LETHAL AND SUBLETHAL EFFECTS OF AMMONIA IN Deuterodon iguape ( Eigenmann 1907 ) , POTENTIAL SPECIES FOR BRAZILIAN AQUACULTURE

The cultivation of the lambari, Deuterodon iguape, supplies much of the Brazilian market with live bait for sport fishing. The high densities used to maximize production can increase the concentration of ammonia. In order to evaluate the sublethal and lethal effects of different concentrations of ammoniacal nitrogen (non-ionized ammonia plus ionized ammonia), D. iguape were exposed to this xenobiotic. The LC50 values for 24, 48, 72, 96 h of ammonia-N were 6.17, 5.57, 3.88 and 2.90 mg L-1 at 23°C. The LC50 values of 24, 48, 72, 96 h of NH3-N (non-ionized ammonia with nitrogen) were 0.015; 0.013; 0.009; 0.007 mg L-1. The specific oxygen consumption increased at the ammonia-N concentrations tested. The values for the concentrations of 0.1; 0.25; 0.5 and 1.0 mg L-1 were: 0.25, 0.33; 0.31 and 0.44 mlO2 g-1 h-1. At the concentration of 1.0 mg L-1 of ammonium chloride there was a 41.33% increase in the consumption level in relation to the control. Ammonia excretion also increased with increasing ammoniacal nitrogen concentration. The fish excreted, on average, 0.0320, 0.0364, 0.0368 and 0.0370 mg/g/h of ammonia. Comparing these results with the means of ammonia excretion of the control (0.0 mg L-1), it was observed that these values represent an increase in the excretion of 146%, 177%, 189% and 184%, respectively. After 24 h of exposure to ammonia, the O:N ratio decreased by 43.46%. Our results indicate pronounced metabolic effects and increased toxicity with increasing ammonia concentrations. We recommend avoiding concentrations higher than 0.25 mg L-1 NH3-N in the culture water.


INTRODUCTION
There are few studies on the toxicity and sublethal effects of ammonia for neotropical freshwater fish species in Brazil, justified by the predominant cultivation of exotic species, mainly Nile tilapia (Martinez et al., 2006).For this reason studies that approach the effect of the ammonia in neotropical fishes that have potential for aquaculture are very important.
Several authors have agreed to call NH 4 + ionized ammonia, and NH 3 non-ionized ammonia, and the sum (NH 4 + + NH 3 ) is called total ammonia (Arana, 1997).The increase of the concentration of total ammonia in crop systems is one of the most damaging factors in production, being able to reach sublethal or lethal levels in static or recirculation systems (Medeiros et al., 2016).Therefore, it is fundamental to determine the toleration limits of commercially valuable aquatic organisms, in relation to this compound for the technical and economic viability of intensive crops (Damato and Barbieri, 2011).
Among the native species with potential for Brazilian aquaculture, some small species of the Characidae family, popularly known as lambaris, have received more attention due to zootechnical characteristics favorable for production in captivity (Silva et al., 2011a).The lambari Deuterodon iguape is an endemic species of small rivers and coastal streams of the Atlantic forest with great market potential, both for human consumption and for use as bait in sport fishing (Silva et al., 2011b).However, there is no information in the literature on the effects of ammonia at different concentrations for this species, although it has already been widely cultivated in Brazil.
Exposure to ammonia in an aquatic environment produces many physiological changes in fish, including changes in their metabolism (Cavero et al., 2004, Barbieri andDoi, 2012).The metabolic rate of an organism is a useful and sensitive indication of their daily energy consumption.Therefore, in aerobic organisms, quantifying the rate of oxygen consumption may be directly associated with the amount of energy released from the food substrate oxidation.Based on the amount of oxygen consumed by an animal over a period of time, it is possible to calculate the energy spent during the same period to maintain its vital processes (Bosisio et al., 2017).
Particularly in Brazil, the cultivation of D. iguape was intensified due to the market demand and the availability of water bodies and their breeding facility.In an intensive culture system, ammonia is a common pollutant resulting from the excretion of domesticated animals and the mineralization of organic debris such as unconsumed food and feces (Medeiros et al., 2016).The accumulation of ammonia can reduce growth, increase oxygen consumption and N-ammonia excretion, and even cause increased mortality (Barbieri and Bondioli, 2013).
In order to guarantee the success of the intensive production of D. iguape, determining the toxicity and effects of ammonia are of fundamental importance to verify the sensitivity of this species to parameters of water quality, such as nitrogen residues, since these are limiting factors for the survival and growth of confined fish.Therefore, the objective of the present study was to estimate the toxicity and the ammonia effect on oxygen consumption and ammonia excretion of D. iguape, increasing the knowledge concerning the culture of this fish that has excellent market potential in Brazil.

MATERIALS AND METHODS
This research is in accordance with the ethical principles in animal experimentation adopted by the Brazilian College of Animal Experimentation (COBEA) and was approved by the Animal Experimentation Ethics Committee of the Fisheries Institute (process number 06/2016).

Acute toxicity assays
The acute toxicity of ammonia-N and NH 3 -N in lambaris D. iguape was evaluated in the laboratory of the Fisheries Institute (SP) exposed to different concentrations of these chemicals for a period of up to 96 h.A total of 120 fish were used, with a mean body weight of 6.5 ± 0.6 g, obtained from a local fish farmer.
During the experiment, dechlorinated supply water, previously filtered using a 0.25 μm pore cellulose ester filter was used with the aid of a vacuum pump.Three replicates of groups of 5 individuals were exposed to each of the following concentrations of ammonia-N: 0.0; 0.1; 0.5; 1; 2.5; 5; 10 and 15 mg L -1 .The experiments were carried out in glass tanks with a volume of 50 liters each in a 3/7 LETHAL AND SUBLETHAL EFFECTS OF AMMONIA IN Deuterodon iguape… static system, aerated by blowers connected to porous stones.The photoperiod was kept in the 12/12 light/dark cycle.
The 10,000 mg L -1 stock solution of ammonia-N at was prepared with ammonium chloride (Synth, Brazil), and then diluted to the desired concentrations of ammonia-N.The actual ammonia-N concentration for the test solution ranged from 0.1 to 15 mg L -1 .Experimental concentrations of ammonia-N were 0.1, 0.5, 1, 2.5, 5, 10 and 15 mg L -1 .Actual concentrations of ammonia-N in the test solutions were measured using the method described by Greenberg (1995).The concentration of NH 3 -N (non-ionized ammonia) was calculated according to the equations of Khoo et al. (1977) based on pH 6.70 and temperature 23 °C.NH 3 -N concentrations were 0.0002, 0.001, 0.002, 0.006, 0.012, 0.024 and 0.036 mg L -1 , respectively (Khoo et al., 1977).
During the experiment, the fish were fed a diet of 40% crude protein and 12% lipids, once daily (4:00pm) at 5% of their body weight per day.The water temperature was maintained at 23 ± 0.5 °C, the dissolved oxygen was at 6.8 ± 0.5 mg L -1 , the pH ranged from 6.70 to 6.75, and these parameters were measured with the YSI Pro Plus multiparameter equipment (USA).
Dead fish were removed daily from the tanks and counted after 24, 48, 72 and 96 h of exposure, death was presumed when the fish were immobile with paralyzed opercula and showed no response to mechanical stimuli.

Routine metabolism
Another fifty lambaris (D. iguape), 6.57 (± 0.4) g, were used for measurements of routine metabolism using sealed respirometers (Barbieri et al., 2015).The respirometers were manufactured in our laboratory, with an acrylic tube and PVC lids.Ten fish were exposed to the following concentrations of ammonium chloride (0, 0.05, 0.1, 0.5 and 1 mg L -1 ) for 24 hours.After this period the oxygen consumption was measured.The pH and oxygen concentration of the test solution at the different concentrations of ammonium chloride were also measured, and the range of pH values was between 6.75 to 6.90.The range of oxygen values was between 6.35 to 6.55 mg L -1 .
Prior to initiating the experiments, the animals were kept on the respirometer with continuous circulation of water for at least 90 minutes to attenuate the handling stress.Then the water supply was suspended and the vent was closed so that the fish could consume the oxygen in the known volume of water for a period of one hour.The respirators were protected by a barrier to isolate the animals from movement in the laboratory.The difference between the oxygen concentrations determined at the beginning and at the end of the confinement was used to calculate the consumption during the period.To minimize the effect of low oxygen concentration and accumulation of metabolites on metabolism, the duration of the experiment was regulated so that the oxygen concentration at the end of the experiments never fell below 70% of its initial concentration.Dissolved oxygen was determined by the Winkler method (Winkler, 1888).
To obtain the desired concentration of ammonia, the required volume of the main substance (10,000 mg L -1 ammonia-N stock solution prepared with ammonium chloride) was calculated for each volume of aquarium and adjusted with an aid (micropipet) at the end of acclimatization.Once the substance was added to the aquarium, after 15 minutes the fish were placed and kept for 24 h until the metabolism experiments were started.After this procedure, the water was sampled at the beginning and end of the oxygen consumption analysis.Total ammonia nitrogen (TAN) was considered as values of ammonia-N (non-ionized plus ionized ammonia) and NH 3 -N (non-ionized ammonia) (mg L -1 ).

Statistical analyses
The median lethal concentrations (LC 50 with 95% confidence limits) of ammonia-N and NH 3 -N were calculated at 24, 48, 72 and 96 h using the Trimmed Spearman-Karber method (Hamilton et al., 1977).
The specific oxygen consumption and ammonia excretion data were analyzed for normality of distribution using the Shapiro-Wilk test and homoscedasticity of variances using the Levene test.Since the results were normal and homoscedastic, the differences between the means of the treatments were evaluated through analysis of variance (ANOVA), with a significance level of p<0.05.

Mortality
The percentage mortality of D. iguape exposed to ammonia-N and NH 3 -N at each 24-hour interval is shown in Table 1.No fish died in the control.As expected, the higher the concentration of ammonia to which the fish were exposed, the higher the observed Table 1.Mortality rate (%) of Deuterodon iguape exposed to various ammonia-N and NH 3 -N concentrations at different exposure times and its median lethal concentration (LC 50 with 95% confidence limits) calculated by Spearman-Karber Estimates, for the temperature of 23°C.

Exposure time (h)
Ammonia-N LC 50 of ammonia-N (mg L -1 ) LC 50 of NH 3 -N (mg L -1 ) (mg L -1 ) 0.0 0.1 0. mortality.The actual concentrations of ammonia in the test solutions were within 5% of the nominal concentrations.No fish died in solutions of 0.1 and 0.5 mg L -1 of ammonia when exposed for 96 hours.The survival rates of D. iguape were significantly different for the different interactions: ammonia-N concentration and exposure time.

Mean lethal concentration
The LC 50 values of ammonia-N and NH 3 -N at different exposure times for D. iguape are shown in Table 1.At 24, 48, 72 and 96 h, the N-LC 50 ammonia values were 6.16, 5.57, 3.88 and 2.90 mg L -1 , and NH 3 -N were 0.015, 0.013; 0.009 and 0.007 mg L -1 at 23 °C, respectively.The LC 50 values of ammonia-N and NH 3 -N noticeably decreased during the first 72 hours.

Oxygen consumption
Fish exposed to ammoniacal nitrogen for 24 hours showed an increase in the specific oxygen consumption.The specific oxygen consumption at all of the concentrations of ammonia-N increased.The oxygen consumption values for the concentrations of 0.1; 0.25; 0.5 and 1.0 mg L -1 of ammonia were, 0.25, 0.33, 0.31 and 0.44 mlO 2 g -1 h -1 , respectively.With the concentration of 1 mg L -1 of ammonium chloride there was a 41.33% increase in the metabolic level in relation to the control.Using the Tukey statistical test (p <0.05), it was found that only at the concentration of 1 mg L -1 ammonia-N was a statistical difference in relation to the control (Figure 1).

Ammonia excretion
Ammonia excretion increased with increasing ammoniacal nitrogen concentration.The samples were incubated at 23 °C (Figure 2) and excreted 0.032, 0.0364, 0.0368 and 0.037 mg/g/h of ammonia on average at concentrations of 0.1, 0.25, 0.5 and 1.0 mg L -1 of ammonia for 24 hours.Comparing these results with the means of ammonia excretion of the control (0.0 mg L -1 ), it was observed that these values represent an increase in the metabolic level of 146%, 177%, 189% and 184% in relation to the control.Using the Tukey statistical test (p <0.05), the mean values of ammonia excretion for 0.25, 0.5 and 1.0 mg L -1 ammonia-N concentrations were different in relation to the control.

O:N ratio
The O:N ratio for D. iguape under exposure to different concentrations of ammonia-N decreased significantly with increasing ammonia concentration.After 24 h of exposure to ammonia-N, the O:N ratio decreased by 43.46% (Figure 3).

DISCUSSION
Toxicity tests are desirable in cultures of aquatic organisms, since only chemical and physical analyses of water are not sufficient to estimate their sublethal and lethal effects on these organisms (Key et al., 2007).The toxicity of ammonia-N to fish has already   (Brinkman et al., 2009;Bergerhouse, 2011;Damato and Barbieri, 2011;Miron et al., 2011).The results of these studies confirm that ammonia-N and NH 3 -N are toxic for fish as well as for D. iguape, an ecologically and economically important organism in Brazil.

5/7 LETHAL AND SUBLETHAL EFFECTS OF AMMONIA IN Deuterodon iguape…
Toxic effects of ammonia on various freshwater teleost fishes have already been studied by Brinkman et al. (2009) and Bergerhouse (2011).In different neotropical fish species, Martinez et al. (2006) showed that ammonia-N LC 50 (24 h) was 5.21 mg L -1 for Astyanax altiparanae, 6.88 mg L -1 for Piaractus mesopotamicus, and 5.82 mg L -1 for Prochilodus lineatus.Studies on the effects of ammonia on A. altiparanae, P. mesopotamicus and P. lineatus demonstrated that the LC 50 (24 h) of ammonia-N was 5.21 mg L -1 for A. altiparanae, 6.88 mg L -1 for P. mesopotamicus and 5.82 mg L -1 for P. lineatus.The values reported for LC 50 (24h) of NH 3 -N varied from 0.66 mg L -1 for A. altiparanae, 0.85 mg L -1 for P. mesopotamicus and 0.74 mg L -1 for P. lineatus.
The lambari D. iguape showed sensitivity to ammonia, presenting the lowest LC 50 value of 6.16 (5.33 -7.10) mg L -1 for ammonia-N and 0.015 (0.013 -0.017 mg L -1 ) in 24 hours of exposure.It should be noted, however, that the fish tested were adults.According to Walker et al. (1996), there is evidence that larger fish are less susceptible than smaller specimens.The effects of size according to Martinez et al. (2006) may be the result of the following factors: i) smaller individuals of a given species present greater body surface area for absorption, in relation to body mass; ii) small individuals have higher respiratory rates.
The toxicity of N-ammonia in D. iguape, as expected, increased with time of exposure.The tolerance of D. iguape to ammonium-N notably decreased by 9.57%, 37.01% and 52.92% after 48, 72 and 96 h compared to the LC 50 in 24 hours.Similar results with another lambari species were obtained by Martinez et al. (2006), who observed that the tolerance of A. altiparanae decreased with time of exposure.LC 50 (24h) for freshwater fish in general is 0.82 mg L -1 NH 3 (Person-Le Ruyet et al., 1995), and for D. iguape, in the present study, it was 0.015 mg L -1 .Medeiros et al. (2016) exposed juveniles of Amphiprion ocellaris to six concentrations of ammonia ranging from 0.23 to 1.63 mg L -1 NH 3 -N.The values of LC 50 -24, LC 50 -48, LC 50 -72 and LC 50 -96h were estimated at 1.06, 0.83, 0.75 and 0.75 mg L -1 , respectively.Histopathological changes were more evident according to the increase of ammonia in fish exposed to 0.57 mg L -1 of NH 3 -N.
Studies concerning the effects of ammonia-N on the respiration of teleost fish showed an increase in the rates of oxygen consumption as the ammonia concentration increased (Smart, 1978;Arana, 1997;Ip and Chew, 2010;Barbieri and Doi, 2012;Zeitoun et al., 2016).Exposure to different concentrations of non-ionized ammonia caused a 3.3-fold increase in oxygen consumption in Salmo gairdneri (Smart, 1978).In the present study, a 1.41-fold increase in the oxygen consumption of D. iguape was recorded, with a41.33% increase in the metabolic level.Despite the regulatory capacity of the fish, the rate of oxygen consumption was actually increased after D. iguape had been exposed to high concentrations of ammonia-N.Similar results were also observed for Hyphessobrycon callistus (Damato and Barbieri, 2011).Respiratory compromise in fish resulting from exposure to stressors has also been studied (Brydges et al., 2009;Bilberg et al., 2010), and it has been concluded that oxygen consumption usually increases when fish are acutely exposed to stressors.Zeitoun et al. (2016), studying the effect of NH 3 -N on Oreochromis niloticus, also found an increase in oxygen consumption and opercular ventilation rate in fish exposed for 48 hours.
The ammonia excretion of D. iguape acclimatized at 23 °C varied as a function of the increase of ammonia concentration.There was an increase in excretion when exposed to a concentration above 0.25 mg L -1 .Ammonia is one of the final products of catabolism, especially of amino acids.The increase in ammonia excretion reflects the increase in amino acid catabolism (Damato and Barbieri, 2011), even at the highest concentrations used in this work.Fish can eliminate metabolic ammonia through gill diffusion, active transport with sodium, and by transforming ammonia into less toxic products such as urea.The increase in ammonia concentration reduced its excretion in Carassius auratus (Arana, 1997) and Salmo salar (Fivelstad et al., 1993), different from our results in which excretion increased after 24 hours of exposure.In studies with Rachycentron canadum, Barbieri and Doi (2012) also reported an increase in ammonia excretion for this fish as it increased its concentration in the medium.
The O:N ratio has been widely used to determine the nature of oxidized metabolic substrates in aquatic animals under conditions of stress.An increase in the O:N ratio represents an increase in the catabolism of lipids and carbohydrates (Zhang et al., 2017).In the present study, the O:N ratio decreased from 28.2 to 12.4 with increased exposure to ammonia.These data demonstrate the changes in strategies of energy use (Zhang et al., 2017).Under a higher concentration of ammonia, more proteins were oxidized to maintain the metabolism of D. iguape.Both the increase in ammonia excretion and the consumption of oxygen in D. iguape, shows that this fish increased metabolism by expending energy to maintain homeostasis.This energy could be used for example for other metabolic functions such as growth; which is very important in aquaculture.

CONCLUSION
This study aimed to determine the acute toxicity of ammonia and its effects on the metabolic rates of D. iguape.Our results indicated distinct metabolic effects and increased toxicity with increasing ammonia concentrations.We recommend avoiding concentrations higher than 0.25 mg L -1 NH 3 -N in the culture water.Future work should be carried out with the objective of investigating the chronic effects on metabolism, growth and reproduction, and health status, which are also important parameters for the commercial cultivation of D. iguape.

Figure 1 .
Figure 1.Specific oxygen consumption of Deuterodon iguape in relation to the Ammonia concentration.The columns represent the means (n=10) and the bars are their respective standard deviations.The asterisk indicates the groups that presented statistical difference in relation to the control.

Figure 2 .
Figure 2. Ammonia excretion of Deuterodon iguape in relation to the increase in ammonia concentration.The columns represent the means (n=10) and the bars are their respective standard deviations.The asterisk indicates the groups that presented statistical difference in relation to the control.