CRX directs photoreceptor differentiation by accelerating chromatin remodeling at specific target sites

Recent technological advances have delivered the genome-wide targets of many important transcription factors (TFs). However, increasing evidence suggests that not all target sites mediate regulatory function, raising the questions of how to determine which sites are active, what are the epigenetic consequences of TF binding at these sites, and how the specificity is coded. To address these questions, we focused on CRX, a disease-associated homeodomain TF required for photoreceptor gene expression and development. Since CRX binds more than 6000 sites across the genome in the retina, we profiled chromatin landscape changes at each binding site during normal development and in the absence of CRX and interpreted the

Together, our results suggest that CRX acts only at select, uniquely-coded binding sites to accelerate chromatin remodeling during photoreceptor differentiation. This study emphasizes the importance of connecting TF binding with its functional consequences and provides a framework for making such a connection using comparative analyses of available genomic datasets. Finally, this study prioritizes sets of non-coding DNA sites for future functional interrogation and identification of mutations associated with retinal disease.

CRX is required for chromatin remodeling at only a subset of its binding sites, which undergo retina or neuronal specific activation during photoreceptor differentiation. Genes near these "CRX Dependent" sites code for proteins important for photoreceptor physiology and function, and their transcription is significantly reduced in Crx deficient retinas. In addition, the nucleotide and motif content distinguish these CRX Dependent sites from other CRX-bound sites. Conclusions: Together, our results suggest that CRX acts only at select, uniquely-coded binding sites to accelerate chromatin remodeling during photoreceptor differentiation. This study emphasizes the importance of connecting TF binding with its functional consequences and provides a framework for making such a connection using comparative analyses of available genomic datasets. Finally, this study prioritizes sets of non-coding DNA sites for future functional interrogation and identification of mutations associated with retinal disease.