Gene activation and repression by retinoid receptors during vertebrate embryogenesis

Retinoic acid is important for many aspects of vertebrate development, including patterning the neural tube, retina, limb, and heart. Localization studies do not provide specific information about where retinoic acid and the retinoid receptors work together to activate transcription. I have addressed this problem by using retinoic acid response elements to drive fluorescent reporter gene expression in zebrafish embryos. Presumably a fluorescent signal will be produced where retinoid receptors and retinoic acid colocalize. The resulting transgenic zebrafish lines express the transgene in the notochord, neural tube, retina, pronephric ducts, somites, and heart in spatially and temporally regulated patterns, consistent with predicted activity of retinoic acid. Treatment with a retinoic acid synthesis inhibitor results in the elimination of the transgene expression. The inhibitor, diethylaminobenzaldehyde, also phenocopied the effects of retinoic acid deprivation, particularly those seen in retinaldehyde dehydrogenase-2 mutants. Together these data indicate regions where retinoic acid activates transcription and is required for normal patterning. I have also studied retinoid receptor-meditated repression by cloning a corepressor that is orthologous to the mammalian corepressor, silencing mediator of retinoic acid and thyroid hormone receptors. This molecule interacted with zebrafish retinoic acid and thyroid hormone receptors, and like mammalian corepressors, addition of ligand eliminated the interaction. The zebrafish corepressor was expressed in a restricted pattern, primarily the anterior nervous system and notochord. This expression domain is complementary to regions of active retinoid signaling. These data indicate that where expressed the corepressor can mediate repression because retinoid receptors are available, but retinoic acid is not. The importance of repression and absence of retinoic acid in these tissues is supported by the sensitivity of these regions to excessive doses of retinoic acid. Combined the aspects of this dissertation confirm restricted regions of retinoic acid signaling exist in vertebrate embryos, and are essential for normal development. The corepressor expression pattern indicates regions that lack retinoic acid might be important for repression by retinoic acid receptors. The transgenic line, inhibitor and corepressor clone all represent tools for enhancing our understanding of the positive and negative effects of retinoic acid receptors on gene transcription.