GROWTH PERFORMANCE AND CHEMICAL COMPOSITION OF PACUS ( Piaractus mesopotamicus ) FED WITH VEGETABLE OIL SOURCES AND ALPHA-LIPOIC ACID *

The objective of this study was to evaluate the effects of different oil sources and alpha-lipoic acid levels on pacus performance, chemical composition of fillets and histological changes in the liver. Were used 480 pacu juveniles (3.35 ± 0.78 g), distributed in 24 experimental units (n = 20). Six treatments were delineated in a completely randomized design (CRD), in a 3x2 factorial scheme, with three sources of oil (soybean, sesame and linseed) and two levels of alpha-lipoic acid (0 and 0.1 %), with four replications. Data were analyzed statistically by analysis of variance (ANOVA) and Tukey’s test (5%). Diets containing linseed oil as the only lipid source provided better indices of final body weight and individual weight gain in pacu juveniles. Crude protein content was higher in fish fillets that fed diets without alpha-lipoic acid. Liver’s histology was not affected by the experimental diets. In conclusion, linseed oil improved final body weight and individual weight gain on pacus, and alpha-lipoic acid did not affect the performance of the animals, however its absence promoted higher levels of crude protein in fish fillets.


INTRODUCTION
The addition of lipids, from vegetable or animal sources, in fish diets is common.In general, vegetable oils, especially soybean oils, are used, and the quality of these products is assessed primarily in relation to nutritional quality (DUNFORD, 2004).Besides playing an important rule in food quality, lipids are extremely important for living organisms in general (PORTZ and FURUYA, 2013), as they participate in the formation and maintenance of various tissues, organs and structures, as well as being sources of essential fatty acids (SARGENT et al., 2003).In addition, the amount and/or quality of the lipids in the diets may influence the structure of the hepatocytes of the fish livers (CABALLERO et al., 1999) and even the hepatosomatic index of the animals (CHEN et al., 2013).
For fish, lipids are the main source of energy, releasing more available energy than proteins and carbohydrates (GLENCROSS, 2009).There are several factors that may influence the optimum level of lipids needed in fish diets, such as the species and environmental conditions to which they are adapted (fresh water or salt water, water temperature), food habits (carnivorous or omnivore/herbivore), development phase (larvae, juveniles, adults or breeding), and the amount of proteins and carbohydrates present in the diet due to the interactions between lipid, carbohydrate and protein metabolism (GARCIA et al., 2013), and the lipid source to be used (ZHOU et al., 2014).
The vegetable oils that are used in fish feed are soybean oil and linseed oil.Soybean oil, which is rich in linoleic acid (MCCLEMENTS and DECKER, 2010), can be used as a source of n-6 fatty acid, and linseed oil, which is rich in linolenic acid (MCCLEMENTS and DECKER, 2010), can be used as source of n-3 fatty acid (REGOST et al., 2003).Sesame oil is known to exhibit high oxidative stability due to the antioxidant properties of bioactive compounds present in its composition, such as sesamin and sesamol (KOCHHAR, 2002).High concentrations (5.8 g kg -1 ) of sesamin in salmon diets are capable of affecting the energy metabolism of fish, especially the metabolism of liver and white muscle (WAGNER et al., 2014).
Among native species, pacu (Piaractus mesopotamicus) has stood out in recent years as a species of great economic interest in Brazil.As well as its excellent characteristic of adapting to the climatic conditions, it has omnivorous food habit, with strong tendency to herbivorous, as it is able to feed on fruits, crustaceans, organic debris, small fishes and molluscs (URBINATI et al., 2005).
In pacu juveniles fed with diets supplemented with ALA (1,000 mg kg -1 of feed) and/or ascorbate (500 mg AA kg -1 of feed), eicosapentaenoic acid (EPA, 20:5n-3) levels increased significantly in the muscles, but not significantly in the brain, suggesting that the brains are influenced very little by supplementation of this substance in the diet, a fact that can be explained by a possible defensive reaction of these animals to environmental changes (TRATTNER et al., 2007).
In view of the importance of the use of vegetable oil sources and bioactive components, such as alpha-lipoic acid, in fish nutrition, this study aimed to evaluate the effects of adding of the three sources of vegetable oils (soybean, sesame and linseed) and the two concentrations of alpha-lipoic acid on the performance and chemical composition of fillets and histological changes of the liver of pacus.

MATERIAL AND METHODS
The experiment was approved by the Research Ethics Committee of the Faculty of Animal Science and Food Engineering (N° 5451281015).The experimental period lasted for 90 days, using 480 pacu juveniles (3.35 ± 0.78 g) from a commercial fish farm.

Local
The trial was conducted at the Paulista Agribusiness Technology Agency (APTA, Pirassununga-SP).The analyses of the chemical composition of the diets and muscles and the Bol.Inst.Pesca, São Paulo, 44(vol.esp.): 74 -84, 2017 histology of the livers were carried out at the Faculty of Animal Science and Food Engineering, University of São Paulo (FZEA/USP).

Experimental diets
The analysis of the chemical composition of the ingredients and experimental diets were performed in triplicate, according to the AOAC (2000) methodology: dry matter was obtained after drying the samples in an oven at 105 ºC until constant weight; crude protein was obtained by the micro-Kjeldahl method, after acid digestion (total nitrogen x 6.25); ether extract was determined by extraction by heating the petroleum ether in a Soxhlet apparatus for six hours; mineral content was obtained by incineration in a muffle furnace at 550 °C for four hours; and gross energy was determined in a calorimetric pump.
After previous analysis of the chemical composition of the ingredients, six isoproteic (22% crude protein -CP) and isoenergetic (4,200 kcal.kg 1 of GE; gross energy) diets were formulated as presented in Table 1.
In order to prepare the experimental diets, the ingredients were ground, weighed, homogenized and submitted to the extrusion process (2 mm granulometry) using an extruder with a processing capacity of 30 kg hour -1 (NX 30 Imbramaq, Ribeirão Preto-SP, Brazil) at Fisheries Laboratory of ESALQ -USP (Piracicaba-SP).After extrusion, soybean, linseed and sesame oils, as well as alpha-lipoic acid, were added in the proportions of each treatment and homogenated.After drying in a forced ventilation oven (55 °C for 24 h), the diets were conditioned in identified plastic containers and stored in a cold room (4 °C) until use.

Fish and experimental conditions
Prior to the beginning of the experiment, fish were acclimatized to experimental conditions for seven days and fed with a commercial diet (Pirá Alevinos 1.7 mm, Guabi Nutrição Animal, Campinas-SP, Brazil) with 40% crude protein, which had been ground in a meat grinder, and the granules retained in a 1 millimeter mesh were selected.After the acclimatization period, fish were individually weighed (3.35 ± 0.78 g) and randomly distributed in 24 tanks with a capacity of 50 L, in a recirculating system with thermostatic heaters and a biological filter.Each tank corresponded to an experimental unit.
The parameters of water quality evaluated were temperature (T), dissolved oxygen (DO), hydrogenation potential (pH), nitrite (NO2 -) and toxic ammonia (NH3).Temperature was measured with a digital thermometer (Incoterm, Porto Alegre-RS, Brazil) with minimum and maximum daily temperature records.The other water parameters were evaluated weekly throughout the experimental period.Dissolved oxygen was measured with a digital oximeter (YSI Pro2, Yellow Springs Instruments, Ohio, USA), and pH, nitrite and toxic ammonia were measured daily using commercial kits (Labcon Test).The animals were fed ad libitum until apparent satiation twice a day (8:00 a.m. and 5:00 p.m.) for 90 days.
Two individual biometrics were performed, one before starting the administration of the experimental diets (Day 0) and another at the end of the experimental period (Day 90), in order to measure the parameters of growth performance of the pacus.In the first biometry, the fish were anesthetized with eugenol (20 mg L -1 ).In the second biometry, the fish were immersed in a lethal dose (250 mg L -1 ) of benzocaine (BRASIL, 2013).Six fish per experimental unit were filleted and the fillets were frozen separately in polyethylene packages and stored in a freezer (-20 °C) for further analysis.

Chemical composition of pacu fillets
For each experimental unit, three fish were randomly collected, filleted, stored in polyethylene packages in a freezer (-20 °C) and kept frozen until the analysis.Analyses of the chemical composition of fish fillets were carried out in triplicate according to the AOAC (2000) methodology: moisture, crude protein, ether extract and mineral matter, as previously described.

Liver histology
At the end of the experimental period, livers from three fish per experimental unit were collected and stored frozen at -20 °C for further analysis.After thawing, the samples were fixed in Bouin for 24 hours and then stored in 70% alcohol.The blocks were made with paraffin and histological sections were made using a microtome, with a thickness of 5 μm.In order to evaluate the morphology of the structures under light Bol.Inst.Pesca, São Paulo, 44(vol.esp.): 74 -84, 2017 microscopy and to measure the area of the epithelial mucosa, histological sections were dewaxed, rehydrated according to routine histological methods and submitted to staining with hematoxylin and eosin (BANCROFT and GAMBLE, 2008).The slides were photographed using a light microscope (Carl Zeiss, Germany) coupled to a digital camera and analyzed with the aid of Axio Vision 4.1 software.

Statistical analysis
The data were analyzed using Statistical Analysis System (SAS, 2011), version 9.3, by analysis of variance ANOVA and significant differences were verified by Tukey's Test with a significance of 5%.Means within each factor (oil sources and ALA levels) were compared by ANOVA and Tukey's Test (5%).In the analyses of variance, SAS procedure PROC GLM was used.
All assumptions of the statistical model were verified using SAS/LAB of SAS (R).When the assumption of homogeneity of variance was rejected, SAS procedure PROC GLIMMIX was used by associating different variances with each treatment.When the assumptions were satisfied, SAS procedure PROC GLM was used.

Growth performance
Growth performance data are presented in Table 2, and there were no differences (P>0.05)among the treatments evaluated.Although no differences were observed between the treatments for growth performance (P>0.05),there was a difference between FBW and IWG within each factor (oil sources and ALA levels) (Table 3).

Chemical composition of pacu fillets
Analyses regarding the chemical composition of fish fillets can be seen in Table 4.
Although the chemical composition of the fillets did not present differences between the treatments (P>0.05),there was a significant difference between the means of CP within the factor of ALA levels.
The type of oil used in the diets did not influence (P>0.05) the CP levels of the fillets (16.34%CP mean).However, the use of ALA in the diet resulted in fillets with higher levels of CP (16.52%) than the 16.16% of CP of fillets of fish fed diets without ALA inclusion.

Liver histology
Evaluating the histology of the livers of fishes that received commercial diet, it was observed that the hepatocytes had a polyhedral shape and normal arrangement of the nucleus, nucleolus and cytoplasm, with a rounded and central nucleus (which could be displaced to the periphery of the cell in some cases), well-evident nucleolus and clear cytoplasm (Figure 1), in accordance with what was observed by SOUZA et al. (2001) andFUJIMOTO et al. (2008) in their control groups.Also, it was possible to see the portal vein, which is responsible for draining blood into the liver, the cord-like arrangement of hepatocytes and sinusoids (coated by endothelial cells), which are irregularly distributed between hepatocytes (Figure 1).

Photomicrographs
for the histological analysis of the livers of fish that received the experimental diets without alpha-lipoic acid (T1, T2 and T3) and supplemented with ALA (T4, T5 and T6) can be seen in Figures 2 and 3, respectively.

Growth performance
In the biometry performed at the 90 th day, a high heterogeneity was observed among fish weight within each experimental unit.The heterogeneity observed in the pacu juveniles may be due to the hierarchical relationship between the animals, where the larger animals consumed more than the smaller ones.
In a study with pacus consuming diets containing 22 and 25 % CP, IWGs (39.9 to 57.8 g) were similar to those obtained in this study (ABIMORAD and CARNEIRO, 2007).Thus, it is possible to verify that lower CP levels in pacu diets can promote weight gain approximately equal to that obtained with diets with higher protein levels, and may suggest lower protein requirements for this species.
In some studies with a similar duration time, higher values of diet intake were observed, such as in FERNANDES al. (2001), where feed intake varied from 151.6 to 157.9 g, and in MUÑOZ-RAMÍREZ and CARNEIRO (2008), where it ranged from 104.4 to 129.6 g.
The results of feed conversion ratio in this study were similar to the values found by MUÑOZ-RAMÍREZ and CARNEIRO (2008), with results varying between 1.33 and 1.58.However, FCR values obtained for pacu were better than those found by FERNANDES et al. (2001), who obtained FCR ranging from 3.34 to 4.14; ABIMORAD and CARNEIRO (2007), whose FCR values were 2.7 to 3.0; SANTOS et al. (2009), with results between 3.07 and 3.60, even using diets with 28% CP; and by SIGNOR et al. (2010), who found FCR from 2.88 to 3.16 in pacu juveniles fed diets containing higher protein levels (25, 30 and 35%) than the levels addopted in the present study.
Fish fed diets with LO (T3 and T6) showed a average final weight, while those fed diets with SBO presented the lower final weight.Individual weight gain followed the same trend, with better results to the fishes fed diets T3 and T6.
The presence or absence of alpha-lipoic acid in the diets had no effect (P>0.05) on the mean final weight of the animals.However, in a study with pacu juveniles fed diets containing ALA (100 mg 100 g -1 ) and ascorbic acid (50 mg 100 g -1 ), there was an influence (P<0.05) of this compound on the final weight of the fish.Fish consuming diets with ALA and without ascorbic acid had a lower average weight (51.7 g), but for the treatments with ALA associated with ascorbic acid the index was improved (58.6 g) (TRATTNER et al., 2007).

Chemical composition of pacu fillets
The experimental treatments had no effect (P>0.05) on the chemical composition of the fillets.When feeding pacus with diets containing different CP levels (18.5 to 28.3 %), KLEIN et al. (2014) observed that the diets with lower CP content decreased the humidity (69.04%) and increased the ether extract (11.88%) of fish fillets, whereas diets with higher CP content had an opposite effect, resulting in fillets with higher humidity (72.65%) and lower ether extract (8.00%).
The source of oil used in the diets did not influence (P>0.05) the CP levels of the fillets, but the ALA caused a difference (P<0.05) between the means of CP.It was observed that fish fed diets without ALA presented fillets with higher CP levels.

Liver histology
The livers from pacu juveniles were light brown in color, typical of herbivorous fish (BERNET et al., 2004).The shape and constitution of the liver observed were the same as those described for several teleostean fish species studied (HINTON et al., 1972;HINTON and POOL, 1976;ROCHA et al., 1994;BRUSLÉ and ANADON, 1996;ROCHA et al., 1997;VICENTINI et al., 2005;FUJIMOTO et al., 2008), with no particular characteristics in the treatments.
Although the liver`s structure varies according to species, sex, age, diet or even environmental conditions (GENTEN et al., 2009), some changes observed through histological analysis may be indicative of animal stress or inadequate nutrition.CHEN et al. (2013) evaluated the effects of diets containing different levels (0.00, 0.10, 0.32, 0.63, 0.98, 1.56 and 2.04 % of dry weight) of linolenic acid (LNA, 18:3n-3) for Nile tilapia (Oreochromis niloticus) and observed that fish fed diets with 0.10% of LNA increased the lipid content stored in hepatic tissue and showed the highest hepatosomatic index.
In this study, histological analysis of the livers of the fish that received the diet supplemented with alpha-lipoic acid (T4, T5 and T6), indicated the presence of blood congestion at some points.However, it is important to point out that better health conditions for animals are not characterized by the complete absence of histopathological alterations; so it is normal to verify some changes, such as mild inflammatory reactions, small structural modifications (BERNET et al., 2004;BARBIERI and BONDIOLI, 2103) or even the appearance of moderate blood congestion, as observed in the present study.
The nucleus of the hepatocytes in all groups of fish analyzed was centralized, indicating that the cell had normal metabolic activity.The presence of cytoplasmic vacuolations increases the volume of hepatocytes, displacing the nucleus to the periphery of the cell, and indicates the existence of regions with a probable concentration Bol.Inst.Pesca, São Paulo, 44(vol.esp.): 74 -84, 2017 of lipids and glycogen, or the combination of toxic agents with intracytoplasmic lipids (SANTOS et al., 2004).
It is important to note that the diet may influence the structure of the liver and can, depending on the quality of the food, cause lesions in the hepatic tissue (ROCHA et al., 2010).It was evidenced that the experimental diets in this study did not cause major alterations in the liver histology of fish fed diets with different lipid sources and levels of alpha-lipoic acid.
In cases of inadequate nutrition, it is possible to observe some characteristics in liver histology such as alterations of basic structures, with nuclei located in the hepatocyte peripheries, cytoplasmic vacuolization and cell necrosis (RAŠKOVIĆ et al., 2011).As no significant changes were observed in pacu livers, it can be stated that the experimental diets were nutritionally adequate, besides not causing damage to fish health.

CONCLUSIONS
Linseed oil improves final body weight and individual weight gain of pacus juveniles.The addition of alpha-lipoic acid in the diets does not alter the growth performance of the animals, and, in the absence of this acid, pacu fillets accumulate higher levels of crude protein.None of the vegetable oil sources (soy, sesame and linseed) evaluated caused changes to pacu liver cells, although the addition of alpha-lipoic acid promoted minor changes.

Table 3 .
Mean values of final body weight (FBW) and individual weight gain (IWG) of pacu juveniles fed different experimental diets, within each factor (oil sources and ALA levels).