As described above, an important role of lipids is as a source of energy-rich fatty acids. Dietary lipids are a major provider of energy in all fish, especially carnivorous species, which includes most marine fish, where carbohydrate plays little role as an energy source due to its low abundance in natural diets. In contrast, herbivorous/omnivorous species such as carp (Cyprinus carpio) consume plant material containing easily digestible sugars and also complex indigestible polysaccharides such as cellulose, chitin, and lignin, which the fish cannot digest easily without the assistance of specific gut flora (Watanabe, 1982; Smith, 1989). However, the extent to which fish in general, particularly marine species, can utilize dietary carbohydrate is debatable.
One possible result of adaptation to an almost totally carnivorous habit is that marine fish in general may have a limited ability to digest and utilize dietary carbohydrate (Smith, 1989), in which case the energy requirements of most marine fish will be met primarily by dietary lipids and protein. Thus, the diets formulated for the aquaculture industry contain predominantly protein and lipid, with small amounts of vitamins and minerals. Easily digestible, water- soluble mono- or disaccharides such as glucose and fructose are obviously not an option in fish diets. However, for many species, polysaccharides such as starch and dextrin can be included in formulated diets at levels of up to approximately 10% (and higher in some species), although digestibility normally has to be improved by processing or cooking the ingredients prior to inclusion in the diet (Smith, 1989).
As protein is generally the most expensive of the basal components, the major issue regarding the dietary formulations for fish has been to supply the minimum protein requirements for optimal or maximal growth with an appropriate balance of other nutrients to supply the required energy. This balance is termed the ratio of digestible protein:digestible energy (DP:DE).
Some protein is inevitably utilized for energy through direct oxidation of amino acids via the tricarboxylic cycle, or after conversion of amino acids to glucose via gluconeogenesis, but these processes can be minimized by in- cluding an abundance of energy-providing nutrients in the diet. Lipid as the primary energy-providing nutrient in fish diets has a major protein-sparing effect in many fish species (Wilson, 1989; Sargent et al., 1989). Because of the metabolic interactions among protein, lipid, and carbohydrate de- scribed above, definition of the exact dietary lipid requirements for fish is not particularly meaningful. However, it has long been known that, in gen- eral, lipid in the range of 10–20% of the dry weight of the diet is sufficient to allow protein to be effectively utilized for fish growth without depositing excessive lipid in the fish tissues (Cowey and Sargent, 1979; Watanabe, 1982;
Sargentet al.,1989). Nonetheless, the precise amount of lipid required de- pends on the dietary protein level and, in some cases such as carp, also on the dietary carbohydrate level (Watanabe, 1982; Sargent et al.,1989).
More detailed accounts of the nutritional energetics of fish and the role of protein as an energy source and its interaction with other dietary compo- nents, including lipids, are provided elsewhere in this book. Therefore, this section focuses primarily on the use of so-called “high-energy” diets, which have become increasingly widespread in aquaculture. These diets are, in reality, “high-oil/fat” diets. They can have consequences for the farmed fish by altering its lipid and fatty acid metabolism, with implications for its health and welfare and, also, for the consumer in terms of the quality and taste of the end product (Bellestrazzi and Mion, 1993) and its health-promoting properties (Sargent and Tacon, 1999).
4.4.1. High-Energy (Fat) Diets
High-energy (fat) diets are formulated to increase the growth perfor- mance of the fish for a given amount of feed by maximally exploiting the protein-sparing effect of high-energy lipid and allowing as much of the dietary protein as possible to be converted into muscle protein. Al- though protein sparing by dietary lipid is well documented, the limits to its effectiveness have not been accurately defined for any fish species so that recent dietary formulations have tended to continue the upward trend in dietary lipid. In many cases this has successfully increased weight gains. How- ever, the use of increasingly high-lipid diets is not without controversy, and
it is debatable whether the protein-sparing effect alone accounts for all of the weight gains observed.
A problem in considering the effects of high-energy diets in aquaculture is that what precisely constitutes a high-energy or high-fat diet is all too often undefined. Few data are available in the literature, and because commercial pressure to increase growth rates and reduce production times is the main rationale behind high-energy diets, most information concerns studies on the four main species farmed in Europe: Atlantic salmon, rainbow trout, gilthead sea bream, and sea bass (Dicentrarchus labrax). In rainbow trout, dietary lipid at 21% increased growth over that of fish consuming diets with 8 and 11% dietary lipid (Luzzanaet al.,1994). In brown trout (Salmo trutta) a diet containing 29% lipid increased the growth rate compared to a diet containing 21% lipid (Arzel et al., 1993). In Atlantic salmon, even higher lipid levels have been investigated and diets containing 47 and 38%
lipid both increased growth in comparison to a diet containing 31% lipid (Hemre and Sandnes, 1999). In a recent study with sea bass, a diet of 19%
lipid showed increased growth compared with diets containing 11 and 15%
dietary lipid (Lanariet al.,1999). A limit to the growth-promoting effect of high dietary lipid was indicated in the study on sea bass by Peres and Oliva- Teles (1999) with sea bass, where the growth rate was higher in fish on a diet containing 24% lipid compared to fish on diets with lower (12 and 18%) and higher (30%) lipid levels. An indication that the growth-promoting effects of dietary lipid may vary with the stage of development is the observation that increasing the dietary lipid from 12 to 20% did not alter the growth of sea bass larvae (Salhiet al.,1994). However in sea bass fingerlings, growth and protein sparing were both improved by increasing the dietary lipid from 9 to 15% (Vergaraet al.,1996). The importance of optimising ration size when feeding high-energy diets to avoid unwanted adiposity was also investigated in sea bream (Companyet al.,1999).
4.4.2. Dietary Lipid Levels and Tissue Lipid Levels
It has long been known that there is a strong relationship between the dietary lipid levels and the levels of lipid in the carcass of fish (Cowey, 1993).
Deposition of excess lipid in the carcass will clearly be a more serious prob- lem in those species that tend to store lipid in the flesh, although it ap- pears that flesh lipid levels are increased in most species when fed high-fat diets. Many recent studies have shown that a potential and, perhaps, detri- mental effect of high-fat diets is the deposition of excess lipid in the flesh.
Thus, feeding diets with high lipid levels has been shown to increase flesh lipid levels in freshwater fish, including catfish (Stowell and Gatlin, 1992;
Anwar and Jafri, 1995) and silver perch (Leiopotherapon bidyanus) (Anderson
and Arthington, 1989), in marine fish such as turbot (Stephanet al.,1996), Atlantic mackerel (Scomber scombrus)(Hemreet al.,1997), sea bass (Catacutan and Coloso, 1995), and red drum (Sciaenops ocellatus) (Williams and Robin- son, 1988), and in salmonids including brown trout (Arzel et al., 1993), rainbow trout (Luzzanaet al.,1994; Weatherupet al.,1997; Diaset al.,1999), chinook salmon (Oncorhychus tshawytscha) (Silveret al.,1993), and Atlantic salmon (Bellet al.,1998; Hemre and Sandnes, 1999).
High-energy feeds have recently been used widely in the salmon indus- try, where the upper level for dietary lipid in the 1970s was about 20%, whereas in the late 1990s this level had almost doubled, with Einan and Roem (1997) suggesting a diet with a lipid level of 35% to be optimal for growth of Atlantic salmon. However, the Atlantic salmon industry, particu- larly the processing and retailing sectors, now recognizes the problem of deposition of excessive dietary lipid in the flesh as a major issue relating to carcass and product quality (Hillestadet al.,1998). Increased flesh lipid in salmon can lead to problems of so-called “free oil,” reduced visualization of carotenoid pigment, and subsequent consumer and processor resistance to the oily texture, poor smoking performance, and potential rancidity prob- lems (Bell et al.,1998; Refsgaard et al., 1998; Hillestad et al.,1998). It is particularly interesting that, in salmon consuming a diet of a given oil con- tent, the level of lipid in the flesh can vary considerably between populations and between individuals in the same population (Bellet al.,1998; Refsgaard et al.,1998). For example, it was observed that the lipid levels in individual salmon receiving a diet containing 28% lipid could vary from less than 5%
to more than 18% of the wet weight of the flesh (Bellet al.,1998). It is not clear to what extent environmental and/or genetic factors contribute to the observed biological variation (Olsen and Skjervold, 1995; Bellet al.,1998).
4.4.3. Dietary Lipid Levels and Essential Fatty Acids As well as providing energy, dietary lipid is also important as the source of essential fatty acids (EFA). The precise nature of the EFA and their absolute dietary requirements vary with species and are described in Section 4.5.
However, it is also apparent that the quantitative requirement for EFA may vary with the total dietary lipid level, and this may also vary with the stage of development (Izquierdo, 1996). For instance, the requirement for n-3 HUFA appeared to increase as the level of lipid in the diet increased in both red sea bream (Pagrus major) fingerlings (Takeuchi et al.,1992a) and yellowtail (Seriola quinqueradiata) fingerlings (Takeuchiet al.,1992b).
However, in contrast, there was no apparent variation in then-3 HUFA re- quirement of larval gilthead sea bream as the dietary lipid level increased from 12 to 20% (Salhiet al.,1994).
4.4.4. Dietary Lipid Levels and Antioxidant Requirements Feeding very high levels of unsaturated lipid can increase the prooxidant stress on fish consuming the diets. Thus, as the lipid content of the diet in- creases, the dietaryn-3 HUFA levels also increase and the resulting increased unsaturation index of the diet must be balanced by an increasing dietary antioxidant content, especially vitamin E (tocopherol). It was shown in At- lantic salmon that, as the dietary lipid increased, vitamin E levels in the flesh decreased (Hemre and Sandnes, 1999). It has also been shown in tilapia that the vitamin E requirement increased as the level of dietary lipid increased (Satoh et al.,1987). For instance, in blue tilapia (Oreochromis aureus) the vitamin E requirement at 3% dietary lipid was 10 mg/kg diet, whereas at 6% dietary lipid the vitamin E requirement was 25 mg/kg, when both diets contained 120 mg/kg butylated hydroxyanisole (BHA) (Roemet al.,1990).
The flesh vitamin E level has been shown to influence the development of rancid taste after slaughter in both Atlantic salmon (Waagboet al.,1993) and cultured mackerel (Hemreet al.,1997). Muscle homogenates from rainbow trout and sea bass fed high-fat diets were more susceptible to lipid peroxida- tion than fish fed low-fat diets (Diaset al.,1999). The relationship between dietary lipid and fatty acid contents and dietary antioxidant requirements is discussed more fully in Section 4.7.
4.4.5. Dietary Lipid Levels, Lipogenesis, and Fatty Livers The level of dietary lipid has other effects on lipid metabolism in fish including modulation of lipogenesis. It has long been established that in- creased dietary lipid levels depressde novofatty acid synthesis through inhi- bition of several enzymes involved in hepatic lipogenesis, including acetyl coenzyme A carboxylase, fatty acid synthetase, and NADPH-generating en- zymes such as glucose-6-phosphate dehydrogenase and 6-phosphogluconate dehydrogenase (Sargentet al.,1989). The perceived view was that, in com- parison to mammals, where dietary lipid as low as 2.5% can cause inhibition of lipogenesis, inhibition of lipogenesis in fish was obvious only with diets containing in excess of 10% lipid (Sargent et al., 1989). With almost all species a diet would not be considered “high energy” with a lipid level lower than 10%, so it can safely be assumed that high-energy diets inhibit lipogen- esis in fish. For instance, in hepatopancreas from carp fed a diet containing 18% lipid, there were decreased levels of lipogenic, gluconeogenic, and amino acid-degrading enzyme activities in comparison with fish fed diets containing 9% lipid (Shimenoet al.,1995). The activities of various hepatic enzymes including fatty acid synthetase, glucose-6-phosphate dehydroge- nase, and malic enzyme were reduced by decreasing the ratio of digestible
protein to digestible energy (DP/DE) through an increase in dietary lipid in rainbow trout and sea bass (Diaset al.,1999). These enzyme activities were also decreased in yellowtail liver (Shimenoet al.,1996) and carp hepatopan- creas (Kheyyaliet al.,1989). Lipogenesis was also decreased in rainbow trout liver in response to increased dietary ratios of lipid to carbohydrate (Corraze et al.,1993a; Braugeet al.,1995). At an enzyme kinetic level, a high-fat diet in rainbow trout was shown to cause a significant decrease in the specific activity, catalytic efficiency, andVmax of hepatic glucose-6-phosphate dehy- drogenase, whereas theKmwas unaffected (Sanchez-Muroset al.,1996).
High-fat diets have also been shown to affect liver histology in ways con- sistent with the development of fatty liver pathology. Thus, increased fatty infiltration of liver has been reported in sunshine bass (Morone chrysops× M. saxatilis) fed a diet containing 16% lipid in comparison with a diet con- taining 13% lipid (Gallagher, 1996). Compared to fish fed diets containing 15 and 22% lipid, sea bream fed diets containing 27% lipid showed foci of swollen hepatocytes characterized by displaced nuclei and large lipid droplets in the cytoplasm (Caballeroet al.,1999).
4.5