DEVELOPMENT OF A HIGH-THROUGHPUT IN VIVO DRUG SCREENING ASSAY USING A FSGS-LIKE ZEBRAFISH MODEL

Abstract BACKGROUND AND AIMS Focal segmental sclerosis (FSGS) is a severe histopathological condition which often results in end-stage renal disease. Loss of podocytes and severe changes of podocyte foot process morphology accompanied by proteinuria are the hallmarks of FSGS. Since no therapeutic agents are available until now, further efforts must be made to identify new drugs to treat this disease. The zebrafish larva is a widely used kidney research model as the morphology and function of zebrafish and mammalian glomeruli are very similar. Since our group has already established a FSGS-like zebrafish injury model based on the nitroreductase-metronidazole cell ablation system, we have developed a high-throughput based in vivo drug screening assay. METHOD We used a zebrafish screening strain that expresses mCherry and the bacterial enzyme nitroreductase in podocytes. Additionally, a circulating vitamin-D-binding eGFP fusion protein with a size of 78 kDa is expressed and serves as a read-out for proteinuria. Incubation with 80 µM metronidazole for 24 h at 4 days post-fertilization (dpf) induces an FSGS-like phenotype in these larvae. A library of 138 epigenetic drugs were screened and the results are presented herein. Groups of 12 larvae received injury induction as well as drug treatment and were transferred individually into 96-well plates prepared with custom agarose molds. Larvae were oriented laterally and imaged with a high-content screening microscope (Acquifer Imaging Machine). The eGFP signal in the caudal vasculature as well as the glomerular mCherry signal served as readout systems. At 6 dpf, the same larvae and regions were imaged again. Custom written FIJI codes automatically segmented and measured the fluorescence intensities of both readouts and timepoints. For each larva, intensity ratios were determined and the ratios of each treatment group served as the input for statistical analysis. RESULTS Treatment for 24 hours with 80 µM metronidazole resulted in a measurable decline of the eGFP signal in the vasculature and of the podocyte mCherry signal compared to the vehicle (DMSO) group. Simultaneous treatment of metronidazole with the 138 drugs in the epigenetics library resulted in beneficial outcomes in varying degrees (9 drugs were positive, 36 compounds were lethal). CONCLUSION This new in vivo semi-automated high-throughput compatible drug screening assay allows rapid identification of FSGS protective drugs. Here we screened 138 epigenetic drugs, which led to the identification of 9 potential drugs which have had a beneficial effect on podocytes and/or the glomerular filtration barrier.