PROJECT SUMMARY One of the most critical functions of the vitamin A (retinol) metabolite retinoic acid (RA) is control of eye development. In mouse, RA synthesis occurs early in the optic field with expression of retinol dehydrogenase-10 (RDH10) and all three retinaldehyde dehydrogenases (ALDH1A1, ALDH1A2, ALDH1A3) that convert retinol to RA. RA diffuses to tissues throughout the optic placode, optic vesicle, and adjacent mesenchyme to stimulate folding of the optic vesicle to form the optic cup by E10.5. At E12.5-E14.5, RA is needed for further morphogenesis of the optic cup and surrounding perioptic mesenchyme; loss of RA leads to microphthalmia. RA functions by binding to nuclear RA receptors at RA response elements (RAREs) that either activate or repress transcription of key genes. Binding of RA to RA receptors regulates recruitment of transcriptional coregulators such as nuclear receptor coactivator (NCOA) and nuclear receptor corepressor (NCOR), which in turn control binding of the generic coactivator p300 and the generic corepressor PRC2. However, a major unsolved problem is what are the key genes controlled by RA during development of the eye; only two candidate direct RA target genes are known (Pitx2 and Foxc1). As loss or gain of RA activity alters expression of thousands of genes (perhaps many due to post-transcriptional effects), it remains difficult to identify genes that are direct transcriptional targets of RA. In our Preliminary Studies we addressed this question by comparing ChIP-seq and RNA-seq for tissues from Aldh1a2-/- embryos lacking RA synthesis, thus identifying genes with altered expression when RA is missing that also have nearby RA-regulated deposition of H3K27ac (gene activation mark) or H3K27me3 (gene repression mark) associated with RAREs. Such RARE enhancers/silencers were identified near genes already known to be required for embryonic development, thus validating our approach. CRISPR knockouts for several predicted new direct RA target genes verified their requirements for development. Here, we plan to use this approach to identify RA target genes and PITX2 target genes during eye development by comparing ChIP-seq (H3K27ac & H3K27me3) and RNA-seq for wild-type vs RA-deficient optic vesicle and eye, and wild-type vs Pitx2 knockout eye. We will also identify RA-regulated enhancers and silencers in the eye to uncover the mechanisms through which RA regulates Pitx2, Foxc1, or other genes. Our studies will provide vital information on the mechanisms utilized by RA and PITX2 to control transcription in the eye and will identify gene regulatory networks during eye formation. This knowledge will help determine how eye defects occur, identify new genes or enhancers/silencers that may be mutational targets causing human eye defects, and improve strategies to treat eye defects.