Dourado, Salminus brasiliensis (Cuvier, 1816), from Characidae family, is broadly distributed at Pantanal, Paraná, Uruguay, São Francisco watersheds as well as at Lagoa dos Patos associated-basin (Morais and Schubart, 1955; Gomes et al., 2003). It is a carnivorous fish, found    in    lotic    environmental    and   performing ascending reproductive migrations (Streit et al., 2007). Zaniboni and Schulz (2003)  assigned the dwindling of this specie in nature to riparian deforestation, fishing and capture of young specimens, drainage of adjacent lagoons and alterations of hydrological regime caused by dams’ construction,  water contamination  and  nonnative species introduction. Hence, S. brasiliensis is candidate for artificial reproduction programs for aquaculture preservation purposes.

More recently, some studies have demonstrated environmental factors-induced cortisol releasing to play crucial role in physiological response during the onset of sexual development and increasing steroids levels (Solomon-Lane et al., 2013; Nozu and Nakamura, 2015); however, the exact mechanism involved in cortisol signaling is still unclear (Goikoetxea et al., 2017). During induced reproductive process, the fertilization rate of S. brasiliensis is considered limited, which can be associated with inadequate procedures, commonly applied to exotic species reproduction (Zaniboni-Filho  and Weingartner, 2007). Eggs quality and viability are directly related with reproductive stress (Zanoni et al., 2016).

Handling of fish breeders during hormonal induction leads to morphological, biochemical and physiological alterations that can be characterized as a stress condition (Eslamloo et al., 2014). In this context,  fish stress monitoring through attendance of physiological conditions is an important and valuable tool for aquaculture. The studies of Mazeaud et al. (1977),  Barton and Iwama (1991)and Schreck (2010)proposed that fish physiological stress response initiates with stressor  agent perception, which leads to hypothalamic-pituitary-interrenal axis catecholamines releasing. The elevated levels of these hormones in turn result  in a secondary physiological response that influence other organs and systems causing osmotic status, immunological status and bioenergetic use to change, leading to important alterations of reproductive capacity . 

Hematological parameters are important tools to evaluate fish physiological status, may serve as an indicator of stress and reflects the associated pathological alterations  (Fazio et al., 2012; Zanoni et al., 2016). To recognize these alterations and to compare them with reference values for a determined species allow to quantify the duration and the severity of stress (Fazio et al., 2015). In addition, hematological profile has been used as biomarker for aquaculture (Tavares-Dias et al., 2007)and able to validate stress conditions in fish. Blood sample analysis predict pathological alterations in an organism since hematopoietic process is under influence of several biological and environmental factors (Tavares-dias et al., 1999). Hematocrit alterations such as red blood cells numbers and hemoglobin levels (Graham et al., 1985), are normally investigated in this context.  The RBC elevation results from the releasing of young erythrocytes caused by spleen contraction (Caldwell and Hinshaw, 1994). Yet, there is a considerable variation in the results found by different authors.

Among blood figurative elements, leukocytes play crucial role in nonspecific immunity, and changes in its value can be attributed to the fishes’ welfare (Misra et al., 2006).   Particularly,   leukocyte   profile  is   useful  as  a physiological indicator of stress. Plasma glucocorticoids disturbance can increase neutrophil numbers (neutrophilia) and reduce limphocyte concentrations (lymphocytopenia), and as result, the proportional neutrophil to limphocyte ratio is positively associated to the intensity of stressor agent as well as to the levels of circulating glucocorticoids. Some evidences also point to infections and diseases caused by cortisol excess to influence neutrophil to lymphocyte ratio (Davis et al., 2008).

In order to put forward conservational aquaculture, there is an increasing need to achieve higher breeder survival indexes and improve quality of their eggs. During induced reproduction, maintenance of physiological homeostasis and minimization of stress procedures could lead to an increased number of healthy fish fry. The aim of this study was to evaluate the fertilization rate, survival, cortisol and glucose concentrations; and, hematological parameters from dourado breeders underwent two different induced reproduction procedures; semi-natural and extruded methods.

For both procedures, the following reproductive efficiency indexes were calculated:

To measure the concentration of cortisol, 1 mL of blood was collected by venipuncture in caudal region using 3 ml syringes embedded in EDTA 10% and 28 x 12 needles . Female blood samples of the two experimental groups were collected soon after spawning and control group blood sample was collected soon after fish were captured. Cortisol concentration was determined by ELISA using commercial kit (EIA, 55050, Human-Kit; Cayman Chemical, Ann Arbor, MI, USA) and glucose levels were determined with digital glucometer ACCU CHECK active®. 1 mL of blood was also collected and added to tubes with EDTA 10% for hematological analysis that were carried out at the Clinic Pathology Laboratory of Universidade Estadual do Norte do Paraná – UENP, Paraná, Brazil. Blood smears, in duplicate, were stained with May–Grunwald–Giemsa dye (Rosenfeld, 1947). The indirect total and differential leucocyte count were performed as proposed by Hrubec and Smith (2000). Our results were evaluated by one-way ANOVA and complemented with Tukey, p≤0.05, while data are expressed as mean±SD.

In the semi-natural procedure, females produced 8 L of hydrated eggs, with 46 eggs in each mL, resulting in 368,000 eggs and in extruded procedure, the result was 636.000 eggs, in a final volume of 12 L with 53 eggs/L as shown in Table 1. In both procedures, 100% of dourado females responded to the treatment with pituitary extract. According to Bromage et al. (1994), eggs releasing at the time of spawning is crucial for egg quality; in such way, the release of eggs before or after this event can lead to low fertilization rates and low quality fish larvae (Springate and Bromage, 1985). 

Fertilization rate can be an indicative for eggs quality, and immature eggs display small perivitelinic space caused by incomplete cortical activation which in turn can compromise fertilization process and embryogenic development (Kjørsvik et al., 1990; Zanoni et al., 2016). Our results show fertilization rate to be higher in semi-natural method compared to extruded method as described in Table 1. This result corroborates David et al. (2002) , who obtained fertilizations rate of 25.8% for extruded and 94.5% for semi-natural spawning of Leporinus macrocephalus, and also by Zanoni et al. (2016)with Brycon orbignyanus, who described fertilization rates of 87.9 and 8.17%, for semi-natural and extruded procedures, respectively.

The percentage of females that survived the reproduction procedures was 100 and 62.5% for semi-natural and extruded, respectively. Notably, 24 h after the extrusion processes, all the females of the last group died. The mortality rate after extruded procedure for S. brasiliensis is normally high. Sato et al. (1997) related 83.3% death of extruded breeders, which showed several lesions, body hiperemia, bacteria and fungi secundary infections. All these symptoms can be caused by loss of scales during fish manipulation (Elaine et al., 2002), and as in our case could, in part, explaining our findings pointing the first caution with extruded procedure.

Barton (2002)proposed ideal levels for corticosteroids to be under 5.0 ng/mL of blood, although in fishes, this range between 30-40 ng/mL. In Figure 1, it can be seen that cortisol concentration in control, semi-natural and extrude groups were 11.27±4.26, 22.74±12.93 and 34.9±11.26 ng/mL respectively. These results are in accordance with previous studies from our group and with other authors who have not found differences in this hormonal parameter in fish submitted to stressor conditions (Gomes et al., 2003; Hoshiba, 2009; Zanoni et al., 2016). However, in semi-natural group, the lower cortisol level and higher matrix survival can be attributed to an improved recovering period since the animal were displaced in an external circular tank with bigger space and increased water flow which is similar to what is found in natural environments (Boesoaard et al., 1993; Milligan et al., 2000). 

Our outcomes for fertilization and cortisol levels demonstrate the stress effects for eggs quality, which can be explained by cortisol influence on fish reproductive glands.  Nepomnaschy et al. (2006) and Whirledge and Cidlowski (2010)showed the necessity of glucocorticoids levels for a normal gonadal function. Elevated levels of cortisol have a negative impact over fertility rates (Carragher and Sumpter, 1990); furthermore, stress itself can  suppress  reproductive  hormones   (Pickering et al., 1987; Pankhurst, 1994; Haddy and Pankhurst, 1999). Extruded procedure increased glucose circulating levels compared to control and semi-natural  groups,  as  it  can be seen in Figure 2. Similar response was observed by Gomes et al. (2003)and (Brandão and Levy, 2006), for Arapaima  gigas, and Gomes et al. (2003)for Colossoma macropomum. Glucose levels elevation is considered as an adaptive response during the stress response and is related to cortisol and others catecholamine’s levels. Taken together, cortisol and glucose levels can be considered good predictors of stress in S. brasiliensis.

According to Lucelle et al. (2004), hematological evaluation of fishes during fish farming can reflect its physiological status. Stress conditions increase cortisol concentration modifying physiologic and metabolic status, which can be demonstrated by erithrocytes elevation and mean corpuscular volume (VosylienÄ—, 1999). Alterations found in erythrogram helps the identification of anemic conditions, and some alteration seen in leucograms is useful for infectious process and homeostasis imbalance diagnostics. Data presented in Table 2 demonstrate  that there is no statistically  difference for hemoglobin concentration, hematocrit and number of red blood cells comparing experimental groups. Similarly, Abreu and Urbinati (2006), for Brycon amazonicus,Martins et al. (2004), for Oreochromis niloticus and Pimpão (2006), for Ancistrus multispinis, found no difference in hematocrit in fish under stress conditions. Number of red blood cells and hemoglobin concentration, otherwise, differ from Das and Mukherjee (2003)and Adhikari et al. (2004), who demonstrated reduction in these parameters for Labeo rohita exposed to stressor agents. It can be assumed that, for dourado, semi-natural and extruded methodologies do not alter eritropoietic process, besides that, for the same species, different results for hemoglobin concentration and hematocrit can be caused by blood samples collection techniques, including anticoagulants and anesthetic use, time elapsed between collection and laboratorial analysis, among others (Tavares-Dias and Sandrim, 1998). More studies are needed to the understanding of these differences.

Teleosts blood composition depends on physiological and ecological factors such as sex, gonadal development stage, stress, infection and environmental conditions. Leukocytes number vary among families, genus and species, and it is related with the surrounding environment (Tavares-dias and Moraes, 2004) There was no difference in total leukocytes when comparing both methods as described in Table 3. According to Abreu and Urbinati (2006), feeding Brycon cephalus breeders with different concentrations of vitamin C submitted to air exposure, and Pimpão (2006), studying Ancistrus multispinis exposed to deltametrin, found similar results.  For the most of fishes’ species, leukocytosis can be found as soon as a stressor agent is present in attempt to recover from homeostasis disruption, while leukopenia is generally attributed to diseases that affect immunological system (VosylienÄ—, 1999). In this study, leukocytes total number was unaltered despite cortisol levels alterations. Nevertheless, monocytes number was higher in both experimental groups compared to control as shown in Table 3. This cell type acts in inflammatory responses involving phagocytosis process (Dalmo and Bøgwald, 2008;  Salvador  et  al.,  2013).  The  result demonstrated here point the reproductive induction process in Dourado to be considered an inflammatory stimulus, recruiting monocytes for local tissue, since they are considered primary cells for antigens presenting in teleosts.

Neutrophils, lymphocytes and thrombocytes numbers did not differ among groups as seen in Table 3. Stress leads to a lymphocyte redistribution, mainly in lymphoid organs, lowering their circulating levels or, the high levels of cortisol in response to a stressor agent can induce lymphocytes death (Benfey and Biron, 2000). Wojtaszek et al. (2002)confirmed this last alternative by demonstrating accentuated lymphopenia in Cyprinus carpio 24 h after cortisol inoculation. Other authors however have related higher number of neutrophils after stress conditions in Salmo trutta e Salmo gairdneri (Johansson-Sjobeck and Larsson 1979; Pickering et al., 1987). Thrombocytes number found in our experiment are in accordance with Benfey and Biron (2000). These researches studying Oncorhynchus mykiss and Salvelinus fontinalis under feedlot and manipulation stress, have verified a slight, but non-statistically significant increase in the number of thrombocytes in fish. This cell type is produced at kidneys and spleen in teleost fishes and play diverse functions, being in a constant flux between circulation and organs (Tavares-dias and Moraes, 2004). Lower number of thrombocytes in circulation could be related to the moment of blood sample collection.

Our result demonstrates S. brasiliensis manipulation by utilizing semi-natural procedure or by extrusion causes similar stress characteristics. The improved egg quality obtained by semi-natural procedure is relevant and needed to point out. Taken into account, semi-natural proceeding should be preferred in order to achieve a larger number of healthy larvae

The authors have not declared any conflict of interests.