High-throughput Brain Activity Mapping Based Drug Screening for Prediction of Neuropharmacology


Student thesis: Doctoral Thesis

View graph of relations



Awarding Institution
Award date27 Mar 2017


High-throughput screening (HTS) is a powerful technology in biomedical research and drug discovery, which enables efficient and quick testing a large number of phenotypes. Currently, most HTS assays are based on biochemical reactions, or cell cultures, but the results often fail to translate to animal models, especially for pharmaceutical development targeting neurological diseases. Even though HTS using whole organisms (e.g. zebrafish, C. elegans, Drosophila, etc.) becomes increasingly prevalent, the complexity for handling living small animals still restricts the assay throughput, and the phenotypes are typically limited to morphological or behavioral changes, which are not sufficient for evaluating drug leads due to the lack of readouts for direct analysis of physiological effects on the nervous system with both molecular and systematic resolutions.

This dissertation presents the development and applications of a microfluidic system that allows in vivo HTS using larval zebrafish. Specifically, to enable automatic manipulation, immobilization and orientation of zebrafish larvae in a gel-free and anesthetic-free manner, we developed a platform that consist of a series of microfluidic arrays and a robotic loading/dispensing component, so that we could perform real-time analysis in dorsal-up orientated larvae with a focus on their brain activity under the influence of acutely applied chemical or biological stimuli.

Of particular interest to nervous system, the technology was unitized to analyze the whole-brain activity of living larvae in an automatic and high-throughput format. This platform enables evaluation of a compound’s therapeutic potential with functional phenotyping based on the rich and sophisticated brain activity maps (BAMs) that are derived from drug-treated zebrafish larvae. From a blind screen of a library of 179 clinically used drugs, we found intrinsically coherent drug clusters that are significantly associated with the Anatomical Therapeutic Chemical classification. Using the BAM-based clusters as a functional classifier, we successfully predicted highly potent anti-epileptic candidate drugs from 121 non-clinical compounds, and also resolved novel insights into the pharmacology for the development of new anti-seizure drugs by targeting the epigenetic regulators.

We believe that this study provides an innovative and expandable technology for HTS using larval zebrafish and also a novel perspective for drug development without much prior art of compounds’ chemical or molecular basis.