Protein Requirements of Fattening Nile Tilapia ( Oreochromis niloticus ) Fed Fish Meal-Free Diets

: This study aimed to determine the protein requirements of the fattening phase for Nile tilapia ( Oreochromis niloticus ) fed fish meal-free diets. A total of 75 Nile tilapia were maintained in a water recirculation system, and five isoenergetic diets were formulated with increasing protein levels encompassing three repetitions each. The findings revealed that protein levels significantly affected ( p < 0.05) certain Nile tilapia performance, yield, and composition parameters. The determined parameter values clearly indicated that Nile tilapia can be fed fish meal-free soybean meal and corn-based diets. Furthermore, the metabolic plasticity of this species concerning dietary protein concentrations was also demonstrated, with adequate performance results achieved in treatments containing from 267 to 294 g/kg digestible protein (DP), or 298 to 327 g/kg crude protein (CP), where the balance between essential and non-essential amino acids and energy resulted in adequate performance correlated to satisfactory feed conversion values and filet yields and composition. A DP concentration of 267 g/kg (298 g/kg CP) is recommended when offering corn and soybean meal-based diets during the Nile tilapia fattening phase to fish weighing between 400 and 700 g.


Introduction
Aquaculture plays a fundamental role in ensuring global food security and is the food production sector that has expanded the most in the last 50 years, with total fish production growth estimates of over 15% by 2030 [1].In this sense, Nile tilapia (Oreochromis niloticus) stands out as the third most cultivated species, contributing to 9% of the 2020 global production [2].About 92% of the worldwide production of this species employs commercial feed [3], representing around 70% of total production costs [4].
Among basic tilapia feed nutrients, protein stands out as a fundamental dietary component, significantly affecting Nile tilapia performance [5].Inadequate digestible protein (DP) levels can result in several issues, such as reduced feed consumption, consequently compromising fish growth [6].
Protein is an essential component in the construction and maintenance of cellular structures, tissues, and organs, playing a crucial role in metabolic processes [7].When ingested, proteins undergo a hydrolysis process [8], and part of the derived amino acids are absorbed, following several possible metabolic pathways.The nitrogen resulting from oxidative amino acid processes is then excreted into the water in the form of ammonia (NH 3 ) [9], which can cause significant environmental impacts, contributing to issues such as eutrophication and other negative water quality effects [10,11].Therefore, fish diet formulations should be adjusted to meet the specific needs of each species, ensuring an adequate nutrient balance to optimize animal growth and health and minimize negative environmental impacts.
Recently, several studies have determined the optimal levels of dietary protein for Nile tilapia at different stages of development: 318 ± 6.50 g/kg [12][13][14][15][16], 292 ± 18.19 g/kg [17][18][19][20][21], and 255 ± 16.5 g/kg [6, 22,23] for tilapia with an average final weight of 12 ± 2.5 g, 57 ± 2.39 g, and 415 ± 3.37 g, respectively.These requirements are lower than those presented in the NRC (2011) [7], where fish with an average weight of less than 20 g, from 20 to 200 g, and from 200 to 600 g have a requirement of 400, 340, and 300 g/kg of crude protein, respectively.In this sense, it is worth noting that the evolution of the tilapia production chain, combined with the use of new strains with higher productivity, demands new studies to better meet the nutritional requirements of this species.In addition, there is a lack of studies on protein requirements in the fattening phase for fish with an average weight above 500 g [24].
To meet these nutritional requirements, fish meal has been recognized as the most appropriate protein source option for this species, due to its high nutritional value and palatability [25].However, in addition to its production having plateaued for some years, reduced fish meal availability was noted between 2018 and 2020, contradicting market demands.This decline, driven by rapid aquaculture production growth and use by other animals, has led to significant fish meal price increases [2].
In this sense, plant protein sources have emerged as a promising alternative, capable of partially or fully replacing fish meal in Nile tilapia diets.This approach not only provides satisfactory results in terms of performance and immune response enhancement but also contributes to reduced total feed costs [26].
In general, studies investigating the protein needs of tilapia employing fish meal-free feed [6, [27][28][29][30][31] have reported positive results compared to other studies assessing similar developmental stages but employing diets containing animal ingredients.The developmental stage of the fish directly influences their protein requirement, with a tendency for this requirement to decrease with the increasing weight and age of the fish [7,8,32,33].However, when evaluating studies conducted on the protein requirement of Nile tilapia, Meurer et al. (2024) [24] identified that these studies have focused on the early developmental stages of this species, with the few studies conducted during the fattening phase (with a final average weight above 500 g) showing a tendency for an increase in the protein requirement compared to the growth phase of these fish.Additionally, although it has been scientifically proven that it is possible to use fish meal-free feed in Nile tilapia feeding, no study has been conducted on this developmental phase.Therefore, this study aimed to determine Nile tilapia fattening phase protein requirements employing fish meal-free diets.

Materials and Methods
This study was carried out at the Aquaculture Technology Laboratory (Lataq) belonging to the Federal University of Paraná (UFPR), located at the Advanced Jandaia do Sul Campus (PR, Brazil).The experimental procedure protocol was approved by the UFPR Palotina Sector Ethics Committee on the Use of Animals (Protocol n. 05/2021-CEUA/Palotina).

Fish Subjects and Experimental Design
The experiment lasted 84 days, beginning at the end of December 2022.A total of 75 adult Nile tilapia (Genetically Improved Farmed Tilapia-GIFT lineage) with an average initial weight of 412 ± 8.06 g were acclimatized for 60 days.After this period, the fish were anesthetized by immersion in a 50 mg/L eugenol solution, weighed, and randomly distributed into five treatments encompassing three repetitions each.
A recirculating aquaculture system (RAS) consisting of 15 circular 1000 L tanks installed in a cultivation greenhouse was used.The tanks were connected to a 2000 L mechanical filtration tank and a 30,000 L biofiltration tank.Each 1000 L tank was considered an experimental unit.A water recirculation rate of 12 times a day was maintained in each experimental unit during the experiment.The 1000 L tanks were aerated using a porous hose connected to an air system that was connected, in turn, to a 0.5 hp radial blower, with an individual water inlet and outlet.
All experimental units were equipped with a self-siphoning system at the bottom of the tank, with a centralized slope of 5% and a central outlet for the removal of feces and other waste.The mechanical filtration and biofiltration tanks were siphoned weekly (500 L) to remove residues.The biofiltration tank was partly occupied by Eichornia crassipes, which were removed as necessary, maintaining between 50% and 80% total coverage.
The fish meal-free diets were prepared by finely grinding soybean meal and corn through a 0.7 mm sieve and subsequently mixing these components with other ingredients, according to each formulation (Table 1), using a mixer with a mixing capacity of 63 L (27.5 rpm, 10-min mixing cycle).The feed processing incorporated 175 g/kg of water, added homogeneously, followed by processing in an extruder (Ex laboratory, Exteec ® Máquinas, Ribeirão Preto, Brazil) employing a 3 mm matrix, obtaining 5.5 mm extruded pellets after the drying process.Soybean oil was added by spraying.
The produced feed was dried at 55 • C in an air recirculation oven (MA035/1152, Marconi, SP, Brazil) for 24 h, cooled, and stored under refrigeration.This procedure provided 100% buoyancy extruded pellets, as determined in buoyancy and extruder integrity tests from zero to sixty min.
Table 2 lists the water quality parameter results of the experimental systems during the Nile tilapia fattening phase.The data are categorized into five treatments and encompass several variables.Total ammonia, nitrite, and dissolved orthophosphate (PO 4 3− ) concentrations were not significantly different (p < 0.05) between treatments (Table 2).All physicochemical water parameter values were similar between treatments and within the appropriate standards established in the literature [38].

Nile Tilapia Performance and Body and Fillet Composition
At the end of the experiment, all fish were fasted for 24 h, anesthetized by immersion in a 120 mg/L eugenol solution, counted, and weighed.Weight gain (WG), daily weight gain (DWG), feed conversion ratio (FCR), specific growth rate (SGR), and protein efficiency rate (PER) were calculated.Six fish per treatment (two individuals per experimental unit chosen at random) were dissected to determine carcass yield (CY), fillet yield (FY), the hepatosomatic index (HSI), visceral fat (VF), and the viscerosomatic index (VSI).
One whole fish, two carcasses, and the fillets of two other fish were randomly chosen from each experimental unit to determine bromatological compositions.All analyses were carried out according to the AOAC methodologies (2005) [34] for dry matter (DM; method no.930.15), crude protein (CP; method no.984.13), ether extract (EE; method no.920.39), and mineral matter (MM; method no.924.05).The respective values of each parameter were then determined for whole fish, carcasses (fish containing no viscera), and fillets.

Statistical Analyses
The obtained data were subjected to an analysis of variance (ANOVA).When a significant effect (p ≤ 0.05) was identified, the Tukey test was applied to discriminate the means.A regression analysis was also performed to determine which model best fit the data.The data were adjusted using the combination of linear-plateau regression (broken-line) and quadratic polynomial regression models when the regression models were significant or presented a high coefficient of determination (R 2 > 70).This adjustment was performed to assess the optimal protein requirement, with the first point defined by the intersection of the quadratic regression line when crossing the plateau of the broken line, according to the model described by [39].Statistical analyses were performed with the RStudio 4.3.1 software [40].

Zootechnical Performance
The zootechnical performance results of Nile tilapia fed fish meal-free diets containing increasing protein levels during the fattening phase are presented in Table 3.The offered protein levels significantly influenced (p < 0.05) WG, DWG, SGR, FCR, and PER.Better WG, DWG and SGR results were observed in the T27 and T29 treatments.The worst result was noted for the T22 diet, and the T24 and T32 diets did not differ from each other or from the other treatments.A linear PER reduction was observed y = 3.55 − 0.0057x, R 2 = 0.94 ) due to increasing dietary protein levels.  Final weight. 4Weight gain = (final average weight − initial average weight).5 Daily weight gain = ((final average weight − initial average weight)/experimental days). 6Feed consumption.7 Specific growth rate = ((ln final average weight − ln initial average weight)/(experimental days) × 100). 8Protein efficiency rate = ((weight gain (g)/protein ingested (g)). 9Feed conversion ratio = (feed intake (g)/weight gain (g)).a-d Means within row with different superscripts differ (p < 0.05).
The FCR (Figure 1) exhibited a quadratic effect depending on the dietary protein increase, with maximum points at 285.76 g/kg DP (Figure 1a) and 319.07 g/kg CP (Figure 1b).When analyzing the FCR using the combination of linear-plateau regression (broken-line) and quadratic polynomial regression models, the intercept points of the lines indicate values of 267.36 g/kg DP and 298.75 g/kg CP.
The FCR (Figure 1) exhibited a quadratic effect depending on the dietary protein increase, with maximum points at 285.76 g/kg DP (Figure 1a) and 319.07 g/kg CP (Figure 1b).When analyzing the FCR using the combination of linear-plateau regression (brokenline) and quadratic polynomial regression models, the intercept points of the lines indicate values of 267.36 g/kg DP and 298.75 g/kg CP.

Body Performance
Nile tilapia body yield parameters in fish offered fish meal-free diets containing increasing protein levels in the fattening phase are presented in Table 4.The CY of Nile tilapia was significantly influenced (p < 0.05) by dietary protein levels, with the best performance obtained in T29 and the worst, in T22.No difference between the other treatments were observed, and no significant differences (p > 0.05) were observed for HSI, VF, VSI, TL, SL and TH. 1 Coefficient of variation. 2Fillet yield = (fillet weight (g)/body weight (g)). 3Carcass yield = ((carcass weight/body weight) × 100). 4Hepatosomatic index = ((liver weight/body weight) × 100). 5Visceral fat = ((visceral fat weight/body weight)/100). 6Viscerosomatic index = ((viscera weight/body weight)

Body Performance
Nile tilapia body yield parameters in fish offered fish meal-free diets containing increasing protein levels in the fattening phase are presented in Table 4.The CY of Nile tilapia was significantly influenced (p < 0.05) by dietary protein levels, with the best performance obtained in T29 and the worst, in T22.No difference between the other treatments were observed, and no significant differences (p > 0.05) were observed for HSI, VF, VSI, TL, SL and TH.The FY (Figure 2) exhibited a quadratic effect depending on the dietary protein increase, with a maximum value of 281.45 g/kg PD and 314.19 CP.When analyzing the same parameter using a combination of linear-plateau regression (broken-line) and quadratic polynomial regression models, the intercept points of the lines indicated a value of 256.78 g/kg DP and 286.20 g/kg CP (Figure 2a,b).
The FY (Figure 2) exhibited a quadratic effect depending on the dietary protein increase, with a maximum value of 281.45 g/kg PD and 314.19 CP.When analyzing the same parameter using a combination of linear-plateau regression (broken-line) and quadratic polynomial regression models, the intercept points of the lines indicated a value of 256.78 g/kg DP and 286.20 g/kg CP (Figure 2a,b).

Bromatological Composition
Table 5 presents the bromatological moisture, crude protein, mineral matter, and ether extract values of whole Nile tilapia, carcasses, and fillets offered increasing protein levels.A linear decrease (p < 0.05) was observed for the carcass ether extract parameter ( = 155 − 2281, ² = 0.60) with increasing protein levels.The same parameter evaluated by the means test revealed that the highest value was obtained in T22 and the lowest value was obtained in T32.The other treatments did not differ from each other, and the other bromatological parameters were not influenced (p > 0.05) by the dietary treatments.

Bromatological Composition
Table 5 presents the bromatological moisture, crude protein, mineral matter, and ether extract values of whole Nile tilapia, carcasses, and fillets offered increasing protein levels.A linear decrease (p < 0.05) was observed for the carcass ether extract parameter (y = 155 − 2281x, R 2 = 0.60) with increasing protein levels.The same parameter evaluated by the means test revealed that the highest value was obtained in T22 and the lowest value was obtained in T32.The other treatments did not differ from each other, and the other bromatological parameters were not influenced (p > 0.05) by the dietary treatments.

Discussion
Proteins, alongside peptides, are composed of amino acids.Thus, protein requirements, whether crude or digestible, should be ultimately translated in terms of amino acid requirements.Amino acids can be essential or non-essential, with their essentiality correlated to the possibility of a given individual's capacity to synthesize these elements in adequate amounts, which may vary according to growth phase or a specific momentary physiological situation.The importance of assessing protein levels should, thus, be verified, as amino acid requirements should be the preferred parameters.However, protein assessments are made easier by the fact that their analyses are much cheaper than amino acid assessments and can be performed easily in feed mills.Furthermore, specifically concerning this study, soybean meal composition values vary only slightly [36].Thus, when compared to other ingredients, feeds formulated with these ingredients allow for adequate indirect amino acid-level estimates.
Nile tilapia performance is directly influenced by dietary protein levels, which are, in turn, associated with several biochemical and physiological processes that directly affect fish growth, development, and health.Proteins play crucial roles in tissue synthesis, metabolic regulation, immune responses, and other vital fish organism functions [41].Therefore, if fish are offered inadequate dietary protein amounts or imbalanced essential amino acids, protein synthesis can be compromised, negatively affecting performance indices [7,8,42].This was evidenced herein by the data analysis according to the Tukey test or the linear-plateau (broken-line) and quadratic polynomial regression models (Table 3, Figures 1 and 2), which revealed that both the lowest and the highest protein levels were associated with Nile tilapia performance declines.This non-linear response highlights the need for precise dietary formulations, highlighting that adequate protein levels are essential to optimize Nile tilapia growth and body composition.This, in turn, suggests an optimized Nile tilapia response to a specific DP concentration, indicating the importance of adequate diet formulations to maximize fish growth rates during the fattening phase.
As stated previously, our findings indicate that varying the dietary protein concentrations significantly influences Nile tilapia performance indices (Table 3).The T27 treatment resulted in the highest values for several of the assessed variables.These results are in line with previous studies indicating tilapia sensitivity to diet composition variations, especially concerning DP [22,29,31,[43][44][45].
The relationship between the WG, SGR, and FCR is complex and interdependent.A direct and proportional relationship was observed between the SGR and FCR, where better results in these indices resulted in a proportional increase in the WG relation to the initial weight.This study demonstrated that although there were no significant differences in protein levels between the T24 and T32 treatments, better WG and SGR were observed in the T27 treatment, followed by a decline from the T29 treatment.In the same way, the best FCR indices were provided by T27, with a worsening of higher protein levels.This trend was corroborated by Liu et al. (2017) [46], who, when evaluating Nile tilapia with a final weight average of 550 g, determined 293.1 g/kg of CP for the maximum WG, noting a reduction after this level.Similarly, Carneiro et al. (2020) [6] observed the best WG and SGR in fish fed 240 g/kg of DP, significantly influencing the WG (p < 0.05) and the efficiency of DP utilization.
The increase in protein levels resulted in a decrease in the PER (Table 3), reflecting the efficiency of converting the protein consumed into body weight gain.Additional studies [31,43,[46][47][48][49] support this observation, suggesting that isoenergetic diets with high protein levels use protein as an energy source, justifying the linear reduction in the PER.
The FCR is a crucial parameter, correlating the WG and feed intake, and is essential to evaluate the effectiveness of the diet.Based on the quadratic polynomial model, the best FCR was estimated at 1.82, corresponding to a requirement of 285.76 g/kg DP (Figure 1a).The combined approach of the broken line and quadratic polynomial models is preferable for a more accurate analysis of protein requirements, as suggested by Lamberson and Firman (2002) [50] and Sakomura and Rostagno (2016) [39].Thus, DP requirements ranging from 267.36 to 304.20 g/kg were determined, indicating that DP levels outside this range negatively impact the FCR.Insufficient protein in the diet compromises growth [8], while excess protein increases metabolic costs [51], contributing to environmental pollution due to excess nitrogen excreted [9], in addition to increasing the cost of the diet, since protein is the ingredient with the greatest economic value [5].
The literature indicates that the protein requirement of fish is influenced by the growth stage, being higher in the initial phase due to the greater metabolic intake [7,8,32,33,52,53].Studies with Nile tilapia at similar stages of development indicate that the FCR of this study is slightly above the values found in the literature, despite presenting a lower protein requirement [22,[54][55][56].This variation can be attributed to the initial weight of the fish in the comparative studies.For example, Green et al. (2019) [22] determined a requirement of 277 g/kg DP with an FCR of 1.40 in a study with tilapia ranging from 32 to 545 g.Fernandes Junior et al. (2016) [55] estimated a requirement of 289 g/kg DP for an FCR of 1.45 in fish from 148 to 800 g, while Costa et al. (2009) [54], working with tilapia from 80 to 1000 g, found a protein requirement of 320 g/kg CP for an FCR of 1.63.On the other hand, Koch et al. (2017) [57], when evaluating tilapia from 450 to 800 g, observed an FCR of 1.72 with a requirement of 320 g/kg PD Concerning CY (Table 4), treatment T29 presented higher values than the other treatments.This finding is similar to that observed by Youssef et al. (2023) [58], who reported the highest CY in fish (7 to 45 g) fed a diet containing 280 g/kg CP.Similarly, Furuya et al. (2005) [59], when evaluating the impacts of reducing protein levels in Nile tilapia diets (5 to 125 g) through the concept of ideal protein, observed a quadratic effect on CY, with the maximum estimated point in fish fed diets with 277 g/kg of DP.This factor may be related to a better amino acid balance in these diets, which can optimize the utilization of the nitrogen fraction by fish, improving efficiency in body protein synthesis.This, in turn, contributes to better muscle growth and a higher carcass yield [59,60].
Another important parameter comprises FY (Figure 2), as the fillet is the most noble Nile tilapia product.Herein, the best DP value was determined as 256.78 g/kg according to the broken-line analysis.Fernandes Junior et al. ( 2016) [55] emphasized the importance of cost-benefit analyses and encouraged a choice of ideal protein levels based on the desired final Nile tilapia product.These authors evaluated five DP (240, 260, 280, 300, and 320 g/kg DP) and two digestible energy (DE) (13.4 and 14.65 MJ DE kg −1 diet) levels, reporting the highest profitability at 300 g/kg DP/13.40MJ kg −1 when aiming at fillet production.In contrast, [29] reported fillet yields for Nile tilapia weighing from 30 g to 130 g as varying according to weight, with heavier fish resulting in greater fillet yields.
Higher protein levels did not influence either the HSI, which may be indicative of hepatic glycogen synthesis and reserves, or the VF, indicative of lipid synthesis and reserves.Both parameters result in excess dietary energy or nutritional imbalances.These findings are similar to those reported by Carneiro et al. (2020) [6], who also assessed Nile tilapia fish meal-free diets and observed no statistical differences for the HSI and VF concerning different protein levels.
The VSI also was not affected by the different DP treatments, although Youssef et al. (2023) [58] point out that reducing DP levels results in higher HSI and VSI values, due to increased starch levels, which may be deposited in the liver and viscera as fat.The TL, SL and TH morphometric measurements (Table 4) also were not influenced by the treatments.Such values are critical, as morphometric indices favor the industrial yield of tilapia fillets [61].
Dietary protein quantity and quality directly affect the body's ability to synthesize new proteins, including fish muscle proteins.A diet containing limited protein levels may restrict the availability of essential amino acids required for adequate protein synthesis [62,63].Thus, the increased body protein deposition reported by several studies [6,18,45,46,64,65] may be associated with an increased protein synthesis capacity.
In the present study, composition in terms of moisture, protein, and mineral matter in whole Nile tilapia, carcasses, or fillets (Table 5) was not influenced by the DP treatments, corroborating previous studies [19,22,23,66].A linear reduction was, however, observed with increasing dietary DP levels, corroborating previous studies identifying higher lipid contents in response to reduced DP levels [22,48,67].This may be associated with the possible conversion of excess carbohydrates, particularly glucose, into lipids through the lipogenesis metabolic process [68].
The results of the present study on the protein requirement of tilapia in the fattening phase with fish meal-free diets are pioneering, as identified by Meurer et al. (2024) [24].However, for younger stages, there is some divergence in the literature between studies that used or did not use corn soybean meal-based diets.Studies on fingerlings (0.5 to 25 g) have reported a requirement of 319 ± 5.06 g/kg CP in recirculating aquaculture systems (RAS) and 340 ± 10.15 g/kg CP in excavated tanks or net tanks [24].Bomfim et al. (2008) [28] and Furuya et al. (2000) [27] determined a requirement of 280 and 320 g/kg CP in RAS systems and excavated tanks, respectively, in the fingerling phase, but using fish meal-free diets.Regarding the juvenile phase (25 to 200 g), while the protein requirement determined in the RAS system was 300 g/kg DP [69], Carneiro et al. (2017) [29], with fish meal-free diets, determined a requirement of 283 g/kg.In the growth phase (200 to 500 g), the protein requirement of 245 g/kg DP determined in the RAS system [43] was similar to the requirement of 240 g/kg DP [6] and 243 g/kg DP [31], determined with fish meal-free diets in RAS and net tank systems, respectively.These results show that the use of ingredients with different nutritional profiles, such as soybean meal and fishmeal, provides a distinct protein composition in the diets, which can influence the protein requirement of the animals.
According to the parameters evaluated herein, Nile tilapia can indeed be fed a soybeanand corn meal-based fish meal-free diet.The species' metabolic plasticity concerning dietary protein concentrations was also verified, with adequate performance results noted for DP treatments containing 268 to 294 g/kg, or 298 to 327 g/kg CP.In this sense, the balance between essential and non-essential amino acids and energy results in adequate Nile tilapia performance correlated to good feed conversion values, fillet yields, and composition.

Conclusions
A DP concentration of 267 g/kg (298 g/kg CP) is recommended for Nile tilapia that are fed a corn soybean meal-based diet and that weigh between 400 and 700 g in the fattening phase.

Figure 1 .
Figure 1.Feed conversion of Nile tilapia fed diets containing different concentrations of DP (a) and CP (b) during the fattening phase, analyzed using the combination of linear-plateau regression (broken-line) and quadratic polynomial regression models.

Figure 1 .
Figure 1.Feed conversion of Nile tilapia fed diets containing different concentrations of DP (a) and CP (b) during the fattening phase, analyzed using the combination of linear-plateau regression (broken-line) and quadratic polynomial regression models.

Figure 2 .
Figure 2. Fillet yield of Nile tilapia fed diets containing different concentrations of DP (a) and CP (b) during the fattening phase, analyzed using a combination of linear-plateau regression (brokenline) and quadratic polynomial regression models.

Figure 2 .
Figure 2. Fillet yield of Nile tilapia fed diets containing different concentrations of DP (a) and CP (b) during the fattening phase, analyzed using a combination of linear-plateau regression (broken-line) and quadratic polynomial regression models.

Table 1 .
Experimental diets and ingredients composition with increasing levels of protein for Nile tilapia during the fattening phase.

Table 2 .
Water quality parameters determined in experimental fattening phase systems for Nile tilapia offered diets containing increasing protein levels.

Table 3 .
Performance parameters of Nile tilapia fed different levels of protein in the fattening phase.

Table 4 .
Body performance parameters of Nile tilapia fed different levels of protein in the fattening phase.

Table 4 .
Body performance parameters of Nile tilapia fed different levels of protein in the fattening phase.

Table 5 .
Body and fillet chemical compositions of Nile tilapia fed different levels of protein in the fattening phase.

Table 5 .
Body and fillet chemical compositions of Nile tilapia fed different levels of protein in the fattening phase.