Abstract

The aim of this research was to evaluate neotropical fish sensitivity (Piaractus mesopotamicus, Phallocerus caudimaculatus, Hyphessobrycon eques, and Brachydanio rerio) to a reference substance (potassium chloride); to estimate the lethal concentration (LC₅₀; 96 h) for glyphosate, formulated as Rodeo^® alone and in association with 0.5 and 1.0% Aterbane^® BR surfactant and to evaluate the histopathology of the gills, liver, and kidney from the fish after acute exposure. P. caudimaculatus and H. eques are good bioindicators like B. rerio because they have similar sensitivity. The LC₅₀;96 h for glyphosate alone and in association with 0.5% Aterbane^® BR was similar (>975.0 mg L^-1) for all the fish. Aterbane^® BR alone was the most toxic substance to P. caudimaculatus (5.81 mg L^-1 LC₅₀;96 h) and glyphosate associated to 1.0% Aterbane^® BR was more toxic to H. eques (411.91 mg L^-1 LC₅₀;96 h). The glyphosate alone and in association with Aterbane^® BR was classified as practically non-toxic, whereas Aterbane^® BR alone was considered moderately toxic for the tested organisms. The histopathological effects caused by glyphosate exposure on gills, liver, and kidneys are reversible, except for the liver necrosis on P. caudimaculatus. H. eques, P. caudimaculatus, and P. mesopotamicus present great potential to be used as standard organisms for herbicides monitoring and the use of glyphosate without surfactant addition is enough to cause histological alterations on H. eques and P. caudimaculatus, which makes them possible to be applied on environmental monitoring studies as biomarkers.

Key words: Bioindicator, pesticide, histology, biomarker, gill, herbicide.