Effect of Diets With Different Lysine Levels on the Hemograms of Juvenile Tambaqui Colosssoma Macropomum (CUVIER, 1818)

Blood transports lysine and other nutrients derived from the diet and ensures good health and greater productivity for the sh. Therefore, this study aimed to verify the hematological behavior of juvenile tambaqui Colossoma macropomum nourished with levels of 9.72, 12.84, 15.96, 19.08 and 22.20 g.Kg -1 of total lysine, corresponding, respectively, to the supplementation levels with L-lysine: 0.00, 4.00, 8.00, 12.00, 16.00 and 20.00 g.Kg -1 of L-lysine. The blood of the sh was collected in the initial and nal periods by puncture of the caudal vein. The variables were validated by ANOVA and Tukey's test (p< 0.05). No differences were found for hemoglobin, hematocrit, CHCM, leukocytes of the eosinophil and monocyte types, and thrombocytes (p> 0.05). The changes observed in total leukocytes were not attributed to diets (p= 0.00). Means of CMV, HCM, lymphocytes and neutrophils, varied over the initial period, but not in regard to different lysine levels (p< 0.05). Thus, it is concluded that the varying lysine levels of the diets did not compromise the hematological parameters analyzed. regarding Given the above, the objective of this study was to investigate the inuence on the blood parameters when feeding juvenile tambaqui diets with different L-lysine levels.


Introduction
Hematological analysis in monitoring the sh physiology is a consolidated methodology, with simple sampling, and promotes reliable results regarding the animal's health (Aride et al. 2016(Aride et al. , 2018Castro et al. 2020;Lima et al. 2020;Nascimento et al. 2020;Oliveira et al. 2016Oliveira et al. , 2017Burgos et al. 2019).
Multidisciplinary hematology as prognostic device in environmental and xenobiotic stress-induced response in sh (Fazio et al. 2013;Vajargah et al. 2019;Aliko et al. 2018). It can also indicate when it is necessary to provide interventions that solve or mitigate any unfavorable condition during cultivation (Pereira et al. 2016). Blood count analyzes the levels of erythrocytes (erythrogram), leukocytes (leukogram), and thrombocytes (thrombogram) (Sula et al. 2020;Prado et al. 2016) and becomes more e cient when associated with the analysis of the chemical composition of the blood.
The components of blood are distributed in the following proportions: 3% is made up of hormones and enzymes, 7% is proteins, and 90% is water (Ranzani-Paiva et al. 2013), in the aqueous tissue, erythrocytes, leukocytes, and thrombocytes . Since the cellular concentration has the functions of protecting the biological system, eliminating residues from metabolism, and transporting gases and nutrients, it can be used to measure responses to exogenous functional changes, such as temperature, pollutants and stress, and endogenous changes, such as hormonal action, weight, length variations and nutritional conjunction resulting from the elements absorbed in the diet (Aride et al. 2016(Aride et al. , 2018Castro et al. 2020;Oliveira et al. 2016, Seriani et al. 2011Ranzani-Paiva et al. 2013).
The need for specialized diets to enhance the productive gains in sh farming has evolved proportionally with the increase in sh production (Nascimento et al. 2020). L-lysine is a typi ed crystalline amino acid for protein synthesis in tissues, which conditions greater growth, weight gain and other zootechnical performance indices (Richter et al. 2020). It also acts in the production of L-carnitine which contributes to lipid oxidation and helps the immune system by preventing degeneration response mechanisms (Grisdale-Helland et al. 2011;Furuya et al. 2013;Hua et al. 2019). Fish farming requires the sh's health to be protected by the body's defenses in order to make better use of the diet (Aride et al. 2016(Aride et al. , 2018Nascimento et al. 2020;Hamed et al 2020;Van Doan et al. 2019;Hossein et al. 2020;Gholampour et al. 2020;Rashidian et al. 2019Rashidian et al. , 2020 Active in the protection of vital processes against infectious disorders and stress, leukocytes are mediators of the innate immune system, and can present different types of granulocytic and agranulocytic cells according to the species of sh investigated (Ngugi et al. 2015;Lemos et al. 2018;Oliveira et al. 2016). Thrombocytes are small cells, ovoid, elliptical or fusiform, which act in the blood clotting process ). Although they are not leukocyte cells, thrombocytes are inserted as an active blood component in the organic defense of freshwater sh (Ishikawa et al. 2008;Oliveira et al. 2016).
Erythrocytes in the blood of teleost's represent on average 45% of the total blood volume, have a central nucleus and structures with oval to elliptical conformations, in addition to homogeneous cytoplasm containing vacuoles from the depreciation of cellular organelles, where hemoglobin is located (Ranzani-Paiva et al. 2013). Hemoglobin is usually con gured by 96% globins and 4% heme, a prosthetic group formed by iron and porphyrin groups that gives protein its red color (Prado et al. 2016). This amino acid macromolecule in uences the color, size and morphology of erythrocytes, is specialized in gas displacement, has a serum concentration which is dependent on the sex, age and nutritional status of the sh. Its alterations may characterize anemias and other dysfunctions (Tavares-Dias et al. 2009;Drumond et al. 2010;Prado et al. 2016;Signor et al. 2017). For these reasons, we may con rm that the FBC is applicable both in studies regarding the health of the sh and in the analysis of nutritional responses (Tavares-Dias et al. 2009;Signor et al. 2017;Aride et al. 2018).
The requirements for lysine and other nutrients for native sh species already established in Brazilian sh farming such as tambaqui Colosssoma macropomun (Cuvier, 1818) are still unclear (Chagas and Val 2003). Tambaqui is highly recommended for sh farming, has an omnivorous eating habit, ease in adapting to the farming systems used and, in a commercial sense, is well-appreciated in northern Brazil (Araújo- Lima and Gomes 2005;Rodrigues et al. 2014). Given the above, the objective of this study was to investigate the in uence on the blood parameters when feeding juvenile tambaqui diets with different L-lysine levels.

Ethical animal use
The experiment was developed in accordance with the rules of ethical principles for animal experimentation approved by the National Council for the

Initial considerations: acquisition, location and acclimation of sh
The experiment with L-lysine diets was carried out at the Aquatic Organisms Production Laboratory at Nilton Lins University, Manaus -AM, using juvenile tambaqui Colossoma macropomun with average initial biometric data of 33.54 ± 1.9 g and 11.72 ± 0.36 cm, from the Balbina Fish Farming Station (Aquaculture Training, Technology and Production Center), Presidente Figueiredo -Amazonas state, Brazil. The sh (N= 180) were randomly stored in polyethylene tanks (10 sh/tank) (310 L -1 ), with constant oxygenation and a biological lter. The siphoning and replacement of water in the tanks (20%), the dissolved oxygen levels (mg. L -1 ), temperature (°C), pH and nitrite (mg. L -1 ) were were measured daily with a multiparametric device (Horiba® G-50, Kyoto, Japan). The sh received a commercial diet (32% crude protein) during the acclimatization period.
The water temperature, dissolved oxygen content, and pH were respectively 26.25 ± 0.06 °C, 4.95 ± 0.32 mg.L -1 , and 5.74 ± 0.07, while the nitrite content remained between 0.00 and 1.0 mg. L -1 . All parameters monitored to ensure that the quality of the water in the tanks remained within the recommendations for production and well-being of tambaqui C. macropomum, in order to eliminate possible interference on the information obtained regarding nutrition with different lysine levels and the respective parameters evaluated (Schimittou 1993;Araujo-Lima and Gomes 2005;Aride et al. 2007;Rebouças et al. 2014).
After being ground in a knife mill (Tecnal ® , Piracicaba, Brazil), the ingredients were mixed with the crystalline amino acids (Table 1), homogenized with the addition of water (20%) at a temperature of 50 °C. The mixture was extruded, dried in a forced ventilation oven (55 °C for 24 hours), and supplied four times a day (08:00, 11:00, 14:00 and 17:00 hours), until the apparent satiety of the juvenile tambaqui C. macropomum.

Sample collection, biometrics and blood parameters
At the beginning and at the end of 90 days of the experiment, nine sh in total (N= 9) were captured with nets (n= 9 sh per diet; N= 54) after fasting for 24 hours. The juvenile tambaqui collected were anesthetized with benzocaine (100 mg.L -1 ), for blood collection (1 mL per sh) by puncturing the caudal vessel, mediated by 1 mL syringes, 13 x 4.00 mm needles, containing 15 μL of the anticoagulant Heparin sodium 5,000 IU (150 IU.mL -1 of blood) diluted in 0.65% saline solution (1:50) (Pádua et al. 2012;Oliveira et al. 2012).
The samples, preserved at -4 °C throughout the collection process, were used in the analysis of the components of the complete blood parameters (erythrogram, leukogram and thrombogram). Fish were measured biometrically for weight (g) and length (cm) using a digital scale (accurate to 0.001g; Gehaka ® , São Paulo, Brazil) and an ichthyometer.
The indirect method was used to check total leukocytes, observing 2000 erythrocytes with concomitant identi cation of the amount of leukocytes and thrombocytes in homogeneous elds of blood extension (Ranzani-Paiva et al. 2013). The information obtained was related to the erythrocyte count in a Neubauer chamber (µL of blood) and inserted in the formulas, according to Tavares-Dias and Mataqueiro (2004).

Statistical analysis
The experiment consisted of a completely randomized design (DIC), with seven treatments (6 elaborated diets + initial commercial diet) in triplicate. The data were veri ed by Analysis of Variance (ANOVA) (p< 0.05), Tukey's multiple comparison test using the statistical software R ® (r-project, Auckland, New Zealand), version 3.5.3.

Results
The nal weight of the juvenile tambaqui C. macropomun was more expressive in sh fed with 15.96 g. Kg -1 and 22.20 g. Kg -1 of lysine, and showed a difference between all other treatments (p= 0.01). The greatest lengths were recorded in sh fed with 15.96 and 22.20 g. Kg -1 of lysine, with a difference only for sh from the initial period (p= 0.01; Table 2).
The varying levels of lysine added to the diet did not change the mean hemoglobin (p= 0.430), hematocrit (p= 0.430) and MCHC (p= 0.950) concentrations. The in uence of the diet was observed on the amount of erythrocytes in sh fed with 9.72 g. kg -1 of lysine (p <0.05), since these sh presented a greater amount of erythrocytes, with variation in relation to the initial diet ( Table 3). The MCV and MCH indices for the sh in the initial period expressed variation in the mean values in relation to the treatments with lysine, the exception occurred in the MCH of the level of 12,84 g.kg -1 (p <0.05) ( Table 3).
No difference was found between the averages of thrombocytes in juvenile tambaqui (Table 4). The average of the total leukocytes from the levels of 6.60 g.kg -1 and 15.96 g.kg -1 of lysine differed from the averages from levels 9.72, 19.08 and 22.20 g.kg-1 of lysine (p= 0.02) ( Table 4).
In the analysis of differential white blood cell counts, lymphocytes, eosinophils, monocytes and neutrophils were found. There were no statistical changes caused by the addition of lysine to the diet in eosinophils and monocytes (Table 5). The percentage of lymphocytes in sh in the nal period decreased in relation to the values found in juvenile tambaqui fed diets in the initial period (p= 0.00). Diets with lysine levels above 12.84 g.kg -1 generated an increase in the neutrophils of the sh (p= 0.00; Table 5).

Discussion
The length and nal weight of the tambaqui increased proportionally to the addition of lysine in the diets up to the level of 15.96 g.kg -1 of lysine, where a mean growth of around 119% was registered in relation to the initial weight. This behavior was also detected by Costa et al. (2020) for other parameters of animal performance such as weight gain and apparent feed conversion, of tambaqui fed with different levels of lysine.
Biometric data on the in uence of the crystalline amino acid can be better understood when related to information on the sh's physiological and metabolic variables. Zhou et al. (2007) and Ruchimat et al. (1997) related the performance responses, respectively, of cobia or bijupirá Rachycentron canadum (Linnaeus 1766) and yellowtail Seriola quinqueradiata (Temminck and Schlegel 1845), to the hematological condition of the sh, and did not identifying losses at the ideal level of L-lysine in sh health. However, the excess of L-lysine can be catabolized and inserted in the energy metabolism, providing lipid reserves in the visceral region of the animal's body (NRC 2011) and insu cient amounts limit the synthesis of L-carnitine and lipid transport into hepatocytes (Furuya et al. 2013).
According to Tavares et al. (2009), the amount of erythrocytes expected for tambaqui C. macropomum weighing between 1.63 to 369.50 g and length from 26 to 45.6 cm is between 1.25 and 2.96 (10 µL -1 ), with a reference range comprising values from 1.62 to 3.38 (10 µL -1 ). Although the response observed in the erythrocytes of juvenile tambaqui on diets with 9.72 g.kg -1 of lysine (p= 0.02) approaches the maximum detected for the species by Tavares et al. (2009), cannot be considered polycythemia, which results from the elevation of plasma osmolality, blood chemical components and stressful situations (Clauss et al. 2008). This functional disorder is more easily validated when the hematocrit percentage exceeds 45%, which did not occur for sh treated with 9.72 g of lysine.kg -1 (hematocrit= 32, 72%), likewise, for sh from other treatments analyzed (Clauss et al. 2008;Prado et al. 2016).
In hematological studies with juveniles of cobia R. canadum on diets with L-lysine, contrary to what was observed for tambaqui, the hematocrit means differed (mean hematocrit: 30.97%), however, they were similar to the percentage veri ed for the Amazonian species (average hematocrit: 31.65%), which did not occur with erythrocytes (RBC: 3.53 to 3.86 × 10 9 .ml -1 ) (Zhou et al. 2007). The absence of a difference between the hemoglobin concentrations also corroborated what was detected in this study.
The mean corpuscular volume (MCV) expresses the link between hematocrit and the abundance of erythrocytes, denoting the amount of red series cells present in the animal's blood, and can be altered by conditions such as diet, age, and cell maturity of sh of different species or of the same species  (112.7-192.69) for tambaqui as described in the literature (Tavares et al. 2009). However, the MCV of sh in the initial period was higher than the expected maximum volume of 192.69 fL, generating uneven cells and con guring erythrocyte anisocytosis, which initially suggests physiological disorders, such as anemia, that may originate in nutrition. However, the anemic prognosis was ruled out because high concentrations of erythrocytes in immature forms were not found in the blood of the analyzed sh and the biometric data were not affected ).
The maturation of sh erythrocytes occurs naturally in the circulating medium, however, when anemic, the unripe erythrocytes have a greater concentration, making it di cult to interpret the parameters of the erythrogram ). Furthermore, the possibility of nutritional de ciency in the initial period was not expressed in sh in the nal period, given that, if real, it would be circumvented by diets containing L-lysine. The amino acid tends to bind with other compounds such as minerals and micro minerals, and increases the villi of the intestinal mucosa, optimizing the transport and absorption of other nutrients necessary for animal maintenance and production.
Information on mean corpuscular hemoglobin (MCH), even when calculated using hemoglobin data, is obtained as a function of the average value of sh erythrocytes. This elucidates the behavior of data aligned with the pattern observed in the MCV, where the analysis of sh from the initial period (MCH: 81.65 pg) was higher than sh fed with lysine in the nal period (mean nal MCH: 45.54 ± 5.69 pg).
Tavares-Dias and Sandrin (1998) veri ed MCH of tambaqui weighing between 500 and 700 g and identi ed amplitude variation of between 32.2 and 51.3 pg and a mean value of 41.40 pg, which is close to the MCH average found in this work for sh fed with L-lysine. The high levels of hemoglobin observed in the blood of juvenile sh indicates a nutritional demand that was not met in the initial period, but was appears to have been solved by the addition of L-lysine in the diets, according to the MCH of the nal period.
The interpretation of the MCH and MCV values requires careful analysis because the calculation base (total erythrocytes) may present a margin of error that in uences the nal response (Tavares-Dias and Sandrin 1998). According to Tavares-Dias and Sandrin (1998), the RBC that expresses greater precision is the MCHC (mean MCHC: 36.34 ± 1.06 (g.DL -1 ), which in the current study did not uctuate between the means analyzed (p> 0.05).
In the observation of neutrophils and lymphocytes, as well as eosinophils and monocytes for sh that received experimental diets, it is clear that the diets with lysine did not alter the different circulating leukocytes. When verifying the different types of leukocytes for all treatments (initial diet + diets with lysine), sh do not fed the experimental diets showed a higher percentage of lymphocytes and a lower percentage of neutrophils. The increase in the percentage of neutrophils, or neutrophilia, followed by a reduction in the amount of lymphocytes, or lymphopenia, suggests a stress condition that would be proven if a reduction in animal performance was detected (Urbinati et al. 2015). Such condition was not observed, since biometric analyses showed an increase in the weight and length of juvenile tambaqui up to the highest productive level due to the L-lysine addition, which shows the absence of impediment to the manifestation of the tested amino acid on the performance of the sh.
Evaluating the changes that occurred between the means of each differentiated leukocyte, it was observed that the neutrophils from treatments with diets containing lysine above 12.84 g. kg -1 exceeded that of sh which received 9.72 g.kg -1 and those which did not receive lysine (treatments with lysine 9.72 g.kg -1 and initial diet). The highest percentage was recorded at the lysine level of 15.96 g. kg -1 . Tavares-Dias et al. (1999) analyzed lymphocytes in tambaqui blood (average live weight: 584.6 g; diet with 28% crude protein -CP) and registered 1566.2 ± 754 µL of neutrophils for a total of 2663.3 ± 1288 µL leukocytes, which represents about 58.80% of neutrophils. This percentage is close to the neutrophil values found in the highest levels of lysine in our study with tambaqui.
In the studies by Tavares-Dias et al. (1999) with tambaqui, neutrophils were the most abundant circulating leukocytes. In teleosts, such cells raise the sh's hypersensitivity to adaptations to the environment (Tavares-Dias et al. 1999), which is the same as was found in this study. In general, the highest percentage of leukocytes in the blood of Brazilian freshwater teleosts corresponds to lymphocytes (LF) (Tavares-Dias et al. 1999), as detected for the Amazonian pirarucu Arapaima gigas (Schinz 1822) by Drumond et al. (2010) (LF: 89.9 ± 6.3% for fry and 55.6 ± 9.9% for juveniles). However, for parasitized tambaqui juveniles (live weight: 35.53 ± 2.77 g and 62.71 ± 2.75 g), fed with a commercial diet (CP: 36%) and evaluated for the effect of different bio lters under different densities, Lima et al. (2019) recorded higher concentrations of lymphocytes (65.4 to 84.9 x 10 3 µL -1 ).
The absence of a pattern expressed by the means of the total leukocytes does not allow a clear association with the effect of the diet. The same depatterning was observed by Zhou et al. (2007) for total leukocytes of cobias (R. canadum) under nutrition with different levels of lysine (4.90 ± 0.20 to 6.87 ± 0.21 × 10 6 .mL -1 ). Zhou et al. (2007) stated that such differences were probably caused by nutritional stress and not by ine cient action by different concentrations of the amino acid. The average of the values analyzed in this work with tambaqui (30401 ± 12560 µL) did not differ from the average veri ed by Pádua et al. (2013) for tambaqui (C. macropomum; average weight: 357.5 ± 94.5 g) used as a control in studies on the effect of anesthetics on hematological parameters (Total leukocytes: 32.2 ± 6.1 × 10 3 µL -1 or 32200 µL), and for juvenile pirarucu (A. gigas; total leukocytes: 25091 ± 6121 µL) farmed in a semi-intensive system in the Amazon.
Leukocytes have recognized importance in the non-speci c immunity of sh organisms, and act in in ammatory processes of the tissues, and are thus, indicators of the animal's health (Ishikawa et al. 2008;Faggio et al. 2015Faggio et al. , 2014aFaggio et al. , 2014bFazio et al. 2015).
Biotic and abiotic factors can interfere with thrombocyte concentrations (Tavares-Dias et al. 2009). Despite participating in coagulation, the clinical translation of the functions and behavior of sh thrombocytes have not yet been fully elucidated. However, it is known that some neotropical sh, such as traíra Hoplias malabaricus (Bloch 1794), can express a reversible reduction of thrombocytes, or thrombopenia, through prolonged fasting (Rios et al. 2005;Clauss et al. 2008). Considering that the tambaqui analyzed had fasted for 24 hours, the occurrence of thrombopenia due to food restriction is a possibility that, in order to be con rmed, require further hematological analyses under fasting, followed by re-feeding.
The study concluded that different levels of lysine added to the diet did not in uence the parameters veri ed in the blood count of juvenile tambaqui. Thus, lysine does not promote physiological changes, and can be used routinely without problems of well-being in tambaqui C. macropomum.