Gelatin in replacement of bovine heart in feed training of Lophiosilurus alexandri in different water salinities

The aim of this study was to evaluate commercial gelatin in the total replacement of bovine heart in feed training of “pacamã” Lophiosilurus alexandri in different water salinities. A completely randomized experimental design, in a 2 × 3 factorial arrangement, was performed using two types of moist ingredients (bovine heart and gelatin) and three water salinities (0.0; 2.0 and 4.0 g of salt L) with three replications. Juveniles (2.39 ± 0.08 cm standard length and 0.20 ± 0.03 g of weight) were conditioned to accept commercial diets by the technique of the gradual transition of ingredients. At the end of 36 days no differences were observed to weight gain, length gain and specific growth rate. The feed training efficiency was better (P < 0.05) with the gelatin use, 100.0%. There was a negative effect of salinity on the survival rate and management efficiency in the concentration of 4 g of salt L, with values of 58.6 ± 12.0 % and 58.0 ± 12.0 %, respectively. Lophiosilurus alexandri juveniles could be feed-trained to accept commercial diets with gelatin in the total replacement of bovine heart in freshwater or salinity of 2 g of salt L.


INTRODUCTION
The production of carnivorous fish is lower than the production of omnivorous fish species, which can be explained by the high cannibalism rate.Due to this, it is necessary to perform the feed training of these species (Cyrino and Kubitza 2003).
Various feed training techniques have been studied, emphasizing the gradual transition of ingredients (Kubitza and Lovshin 1997, Kasai et al. 2011, Luz et al. 2011).This transition constitutes in the replacement of the moist ingredients (normally bovine heart) for dry diet.However, this technique has some disadvantages, such as the preparation and storage of the diets, and the instability of the pellets, which impairs water quality (Aubin et al. 2009).Thus, is a need to find alternatives for the product normally used as moist ingredient.
Commercial gelatin has been mixed with commercial diets during the feed training of ANA LUCIA SALARO et al. carnivorous fish, but only empirically by fish farmers in the production of "surubins" (Pseudoplatystoma sp.).The use of gelatin showed promising results when used in the feed training of juvenile Hoplias lacerdae (Salaro et al. 2012a).
Another management used in fish production is the salinization of water (Fashina-Bombata andBusari 2003, Luz andSantos 2010).For some species of tropical freshwater, low salinity water (2 to 4 g of salt L -1 ) can be beneficial for the survival and growth of fish.The salinity of the water stimulates feeding and reduces the metabolic rate and feed conversion rate (Altinok andGrizzle 2001, Luz et al. 2008), in addition to decreasing the energy expenditure with the osmoregulation of animals (Altinok andGrizzle 2001, Boeuf andPayan 2001).
The "pacamã" Lophiosilurus alexandri, is a carnivorous species, of nocturnal habit and easy handling (Tenório et al. 2006).It can reach over 8 kg in nature.This species has been used in restocking programs of the São Francisco River (Guimarães-Cruz et al. 2009).Lophiosilurus alexandri has a firm meat, absence of intramuscular bones and is much appreciated by the culinary sector.Previous studies have demonstrated the positive effect of salinity on larval breeding of this species, when using Artemia nauplii in the feeding (Luz andSantos 2008a, Santos andLuz 2009), which indicates the need for further study on water salinity in the remaining stages of the production of this species.
The objective of this study was to evaluate commercial gelatin in the total replacement of bovine heart in the feed training of Lophiosilurus alexandri in different water salinities.

MATERIALS AND METHODS
This experiment was approved by the Ethics Committee of Department of Animal Science, UFV (Protocol N 19/2014).

experimental desiGn and fisH GrowtH
After absorption of the yolk sac, the Lophiosilurus alexandri larvae were kept in freshwater and fed with Artemia nauplii for a period of 30 days.After this period, the juveniles with standard length of 2.39 ± 0.08 cm and weight of 0.20 ± 0.03 g were stocked (5 fish L -1 ) in 18 aquariums (35 × 30 × 14 cm) containing five liters of water, a biological filter, constant aeration and controlled temperature (27.0 ± 1.0 ºC) through the use of a heater and thermostat.The aquariums were covered with a screen of 2 mm in diameter to prevent fish from escaping.The aquariums were kept in a 12 h of photophase by fluorescent lamps (60 W).
A completely randomized experimental design, in a 2 × 3 factorial arrangement was performed using two types of moist ingredients (bovine heart and gelatin) and three water salinities (0.0; 2.0 and 4.0 g of salt L -1 ) with three replications.A commercial common salt (Cisne®, Brazilian Industry -390 mg sodium g -1 of salt, 25 μg iodine g salt -1 ) was used to create the different salinities.The salt was dissolved in the water of the aquariums prior to distribution of the fish.The fish were transferred directly from freshwater to the different salinity waters.
The fish were conditioned to accept commercial diets by the technique of gradual transition of ingredients in accordance to Salaro et al. (2012a), with modifications to the transition time of the last diet.The diets were made using commercial diets (41.0%crude protein, 8.0% ether extract, 9.5% ash, 4490.0 kcal kg -1 gross energy) as dry ingredients and bovine heart or commercial gelatin powder (Gelita®) as moist ingredients.The gelatin was dissolved in water at 50 °C and mixed with ration powder in accordance to Salaro et al. (2012a) (Table I).All diets were stored in a freezer (-18 °C) and thawed daily to make the pellets (1 mm) just before the feeding of the fish.
The chemical composition of feed ingredients (Table II) was analyzed following the Association of Official Analytical Chemists methods (AOAC 1999).The crude protein (total nitrogen 6.25) was measured using the Kjeldahl method after acid digestion.The total lipids content was estimated using the Soxhlet apparatus method.The moisture was determined by drying the sample at 105 °C for 24 h to a constant weight.The ash was determined by incinerating the dried sample in a muffle furnace at 550 °C for 12 h.These analyses were performed in the Food Analysis Laboratory, Department of Animal Science, Federal University of Viçosa, Brazil.The Lophiosilurus alexandri juveniles were hand fed to apparent satiety, three times a daily, at 08:00, 13:00 and 18:00 h.During the feeding of the fish, the aquariums were verified and the dead fish were removed and quantified.Every day, at 18:30 h, 100% of the water volume of each aquarium was changed.During this procedure, which lasted about 30 sec, juveniles were caught with a net and immediately transferred into another clean aquarium under the same experimental conditions (temperature and salinity water).This procedure is routine in the feed training of carnivorous fish in the Fish Nutrition Laboratory of the Department of Animal Biology of the Federal University of Viçosa.This management is performed with the aim of maintaining water quality due to the characteristics of the ingredients used in feed training.The water oxygen concentration was maintained above 6.50 ANA LUCIA SALARO et al.
± 1.0 mg L -1 by a supplemental aeration system.The pH was 8.10 ± 0.56 and the total ammonia was lower than 0.02 mg L -1 .Water parameters were measured with Multiparameter Meter (model HI 9828, Hanna Instruments, Brazil).

BiometriC index
At the end of the experiment (36 days), all fish were counted, weighed on a precision scale (model MB45 Toledo ® 0.001 g) and their standard lengths were measured with a digital caliper (model series 500 Absolute Coolant Proof ® of 0.0001 m) for the calculation of weight gain (WG), length gain (LG), specific growth rate (SGR), survival rate (SR), the feed training efficiency (FTE), and the management efficiency (ME).
Specific growth rate (SGR) of fish were calculated according to the expression: SGR = 100 × (lnWf− lnWi)/Δt, where Wi and Wf are the initial and final weights (in grams).The feed training efficiency (FTE) was calculated following the expression: FTE = NCF/NLF x 100, where NCF = number of conditioned fish (fish with body weight 20% greater than the initial average weight) and NLF = number of live fish.The management efficiency (ME) was calculated following the expression: ME = FTE x SR/100.

statistiCal analyses
Data were analyzed using Factorial Anova, and means were compared by SNK test, at 5% significance.The percentage results were arcsine transformed but only the percent results are presented.

RESULTS
No differences were observed between the moist ingredients (bovine heart or gelatin), water salinities and the interaction between them (P > 0.05) for weight gain, length gain and specific growth rate.The feed training efficiency was better for the fish conditioned with gelatin (P < 0.05).Salinity effects were observed on survival rate and the management efficiency (P < 0.05), the lower rates of both parameters were verified at 4 g of salt L -1 (Table III).

DISCUSSION
The lack of difference in weight and length gains and the specific growth rate of fish, using gelatin as a replacement of bovine heart, can probably be explained by the greater stability of the pellets on the water, facilitating the seizure and intake of food by fish.These results show the possibility of using gelatin as a moist ingredient in feed training of Lophiosilurus alexandri as found by Hoplias lacerdae (Salaro et al. 2012a).Thus, studies using gelatin for others carnivorous species should be performed.
The best feed training efficiency of fish, achieved by those fed diets containing gelatin, indicates the good palatability of these diets for Lophiosilurus alexandri.However, for Hoplias lacerdae, the use of gelatin afforded lower feed training efficiency than that observed with bovine heart, requiring supplementation with fish meal to obtain the feed training efficiency equivalent to diets with bovine heart (Salaro et al. 2012a).These results reinforce that palatability of the diet is species-specific (Kasumyan and Tinkova 2013).
The lower survival rate of fish in salinity 4 g of salt L -1 may be related to the development phase of the animals (Fashina-Bombata and Busari 2003) and the time of exposure to water salinity.Larvae of Lophiosilurus alexandri with 12 days of age support salinity up to 6 g of salt L -1 for 96 h (Luz and Santos 2008b).It is possible that the lower survival rate is also associated with stress related to constant changes in diets (Barcellos et al. 2000) and the daily changes of water aquariums (Salaro et al. 2012b).Fish more susceptible to stress are less well suited to tolerate multiple stressors and require a long time period to adapt to saline water (Fevolden et al. 2003).The lower survival rate could also be related to an osmotic imbalance of the fish at 4 g of salt L -1 .The increasing salinity of the water can lead to a number of metabolic and physiological changes (Boeuf andPayan 2001, Luz et al. 2008).This fact was also previously observed for juvenile of Hoplias lacerdae during feed training in different water salinities (Salaro et al. 2012b).However, the use of specific salinities can reduce stress and decrease susceptibility to diseases, as well as preserve and control pathogens (Altinok andGrizzle 2001, Garcia et al. 2007).For Hoplias lacerdae, salinity up to 5 g of salt L -1 attenuated stress during feed training (Salaro et al. 2012a).
In conclusion, Lophiosilurus alexandri juveniles can be feed-trained to accept commercial diets with gelatin as a replacement for bovine heart in freshwater or salinity of 2 g of salt L -1 .ANA LUCIA SALARO et al.