Growth and hematology of pacu subjected to sustained swimming and fed different protein levels

The objective of this work was to evaluate the effect of sustained swimming and dietary protein levels on growth and hematological responses of juvenile pacu (Piaractus mesopotamicus). A completely randomized design was used in a 3x2 factorial arrangement, with three levels of dietary protein (24, 28, and 32% crude protein), two rearing conditions (sustained swimming or motionless water), and 15 replicates. Fish were subjected to sustained swimming at the velocity of two body lengths per second (2 BL s‐1), for 45 days. The level of dietary protein and the swimming conditions affected the performance, growth, and hematological profile of pacu. Swimming conditions influenced nutritional factors, increasing daily weight gain, specific growth rate, number of erythrocytes, mean corpuscular volume, and mean corpuscular hemoglobin. Fish under sustained swimming and fed with 24% crude protein showed better growth performance, with higher specific growth rate (4.11±0.88) and higher daily weight gain (2.19±0.47 g per day). Sustained swimming can increase the productive performance of pacu and simultaneously reduce dietary protein levels.


Introduction
Pacu (Piaractus mesopotamicus) is an omnivorous fish species with high growth rates, rusticity, and easy adaptation to artificial feeding.It lives in the Paraná, Paraguay, and Uruguay Rivers (Saint-Paul, 1986), and is a relevant fish species in Brazil.
Currently, a challenge for the fish farming industry is to reduce production costs by improving culture techniques and developing more feasible and efficient diets for different species.Protein, used for promoting tissue growth and meeting energetic demands, is the most expensive diet component (Wilson, 2002).Efforts have been made towards reducing dietary protein content, especially animal protein, due to economic and environmental reasons (Tacon & Metian, 2008).In addition, several works have reported metabolic impairments in consequence of high protein levels (Singh et al., 2006;Portz & Furuya, 2012).
Different strategies have been proposed to decrease protein content in diets.One of them is protein sparing, in which a higher concentration of nonprotein sources of energy is used to meet physiological demands (Bicudo et al., 2012).Subjecting fish to exercise can lead to the protein sparing effect, make carbohydrates and lipids more efficient in meeting energetic demands, and redirect protein to an anabolic process (Arbeláez-Rojas & Moraes, 2010;Felip et al., 2013).These adaptions result in faster growth, better feed conversion ratios, increased survival rates, and decreased aggressive behavior (Moraes et al., 2009;Hackbarth, 2010).Sustained swimming also acts on blood flow, blood capillary diameter, and heart beating frequency (Sandblom et al., 2005).
There are studies on sustained swimming and animal nutrition in tropical freshwater fish (Hackbarth, 2010;Arbeláez-Rojas et al., 2011).However, few of them have shown beneficial effects of it on fish growth and metabolism.In general, each of these variables is studied independently, but it is reasonable to assume that the association of an ideal swimming activity and a well-balanced diet will result in more satisfactory responses on fish performance than sustained swimming or diet alone.One of the recurring concerns in the commercial breeding of fish is the reduction of dietary crude protein (CP) levels, since it would reduce production costs and the undesired effects of nitrogen excretion.
The objective of this work was to evaluate the effect of sustained swimming and dietary protein levels on growth performance and on hematological responses of juvenile pacu.

Materials and Methods
The experiment was carried out from February to April 2009 at the Universidade Federal de São Carlos, SP, Brazil (22 o 1'4"S, 47 o 23'57"W, at 860 m altitude).A completely randomized design was used, in a 3x2 factorial arrangement, with three levels of dietary protein (24, 28, and 32% CP), two rearing conditions (sustained swimming or motionless water) and replicates.Each fish was considered as an experimental unit.
Juvenile pacu were obtained from the São Geraldo commercial fish farm, located in Sertãozinho, SP, Brazil.The fish were held in 2,000 L tanks for one month in order to be acclimatized to the experimental conditions.After this period, 90 fish were anaesthetized with 40 mg L -1 of eugenol (Inoue et al., 2003) and individually tagged with a microchip, implanted in the abdomen.After seven days of recovery, fish were anaesthetized, weighted (23.9±4.7 g), gauged (10.6±0.77cm), and randomly distributed into six 200 L circular fiber tanks.Twenty-four hours after the biometry, the exercise protocol was performed in three tanks at the swimming velocity of two body lengths per second (BL s -1 ), in accordance with Moraes et al. (2009).The water velocity was generated by a NXDP4 pump (Grundfos do Brasil, Ltda., São Bernardo do Campo, SP, Brazil) as previously reported (Arbeláez-Rojas et al., 2011), and the flow was regulated every other day during the entire experimental period (45 days).Water speed was checked with a 2030 series mechanical flow meter (General Oceanics Inc., Miami, FL, USA).
The following water parameters were checked daily: dissolved oxygen (6.01±0.39mg L -1 ), temperature (27.9±1.8°C), and ammonia (0.05±0.001 mg L -1 ).Besides first biometry, two other biometric evaluations were made on the 22 nd and 45 th days of the experimental period.
32%), one level of total lipids (15%), and one of carbohydrate (25%).The concentration of total lipids and carbohydrate was based on previous studies developed in the laboratory and posteriorly published by Hackbarth (2010).Fish were fed three times a day to apparent satiation, during the entire experimental period.
At the end of the experimental span, eight fish from each tank were randomly sampled, and 1 mL of blood was withdrawn from the caudal vein with a heparinized plastic syringe.Hematocrit (Ht), hemoglobin concentration (Hb) (Drabkin, 1948), number of red blood cells (RBC) (Lima et al., 1969), mean corpuscular volume, and mean corpuscular hemoglobin were determined.Growth performance and feed utilization were determined and calculated as follows: daily weight gain (DWG; g per day) = (mean final weight -mean initial weight)/[mean initial weight x time (days)]; feed conversion ratio (FCR) = feed intake (g)/weight gain (g); and specific growth ratio (SGR) = 100 x (ln final weight -ln initial weight)/time (days).
Data were analyzed by two-way analysis of variance, and means were compared by Tukey's test, at 5% probability.Data were analyzed using the SAS software, v.8 (SAS Institute, Cary, NC, USA).

Results and Discussion
Both dietary protein levels and sustained swimming affected growth performance and the hematological profile of pacu.Fish fed with 24 and 32% CP showed the highest DWG, whereas the highest SGR was observed in fish fed with 24% CP (Table 2).This indicates that the 24% CP group -in which the protein anabolism was more noticeable -had a faster growth.According to several studies, the best CP values for pacu are between 26 and 27% (Abimorad et al., 2007;Bicudo et al., 2009), which supports the result obtained here.Considering that lower dietary protein reflects both lower diet costs and nitrogen wasting, the diet containing 24% CP is a sound alternative to feed pacu subjected to swimming at 2 BL s -1 .
The distinct values (Table 1) of gross energy (GE) and crude fiber (CF) did not change FCR (Table 2), showing that these conditions do not interfere in food intake of exercised pacu.Although high concentration of dietary fiber reduces food intake, increasing bolus size (Montagne et al., 2003) and satiation (Hansen & Storebakken, 2007), the fiber used as a diet component (carboxymethyl cellulose) did not affect FCR in all groups.However, diets with more than 11% of microfine cellulose reduce the growth of pacu (Rodrigues et al., 2010).It is possible that swimming attenuated the effect of high crude fiber concentration, since all fish exhibited good FCR values.
Independently of the dietary protein level, sustained swimming at 2 BL s -1 resulted in the best growth performance.Fish held under sustained swimming presented higher DWG and SGR (Table 2), probably as consequence of a more efficient mobilization, due to the exercise, of carbohydrates and lipids, in order to meet energetic demand.At the same time, exercise could have redirected protein to anabolic pathways.Similar results were reported for matrinxã (Brycon amazonicus) kept under sustained swimming for 72 days at 1 BL s -1 .An improvement in FCR was observed, increasing growth and leading to greater weight gain (Hackbarth & Moraes, 2006).In another work with the same species, with sustained swimming span extended to 90 days at 1-1.5 BL s -1 , Arbeláez-Rojas & Moraes ( 2010) observed a 20% increase in growth and weight gain.Sustained swimming at velocities between 1 and 2 BL s -1 improves pacu mean weight, mean length, FCR, and protein efficiency ratio, especially at 2 BL s -1 (Hackbarth, 2010).Therefore, sustained swimming can be considered a stimulator of growth and of carbohydrate and lipid utilization in order to meet the increased metabolic expenditures (Moraes et al., 2009).
Other studies on sustained swimming have shown that it stimulates fish growth and weight gain, bringing Table 2. Growth performance of pacu (Piaractus mesopotamicus) fed with three levels of dietary protein and subjected to sustained swimming (1) .
(2) Swimming velocity of 2 body length per second.DWG, daily weight gain; FCR, food conversion ratio; SGR, specific growth rate.
The dietary protein levels (Table 3) can explain the alterations observed in RBC (in the 32% CP group), in Ht and MCV (24 an 28% CP groups), and in MCH (24% CP group).Hematological alterations reflect the improvement of nutrient and oxygen transport, and of higher tissue oxygen uptake.Changes in MCV should be related to osmoregulatory status, cardiac dynamics, and blood flow, which change due to exercise.Matrinxã subjected to sustained swimming at 1 BL s -1 did not show alteration in hemoglobin, MCV, and MCHC levels (Hackbarth & Moraes, 2006); whereas under sustained swimming at 1.5 BL s -1 , Arbeláez-Rojas & Moraes (2010) reported increased contents of hemoglobin, Ht, and RBC.In both works, the authors concluded that exercise increases metabolic demand, which, in turn, caused a series of hematological adaptions.Hackbarth (2010) reported increased levels of hematocrit, hemoglobin, VCM, and HCM in pacu exposed to 2 BL s -1 and attributed these alterations to exercise.
In the present study, RBC and MCH parameters were also altered by exercise (Table 3).The increased RBC observed in fish fed with 32% CP may be explained by one of the following factors: 1, RBC release into the blood stream from a spleen contraction, where a large number of these cells is stored; or 2, a higher RBC value followed by lower Ht, MCV, and MCH, suggesting the presence of young erythrocytes from erythropoiesis.Studies with other species have also shown interference of CP on hematological variables.Camargo et al. (2005), for example, observed increased RBC, Hb concentration, and Ht values in silver catfish (Rhamdia quelen) fed with 50% CP.Abdel-Tawwab et al. (2010) also found changes in RBC, Hb, and Ht in response to dietary protein, when studying Nile tilapia (Oreochromis niloticus).However, specific mechanisms in fish involving erythropoiesis and dietary CP still remain to be elucidated.
The lower number of erythrocytes observed in fish under sustained swimming was probably offset by the larger volume of these cells.The number of RBC decreased about 13.8%, but an equivalent reduction of hematocrit was not observed.This means that the erythrocyte volume swelled, which is typical of the enhanced oxygen demand provided by aerobic exercise.The increases in MCV and MCH resulted from metabolic adaption to deal with higher oxygen uptake and transport.The same responses were observed in matrinxã subjected to intermittent sustained swimming at 12x12 hours vs. rest (Fabrizzi et al., 2013).It is reasonable to infer that the hematological responses found in the present study may be typical of the species, and that they may also depend on external factors, such as swimming speed and exercise type (Hackbarth & Moraes, 2006;Moraes et al., 2009;Arbeláez-Rojas & Moraes, 2010;Fabrizzi et al., 2013).Furthermore, the observed hematological changes may indicate metabolic adaption in order to meet oxygen demand.
Pacu fed with 24% CP and subjected to sustained swimming reached higher DWG and SGR than non-exercised fish (Table 4).Sustained swimming stimulated the use of other nutrients for energy demands, redirecting protein to an anabolic process.According to Table 3. Hematological parameters of pacu (Piaractus mesopotamicus) subjected to sustained swimming and fed three levels of crude protein (1) .Richards et al. (2002), sustained swimming accelerates body growth through stimulation of protein synthesis and hypertrophy of muscle fibers.Therefore, in pacu exercised at 2 BL s -1 , 24% CP is enough to promote higher growth values.Considering that the present study was conducted with specimens of the same brood, it may be inferred that the treatments are the main responsible for the observed effects.Similar responses have been reported in matrinxã fed with 28% CP and subjected to sustained swimming (Arbeláez-Rojas et al., 2011).
Sustained swimming and lower CP dietary levels in pacu interacted through the blood parameters RBC, MCV, and MCH (Table 4).Fish fed with 24% CP under sustained swimming showed reduction of RBC when compared with sedentary ones.However, an increase of MCV and MCH was observed in fish fed with 24 and 28% CP under sustained swimming.Low dietary protein levels associated with sustained swimming seem to reduce RBC and to increase erythrocyte volume and hemoglobin concentration.There is an inverse correlation between cell size and fish ability to endure sustained swimming.Therefore, pacu adapted to the sustained swimming condition showed reduced erythrocyte size and produced new RBC to meet the physiological demands.
Higher growth rates, ideal FCR, and protein sparing effect make sustained swimming at 2 BL s -1 an excellent support for pacu farming.

Conclusions
1. Sustained swimming and lower crude protein levels improve growth performance of juvenile pacu (Piaractus mesopotamicus) and favor hematologic adaptations in order to maintain the biological demands from the exercise.
2. It is possible to reduce the level of dietary protein in pacu farming if the fish are kept under sustained swimming at two body length per second.Table 4. Interaction between sustained swimming and dietary protein in pacu (Piaractus mesopotamicus) for growth and hematological parameters (1) .

Rearing condition
Dietary protein levels (g kg -1 ) 240 280 320 Daily weight gain (g per day) Sustained swimming (2)   1) Means±SD followed by equal letters do not differ by Tukey's test, at 5% probability.
(2) Swimming velocity of 2 body length per second.

Table 1 .
Formulation and chemical composition of the experimental diets.