ABSTRACT All-trans-retinoic acid (RA) is the main physiologically active derivative of vitamin A, which serves as a ligand for nuclear transcription factors, RA receptors. During development, RA is produced in a quickly changing spatiotemporal pattern to control the expression of precise sets of genes at different developmental stages. Critical RA-sensitive processes during development are RA-concentration dependent, which underscores the importance of the precise control over RA synthesis in a strictly defined and rapidly regulated manner. Biosynthesis of RA includes reversible rate-limiting oxidation of retinol to all-trans-retinaldehyde, followed by irreversible oxidation of all-trans-retinaldehyde to RA. Multiple studies examined the roles of the enzymes catalyzing the oxidation of retinaldehyde and degradation of RA in establishing the dynamic pattern of RA concentration. However, the mechanism regulating the upstream rate-limiting step, which supplies the immediate RA precursor, retinaldehyde, in a precise spatiotemporal pattern remains unknown. It has been established that two proteins, retinol dehydrogenase 10 (RDH10) and short-chain dehydrogenase/reductase 3 (DHRS3), are critical for the control of retinaldehyde levels during development. We have recently discovered that DHRS3 binds to RDH10 and upon binding reduces the output of retinaldehyde by RDH10 by recycling retinaldehyde back to retinol. As a result, the formation of the bifunctional retinoid oxidoreductase complex (ROC) that consists of an oxidative RDH10 and reductive DHRS3 attenuates the RA biosynthesis. Whether this mechanism works in vivo and whether ROC exists in animal tissues is unknown, but if proven to be true, this finding will have a paradigm-shifting effect on our understanding of the mechanisms that regulate embryogenesis through vitamin A. The major hypothesis driving this proposal is that ROC represents a previously unrecognized universally conserved mechanism that can both provide the RA synthesis with robustness (Aim 1) and enable the dynamic changes in RA spatiotemporal pattern by regulating the levels of RA precursor (Aim 2). The hypothesis will be tested using a zebrafish embryogenesis model to take advantage of external fertilization and transparency of zebrafish for intra-vital visualization of RA synthesis and formation of the complex. Successful completion of these studies will advance the field at the conceptual level by demonstrating a mechanistically novel model of producing strictly controlled spatiotemporal gradients of small molecules. These findings will lay the foundation for a better understanding of the mechanisms of congenital diseases associated with dysregulation of RA homeostasis.