Short-term effects of neuroactive pharmaceutical drugs on a fish species: biochemical and behavioural effects

The presence of pharmaceutical residues in the aquatic environment is receiving great attention since significant levels of contamination have been found, not only in sewage treatment plant effluents, but also in open waters. In our study, the toxicity of three anticonvulsant drugs commonly found in the environment (diazepam, carbamazepine, and phenytoin) was evaluated in Lepomis gibbosus (pumpkinseed sunfish). This study focused on oxidative stress parameters, namely: glutathione reductase (GRed), glutathione S-transferases (GSTs), catalase (CAT), and lipid peroxidation (thiobarbituric acid reactive substances, TBARS) in the hepatic, digestive, and gill tissues of exposed animals. Simultaneously, we assessed the effects of these drugs in terms of behavioural parameters, such as scototaxis and activity. Exposure to diazepam caused an increase in GST activities in the gills and an inhibition of GRed in the digestive tract, relative to control, suggesting an antioxidant response. It also caused fish to spend more time swimming and less time in a refuge area (black compartment of an aquarium). Exposure to carbamazepine caused an increase in GSTs and GRed activity in the digestive tract, which is not always consistent with the literature. A significant positive correlation was found between carbamazepine concentration and time spent in motion and a negative correlation with time spent in black compartment. Exposure to phenytoin was responsible for adaptive responses in the activities of CAT and GSTs (in the liver), but it did not elicit any behavioural alterations. Although all three drugs seemed to induce oxidative stress in some organs, peroxidative damage (measured as TBARS concentrations) was not found at the selected range of concentrations. Our results enlighten the need for more research on the ecological consequences of pharmaceuticals in the aquatic environment, especially drugs that interfere with the CNS and behaviour, because the net outcome of these effects may be difficult to predict. This work was partially funded by Fundação para a Ciência e Tecnologia, by project BiOtoMetal (PTDC/AMB/70431/2006). Bruno B. Castro is hired under the programme Ciência 2008 (FCT, Portugal), co-funded by the Human Potential Operational Programme (National Strategic Reference Framework 2007–2013) and European Social Fund (EU). None of the funding sources had any contribution for the experimental design adopted. published