The zebrafish (Danio rerio) anxiety test battery: comparison of behavioral responses in the novel tank diving and light–dark tasks following exposure to anxiogenic and anxiolytic compounds

RATIONALE: Triangulation of approaches (i.e., using several tests of the same construct) can be extremely useful for increasing the robustness of the findings being widely used when working with behavioral testing, especially when using rodents as a translational model. Although zebrafish are widely used in neuropharmacology research due to their high-throughput screening potential for new therapeutic drugs, behavioral test battery effects following pharmacological manipulations are still unknown. METHODS: Here, we tested the effects of an anxiety test battery and test time following pharmacological manipulations in zebrafish by using two behavioral tasks: the novel tank diving task (NTT) and the light–dark test (LDT). Fluoxetine and conspecific alarm substance (CAS) were chosen to induce anxiolytic and anxiogenic-like behavior, respectively. RESULTS: For non-drug-treated animals, no differences were observed for testing order (NTT → LDT or LDT → NTT) and there was a strong correlation between performances on the two behavioral tasks. However, we found that during drug treatment, NTT/LDT responses are affected by the tested order depending on the test time being fluoxetine effects higher at the second behavioral task (6 min later) and CAS effects lower across time. CONCLUSIONS: Overall, our data supports the use of baseline behavior assessment using this anxiety test battery. However, when working with drug exposure, data analysis must carefully consider time-drug-response and data variability across behavioral tasks.