The effect of retinoic acid (RA) on endothelial cells is still controversial and was examined in the present study. In bovine aortic endothelial cells (BAECs), all-trans RA (ATRA) and 9-cis RA (9CRA), but not 13-cis RA (13CRA), induced fibroblast growth factor-2 (FGF-2) production and exhibited a biphasic dose-dependent effect to enhance BAEC proliferation and differentiation into tubular structures on reconstituted basement membrane proteins (Matrigel); both processes were inhibited by FGF-2-neutralizing antibody. The pan RA receptor (RAR)-selective ligand (E)-4-[2-(5,5,8,8,-tetramethyl-5,6,7,8-tetrahydro-2-naphtalenyl)-1-propenyl] benzoic acid and the RARalpha-selective ligand 4-[1-(3,5,5,8,8-pentamethyl-5,6,7,8-tetrahydro-2-naphtyl)-ethenyl] benzoic acid stimulated the production of FGF-2, whereas the addition of the RARalpha-antagonist RO 41-5253 inhibited this effect. In BAECs, the forced expression of RARalpha, but not RARbeta or RARgamma, enhanced FGF-2 production, whereas the RARalpha-dominant negative, Delta403, blocked this effect. Furthermore, RARalpha overexpression directly stimulated BAEC differentiation on Matrigel and potentiated the effects of ATRA in this assay. Finally, ATRA-treated BAECs coinjected with Matrigel subcutaneously in mice induced neovascularization within the Matrigel plug, and ATRA also enhanced angiogenesis in the chicken chorioallantoic membrane assay. In conclusion, RA can stimulate endothelial cell proliferation and differentiation in vitro via enhanced RARalpha-dependent FGF-2 production, and it can also induce angiogenesis in vivo. The full text of this article is available at http://www.circresaha.org.