Research projects

Alterations of the retinal pigment epithelium (RPE) frequently lead to loss of vision. The replacement of RPE cells is therefore a promising therapeutic strategy, although various technical obstacles make it difficult to perform. One of them is the poor knowledge of the gene regulatory networks (GRN) involved in the specification and maintenance of RPE. To fill this conceptual gap, we have used the zebrafish and applied comparative genomic analyzes, RNAseq and ATACseq to non-committed progenitors and cells of the neural retina (NR) and RPE. With this strategy, we have identified clusters of transcription factors, which mark the transcriptional states of RPE and NR over time. 

A first group acts very early promoting initial tissue specification. A second supports the acquisition of the main identity of the two tissues. The identification of differentially open chromatin regions between the two tissues supports this differential and stepwise process and further shows that RPE is more transcriptionally active than NR. Our data allow us to propose that the identity of the NR is more easily generated by "restricting" the fate of non-committed progenitors with minor adaptations of the GRN of the morphogenetic field of the eye. In contrast, the acquisition of RPE identity requires more dramatic transcriptomic and epigenomic changes to "modify" cell fate. We also observe that the developmental schedules of NR and RPE are mutually exclusive, with clusters of transcription factors exerting cross-repression. Using CRISPR / Cas9 mutagenesis for selected genes, we have validated some of the identified transcripts and the proposed model. In parallel, we have used a transgenic line that allowed us to identify RPE cells from their inception, demonstrating that the transition of RPE from neuroepithelium to a squamous one is necessary for the morphogenesis of the optic vesicle. The results we have obtained will allow the elaboration of better protocols to obtain RPE cells suitable for therapeutic use.