The zebrafish model has emerged as a reference tool for phenotypic drug screening. An increasing number of molecules have been brought from bench to bedside thanks to zebrafish-based assays over the last decade. The high homology between the zebrafish and the human genomes facilitates the generation of zebrafish lines carrying loss-of-function mutations in disease-relevant genes; nonetheless, even using this alternative model, the establishment of isogenic mutant lines requires a long generation time and an elevated number of animals. In this study, we developed a zebrafish-based high-throughput platform for the generation of F0 knock-out (KO) models and the screening of neuroactive compounds. We show that the simultaneous inactivation of a reporter gene (tyrosinase) and a second gene of interest allows the phenotypic selection of F0 somatic mutants (crispants) carrying the highest rates of mutations in both loci. As a proof of principle, we targeted genes associated with neurodevelopmental disorders and we efficiently generated de facto F0 mutants in seven genes involved in childhood epilepsy. We employed a high-throughput multiparametric behavioral analysis to characterize the response of these KO models to an epileptogenic stimulus, making it possible to employ kinematic parameters to identify seizure-like events. The combination of these co-injection, screening and phenotyping methods allowed us to generate crispants recapitulating epilepsy features and to test the efficacy of compounds already during the first days post fertilization. Since the strategy can be applied to a wide range of indications, this study paves the ground for high-throughput drug discovery and promotes the use of zebrafish in personalized medicine and neurotoxicity assessment.

Allele number, or zygosity, is a clear determinant of gene expression in diploid cells. But the relationship between the number of copies of a gene and its expression can be hard to anticipate, especially when the gene in question is embedded in a regulatory circuit that contains feedbacks. Here we study this question making use of the natural genetic variability of human populations, which allows us to compare the expression profiles of a receptor protein in natural killer cells between donors infected with human cytomegalovirus (HCMV) with one or two copies of the allele. Crucially, the distribution of gene expression in many of the donors is bimodal, indicative of the presence of a positive feedback somewhere in the regulatory environment of the gene. Three separate gene-circuit models differing in the location of the positive feedback with respect to the gene can all reproduce well the homozygous data. However, when the resulting fitted models are applied to the hemizygous donors, only one model (the one with the positive feedback located at the level of gene transcription) reproduces the experimentally observed gene-expression profile. In that way, our work shows that zygosity can help us relate structure and function of gene regulatory networks. Author Summary: Nearly all mammalian cells, including human cells, have two copies of each chromosome, and thus possess two potentially different copies of each gene (which might be in some cases non-functional or even absent). Naïively one might expect that two identical copies of the gene would lead to the protein being expressed at twice the rate, but many factors can alter this simple calculation. One of these factors is the existence of feedback mechanisms affecting in one way or another the regulatory circuit in which our gene of interest is embedded. Here we study the relationship between the number of gene copies and the expression of a receptor protein that plays a crucial role in the recognition of pathogens by natural killer cells, which are important elements of the innate immune system. Experimental data of virus-infected donors reveals a bimodal expression profile of this receptor, typical of a positive feedback, and a clear difference between donors with one or two copies of the gene. Mathematical modeling allows us to find the likely location of the feedback loop within the gene’s regulatory circuit, by requiring the correct model to reproduce the expression profiles of both types of donors.