Growth, fat content and fatty acid profile of South American catfish, surubim
(Pseudoplatystoma fasciatum) juveniles fed live, commercial and formulated diets
By M. Arslan1,2, K. Dabrowski1and M. C. Portella31School of Environment and Natural Resources, The Ohio State University, Columbus, OH, USA;2Department of Fisheries and
Aquaculture, Ispir H. Polat Vocational School, Ataturk University, Ispir, Erzurum, Turkey;3Aquaculture Center, Sao Paulo State
University, Jaboticabal, SP, Brazil
Summary
South American catfish, barred surubim (Pseudoplatystoma fasciatum) juveniles (117.6 ± 11.8 mg individual weight; 28.3 ± 2.5 mm total length) were fed various diets: one live (Tubifex worms), two commercial (Aglo Norse and Bio Kyowa), and one semi-purified formulated diet (75% peptide based protein) over a 2-week period. Fish fed the Aglo Norse diet showed the highest growth performance, but cannibalism also was very high (42%). Fish fed peptide based formulated diet demonstrated the lowest growth rate, with no cannibalism. The highest survival was achieved with fish fed Tubifex worms (100%). Lipid level in the whole body of the fish fed four different experimental diets did not differ significantly, aver-aging 3.6 ± 0.7%. Fatty acid composition of neutral and phospholipid fractions of whole body lipids of fish reflected the fatty acid composition of the diets. The high level of 20:4n-6 in Tubifexworms resulted in a high level of this fatty acid in the tissue of fish fed this diet. It remains uncertain how high survival and no cannibalism is related to dietary lipids⁄ fatty acids. In all cases, the increasing ratio of n-3 HUFA (highly unsaturated fatty acids)⁄ n-6 HUFA in phospholipid fractions suggested the elongation and desaturation of 18:3n-3 to 22:6n-3 via 20:5n-22:6n-3. Moreover, in respect to the 20:4n-6 levels in the diets, an increase in the concentration of this fatty acid in phospholipid fraction suggests that South American catfish can transform linoleate into arachidonate.
Introduction
South American catfishes (surubim) of the genus Pseudopla-tystomaare important fish for commercial production in Peru, Brazil and Colombia. Excellent flesh quality with very few inter-muscular bones and a high growth rate with the potential to reach 50 kg (Sato et al., 1998) may provide further incentives to raise this species. However, very little information on the nutritional requirements of these species is available, and this situation seems to limit the culture of surubims (Martino et al., 2002a, 2002b). More specifically, limited information is avail-able on larval rearing of these species (Kossowski, 1996). In our laboratory, we recently carried out several experiments on sur-ubim, testing several live, commercial and semi-purified casein-gelatine based formulated diets in early life stages of surubim. Cannibalism was determined to be the most important short-coming in rearing of larval and juvenile surubim (K. Dabrowski, M. Arslan and M. C. Portello, unpublished data).
Intensification of aquaculture requires high quality (viabil-ity) fish juveniles and, in most of the species, juvenile fish
production depends on rearing fish larvae at high densities (Izquierdo et al., 2000), which frequently results in high cannibalism as in the case of surubim. The success of larval and juvenile rearing of teleost fish depends on meeting the nutritional requirements of these ontogenetic stages of fish via adequate, and attractive, food sources. The use of formulated diets as the first exogenous diet for fish larvae frequently results in limited success (Dabrowski, 1986). Lucini and Avendan˜o-Salas (1984) reported successful rearing of South American catfish (Rhamdia sapo) that had an acceptable survival (64%) and growth rates when fed a homogenate mixture of equal proportions of egg white, coagulated blood and liver. In larval and juvenile rearing of catfishes, artificial diets usually led to poor growth and high mortality, including high frequency of cannibalism (Fermin and Bolivar, 1991; Kerdchuen and Legendre, 1994; Hung et al., 1999; Evangelista et al., 2005). It was also agreed that photoperiod or light regime was one of the most important factors in larval and juvenile rearing of catfishes (Piaia et al., 1999). Low light or constant dark environment resulted in high growth and lowered stress and aggressive behavior in comparison to higher light periods or continuous light (Britz and Pienaar, 1992; Hung et al., 2002; Almaza´n-Rueda et al., 2004, 2005).
In the present study, we compared growth, survival, lipid contents and fatty acid profiles of fish fed two commercial diets and one formulated peptide-based diet (Terjesen et al., 2006) against those fed live Tubifex worms as a reference diet. The fish were maintained in darkness in order to avoid stress, aggressiveness and cannibalism.
Materials and methods
Fish origin and experimental design
Larvae of surubim were hatched in Brazil and shipped from the Aquaculture Center, Sao Paulo State University, Jaboticabal to Columbus overnight when surubim were 2 days old. Upon arrival, significant mortality was observed and signs of cannibalism were already recorded. Fish were transferred to a recirculated system maintained at 24.6C. Fish were offered freshly-hatched Artemia nauplii for several days. They were then divided into 24 tanks each of 40 L volume, at densities of 30 fish per tank and offered live, commercial and formulated diets for an experimental period of 10 days prior to the present study. At the completion of this preliminary weaning exper-iment, fish were divided into four class sizes and all offered Artemia nauplii to bring their size to as uniform weight as possible. The present experiment started 7 days later and
2008 The Authors
Journal compilation 2008 Blackwell Verlag, Berlin ISSN 0175–8659
Accepted: May 7, 2008 doi: 10.1111/j.1439-0426.2008.01154.x
continued for 2 weeks. The fish (117.6 ± 11.8 mg weight; 28.3 ± 2.5 mm total length), at 15 individuals per aquarium, in triplicate per treatment, were offered one live, one formu-lated and two commercial diets. The fish were maintained in constant darkness to decrease stress and secure feeding activity.
Experimental diets
A diet containing 75% synthetic dipeptides (PT) (Terjesen et al., 2006) was used as the formulated diet along with two commercial diets, a larval marine fish diet (Aglo Norse Larva Feed; EWOS, Norway, particle size 100–500 lm) (AN) and the Bio Kyowa (B700; Kyowa Hakko Co., Tokyo, Japan) (BK), and one live food which was a diet of Tubifex worms (TB). Tubifex worms were cut into approximately 0.5–1 mm prior to the feeding. Lipid levels were 33.1, 27.7, 21.3 and 16.0% (dry weight) in TB, AN, BK, and PT diets, respectively.
Feeding trial and sampling procedure
Ad libitumfeeding was implemented during the first week and feeds were adjusted at 5% biomass in the second week of the experiment. Fish were counted at one-week intervals and mortality was recorded daily. The difference in fish numbers between each count, subtracting natural mortality, was equated with cannibalism. Biomass was determined at a one-week interval for each aquarium. During the calculation of growth parameters, all cannibalized fish were added to the total number of fish at each count to eliminate the contribution of cannibalism to the growth efficiency estimation. In other words, cannibalized fish were included in total biomass. At the end of the experiment, four fish per tank were anesthetized in ice slurry and then frozen on dry ice and stored at)80C for lipid and fatty acid analyses. Four fish sampled from each tank were combined for the analysis and considered as one replicate (three samples per dietary treatment).
Lipid and fatty acid analyses
Dietary and whole body lipids were extracted according to the procedure of Folch et al. (1957). Whole body lipids were then separated into polar (phospholipids) and neutral (mostly triglycerides) lipids using Sep-Pak silica cartridges (Waters, Milford, MA). Chloroform and methanol were used as the mobile phases for neutral and phospholipids, respectively (Juaneda and Rocquelin, 1985). Fatty acid methyl esters (FAME) were prepared according to Metcalfe and Schmitz (1961) and analyzed as described previously (Czesny and Dabrowski, 1998). The FAMEs obtained were separated by gas chromatography (Varian 3900 GC; Varian, Inc., Walnut Creek, CA) equipped with a flame ionization detector, a capillary column (WCOT fused silica 100 m· 0.25 mm coat-ing CPSIL 88 for FAME, DF = 0.2) and an auto-injector (CP-8410 Autoinjector; Varian, Inc.). Helium was used as a carrier gas at a flow rate of 30 ml min)1. The injector and detector temperatures were 270 and 300C, respectively. Initial temperature of the oven was 175C for 26 min, which increased to 205C at increments of 2C min)1, then held at 205C for 24 min. The individual fatty acids were identified by comparing their retention times to that of a standard mix of fatty acids (Nu-Check-Prep, Inc., Elysian, MN). Prior to transmethylation, nonadecanoic acid (C19:0) was added to the samples (0.8 mg per 50 mg of lipid) as the reference standard
for the quantification of fatty acids. Fatty acids were expressed as percentage of total detected.
Statistical analysis
Results were expressed as mean ± SD (n = 3). Each exper-imental unit (aquarium) was considered a replicate. Data were subjected to analysis of variance (ANOVAANOVA) and subsequent
comparison of means by DuncanÕs multiple range test. Percentage data were arc-sin transformed prior to statistical analysis. Differences were considered statistically significant at P < 0.05.
Results
Growth performance and cannibalism
Already 1 week into the feeding trial statistically significant differences were observed in growths of fish among the dietary treatments (P < 0.05). At that time, fish fed the TB diet showed the highest growth rate, and those fed the PT diet, the lowest (Fig. 1). At the end of the experiment, fish fed the AN diet had the highest growth rate, which was not statistically different from fish fed the TB diet (Fig. 1). However, canni-balism was high in AN-fed fish. All mortality was cannicanni-balism related in this treatment (Fig. 2). Fish fed the PT diet had the lowest growth rate and no cannibalism was noted. Final weight, weight gain and specific growth rate were highest in
800 a 600 700 AN TB a a 400 500 BK a b 200 300 PT Individual weight (mg) c b b 2 1 0 0 100 c Time (weeks)
Fig. 1. Growth response of surubim (Pseudoplatystoma fasciatum) fed live, commercial and formulated diets over 2 weeks. Values are mean ± SD, n = 3. Means with different superscript letter are significantly different (P < 0.05) 50 60 Cannibalism a a 30 40 Natural mortality 10 20 Mortality (%) b 0 PT BK AN TB Experimental diets
Fig. 2. Natural and cannibalism mortality of surubim (Pseudoplatys-toma fasciatum) fed live, commercial and formulated diets over 2 weeks. Values are mean ± SD, n = 3. Means with different superscript letter are significantly different (P < 0.05). Standard deviations are for total mortality
fish fed AN and TB diets; the lowest were in fish fed the PT diet (Table 1). Survival differed significantly among the dietary treatments (P < 0.05). Fish fed TB and PT diets had the highest final survival (Table 1). Fish fed the AN diet had the lowest survival, but there was no significant difference com-pared to fish fed the BK diet.
Lipid and fatty acid composition of the experimental diets
The four experimental diets were different in their lipid and fatty acid composition. The lipid level was the highest in the TB (33.1%, dry weight based) and the lowest in the PT (16.0%) diet. Lipid concentrations in the AN and BK diets were 27.7 and 21.9%, respectively.
In comparison to the other diets, the TB diet had high amounts of 14:0 (13.3%), 20:1n-9 (4.6%), 20:2n-6 (8.3%) and 20:4n-6 [8.6%; ARA (Arachidonic acid)], but low levels of 22:6n-3 [2.2%; DHA (Docosahexaenoic acid)]. It is puzzling that live Tubifex contained 10- to 20-fold higher levels of ARA than any other diet. This resulted in an n-3⁄ n-6 ratio lower than one (Table 2), opposite of what was observed in other diets. In comparison to the other diets, the Aglo Norse diet was characterized by high levels of 18:3n-3 (8.1%) and 20:3n-3 (13.4%). The TB and AN diets were rich in PUFA (poly-unsaturated fatty acids) (47.0 and 53.7%, respectively) but the source of these fatty acids was different. The main contributors of PUFA were n-6 fatty acids in the TB diet, whereas these were n-3 fatty acids in the AN diet. The most abundant fatty acids in the BK diet were oleic acid (24.1%; C18:1, OA) and palmitic acid (21.8%; C16:0); this diet was the richest in DHA (10.2%). The PT diet contained predominately OA (22.5%), linoleic acid (LA, 18.7%) and palmitic acid (17.7%).
Lipid and fatty acid composition of fish
Lipid levels in the whole body of surubim did not differ among the dietary treatments, averaging 3.6 ± 0.7%. Fatty acid compositions of the whole body neutral and phospholipids reflected the fatty acid profile of lipids in diets. With respect to the neutral lipid fractions, saturated fatty acids had cantly higher and monounsaturated fatty acids had signifi-cantly lower concentrations in fish fed the TB diet than in those fed the other three experimental diets (Table 3). The amount of total PUFA differed significantly among fish on different dietary treatments (P < 0.05). Fish fed the AN diet had the highest (51.0%) and those fed the BK diet had the lowest (36.5%) concentrations of PUFA. Although fish fed the TB diet had significantly higher amounts of PUFA than those fed the BK and PT diets, the contributions of n-3 fatty acids to
PUFA was higher in fish fed the BK and PT diets. Fish fed the AN diet had the highest amount of n-3 acids and the main contributor to n-3 PUFA was 20:3n-3 (12.6% of total fatty acids). Palmitic acid (C16:0) was the most abundant fatty acid in fish fed the TB and AN diets, whereas 18:1 was the predominant fatty acid in those fed the BK and PT diets (Table 3). As in the diets, 14:0, 20:1n-9, 20:2n-6 and 20:4n-6 fatty acids were the highest and 22:6n-3 was the lowest fatty acid in fish fed the TB diet in comparison to the fish fed other experimental diets. The high dietary concentration of ARA in Tubifexwas reflected in a fivefold higher ARA concentration in fish body triglycerides in comparison to the other dietary treatments. DHA had the highest amount in fish fed BK and AN diets and the lowest amount in fish fed the TB diet.
In respect to phospholipid fractions of whole body lipids (Table 4), 16:0 was the most abundant fatty acid regardless of dietary treatment. Total saturates were highest in fish fed the
Table 1
Growth performance of surubim (Pseudoplatystoma fasciatum) juveniles (initial size 117.6 ± 11.8 mg weight; 28.3 ± 2.5 mm total length), fed live, commercial and formulated diets over 2 weeks. Values are mean ± SD, n = 3. Means with different superscript letter in a row are significantly different (P < 0.05). Weight gain = [(final weight)initial weight) ·100] ⁄ initial weight. SGR = [(logefinal weight)loge initial
weight)·100] ⁄ duration of the experiment in days
Growth performance Experimental diets TB AN BK PT Final weight (g) 0.62 ± 0.05a 0.70 ± 0.06a 0.48 ± 0.05b 0.23 ± 0.02c Weight gain (%) 448 ± 27a 502 ± 50a 273 ± 61b 103 ± 10c SGR (%) 12.2 ± 0.3a 12.8 ± 0.6a 9.3 ± 1.2b 5.1 ± 0.3c Survival (%) 100 ± 0.0a 57.8 ± 13.9c 77.8 ± 21.4b,c 88.9 ± 10.2a,b
SGR, specific growth rate; TB, Tubifex worms; AN, Aglo Norse larva feed; BK, Bio Kyowa; PT, dipeptide-based diet.
Table 2
Fatty acid composition of live, commercial and formulated diets (% of total lipids detected)
Fatty acids Experimental diets TB AN BK PT 14:0 13.3 5.6 3.9 4.7 16:0 12.7 16.0 21.8 17.7 16:1 (n-7 and n-9) 6.1 4.0 3.6 5.5 18:0 6.3 2.2 4.6 4.0 18:1 (n-7 and n-9) 9.4 12.6 24.1 22.5 18:2n-6 10.8 14.2 11.3 18.7 18:3n-3 2.7 8.1 3.5 3.7 20:0 nd 2.9 1.6 1.3 20:1n-9 4.6 0.1 0.3 0.3 18:4n:3 0.3 nd nd nd 21:0 0.0 2.5 1.3 1.4 20:2n-6 8.3 0.2 0.2 0.2 20:3n-6 3.2 0.1 0.1 0.1 20:3n-3 0.1 13.4 3.2 1.1 20:4n-6 8.6 0.6 0.4 0.8 22:1 0.6 0.4 0.4 0.5 20:5n-3 9.1 7.2 8.2 8.7 22:4n-6 0.9 0.5 0.3 0.4 22:5n-6 nd 0.2 0.3 0.3 22:5n-3 0.8 0.6 0.6 1.4 22:6n-3 2.2 8.7 10.2 6.9 PSaturated 32.3 29.2 33.2 29.1 P Monounsaturated 20.7 17.1 28.4 28.7 PPolyunsaturated 47.0 53.7 38.4 42.2 P n-3 15.2 38.0 25.7 21.7 P n-6 31.8 15.7 12.7 20.4 n-3⁄ n-6 0.5 2.4 2.0 1.1
TB, Tubifex worms; AN, Aglo Norse larva feed; BK, Bio Kyowa; PT, dipeptide-based diet; nd, not determined.
BK and PT diets and lowest in those fed the TB diet (P < 0.05). Monounsaturated fatty acids differed significantly among the dietary treatments (P < 0.05), having the highest and lowest values in fish fed the PT and TB diets, respectively. The highest and lowest PUFA were found in fish fed the TB (55.0%) and PT (41.0%) diets, respectively. The ratio of n-3 fatty acids to n-6 fatty acids was significantly different among
dietary treatments with the highest and lowest values seen in fish fed the AN and BK (3.0), and the TB (0.8) diets, respectively. This finding suggests in fish fed the other experimental diets that PUFA was mostly composed of n-6 (mainly 20:4n-6) fatty acids, whereas in fish fed the TB diet, most of them consisted of n-3 fatty acids. Fish fed the AN and BK diets had significantly higher amounts of DHA (21.6 and
Table 3
Fatty acid composition of neutral lipids of whole body surubim (Pseudoplatystoma fasciatum) juveniles fed live, commercial and formulated diets over 2 weeks. Values are mean ± SD, n = 3 (each replicate is the combination of four fish). Means with different superscript letter in a row are significantly different (P < 0.05) Fatty acids Experimental diets TB AN BK PT 14:0 8.7 ± 0.4a 5.6 ± 0.1b 3.6 ± 0.1d 4.2 ± 0.1c 16:0 15.3 ± 0.8c 16.4 ± 0.2c 22.4 ± 0.3a 20.0 ± 1.2b 16:1 (n-7 and n-9) 4.8 ± 0.2b 4.2 ± 0.1c 4.0 ± 0.2c 5.3 ± 0.3a 18:0 7.6 ± 0.4a 2.2 ± 0.2c 4.2 ± 0.2b 4.5 ± 0.5b 18:1 (n-7 and n-9) 10.8 ± 0.4d 14.3 ± 0.3c 25.6 ± 0.1a 23.0 ± 0.3b 18:2n-6 12.4 ± 0.4b 12.6 ± 0.3b 11.3 ± 0.3c 16.9 ± 0.5a 18:3n-3 4.8 ± 0.3b 8.8 ± 0.1a 3.5 ± 0.2c 3.0 ± 0.1d 20:0 3.3 ± 0.1a 3.3 ± 0.0a 1.8 ± 0.0b 1.4 ± 0.3c 20:1n-9 0.9 ± 0.0a 0.1 ± 0.0b 0.3 ± 0.0b 1.0 ± 0.6a 21:0 0.5 ± 0.0c 2.2 ± 0.0a 1.2 ± 0.0b 1.2 ± 0.2b 20:2n-6 8.1 ± 0.4a 0.3 ± 0.0b 0.3 ± 0.0b 0.3 ± 0.0b 20:3n-6 3.7 ± 0.1a 0.1 ± 0.0c 0.1 ± 0.1c 0.5 ± 0.1b 20:3n-3 0.4 ± 0.0c 12.6 ± 0.2a 1.7 ± 0.7b 1.4 ± 0.2b 20:4n-6 5.1 ± 0.3a 0.8 ± 1.3b 0.9 ± 0.2b 0.3 ± 0.2c 22:1 0.5 ± 0.0b 0.7 ± 0.0a 0.5 ± 0.0b 0.5 ± 0.1b 20:5n-3 4.2 ± 0.1c 4.8 ± 0.2b,c 5.3 ± 0.1a,b 5.3 ± 0.5a 22:4n-6 1.1 ± 0.1a 0.3 ± 0.2b 0.3 ± 0.1b nd 22:5n-3 2.2 ± 0.4a 1.2 ± 0.2c 1.2 ± 0.0c 1.8 ± 0.5b 22:6n-3 5.3 ± 0.4c 9.7 ± 0.2a 11.8 ± 0.0a 9.3 ± 0.5b P Saturated 35.5 ± 1.3a 29.7 ± 0.3b 33.1 ± 0.4b 31.4 ± 1.5b PMonounsaturated 17.1 ± 0.2d 19.3 ± 0.2c 30.4 ± 0.3a 29.8 ± 0.3b P Polyunsaturated 47.4 ± 1.3b 51.0 ± 0.5a 36.5 ± 0.2d 38.8 ± 1.2c P n-3 17.0 ± 0.7d 37.0 ± 0.4a 23.5 ± 0.5b 20.8 ± 1.0c P n-6 30.4 ± 0.7a 14.0 ± 0.2c 13.1 ± 0.6c 18.0 ± 0.4b n-3⁄ n-6 0.6 ± 0.0d 2.7 ± 0.0a 1.8 ± 0.1b 1.2 ± 0.1c
TB, Tubifex worms; AN, Aglo Norse larva feed; BK, Bio Kyowa; PT, dipeptide-based diet; nd, not determined.
Table 4
Fatty acid composition of phospholip-ids of whole body surubim (Pseudo-platystoma fasciatum) juveniles fed live, commercial and formulated diets over 2 weeks. Values are mean ± SD, n= 3 (each replicate is the combina-tion of four fish). Means with different superscript letter in a row are signifi-cantly different (P < 0.05) Fatty acids Experimental diets TB AN BK PT 14:0 2.0 ± 0.2a 2.0 ± 0.0a 1.1 ± 0.1b 1.4 ± 0.4b 16:0 16.4 ± 0.9c 23.6 ± 0.1a 24.8 ± 1.2a 21.8 ± 0.5b 16:1 (n-7 and n-9) 1.4 ± 0.2 1.2 ± 0.0 1.1 ± 0.1 1.8 ± 0.5 18:0 13.1 ± 0.4a 6.9 ± 0.2d 9.3 ± 1.2c 11.7 ± 0.2b 18:1 (n-7 and n-9) 10.6 ± 0.5d 12.4 ± 0.6c 19.0 ± 0.8b 20.2 ± 0.1a 18:2n-6 4.7 ± 0.4c 6.5 ± 0.2b 5.2 ± 0.3c 8.0 ± 0.5a 18:3n-3 1.1 ± 0.1b 5.6 ± 0.2a 1.0 ± 0.3b 1.4 ± 0.2b 20:0 0.7 ± 0.1b 1.4 ± 0.0a 0.8 ± 0.0b 1.0 ± 0.3b 20:1n-9 0.4 ± 0.0a 0.1 ± 0.0c 0.1 ± 0.0b 0.2 ± 0.0b 21:0 0.1 ± 0.0c 0.5 ± 0.0a 0.3 ± 0.0b,c 0.4 ± 0.2a,b 20:2n-6 5.1 ± 0.2a 0.4 ± 0.0b 0.4 ± 0.0b 0.3 ± 0.0b 20:3n-6 3.5 ± 0.2a 0.5 ± 0.1c 0.4 ± 0.1c 1.1 ± 0.2b 20:3n-3 0.3 ± 0.0a 0.1 ± 0.0d 0.2 ± 0.0b 0.1 ± 0.0c 20:4n-6 15.2 ± 0.3a 4.7 ± 1.3b 3.8 ± 0.2b 3.9 ± 0.2b 22:1 0.3 ± 0.0c 0.6 ± 0.0a 0.4 ± 0.0b 0.6 ± 0.0a 20:5n-3 5.2 ± 0.3d 9.4 ± 0.3a 8.2 ± 0.6b 7.2 ± 0.0c 22:4n-6 1.7 ± 0.1a 0.3 ± 0.2b 0.2 ± 0.1b 0.3 ± 0.0b 22:5n-6 1.0 ± 0.1c 0.5 ± 0.0b 0.7 ± 0.0a 0.4 ± 0.0b 22:5n-3 3.6 ± 0.1a 1.7 ± 0.2c 1.6 ± 0.2c 2.7 ± 0.1b 22:6n-3 13.6 ± 0.5c 21.6 ± 0.8a 21.3 ± 0.4a 15.5 ± 1.0b P Saturated 32.3 ± 0.6c 34.5 ± 0.2b 36.3 ± 0.6a 36.3 ± 0.3a P Monounsaturated 12.7 ± 0.5d 14.3 ± 0.6c 20.7 ± 0.7b 22.7 ± 0.6a P Polyunsaturated 55.0 ± 0.9a 51.2 ± 0.6b 43.0 ± 1.3c 41.0 ± 0.8d P n-3 23.8 ± 0.3d 38.4 ± 1.0a 32.3 ± 1.1b 26.9 ± 0.9c Pn-6 31.2 ± 0.7a 12.9 ± 1.6b 10.7 ± 0.3c 14.0 ± 0.8b n-3⁄ n-6 0.8 ± 0.0c 3.0 ± 0.5a 3.0 ± 0.1a 1.9 ± 0.2b
21.3%, respectively) than those fed TB and PT diets. The concentration of ARA was significantly higher than that of DHA and higher (three- to fourfold) in fish fed the TB diet than in those fed the other experimental diets.
Discussion
The Aglo Norse diet seemed to be a well-accepted diet in terms of growth, but cannibalism was extremely high (Table 1; Fig. 2). It has been reported that nutritional factors may affect fish aggressive behavior (Winberg et al., 2001). Catfishes from the genus Pseudoplatystoma transition early on to piscivory (Giaquinto and Volpato, 2001); however, systematic⁄ quanti-tative description of cannibalism in early ontogeny of these species is not available. Some other studies reported that dry diets exacerbated cannibalism in Clarias gariepinus (Hecht and Appelbaum, 1987) and Clarias macrocephalus (Fermin and Bolivar, 1991). It is conceivable that chemical stimulants in feeds (fish proteins) may encourage cannibalism by simply ÔneutralizingÕ substances released from skin by conspecifics. Although chemicals released from injured catfish may elicit an alarm reaction and demotivation of feeding (Giaquinto and Volpato, 2001), substances released from other tissues (liver, muscle) may override these effects. Interestingly, we did not observe cannibalism in fish fed a peptide-based diet despite a Ôfeeding frenzyÕ when feed was introduced. Weaker fish in these groups were not attacked by conspecifics in comparison to other treatments even though fish doubled in body weight in 2 weeks. This finding is very controversial in comparison to other results from catfishes fed formulated diets.
One week into the experiment, fish fed the TB diet had the highest growth rate. At that time fish fed AN and BK diets showed similar growth performance; those fed the PT diet demonstrated the lowest growth. At the end of the experiment growth was slightly lower in fish fed Tubifex worms than in fish fed the AN diet, but with no significant differences. Fish fed the TB diet did not maintain the initial high growth rate (Fig. 1). This may suggest that Tubifex worms offered at 5% biomass at wet weight basis in week 2 were inadequate in terms of the feeding rate. However, there was no mortality in fish fed the TB diet. Tubifex worms were used as the first exogenous feeding of Asian catfish (C. macrocephalus) along with four live (Artemia, Brachionus calyciflorus, Chironomus plumosus and Moina macrocopa) foods and one artificial diet (40% protein). Eight weeks of feeding Tubifex worms resulted in the highest growth rate among the tested diets. Larvae fed the artificial diet showed the lowest growth and did not survive beyond 4 weeks (Evangelista et al., 2005). In Asian catfish (Pangasius bocourti) larvae, Tubifex worms and Artemia nauplii have been observed to promote the highest growth performance (Hung et al., 2002).
Here, fish fed a peptide-based diet showed the poorest growth rate but no cannibalism-related mortality. This diet was accepted very well, but fish fed the PT diet appeared to be underfed, as the diet was consumed within the first few minutes after introduction. In rainbow trout alevins fed diets with synthetic dipeptides as partial sources of amino acids (50% of dietary N), growth was comparable to those of fish fed protein-based control or commercial diets (Terjesen et al., 2006).
Surubim juveniles did not show any differences in whole body lipid levels when fed diets with different lipid concentra-tions. This finding is in agreement with Martino et al. (2002a), who suggested that surubim juveniles can readily utilize different sources and levels of lipids. Considering the high
levels of lipids in the diets used in the present work, it seems that lipid level may be increased up to 20–25% in the diet for this species in early life stages.
Fatty acid compositions of the diets were reflected in the whole body neutral and phospholipid fractions. High levels of dietary 20:4n-6 were accumulated by fish fed the TB diet. This fatty acid is a very important in cell membrane structural lipids (Sargent et al., 1999). No mortality was found in fish fed the TB diet perhaps due to a reduced response to the stress invoked by a high amount of ARA. This phenomenon was suggested in marine fish gilthead seabream (Sparus aurata) by Van Anholt et al. (2004). In our laboratory, we also observed a trend in which cannibalism was decreasing with increased ARA in tissue of South American surubim (Arslan et al., 2008). In line with the concentration of 20:4n-6 in the diets, the levels of this fatty acid increased in the phospholipid factions of body lipids. This finding confirms our previous result in surubim (Arslan et al., 2008) also suggesting the efficient bioconversion of ARA from 18:2n-6.
The level of n-3 HUFA increased in phospholipid factions of whole body lipids in fish fed TB, BK and PT diets. These results demonstrate the elongation and desaturation of 18:3n-3 to 22:6n-3 via 20:5n-3. Moreover, in all cases, the increasing ratio of n-3 HUFA to n-6 HUFA in phospholipid fractions confirms the efficiency of this pathway.
Our findings suggest that Tubifex worms as live food can be a good alternative to Artemia in larval and juvenile rearing of South American catfish. The AN diet as a commercial larval diet provided excellent growth rate, but a high level of cannibalism was observed. Therefore, quantitative growth performance may be somewhat different under conditions where cannibalism will be diminished or eliminated. Although the weight of cannibalized individuals was included in the total weight, the number of fish at each count reflected accurately the losses due to cannibalism, resulting in somewhat inflated mean individual weight not directly related to the experimental diets. Thus, the results of the current study should be interpreted with caution due to the uncertain contribution of cannibalism to the measured growth parameters. Using the diet enriched with AA or a feeding regime which provides AN and TB diets together should be examined to control the cannibalistic behavior of surubim along with an apparent positive function of ARA. Using these two diets together can also secure considerable amounts of n-3 HUFA. Use of a peptide-based diet resulted in no cannibalism in the present study and should be therefore further explored in larval and juvenile rearing of South American catfish.
Acknowledgements
The senior author was in part supported by Ataturk University. Financial support for this project came in part from USAID Grant No. LAG-G-00-96-90015-00 through the Aquaculture Collaborative Research Support Program (CRSP). The authors (MCP) wish to thank for CNPq support (Grant 477812⁄ 03-1) and IBAMA license (0124497BR). The Aquaculture CRSP accession number is 1342. The opinions expressed herein are those of the authors and do not necessarily reflect the views of the US Agency of International Development.
References
Almaza´n-Rueda, P.; Schrama, J. W.; Verreth, J. A. J., 2004: Behavioural responses under different feeding methods and light
regimes of the African catfish (Clarias gariepinus) juveniles. Aquaculture 231, 347–359.
Almaza´n-Rueda, P.; Van Helmond, A. T. M.; Verreth, J. A. J.; Schrama, J. W., 2005: Photoperiod affects growth, behaviour and stress variables in Clarias gariepinus. J. Fish Biol. 67, 1029–1039. Arslan, M.; Rinchard, J.; Dabrowski, K.; Portella, M. C., 2008: Effects of different dietary lipid sources on the survival, growth and fatty acid composition of South American catfish (Pseudoplatystoma fasciatum), surubim, juveniles. J. World Aquac. Soc. 39, 51–61. Britz, P. J.; Pienaar, A. G., 1992: Laboratory experiments on the effect
of light and cover on the behavior and growth of african catfish, Clarias gariepinus(Pisces, Clariidae). J. Zool. 227, 43–62. Czesny, S.; Dabrowski, K., 1998: The effect of egg fatty acid
concentrations on embryo viability in wild and domesticated walleye (Stizostedion vitreum). Aquat. Living Resour. 11, 371–378. Dabrowski, K., 1986: Ontogenetical aspects of nutritional
require-ments in fish. Comp. Biochem. Physiol. A 85, 639–655. Evangelista, A. D.; Fortes, N. R.; Santiago, C. B., 2005: Comparison
of some live organisms and artificial diet as feed for Asian catfish Clarias macrocephalus (Gu¨nther) larvae. J. Appl. Ichthyol. 21, 437–443.
Fermin, A. C.; Bolivar, M. E. C., 1991: Larval rearing of Philippine freshwater catfish, Clarias macrocephalus (Gu¨nter) fed live zoo-plankton and artificial diet: a preliminary study. Bamidgeh 43, 87– 94.
Folch, J.; Lees, M.; Stanley, G. H. S., 1957: A simple method for the isolation and purification of total lipids from animal tissues. J. Biol. Chem. 226, 497–509.
Giaquinto, P. C.; Volpato, G. L., 2001: Hunger suppresses the onset and the freezing component of the antipredator response to conspecific skin extract in pintado catfish. Behaviour 138, 1205– 1214.
Hecht, T.; Appelbaum, S., 1987: Notes on the growth of Israeli sharptooth catfish (Clarias gariepinus) during the primary nursing phase. Aquaculture 63, 195–204.
Hung, L. T.; Tam, B. M.; Cacot, P.; Lazard, J., 1999: Larval rearing of the Mekong catfish, Pangasius bocourti (Pangasiidae, Siluroidei): substitution of Artemia nauplii with live and artificial feed. Aquat. Living Resour. 12, 229–232.
Hung, L. T.; Tuan, N. A.; Cacot, P.; Lazard, J., 2002: Larval rearing of the Asian catfish, Pangasius bocourti (Siluroidei, Pangasiidae): alternative feeds and weaning time. Aquaculture 212, 115–127.
Izquierdo, M. S.; Socorro, J.; Arantzamendi, L.; Herna´ndez-Cruz, C. M., 2000: Recent advances in lipid nutrition in fish larvae. Fish Physiol. Biochem. 22, 97–107.
Juaneda, P.; Rocquelin, G., 1985: Rapid and convenient separation of phospholipids and non phosphorus lipids from rat heart using silica cartridges. Lipids 30, 40–41.
Kerdchuen, N.; Legendre, M., 1994: Larval rearing of an african catfish, Heterobranchus longifilis, (teleostei, clariidae) - a compar-ison between natural and artificial diet. Aquat. Living Resour. 7, 247–253.
Kossowski, C., 1996: Prospects for catfish culture (Siluroidei) in South America. Aquat. Living Resour. 9, 189–195.
Luchini, L.; Avendan˜o-Salas, T., 1984: Preliminary data on larval survival of South American Catfish, Rhamdia Sapo. Aquaculture 42, 175–177.
Martino, R. C.; Cyrino, J. E. P.; Portz, L.; Trugo, L. C., 2002a: Effect of dietary lipid level on nutritional performance of the surubim, Pseudoplatystoma coruscans. Aquaculture 209, 209–218. Martino, R. C.; Cyrino, J. E. P.; Portz, L.; Trugo, L. C., 2002b:
Performance and fatty acid composition of surubim (Pseudopla-tystoma coruscans) fed diets with animal and plant lipids. Aquaculture 209, 233–246.
Metcalfe, L. D.; Schmitz, A. A., 1961: The rapid preparation of fatty acid esters for gas chromatographic analysis. Anal. Chem. 33, 363–364. Piaia, R.; Townsend, C. R.; Baldisserotto, B., 1999: Growth and survival of fingerlings of silver catfish exposed to different photoperiods. Aquac. Int. 7, 201–205.
Sargent, J.; Bell, G.; McEvoy, L.; Tocher, D.; Estevez, A., 1999: Recent developments in the essential fatty acid nutrition of fish. Aquaculture 177, 191–199.
Sato, Y.; Cardoso, E. L.; Sallum, W. B., 1998: Reproduc¸a˜o induzida do surubim da Bacia do Sa˜o Francisco. Encontro Anual de Aqu¨icultura, Belo Horizente, 1 pp.
Terjesen, B. F.; Lee, K. J.; Zhang, Y. F.; Failla, M.; Dabrowski, K., 2006: Optimization of dipeptide-protein mixtures in experimental diet formulations for rainbow trout (Oncorhynchus mykiss) alevins. Aquaculture 254, 517–525.
Van Anholt, R. D.; Spanings, F. A. T.; Koven, W. M.; Nixon, O.; Wendelaar Bonga, S. E., 2004: Arachidonic acid reduces the stress response of gilthead seabream, Sparus aurata L. J. Exp. Biol. 207, 3419–3430.
Winberg, S.; Øverli, Ø.; Lepage, O., 2001: Suppression of aggression in rainbow trout (Oncorhynchus mykiss) by dietary L-tryptophan. J. Exp. Biol. 204, 3867–3876.
AuthorÕs address: Konrad Dabrowski, School of Environment and Natural Resources, The Ohio State University, Columbus, OH 43210, USA.
