Validation of a larval zebrafish locomotor assay for assessing the seizure liability of early-stage development drugs.

INTRODUCTION Seizure liability is an adverse property of new candidate drugs typically detected only in later stage pre-clinical safety studies. Consequently, pharmaceutical discovery needs small scale (microplate-based), rapid throughput screens to 'front-load' such adverse endpoints in order to reduce associated attrition rates. Of the potential methods available, previously published studies have shown that the quantification of seizure-associated locomotion in the larval zebrafish (Danio rerio) offers high potential for development into such a screen. Here, we present methodology and validation data (on 25 compounds) from a larval zebrafish (Zf) convulsant assay, based on the quantification of high speed locomotion after exposure to a range of test compounds. METHODS All assays were undertaken in 7 days post fertilization (dpf), WIK-strain Zf larvae, at 27+/-1 degrees C. The blinded validation test set consisted of 17 positive and 8 negative controls, based on literature evidence for seizure liability. Initially, a Maximum Tolerated Concentration (MTC) assay was undertaken on each compound to identify the maximum concentration not causing general toxicity, sedation or overt neuromuscular effects. Next, the convulsant assay was undertaken on 5 concentrations from the MTC down, plus a dilution water control. Exposed larvae were videotracked for 1 h, using the Viewpoint Videotrack for Zebrafish system, and high speed movements, typically associated with seizure-like locomotor activity, were quantified. RESULTS According to classification criteria proposed by the European Centre for the Validation of Alternative Methods (ECVAM), the data generated appeared to offer "sufficient" predictivity (72% overall), particularly considering the potential for throughput and likely positioning within a safety pharmacology front-loading screening cascade. DISCUSSION Possible reasons for the misclassifications are discussed, and potential improvements to increase sensitivity and specificity outlined. In all, these initial validation data suggest that this assay offers potential as a medium throughput screen aimed at the early drug discovery detection of this complex safety pharmacological endpoint.