In vertebrate retinal photoreceptors, the absorption of light by rhodopsin leads to photoisomerization of 11-cis-retinal to its all-trans isomer. To sustain vision, a metabolic system evolved that recycles all-trans-retinal back to 11-cis-retinal. The importance of this visual (retinoid) cycle is underscored by the fact that mutations in genes encoding visual cycle components induce a wide spectrum of diseases characterized by abnormal levels of specific retinoid cycle intermediates. In addition, intense illumination can produce retinoid cycle by-products that are toxic to the retina. Thus, inhibition of the retinoid cycle has therapeutic potential in physiological and pathological states. Four classes of inhibitors that include retinoid and nonretinoid compounds have been identified. We investigated the modes of action of these inhibitors by using purified visual cycle components and in vivo systems. We report that retinylamine was the most potent and specific inhibitor of the retinoid cycle among the tested compounds and that it targets the retinoid isomerase, RPE65. Hydrophobic primary amines like farnesylamine also showed inhibitory potency but a short duration of action, probably due to rapid metabolism. These compounds also are reactive nucleophiles with potentially high cellular toxicity. We also evaluated the role of a specific protein-mediated mechanism on retinoid cycle inhibitor uptake by the eye. Our results show that retinylamine is transported to and taken up by the eye by retinol-binding protein-independent and retinoic acid-responsive gene product 6-independent mechanisms. Finally, we provide evidence for a crucial role of lecithin: retinol acyltransferase activity in mediating tissue specific absorption and long lasting therapeutic effects of retinoid-based visual cycle inhibitors.