Modeling of stress and anxiety in adult zebrafish (Danio rerio) is increasingly utilized in neuroscience research and central nervous system (CNS) drug discovery. Representing the most commonly used zebrafish anxiety models, the novel tank test (NTT) focuses on zebrafish diving in response to potentially threatening stimuli, whereas the light-dark test (LDT) is based on fish scototaxis (innate preference for dark vs. bright areas). Here, we systematically evaluate the utility of these two tests, combining meta-analyses of published literature with comparative in vivo behavioral and whole-body endocrine (cortisol) testing. Overall, the NTT and LDT behaviors demonstrate a generally good cross-test correlation in vivo, whereas meta-analyses of published literature show that both tests have similar sensitivity to zebrafish anxiety-like states. Finally, NTT evokes higher levels of cortisol, likely representing a more stressful procedure than LDT. Collectively, our study reappraises NTT and LDT for studying anxiety-like states in zebrafish, and emphasizes their developing utility for neurobehavioral research. These findings can help optimize drug screening procedures by choosing more appropriate models for testing anxiolytic or anxiogenic drugs.

The presence of pharmaceutical products in the aquatic environment has been reported in several studies. However, the impact of these drugs on living organisms is still uncharacterized. Here, we investigated the effects of acute exposure to either diazepam or fluoxetine on the stress response in Danio rerio. We showed that diazepam and fluoxetine inhibited the stress axis in zebrafish. Intermediate concentrations of diazepam suppressed the stress response as measured by cortisol levels, whereas fluoxetine inhibited cortisol increase at concentrations similar to those found in the environment. These data suggest that the presence of psychoactive drugs in aquatic ecosystems could cause neuroendocrine dysfunction in fish.