.291 23.06 ±0.674 59.49 ±0.713 5.55 ±0.036 76 Oct-99 73.79 + 0.298 10.42 ±0.362 20.44 ± 0.689 63.02 + 0.629 5.43 ±0.063 57 November 74.58 ±0.173 10.52 ±0.436 19.54 ±0.656 61.40 ±0.871 5.26 ±0.078 60 F8.571 36.597 5.687 24.516 14.176 10.557 P <0.001 <0.001 <0.001 <0.001 <0.001 1. values are mean + S E M 103 Appendix II: Difference in Proximate Analysis Values between sexes in pollock1 Moisture % Ash % Lipid % Protein % Energy (kcal/g) n January Males 77.62 + 0.237 12.20 + 0.425 13.62 ±0.625 68.48 ±0.762 5.03 ±0.040 46 Females 77.21 ±0.282 12.23 + 0.574 14.57 ±0.662 68.72 ±0.532 4.92 ±0.055 36 February Males 76.56 + 0.303 12.46 ±0.735 17.11 ±0.596 63.27 ± 1.385 5.21 ±0.048 30 Females 75.98 + 0.328 10.35 + 0.440 17.78 ±0.535 62.93 ±1.108 5.25 + 0.030 29 March Males 76.65 ±0.239 12.06 ±0.473 17.49 ± 1.088 61.32 ± 1.556 5.11 ±0.052 33 Females 76.69±0.341 11.23 ±0.683 16.51 ±0.739 63.84± 1.451 5.22±0.049 28 July Males 76.47±0.413 12.31±0.728 22.16±0.828 60.65 ±0.794 5.27±0.100 30 Females 76.19 ±0.259 10.81 ±0.403 21.46 ± 1.226 61.42 ± 1.055 5.41 ±0.156 30 Aug Males 75.28 ±0.361 9.80 ±0.553 22.11 ±0.786 63.18 ±0.862 5.37 ±0.070 30 Females 75.64 ±0.269 9.64 ±0.643 22.13 ± 1.008 63.27 ± 1.043 5.20 ±0.045 30 September Males 75.35 ±0.410 12.35 ±0.808 20.99 ±0.883 61.62 ±0.938 5.22 ±0.072 33 Females 75.65 ±0.306 12.13 ±0.661 21.91 ± 1.279 58.91 ± 1.510 5.24 ±0.069 30 Oct-98 Males 72.27 ±0.250 10.26 ±0.383 22.59 ±0.899 60.13 ±0.960 5.53 ±0.044 30 Females 72.79 ±0.350 9.73 ±0.416 23.99 ± 1.031 58.26 ± 1.044 5.55 ±0.065 44 Oct-99 Males 73.86 ±0.476 10.45 ±0.555 21.48 ± 1.055 61.64 ±0.833 5.54 ±0.121 26 Females 73.7 ±0.382 10.39± 0.486 19.58 ±0.892 64.18 ±0.880 5.35 ±0.054 31 November Males 74.26 ±0.256 11.43 ±0.685 19.84 ±0.810 60.95 ±0.989 5.29 ±0.096 30 Females 74.90 ±0.224 9.61 ±0.498 19.25 ± 1.043 61.84 ± 1.448 5.23 ±0.125 30 SexF1>558 1.309 8.652 0.003 0.212 0.387 p 0.253 0.003 0.957 0.646 0.534 Month F8j558 36.944 5.610 24.139 14.509 10.197 p <0.001 <0.001 O.001 <0.001 O.001 Sex-Month F 8 ) 5 5 8 2.059 0.992 0.723 1.285 1.08 p 0.038 0.441 0.671 0.249 0.375 1. values are mean + S E M 104 Appendix H I : Summary of Morphological Data Collected from Ground Pollock and Herring 1 Weight (g) Length (cm) Girth (cm) Gonad weight (g) GSI 2 n January 524.3 ± 11.10 38.6 ±0 .27 18.3 ±0 .20 25.2 ±2.63 4.51 ±0.377 83 Males 5.23.2 ± 12.06 38.6 ±0.28 18.4 + 0.17 18.7 ± 1.37 3.51 ±0.234 46 Females 530.6 ±20.01 38.8 ±0 .49 18.1 ±0.41 34.3 ±5 .5 5.90 ±0.754 36 February 670.0 ± 14.87 41.5 ±0 .30 20.2 ±0.18 44.5 ±4 .48 6.37 ±0.569 59 Males 643.9 ± 15.28 40.9 ± 0.42 19.9 ±0 .21 24.6 ± 2.76 3.67 ±0.354 30 Females 697.1 ±25.12 42.1 ±0 .39 20.4 ±0 .31 64.8 ±6 .87 9.16 ±0.826 29 March 612.9 ± 13.24 40.7 ±0 .29 19.0 ± 0.15 38.0 ±2.98 6.02 ± 0.403 61 Males 597.4 ± 14.05 40.5 ±0 .33 18.9 ± 0.18 25.7 ± 1.55 4.31 ±0.244 33 Females 631.1 ±23.46 41.0 ±0 .50 19.2 ±0.25 52.5 ±5.03 8.04 ±0.653 28 July 527.6 ± 22.20 39.5 ±0 .61 18.4 ±0.23 6.2 ± 0.83 0.96 ±0.119 60 Males 490.7 ± 26.96 38.6 ±0 .87 18.2 ±0 .32 3.1 ±0.64 0.46 ±0.102 30 Females 564.5 ± 34.42 40.3 ±0.85 18.7 ±0.32 8.5 ± 1.20 1.46 ±0.173 30 Aug Males Females 266.4 ± 6.72 272.6 ±7 .90 261.8+ 11.08 31 .9±0.30 32.5 ±0.38 31.5 + 0.46 15.0 ±0 .17 15.1 ±0 .20 14.9 + 0.29 2.1 ±0 .18 2.6 ±0 .31 ' 1.8 + 0.20 0.63 ±0.059 0.65 ±0.104 0.64 + 0.055 61 30 30 September Males Females 678.6 ±23.10 674.2 ± 29.75 683.5 + 36.33 42.2 ±0 .48 42.1 ±0.55 42.3 + 0.82 19.7 ±0.24 19.7 ±0.33 19.7 + 0.34 16.6 ±2 .30 13.7 ±2 .00 19.8 + 4.27 2.34 ±0.301 1.94 ±0.213 2.78 + 0.582 63 33 30 Oct-98 Males Females 658.9 ± 19.11 643.5 + 19.75 696.0 + 37.10 40.8 ± 0.40 40.4 ± 0.42 41.6 + 0.75 19.6 ±0 .20 19.5 ±0.24 19.9 + 0.34 13.7 ± 1.41 12.9 ± 1.58 14.7 + 2.63 1.94 ±0.195 1.93 ±0.218 2.08 + 0.370 76 30 44 Oct-99 Males Females 917.2 + 23.32 876.4 ± 30.92 951.5 + 33.36 47.5 ±0.55 46.5 ± 1.01 48.4 + 0.55 22.5 ± 0.25 22.0 ±0.33 23.0 + 0.34 22.7 ±3 .11 17.9 ±2 .30 26.7 + 5.24 2.42 ±0.357 1.91 ±0.247 2.85 + 0.318 57 26 31 November Males Females 709.0 ±24.37 728.6 ±37.77 689.5 + 31.05 44.1 ±0.54 43.9 ±0 .81 44.3 + 0.73 21.2 ±0.31 21.2 ±0 .41 21.1 + 0.46 35.3 ±5 .57 29.9 ±4 .96 40.7 + 10.22 4.67 ±0.772 3.77 ±0.520 5.56+ 1.449 60 30 30 Herring Males Females 43.2 ± 1.19 42.4 ± 1.27 43.9 + 2.04 14.8 ±0 .13 14.7 ±0.13 14.8 + 0.23 8.0 ±0.09 8.0 ± 0.09 8.0 + 0.17 2.3 ±0 .23 3.6 ±0.28 1.3 + 0.23 5.23 ±0.487 8.36 ±0.521 2.49 + 0.385 68 33 32 1. values are mean ± S E M 2.Gonadosomatic index = gonad weight / body weight xlOO "@en ; edm:hasType "Thesis/Dissertation"@en ; vivo:dateIssued "2002-11"@en ; edm:isShownAt "10.14288/1.0074850"@en ; dcterms:language "eng"@en ; ns0:degreeDiscipline "Food Science"@en ; edm:provider "Vancouver : University of British Columbia Library"@en ; dcterms:publisher "University of British Columbia"@en ; dcterms:rights "For non-commercial purposes only, such as research, private study and education. Additional conditions apply, see Terms of Use https://open.library.ubc.ca/terms_of_use."@en ; ns0:scholarLevel "Graduate"@en ; dcterms:title "Seasonal variation in nutrient composition of Alaskan walleye pollock (Theragra chalcogramma) and its effect on the nutritional status of Steller sea lions (Eumetopias jubatus)"@en ; dcterms:type "Text"@en ; ns0:identifierURI "http://hdl.handle.net/2429/12322"@en .