NILE TILAPIA FED ENZYME COMPLEX ADDED AT DIFFERENT LEVELS AND PROCESSING STEPS 1

Enzyme activity in solid-state fermentation complex added to diets before and after extrusion process and its influence on Nile tilapia (Oreochromis niloticus) performance were evaluated. The following enzymes were analyzed: b-glucosidase, pectinase, xylanase, endoglucanase, amylase, protease, and phytase. The fish were fed five experimental diets: a control diet (without enzyme), two test diets supplemented before extrusion (400 and 800 ppm SSF), and two test diets after extrusion or “on top” (400 and 800 ppm SSF). The experiment had a completely randomized design with four replicates and seven fish per tank. Average initial weight was 10.37 ± 0.95 g, and fish were fed four times a day to satiation for 56 days. The following performance parameters were evaluated: final weight, weight gain, feed conversion, survival, specific growth rate, and body composition. The results were compared using ANOVA and Tukey’s test at p<0.05. All the enzymes showed partial or total losses of activity when subjected to the extrusion process, except phytase. No differences were observed for survival, water quality parameters, and body composition. On-top inclusion provided a better performance, which is observed through increases in final weight, weight gain, and feed conversion. As a result, 800 ppm of enzyme complex SSF should be added "on top", that is after extrusion, to improve tilapia performance and, consequently, contribute to the fish farming financial efficiency.


INTRODUCTION
Tilapia has many favorable characteristics for fish farming purposes such as hardiness and ease of management.It is highly prolific, productive and its artificial reproduction has been consolidated, making it possible to produce fingerlings all year (EL-SAYED, 2006).Also according to the same author, this fish is able to adapt itself to different production systems, diets, as well as feed plankton available on the medium; it also has meat with excellent nutritional qualities and high market value.
Aquaculture has stood out worldwide as a food production system, encouraging culturists and researchers to seek techniques capable of improving profits and productivity.Thus, food additives may be considered quite promising because, according to Campestrini, Silva and Appet (2005), they are intended to preserve, enhance, or modify the food properties without affecting its nutritional value.
Among the alternatives, enzymes can be considered one of the most promising candidates since they enhance the availability of nutrients and energy, resulting in a better performance and reduced waste (JEGANNATHAN;NIELSEN, 2013).Just as occurs with other monogastric animals, fish are unable to metabolize several nutrients due to insufficient production or absence of specific enzymes (TANIGUCHI; TAKANO, 2004), which justifies its inclusion in diets.
During the degradation of a substrate, various enzymes may act concertedly for an efficient digestion.As in the case of the enzyme complex SSF (solid-state fermentation), some enzymes are combined naturally and evenly, such as bglucosidase, pectinase, xylanase, endoglucanase, amylase, protease, and phytase (CHAUYNARONG et al., 2008).
The activity of enzymes added to animal diets can be affected by several factors such as chemicals, pH and mainly by temperature variations (CARLSON;POULSEN, 2003).In fish farming, extruded diets are widely used; such diets are developed through a process involving high temperatures.Thus, in this case, enzymes should be supplemented carefully because, due to the protein composition, partial or total loss of biological activity might occur.
Therefore, the stability of enzymes (complex SSF) during the extrusion process steps and subsequent effects on Nile tilapia (Oreochromis niloticus) performance were evaluated in this study.

MATERIAL AND METHODS
This study was conducted from May to July 2014, under the authorization of the local ethics council.The basal diet was prepared in a feed mill and extruder, using ingredients of vegetable origin.The used enzyme complex Allzyme SSF, supplied by Alltech Inc. (Nicholasville, Kentucky, USA), contains the enzymes β-glucosidase, pectinase, xylanase, endoglucanase, amylase, protease, and phytase.
For the assay of enzyme activity, the basal feed (Table 1) was used in all treatments, differing only in the steps and inclusion levels.Treatments consisted of substituting supplementation with SSD from 4 to 8% of the basal diet, as follows: control (without enzyme), 4000 ppm SSF before extrusion (BE4), 8000 ppm SSF before extrusion (BE8), 4000 SSF after extrusion (AE4), and 8000 SSF after extrusion (AE4).SSF inclusion levels, before and after, were ten times higher than were those used for performance tests.These higher levels are justified by the difficulty in extracting enzymes and quantifying their activity through the currently used methods (ESMAEILIPOUR et al., 2012).
All treatments had jelly in their composition in order to prevent any kind of influence on ingredient basis, differing only at the time of inclusion.
In the treatments with enzymes added before extruded process and control (no enzyme), the jelly was incorporated into the mixture and only differing as regards the SSF inclusion level (0, 4000, and 8000 ppm).Yet for treatments with enzyme inclusion after extrusion process (4000 and 8000 ppm), the jelly was prepared and mixed with the enzyme complex, so being built "on top".All diets were processed with an extruder machine (Inbramaq, MX40), being produced pellet with an average diameter of 2 mm at 90 °C with subsequent drying.Finally, the samples were ground and separated for enzymatic analysis.

First Trial
For determination of enzyme activities, a 0.5-g sample of each diet was macerated with 10 mL distilled water.This material was placed into polyethylene tubes to be centrifuged at 13000 rpm for 10 minutes.Afterwards, the supernatant was pipetted off for enzyme activity determination.All enzymatic analyzes were performed in triplicate.The following enzymes, present in the complex included to the feed, were selected for analysis: b-glycosidase, pectinase, xylanase, endoglucanase (carboxymethyl cellulase), amylase, protease, and phytase.
The β-glucosidase is an exocellulase with specificity for a variety of substrates.This enzyme hydrolyzes the non-reducing end of β-glycosides, releasing glucose.In this assay, we used 100 µL enzyme extract, 100 µL sodium acetate buffer (100 mM, pH 5), and 250 µL solution of 2 mM synthetic substrate.The reaction was conducted for 15 minutes at 50 °C and paralyzed with 500-µL sodium carbonate (0.5M).Then, the solution was taken to a spectrophotometer for absorbance reading at 410 nm and, thus, determination of enzyme activity.
For pectinase, we used 25 µL enzyme extract, 75 µL sodium acetate buffer (100 mM, pH 5), and 400 µL galacturonic acid solution (0.25% w/ v).The reaction was conducted for 30 minutes at 50 °C and paralyzed with 500 µL DNS reagent (3,5 dinitrosalicylic acid) and thereafter incubated in boiling water bath for 5 minutes for color development.After, the solutions were centrifuged in polyethylene tubes at 13,000 rpm for 5 minutes.The solution was taken to the spectrophotometer for absorbance reading at 540 nm and, then, determination of enzyme activity.
The xylanase activity was determined using 80-µL sodium acetate buffer (100 mM, pH 5), 20 µL enzyme extract, and 400-µL Birch wood xylan solution (1.25% w/v).The reaction was conducted for 15 minutes at 50 °C paralyzed with 500 µL DNS and then incubated in boiling water bath for 5 minutes for color development.The solution was taken to the spectrophotometer for absorbance reading at 540 nm and enzyme activity determination.
For endoglucanase (carboxymethyl cellulase), 25 µL enzyme solution was mixed with 400 µL carboxymethyl cellulose (CMC) (1.25% w/ v), being diluted in 75 µL sodium acetate buffer (100 mM, pH 5).This solution was placed in a water bath at 50 °C for 30 minutes and after, the reaction was paralyzed with 500 µL DNS according to Miller (1956).These solutions were placed in boiling water for 5 minutes and subsequently read for absorbance at 540 nm in the spectrophotometer, for enzyme activity assessment.
kit according to Caraway (1959).Regarding protease activity, 150-µL azocasein (2%) was incubated at 37 °C with 125-µL enzyme extract for 30 minutes.Afterwards, 600 µL TCA (10% trichloroacetic acid) was added to the mixture, being allowed to stand for 15 minutes in an ice bath.Thereafter, the mixture was centrifuged for five minutes at 14,000 rpm, with the supernatant separated.An aliquot of the supernatant (600 µL) was transferred to a tube containing 700 µL 0.1 M NaOH for absorbance reading at 440 nm in the spectrophotometer.Then, the protease activity was determined.
For phytase activity, the amount of inorganic phosphate released was measured from sodium phytate substrate.We used 600-µL sodium phytate substrate (1.5 mM) in 100 mM sodium acetate buffer (pH 5.0) and 150-µL enzyme extract.During 30 minutes, the mixture was conducted in a water bath at 50 °C, being the reaction stopped by the addition of 250-µL trichloroacetic acid 10% (v/ v).Afterwards, 1000 µL colorimetric reagent was added to the assay tubes.This reagent was prepared from 10% ammonium molybdate solution (w/ v) in 5 M sulfuric acid solution.The reagent was prepared at the time of use, blending 10% (v/ v) from colorimetric reagent, 5% ferrous sulfate (w/ v), and deionized water.Subsequently, the absorbance readings were carried out in a spectrophotometer at 700 nm, correlating the values with a standard curve from KH 2 PO 4 .

Second Trial
The second trial consisted of an animal performance evaluation.It was conducted at the Laboratory of Aquaculture and Aquatic Ecology, Federal University of Vales do Jequitinhonha and Mucuri (UFVJM), in Diamantina -MG, Brazil.The study was carried out in 2013, for eight weeks (56 days), using Nile tilapia provided by CODEVASF (Janaúba -MG, Brazil).Two hundred Nile tilapia juveniles, GIFT strain, sexually reversed, with average initial weight of 10.37 ± 0.38 g, were used in the trial.In the beginning, initial weight and biomass (g) were measured for each treatment.Fish were distributed in twenty 35-L tanks, being 7 fish.tank - , and 0.20 fish.L -1 , with recirculation system containing an ultraviolet (UV) filter, being under constant aeration, temperature control, natural photoperiod, and bio-filtration conditions.The tanks were cleaned three times a week by siphoning, renewing 30% of the tank volume.
The juveniles were subjected to five diets (treatments): without enzyme (control), addition of 400 and 800 ppm SSF before extrusion (BE400 and BE800), addition of 400 and 800 ppm SSF after extrusion (AE400 and AE800), being distributed in a completely randomized design with four replicates of 7 fish.
A diet based on vegetable ingredients was adopted with the same initial composition as the one used in the enzyme assay, according to the formulation shown in Table 1.The fish were fed ad libitum (until satiation) four times daily (8 am, 11 am, 2 pm, and 5 pm).
Once a week, water quality parameters were measured as follows: temperature, pH, conductivity, dissolved oxygen (YSI Pro Plus multi-parameter probe) and ammonia (indophenol blue method) (APHA, 2012).The amount of offered and remained feed was daily measured using a precision scale (0.01 g) and subjected to estimate feed conversion ratio.
After 56 days of trial, in the only biometric experiment after the initial, all tilapia remained 12 h without food for the emptying of the digestive tract and body composition analysis, being captured and immersed in water plus eugenol (37.5 mg.L -1 ) about 30 minutes for euthanasia and weighing.In this point, we evaluated: survival (%), final biomass (g), biomass gain (g), final weight (g), weight gain (g), feed conversion (g.g -1 ), and specific growth rate -SGR (%.day -1 ).
The carcasses were dried in forced-hot air ovens and taken to the Laboratory of Animal Nutrition, Department of Animal Science (DZO), in UFVJM.Body composition parameters as dry matter (DM), mineral matter (MM), crude protein (CP), ether extract (EE), calcium (Ca), and phosphorus (P) were evaluated according to the methods described by Van Soest et al. (1991).
Data of enzyme activity and performance underwent analysis of variance (ANOVA) at 0.05 probability.The means were compared by the Tukey's test using the Statistical Analysis System software (SAS).Survival data were transformed to arcsine values but the Table shows them in percentage.For water quality parameters, means and standard deviations were calculated to characterize the rearing environment.

RESULTS AND DISCUSSION
The SSF inclusion into the Tilapia feed before extrusion resulted in the partial or total loss of enzyme activity, except for phytase, which had no influence from process temperature.Treatments showed significant differences (p<0.05) for b-glucosidase, amylase, and protease in function of treatments, with a partial loss of activity in the extrusion process.
A significant difference (p<0.05) was observed for pectinase xylanase and endoglucanase regarding the tested levels.The activities of these enzymes were totally lost during processing and there was no difference compared to control (Table 2).
The inclusion of 8000 ppm SSF after extrusion enhanced enzyme activity if compared to those of before-extrusion and control.The supplementation of 8000 ppm SSF before extrusion showed similar behavior as the 4000 ppm after extrusion, which had no difference with the highest inclusion dose after extrusion.  Means following by the same letter are similar by the Tukey's test (p<0.05). 2 IU.g -1 (International unit.gram - ) 3 AU.g -1 (Amylase unit.gram - ). 4 PU.g -1 (Phytase unit.gram - ).
After the enzymes analyses, Nile tilapia performance was assessed for 56 days.Tables 3 and  4 display the water quality and performance parameters, respectively.
There was a difference among treatments (p<0.05) for final weight, weight gain, feed conversion, and SGR.However, no differences were found for final weight and weight gain among the treatments with SSF inclusion after extrusion.Moreover, the addition of 800 ppm SSF before extrusion had no difference with that of 400 ppm after extrusion.Therefore, the fish fed AE800 obtained 40.29% and 18.43% more weight gain (p> 0.05) compared with control and BE800, respectively.For feed conversion, there were differences (p>0.05)among the treatments.It is noteworthy that any of the diets supplemented with SSF presented an improved conversion, with AE800 being the most effective treatment.
SSF inclusion and extrusion processing also influenced SGR (p <0.05), among which DE800 enhanced tilapia growth.Thus, feed conversion and SGR results were better in the treatment DE800, being 31.96 and 22.22% higher compared to the control, respectively.
On the other hand, there was no effect (p>0.05) of the SSF inclusion levels on final biomass, biomass gain, and survival rate.Likewise, the inclusion had no influence (p>0.05) on Nile tilapia body composition (Table 5).The heat in an extruder feed preparation promotes partial denaturation of the SSF enzyme complex.However, it can be added at higher levels to compensate for these losses.Yet, for improved responses, it should be added after extrusion since it provides a better animal performance and a lower amount and complexity of waste.
The b-glucosidase is a type of cellulase that acts in the final cellulose decomposition process by hydrolyzing cellobiose (AJWA; TABATABAI, 1994).We observed that b-glucosidase activity was partially reduced when subjected to extrusion.This result is similar to that found by Passos et al. (2008), who observed a reduction of b-glucosidase activity under high temperatures.Enzymes are proteins and therefore can be partially denatured by the high temperatures of an extrusion process.Gomes et al. (2007) mentioned that pectinase and xylanase are key enzymes involved in cell wall degradation.Pectinase hydrolyzes pectin present in the middle lamella and primary wall of plant cells.Yet xylanase acts on xylan molecules by endogenous and exogenous mechanisms decomposing hemicellulose, which is one of the major components of plant cell walls.The same authors also pointed out that high temperature can cause total loss of pectinase and xylanase activities, as demonstrated in the enzyme activity assay developed in this study.
This enzymatic behavior also corroborates the findings of Uenojo and Pastore (2007), who emphasized a maximum pectinase activity at 50 °C; they also stated that from this temperature the enzyme activity tends to reduce or stop, as observed here.When evaluating xylanases from different sources and their inclusion in wheat based diets, Reis et al. (2001) verified that a few types of xylanase have a peak activity at 75 °C, others at 50 °C, from these optimum temperatures on activity tends to decrease or be lost, as observed in this study.
Endoglucanase is responsible for starting the hydrolysis of cellulose and for random cleavage of internal glycosidic bonds of a cellulose chain (lignocellulose fiber), making them more exposed.Similarly to pectinase and xylanase, endoglucanase requires strict control of process temperature for preventing its denaturation (NELSON;COX, 2014).In this study, we observed a complete loss of endoglucanase activity due to the extrusion process conditions.
Amylase is a digestive enzyme of pancreatic origin which acts on polysaccharides in the chyme to enter the lumen of the duodenum (MOURA et al., 2007).In a study characterizing the corn-malt amylase activity, Biazus et al. (2006) noted a partial reduction of this enzyme activity at temperatures above 90 °C, with an optimum point at 55 o C.These results corroborate those of the present study where amylase activity partly decreased due to the extrusion process.
Concerning protease, Giongo (2006) mentioned a wide variety of physiological functions performed by this enzyme, leading essential metabolic and regulatory functions through catalyzing protein degradation.This author also reported a partial loss of enzyme activity when it is subjected to temperature rise, confirming the results 1

Table 1 .
Composition of experimental diets (natural matter).

Table 2 .
Enzyme activity of the SSF enzyme complex added at different levels and extrusion process steps for Nile tilapia diets.

Table 3 .
Water quality variables obtained during the experimental time.

Table 4 .
Performance of Nile tilapia juveniles fed diets containing SSF enzyme complex added at different feed processing steps

Table 5 .
Body composition of Nile tilapia juveniles fed diets containing SSF enzyme complex added at different feed processing steps.