DETERMINING THE DAILY DIGESTIBLE PROTEIN INTAKE FOR NILE TILAPIA AT DIFFERENT GROWTH STAGES *

This study aimed to estimate the utilization efficiency and optimal intake of digestible protein to maximize weight gain and feed conversion in Nile tilapia juveniles at different development stages. Four trials, each lasting 45 days, were performed using sexually inverted Nile tilapia juveniles with an initial average weight of 2.01 g (phase A), 14.26 g (phase B), 59.96 g (phase C), and 149.11 g (phase D). The experimental design was completely randomized with five treatments and four replicates. The levels of digestible protein ranged from 175 to 425 g kg-1 (phase A), 163 to 390 g kg-1 (phase B), 150 to 360 g kg-1 (phase C), and 138 to 330 g kg-1 (phase D). The maximum responses in terms of weight gain were obtained with protein intakes (digestible protein) of 88, 328, 713, and 855 mg fish-1 day-1, respectively, for phases A, B, C, and D. The maximum feed conversion ratio was obtained with protein intakes of 78, 272, 697, and 793 mg fish-1 day-1, respectively, for phases A, B, C, and D. The protein utilization efficiency was 52, 51, 51, and 50% for phases A, B, C, and D, respectively.


INTRODUCTION
The rapid growth of aquaculture in recent decades has intensified production systems, which have begun using high stocking densities and balanced diets with high nutritional quality (EL-SAYED and KAWANNA, 2008).The correct balance of these diets, especially in relation to the quantity of protein, can increase the nutrient use efficiency (PONTES et al., 2010), reduce the cost of feed supply (HISANO et al., 2015), and help reduce nitrogen excretion into the environment (BOTARO et al., 2007;ABDEL-TAWWAB and AHMAD, 2009;FERNANDES JR et al., 2016;BOSISIO et al., 2017).Therefore, the search for diets with high nutritional value that are environmentally friendly and ensure high profitability depends upon increased knowledge on the species produced, especially in relation to feed management and meeting their nutritional requirements.
Several studies have aimed to determine the requirement of Nile tilapia for protein (BOTARO et al., 2007;MEURER et al., 2007;TRUNG et al., 2011;OLIVEIRA et al., 2014); however, the results obtained in those studies have been conflicting, with values ranging from 240 to 400 g kg -1 , depending on the stage of development.In addition, there have been few studies on the daily recommended requirement of protein in milligrams per fish (LIEBERT et al., 2006;EL-DAHHAR, 2007).
Therefore, the objectives of this study were to determine the daily intake of digestible protein (DP) required to optimize the performance of Nile tilapia in different phases of development, and to determine the protein utilization efficiency.

MATERIAL AND METHODS
All the procedures used in the present study were approved by the Ethics Committee on Animal Use (CEUA), the Faculty of Agrarian and Veterinary Sciences, UNESP (Protocol No. 009999/14).
The experimental design was completely randomized.The growth phases comprised five treatments, with increasing levels of DP (varying from 175 to 425 g kg -1 in phase A; 163 to 390 g kg -1 in phase B; 150 to 360 g kg -1 in phase C; and 138 to 330 g kg -1 in phase D), with four replicates.
The experimental diets were prepared using the "dilution technique", proposed by FISHER and MORRIS (1970).Two diets were formulated for each phase, one with a high protein content and the other, protein-free.The diet with the high protein content was formulated by maintaining the recommended levels of energy, minerals, vitamins, amino acids, and at least 1.3-times the requirement for DP, according to FURUYA (2010).The protein-free diet was formulated to contain the same levels of energy, minerals, and vitamins as the high-protein diet (Tables 1 and 2).The intermediate levels of digestible protein used in each phase were obtained by appropriately blending the two basal diets (high protein content diet and protein-free diet).
The observed values were within the appropriate levels for fish farming (SIPAÚBA-TAVARES and SANTEIRO, 2013).

Ingredients
The fish were fed to apparent satiety four times each day.Performance was evaluated by assessing weight gain (WG), feed conversion (FCR), specific growth rate (SGR), and DP intake (DPI).Protein deposition (PD) was determined through a slaughter comparative study of a sample (10% of each experimental unit) at the beginning and end of each test.The animals were euthanized after 24 h of fasting by dipping in an anesthetic solution containing benzocaine.Fish were then stored in a freezer at -20°C, ground, and dried in an oven at 55°C.The samples were crushed in a ball mill, and sent to the laboratory for protein and dry matter analyses (AOAC, 2016).PD was calculated as follows: Bol. Inst.Pesca, São Paulo, 44(vol.esp.): 54 -63, 2017  3. Proportions of the high protein and protein-free diets and calculated concentration of digestible protein used for each growth phase.
Performance data were analyzed by the SAS program (2008), using analysis of variance (ANOVA).When statistical significance was observed, Duncan test (P < 0.05) was applied to compare means.The broken line model was adjusted according to the procedures described by ROBBINS et al. (2006).The DPI was obtained by meeting the ascending line with a plateau.To verify the adjustment of the obtained equations, the coefficient of determination (R² = SQ model / SQ total) was taken into account.
The results for the PD of fish were submitted to linear regression analysis, considering the DPI corrected for maintenance (cDPI) as an independent variable in the model broken line where: b = 437.5 mg, parameter used to estimate the requirement of DP for maintenance obtained from LIEBERT et al. (2006).PC = body weight.

Diets
The efficiency of protein utilization (k) was obtained by the relationship between the parameters L and R generated by the "broken line" model when adjusted to the PD and the cDPI, and k was calculated by the following equation: The results obtained for WG (mg fish -1 day -1 ) and FCR per day were submitted to linear regression analysis, considering the DPI (mg fish -1 day -1 ) as the independent variable, by the broken line mathematical model (ROBBINS et al., 2006): where: Y and X are the model variables and L, U, and R are model parameters; L (maximum model response), U (slope of the line), and R (intake for maximum response).The estimates of this function are valid if X < R; when X ≥ R, then Y = L.

RESULTS
The level of DP in the diet significantly affected the growth, utilization, and protein intake of tilapia in all phases of development (Table 4).The recommended level of digestible dietary protein for Nile tilapia, considering weight gain, were 350, 325, 300, and 275 g kg -1 for phases A, B, C, and D, respectively.
In phase A, the level of DP increased to 350 g kg -1 , increasing the WG and reducing the FCR, with no difference observed between the fish fed 385 and 425 g kg -1 DP.The SGR responded positively to increased levels of DP up to 385 g kg -1 , with the intake and PD being proportional to the increased protein level (Table 4).In phase B, the increase in the level of DP up to 325 g kg -1 increased weight gain, feed intake, SGR, and reduced the average FCR.The intake of protein and SGR were proportional to the increase in protein (Table 4).
In phases C and D, when the levels of DP were increased to 300 and 275 g kg -1 , respectively, the WG, SGR, and PD also increased, and the average FCR was reduced, since the intake of protein was proportional to the increase in protein level.In phase D, the feed intake was inversely proportional to the increased level of protein (Table 4).Differences in DP intake were observed in all phases for all weight groups, with smaller WG and higher FCR reported for fish fed with diets containing lower levels of DP.The SGR values also declined when fish were fed with diets containing lower levels of DP in each growth phase.The results of the variables in question also reduced linearly with the increase in fish size.In the present study, the SGR values ranged from 0.8 to 3.9% per day for the fishes that received the D1 and D2 diets in all stages of growth.
The maximum response predicted by the mathematical model broken line for WG was estimated at 265, 1085, 1894, and 1966 mg fish -1 day -1 with a DPI of 88, 328, 713, and 855 mg fish -1 day -1 for growth phases A, B, C, and D, respectively.Through the respective equations, the maximum estimated response for FCR was calculated at 0.999, 1.048, 1.485, and 1.542 mg fish -1 day -1 , with an PD intake of 78, 272, 697, and 793 mg fish -1 day -1 for growth phases A, B, C, and D, respectively (Table 5).
With increasing levels of DP in the diet, a linear increase in the PD was observed at different stages of Nile tilapia growth.Thus, by adjusting Bol.Inst.Pesca, São Paulo, 44(vol.esp.): 54 -63, 2017 the model broken line, the appropriate DPI to maximize PD was estimated at 96, 434, 722, and 779 mg fish -1 day -1 in phases A, B, C, and D, respectively.The protein utilization efficiency observed in this study i.e., 52, 51, 51, and 50% for groups A, B, C, and D, respectively, were obtained with the responses within the range where protein was not a limiting factor.Table 4. Performance of Nile tilapia juveniles fed different levels of digestible protein (DP) during developmental stages A, B, C, and D. IW = initial weight; WG = weight gain; FC = feed consumption; FCR = feeding conversion ratio; SGR = specific growth rate; IDP = ingestion of digestible protein, PD = protein deposition.Table 5. Equations with the broken line model for the productive performance variables (weight gain and feed conversion ratio) in Nile tilapia fed diets containing different levels of digestible protein.

Phase A
Weight gain (mg fish

DISCUSSION
The recommended digestible dietary protein required for Nile tilapia, considering weight gain, are close to those reported by ABDEL-TAWWAB et al. (2010), who determined the optimum value of 450 g kg -1 crude protein for tilapia, with 0.5 g and 350 g kg -1 for tilapia fish with 17 to 43 g of live body weight.EL-SAYED and TESHIMA (1991) found that dietary protein requirements decreased with increasing fish weight and age, and ranged from 400 to 450 g kg -1 for fish weighing less than a 1 g, and declined to 300 g kg -1 CP for fish weighing 46-260 g.
The reduced DP intake of fish fed diets with low levels of DP hindered optimal growth of animals, because the high demand of amino acids necessary for protein biosynthesis is indispensable for the deposition of tissues and subsequent weight gain (NRC, 2011;SILVA, 2014).The highest values of FCR for juvenile tilapia were also observed in animals fed diets with low levels of DP.These results are close to those found by FURUYA et al. (2000) and MEURER et al. (2007) (1.4 and 2.4, respectively).This can be explained by the deficiency of essential amino acids and excess digestible energy in relation to PD favoring the imbalance in the diet and consequently reducing weight gain (SILVA, 2014).
The results of this study more accurately describe the protein requirements during different phases of growth, since possible variations within the limits of 138 to 425 g kg -1 DP were tested.Variations in the requirements for crude protein associated with fish weights can be attributed to differences in the requirement of proteins in different weight classes (ABDEL-TAWWAB and AHMAD, 2009).EL-SAYED and TESHIMA (1991) concluded that the dietary protein requirement decreases with an increase in of the weight and age of fish, both for fry and juveniles of Nile tilapia.
It was also possible to confirm that reduced DP in the diets directly affected the PD of fish, and this was independent of the size of the fish that ingested the D1 and D2 diets with the lowest values of PD.Similar data were observed by MEURER et al. (2007), who observed that Nile tilapia weighing 10 to 30 g, fed with DP levels less than 200 g kg -1 , had a lower PD (16.07%) compared with those fed with higher levels of DP (300 g kg -1 ), that presented a PD of 17.81%.Protein intake levels interfere with the maintenance of whole body total protein and with protein synthesis and degradation (MOREIRA and POZZA, 2014).Linear relationships also exist between the ingestion of protein, growth, synthesis, capacity of protein synthesis and RNA activity (MENTE et al., 2011).Increased protein intake, which occurs with diets containing high levels of protein, led to an increase in the concentration of RNA (capacity of protein synthesis) and/or activity of RNA and alteration in the equilibrium between the synthesis and degradation of protein.Thus, diets containing high protein levels led to the production of fishes with high body protein.(MOREIRA and POZZA, 2014).
Similar results regarding protein utilization efficiency were observed by FURUYA et al. (2005) and GONÇALVES et al. (2009), who worked with Nile tilapia from a Thai lineage, and reported values of 519.7 and 506.8 g kg -1 , with average weight ranging from 5 to 125 g, and from 30 to 194 g, respectively.The results of the present study showed that protein utilization efficiency was very close between the different stages, showing that protein utilization in Nile tilapia is similar in all phases of growth, indicating that this species uses protein in the same way throughout its development.Thus, a single value for protein efficiency (51%) is suitable for use in the growth of Nile tilapia, as recommended by TRUNG et al. (2011).
Although several studies have been performed to determine the requirement of crude protein by tilapias (FURUYA et al., 2000;BOTARO et al., 2007), the proper levels required in the diet are not sufficient when consumption is not considered.Thus, protein intake expressed as an absolute value of the body mass of the fish, or weight gain, and then as a percentage of feed will provide a clearer understanding of the efficiency of the species studied (GLENCROSS et al., 2011;LUPATSCH, 2012).

CONCLUSION
The responses obtained from the broken line mathematical model determine recommendations Bol.Inst.Pesca, São Paulo, 44(vol.esp.): 54 -63, 2017 to optimize the performance of Nile tilapia in different stages of growth.To maximize the animals WG and feed conversion, the recommended daily intake of protein is 88 and 78 mg fish -1 day -1 (for fish weighting between 2.01 and 14.36 g), 328 and 272 mg fish -1 day -1 (14.26 to 64.73 g), 713 and 697 mg fish -1 day -1 (59.96 to 147.72 g), and 855 and 793 mg fish -1 day -1 (149.11 to 238.77 g), respectively.Growing Nile tilapia (2 to 250 g) have a protein utilization efficiency of 51%.

Figure 1 .
Figure 1.Graphical representation of the broken line mathematical function for Nile tilapia body protein deposition in growth phases A, B, C, and D receiving increasing levels of digestible protein (DP).(•) Fish response to diets (D1 and D5 in growth stages A, B, C, and D).

Table 2 .
Composition (g kg -1 ) of the high protein and protein-free diets used for Nile tilapia juveniles at each growth phase.