ACUTE TOXICITY OF AMMONIA IN JUVENILE BRAZILIAN SARDINE Sardinella brasiliensis *

A sardinha-verdadeira (Sardinella brasiliensis) é o principal recurso pesqueiro do Brasil que sustenta diretamente a indústria de enlatados, como matéria prima, e indiretamente a pesca dos tunídeos, pois juvenis são utilizados como isca-viva. A UFSC através do Projeto Isca-Viva vem desenvolvendo técnicas de produção de sardinha-verdadeira em cativeiro com vistas ao uso sustentável do recurso, entretanto não existem informações sobre a toxicidade dos compostos nitrogenados no cultivo desta espécie. O objetivo deste estudo foi de estimar a toxicidade aguda de amônia (NH3-N) em juvenis de sardinha-verdadeira. Grupo de cinco peixes (1,04 ± 0,20 g) foram expostos a cinco concentrações de NH3-N (0,60 a 1,47 mg L-1) durante 96 h mais um grupo controle em triplicata. Os testes foram conduzidos em um sistema semi-estático com renovação diária total dos meios experimentais. Os resultados foram apresentados como dados de mortalidade, registrados nas diferentes concentrações testadas, utilizando o software Trimmed Spearman Karber Method. A concentração letal (LC50) após 96 h de exposição a NH3-N foi estimada em 0,74 (0,61-0,90) mg L-1. Os resultados indicam que juvenis de sardinha-verdadeira são relativamente sensíveis à amônia em relação a outros peixes marinhos e atenção especial deve ser dada a este composto em sistemas de cultivo. Palavras-chave: amônia; LC50; toxicidade; peixes marinhos; sardinha-verdadeira Doi: 10.20950/1678-2305.2017v43n1p135


INTRODUCTION
Ammonia is the main nitrogenous excrement of teleost fish and is produced from the catabolism of dietary protein (EDDY, 2005;IP and CHEW, 2010).In the aqueous solution, ammonia exists in two forms: un-ionized ammonia (NH3-N) and ionized ammonium (NH4+-N), and their sum composes the total ammonia nitrogen (TAN).Ammonia is excreted through the gill epithelium either by passive diffusion as NH3-N along a gas partial pressure gradient between fish blood and water, or through active transport as NH4+-N in exchange for sodium (HARGREAVES andKUCUK, 2001, BARBIERI, 2010).The NH3-N/NH4+-N equilibrium is determined by pH and temperature (IP and CHEW, 2010).Ammonia toxicity is expressed by and depends mainly on the presence of NH3-N which is able to easily diffuse across gill membranes due to a lack of charge and its lipid solubility (TOMASSO 1994;BENLI and KÖKSAL 2005).Sub-lethal exposure to high levels of NH3-N can induce changes in behavioral, physiological and biological responses in fish, thus suppressing immunocompetence followed by hyperventilation, hyperexcitability, coma, convulsions and finally death (HARGREAVES and KUCUK, 2001;EDDY, 2005).Ammonia toxicity can be a crucial factor that leads to mass mortality under unfavorable aquacultural conditions, it is therefore important to determine the level supported by any species.Despite the existence of many studies on different fish species concerning acute ammonia toxicity (WEIRICH and RICHE, 2006a;WEIRICH and RICHE, 2006b;COSTA et al., 2008;BARBIERI and DOI, 2012;RODRIGUES et al., 2014), no studies were found on the effects of ammonia exposure on sardine species.
The Brazilian sardine Sardinella brasiliensis is a marine pelagic and schooling species that is the largest commercial fish resource along the southeastern Brazilian Bight (CERGOLE et al., 2002).It is currently captured by various fleets for use in the canning industry and extensively as live bait for skipjack tuna fishery, and its availability varies seasonally due to over-fishing and environmental changes (SANTOS and RODRIGUES-RIBEIRO, 2000).Commercial catches of the Brazilian sardine began in the late 1950s, and had quick growth in the 1960s, reaching a peak of 228 000 tons in 1973 (CERGOLE et al., 2002).
However, in 2011, production was only 75 223 tons (MPA, 2014).Considering its economic importance, one of the efforts that could avoid collapse of the Brazilian sardine population is its production in captivity (CERGOLE and DIAS-NETO, 2011).The "Universidade Federal de Santa Catarina", through the Live-Bait Project is developing actions for the production of Brazilian sardine in captivity.Research has been conducted on reproduction, nutrition and feed management (BALOI et al., 2016); however, there is no information about toxicity of nitrogenous compounds to Brazilian sardine.Therefore, this study was conducted to estimate the acute toxicity of ammonia (NH3-N) on juvenile Brazilian sardine.

Source of experimental fish
The study was carried out at the "Laboratorio de Piscicultura Marinha" (LAPMAR) of the "Universidade Federal de Santa Catarina" (UFSC, Brazil).Juvenile Brazilian sardine were produced at LAPMAR using a methodology described by BALOI et al. (2016).All animal handling procedures were approved by UFSC's Animal Ethics Committees (PP00861).

Acute toxicity assay
Groups of 5 fish (1.04 ± 0.20 g) were exposed to five concentrations of NH3-N ranging from 0.60 to 1.47 mg.L -1 for 96 h, plus a control group where no toxicant was added.The experiments were run in triplicate and test solutions were obtained with reagent grade ammonium chloride (Synth, Brazil).NH3-N concentrations were calculated using the equations of OSTRENSKY et al. (1992) adapted from WHITFIELD (1974) and BOWER and BIDWELL (1978) using the values of TAN, pH, temperature and salinity.The toxicity trial was conducted in 20 L plastic tanks, using a semi-static system where water and test solutions were fully exchanged daily (at 8 am), and food was withheld for 24 hours before and during the trial.
The water quality parameters in each experimental tank were measured daily (8:30 am).Temperature and dissolved oxygen were measured using an YSI Pro20 dissolved oxygen meter (Yellow Springs Instruments, Yellow Springs, USA), the pH was measured with a pH10A meter (Yellow Springs Instruments, Yellow Springs, USA) and salinity was measured using a portable refractometer (Instrutherm RTS-101ATC-03137, Sao Paulo, Brazil).TAN was determined 20 minutes after collection, following the method described by STRICKLAND and PARSONS (1972).Temperature was maintained at 19.2 ± 0.8°C, dissolved oxygen at 7.53 ± 0.04 mg L -1 , salinity at 33.2 ± 0.3 g L -1 and pH at 8.01 ± 0.06.Fish behavior was observed during the trial.Mortality was evaluated twice a day (8 am and 5 pm) and was presumed when fish were motionless on the bottom, exhibited no opercular movement and failed to respond to tactile stimuli.Dead fish were immediately removed from the tanks and cumulative mortalities were recorded every 24 h.

Statistical analysis
Cumulative mortality data were used to calculate median lethal concentration (LC50 with 95% confidence limits) and their respective confidence intervals (95%) using the Trimmed Spearman-Karber Different letters at each column indicate significant difference (P < 0.05) method (HAMILTON et al., 1977).Safe levels were obtained by multiplying 96-h LC50 values by a factor of 0.1 (SPRAGUE, 1971).Comparisons among median toxic lethal concentrations for NH3-N were made using one-way ANOVA followed by a Tukey Test with a significance level of 95%.

RESULTS AND DISCUSSION
The percentage of mortality, median lethal concentrations and the respective confidence intervals of TAN and NH3-N are presented in Table 1.No mortality, abnormal behavior or clinical signs were detected in the control group.The cumulative mortality increased with increasing ammonia concentrations.At the end of 96 h, all fish exposed to concentrations equal to or higher than 1.31 mgL -1 NH3-N died.
A significant increase in ammonia toxicity was observed with exposure time.Lethal concentration (LC50) values ranged from 1.06 mgL -1 of NH3-N (35.00 mg L -1 TAN) at 24 h of exposure to 0.74 mg L -1 NH3-N (25.62 mg L -1 TAN) at 96 h of exposure.
The toxicity of ammonia on fish have been studied in numerous fish species, however it is difficult to compare tolerance among species because it is dependent on the environmental conditions the trials were run (temperature, pH) and also on biological characteristics (weight, age, nutritional condition).Ammonia is acutely toxic for marine fish in the range of 0.54 mg L -1 for black sea bass Centropristis striata (WEIRICH and RICHE, 2006b) and at 2.93 mg L -1 of NH3-N for threespine stickleback Gasterosteus aculeatus (HAZEL et al., 1971), the most tolerant known species.
At concentrations lower than lethal concentrations, ammonia can have so-called "sub-lethal" effects, such as reduced growth, poor feed conversion and reduced disease resistance.SPRAGUE (1971) suggested that keeping the concentration of toxicant under 10% of the estimated acute toxicity should keep fish from sub-lethal symptoms.The safe level calculated for Brazilian sardine was estimated to be 2.53 and 0.07 mg L -1 for TAN and NH3-N respectively.
In our study, we observed that swimming activity reduced with increased ammonia concentrations and exposure time.Lethargic and decreased swimming activities, increased ventilation rate, loss of equilibrium, increased mucous secretion on the skin and gills and continuous swimming at the bottom of the tank are the first signs noticed as abnormal behavior caused by the exposure to nitrogenous compounds (HAMLIN, 2006;RODRIGUES et al., 2007;DAMATO and BARBIERI, 2011).
Ammonia is an environmental toxicant that is especially problematic for aquatic organism and it should be kept below the toxic levels for fish.The maintenance of water quality and knowledge of a species' tolerance limits for water quality are indispensable requirements for any cultivation system.Our results indicate that juvenile Brazilian sardine is relatively sensitive to ammonia compared to other marine fish and the safe levels presented here have important implications for the management of this specie in captivity.