The supply of yellow perch (Perca flavescens) from the Great Lakes has decreased, but demand for perch remains strong (Cox 1995, Engle et al. 1990). Profitable culture of yellow perch, however, is dependent upon a practical diet formulated to meet the specific nutritional requirements of this species. However, few nutritional studies have been conducted with yellow perch, and diets are not standardized.

Brown et al. (1996) reported the results of a series of studies in which yellow perch were fed several practical and experimental diets. Among the fish fed practical diets, weight gain and feed efficiency were higher in fish fed diets formulated to meet the requirements of rainbow trout (Oncorhynchus mykiss) than in fish fed diets formulated to meet the requirements of channel catfish (Ictalurus punctatus). Practical recommendations from that study were 36 g crude protein and 10-15 g lipid/100 g diet.

Brown et al. (1996) also fed several experimental diets to juvenile yellow perch in their studies. One of those diets was a purified diet containing 35 g crude protein/100 g diet (25 g crude protein/100 g diet supplied by crystalline L-amino acids and 10 g crude protein/100 g diet provided by casein and gelatin). That formulation was similar to one used for nutritional studies with hybrid striped bass (Morone chrysops × M. saxatilis) in our laboratory (Brown et al. 1993, Griffin et al. 1992, 1994a, 1994b and 1994c). In the study with yellow perch, fish fed the purified diet exhibited weight gain and feed efficiency that were not significantly different from those of fish fed the best commercial diet (BioDiet Soft-Moist Grower, Bioproducts, Warrenton, OR) (Brown et al. 1996). Therefore, a purified diet, with a majority of crude protein supplied by crystalline L-amino acids, was considered appropriate for quantifying the essential amino acid requirements for this newly cultured species.

The dietary arginine requirement of several species of fish has been established. The requirements range from a low of 1.0 g/100 g diet for channel catfish (Robinson et al. 1981) to a high of 2.8 g/100 g diet for rainbow trout (Ketola 1983). However, optimal dietary crude protein concentrations vary widely for fish, and this affects the comparison. If we compare dietary arginine requirements as a function of dietary crude protein, the values for fish range from 2.6 g/100 g crude protein for rainbow trout (Kaushik et al. 1988) to 6.0 g/100 g crude protein for chinook salmon (Klein and Halver 1970) and chum salmon (Akiyama 1987, cited in NRC 1993). With this wide range of reported requirements, extrapolation of nutritional requirements from established to new aquaculture species seems unfeasible.

Establishment of dietary essential amino acid requirements for animals is often confounded by the various response criteria used to establish the requirement. In many studies with fish, growth and feed efficiency were the responses used to establish essential amino acid requirements (Griffin et al. 1994b, Santiago and Lovell 1988), and serum free amino acid concentrations have been used to verify certain requirements. However, essential amino acid concentrations in serum of fish are not always responsive to varying concentrations in the diet. Arginine has been notable in this case (Kaushik et al. 1988, Robinson et al. 1981). Urea in plasma or serum has been more responsive than free arginine concentrations (Cho et al. 1992, Tibaldi et al. 1994). Also, 6-h postprandial plasma arginine concentrations from fish deprived of food for 3 d and then offered one meal have been used (Cho et al. 1992). Other criteria used in conjunction with weight gain include free arginine concentrations in liver and muscle, oxidation of L-[U-¹⁴C]arginine, urinary excretion of arginine catabolites (Kaushik et al. 1988), and nitrogen and lipid retention (Cho et al. 1992). However, biochemical responses of fish fed graded levels of arginine have not indicated a requirement different from that indicated by weight gain and feed efficiency data in the same study, only confirmation of the growth data. Thus, the best response for determining dietary essential amino acid requirements for growing animals seems to be weight gain (Baker 1983).

Table 1. Composition of basal diet fed to juvenile yellow perch [View Table]

None of the essential amino acid requirements of yellow perch have been reported. The objective of this study was to determine the dietary arginine requirement of juvenile yellow perch fed purified diets.

MATERIALS AND METHODS

A closed recirculating system was used in this experiment. The system was the same as described previously for hybrid striped bass (Griffin et al. 1992). Solid material removal and denitrification were accomplished with two submerged filtration tanks. Water was pumped from a sand filter to each aquarium at a rate of ~1 L/min. The temperature of the system was maintained at 22 ± 1°C throughout the experiment. Water was added to the system each day to compensate for evaporative losses. The light:dark cycle remained at 16 h light:8 h dark throughout the experiment.

Groups of 15 randomly chosen fish were placed in each aquarium. Means for initial body weight ranged from 10.6 to 11.0 g/fish. Fish were acclimated to their new surroundings and their respective diets during the initial 2 wk of the experiment. Dietary treatments were randomly assigned to triplicate aquaria. All fish were weighed every 14 d to adjust the food allotment, which was 3 g/(100 g body wt·d) offered in two equal meals. That amount is near satiation for juvenile perch maintained at 22°C and has been used in previous studies in our laboratory (Brown et al. 1996). All fish were fed their respective diets for 8 wk.

Water quality was monitored daily and did not appear to affect the results of the study. Dissolved oxygen concentrations were not below 7.0 mg/L, ammonia-nitrogen concentrations were not greater than 0.3 mg/L, and nitrite-nitrogen was not greater than 0.2 mg/L at any point during the study.

Table 2. Amino acid composition of basal diet fed to juvenile yellow perch1 [View Table]

Lipid (menhaden oil) and reagent-grade minerals were obtained from commercial suppliers (Zapata-Haynie, Reedville, VA, and Sigma Chemical, St. Louis, MO, respectively). Carboxymethylcellulose, casein, gelatin, cellulose, vitamins and crystalline L-amino acids were supplied by U.S. Biochemical (Cleveland, OH). Vitamins and minerals were added to the diets as nutritionally complete premixes (Griffin et al. 1992).

The basal diet contained 12 g menhaden oil/100 g diet and 20 g carbohydrate (dextrin)/100 g diet. L-Arginine-HCl was added to the basal diet in gradations of 0.2 g/100 g diet, resulting in eight dietary treatments ranging from 0.44 g to 1.84 g/100 g diet. Diets were made isonitrogenous with glutamic acid. Crystalline L-arginine-HCl and L-glutamic acid provided a total of 1.84 g crude protein/100 g diet.

The dry ingredients were initially mixed in a twin-shell V-mixer (Patterson-Kelly, East Stroudsburg, PA). Water and lipid were added and mixed with the dry ingredients following transfer to a Hobart mixer (Hobart Corp., Troy, OH). Diets were then adjusted to pH 7.0 ± 0.2 with saturated NaOH (Wilson et al. 1977) and pelleted. The diets were dried for 24 h at 60°C in a forced-air oven. All diets were stored under air-tight conditions at 20°C until needed.

RESULTS

Table 3. Initial weight, weight gain, feed efficiency and survival of juvenile yellow perch fed graded levels of L-arginine-HCl1 [View Table]

In the latter stages of the study, consumption of feed by groups of fish fed diets deficient in L-arginine-HCl was lower than for fish fed diets with sufficient L-arginine-HCl. However, consumption of feeds during the early stages of the study was similar across treatments. This situation worsened as the study progressed and likely contributed to the lower weight gain observed in fish fed deficient arginine concentrations.

DISCUSSION

A dietary arginine requirement of 4.3 g/100 g dietary protein is similar to values for other species of fish including the Nile tilapia (4.2 g/100 g dietary protein, Santiago and Lovell 1988), channel catfish (4.3 g/100 g dietary protein, Robinson et al. 1981), common carp (4.3 g/100 g crude protein, Nose 1979), Japanese eel (4.5 g/100 g crude protein, Nose 1979), Mossambique tilapia (4.0 g/100 g crude protein, Jackson and Capper 1982), sea bass (3.9 g/100 g crude protein, Tibaldi et al. 1994) and hybrid striped bass (4.0 g/100 g crude protein, Griffin et al. 1994b). Dietary arginine requirements for other species of fish range from 5.0 to 6.0 g/100 g crude protein. Those include the chinook salmon and coho salmon (6.0 and 5.8 g/100 g crude protein, respectively, Klein and Halver 1970), chum salmon (6.0 g/100 g crude protein, NRC 1993) and gilthead sea bream (5.0 g/100 g crude protein, Luquet and Sabaut 1974). Most of the dietary arginine requirements for rainbow trout are between 2.6 and 4.7 g/100 g crude protein (Cho et al. 1992, Kaushik 1979, Kaushik et al. 1988, Kim et al. 1983, Walton et al. 1986), with one value of 5.9 g/100 g crude protein (Ketola 1983). The similar dietary arginine requirements for fish are surprising, given that food habits of the species studied range from omnivorous to strictly carnivorous, salinity tolerances range from strictly freshwater species to species tolerant of both fresh and salt water, and thermal preferences range from strictly coldwater to strictly warmwater. Several of these factors, and others, influence urea cycle activity in fish (Chiu et al. 1986, Randall et al. 1989, Saha and Ratha 1994) and could affect the dietary arginine requirement. Some freshwater teleostean fish do not have a functional urea cycle (Mommsen and Walsh 1989), whereas others clearly synthesize urea in response to both dietary and environmental changes (Chiu et al. 1986, Depeche et al. 1979, Kaushik and Fauconneau 1984, Randall et al. 1989, Saha and Ratha 1994, Tibaldi et al., 1994). There are no clear distinctions in food habits or environmental requirements that allow a clear classification in dietary arginine requirements among species of fish (Morris 1985).

The dietary arginine requirements among species of fish examined thus far are more similar to each other than are those of mammals or birds. Most dietary arginine requirements for mammals and birds are lower than those determined for fish when expressed as a concentration in the dry diet (NRC 1978 and 1994); however, they are similar to values for fish when expressed as a concentration of crude protein. Dietary arginine requirements for the ureotelic rodents range from 2.4 to 5.1 g/100 g protein (NRC 1978), and values for swine range from 0.8 to 2.5 g/100 g crude protein (NRC 1988). Dietary arginine requirements for uricotelic birds range from 4.4 to 6.3 g/100 g protein (NRC 1994). Most dietary arginine requirement values for fish seem to be between those for ureotelic mammals (lower) and those for uricotelic birds (higher).

Chicks fed excess lysine exhibited a lysine-arginine antagonism (Jones 1964, Jones et al. 1967); however, pigs apparently do not exhibit the same antagonism (Baker 1987). Robinson et al. (1981) were unable to demonstrate a similar response in channel catfish fed lysine or arginine in excess of dietary requirements. Fish that were fed diets with four times the dietary requirement of lysine had growth and feed efficiency similar to those of fish fed diets containing lysine at the required level. Similarly, weight gain and feed efficiency of hybrid striped bass were unaffected by lysine or arginine provided at 2.5 times the required level (Griffin et al. 1994b). However, Kaushik et al. (1988) demonstrated that rainbow trout fed high levels of dietary arginine required a concomitant increase in dietary lysine. The basal diet in the present study was formulated to provide 2.68 g lysine/100 g diet. A preliminary study in our laboratory indicated the dietary lysine requirement of juvenile yellow perch was no less than 1.5 g lysine/100 g diet. A dietary lysine-arginine antagonism may have affected our findings, but that possibility seems remote.

Aggressive feeding behavior and acceptable feed conversion observed in yellow perch during this study indicate that a purified diet remains an appropriate choice for determining nutritional requirements. Furthermore, the palatability of a similar purified diet in 5-g and 51-g yellow perch was previously reported (Brown et al. 1996). The acceptance of the diet by juvenile yellow perch in that investigation, as well as in the present study, is somewhat surprising when one considers the difficulty involved in training yellow perch larvae to accept formulated feeds (Hale and Carlson 1972, Heidinger and Kayes 1986, Malison and Held 1992). However, the results of this study indicate that after juvenile yellow perch are trained to accept commercial feeds, they can be easily converted to a purified diet. Therefore, purified diets should be useful in further nutritional requirement studies, which will be necessary for the continued development of diets for yellow perch.