Physiological occurrence, biosynthesis and metabolism of retinoic acid: evidence for roles of cellular retinol-binding protein (CRBP) and cellular retinoic acid-binding protein (CRABP) in the pathway of retinoic acid homeostasis.

This article will address recent work on the physiological occurrence, biogenesis and metabolism of retinoic acid and summarize the data that retinoic acid is synthesized in situ in multiple tissues and cell types via enzymes or enzyme complexes that are distinct from the alcohol dehydrogenases. There is now considerable evidence that retinoic acid is an activated metabolite of retinol that supports the systemic functions of vitamin A in vivo. Many studies in vitro, for example, have shown that retinoic acid is the most potent naturally-occurring retinoid with an ED-50 in the range of 1 pM to 10 nM, depending on the assay system. This is below the tissue concentrations of retinoic acid which range from approximately 20-600 nM. Retinoic acid synthesis from retinol in the dog kidney cell line MDCK maintained in serum-free medium is inhibited by the prostanoid, PGE, and the phorbol ester, TPA. In tissues, one pathway of retinoic acid synthesis begins with apo-CRBP stimulating retinyl ester hydrolysis by a microsomal, cholate-independent retinyl ester hydrolase to form holo-CRBP. The holo-CRBP itself is used as substrate by an NADP-dependent, microsomal retinol dehydrogenase to generate retinal, which is converted into retinoic acid by a cytosolic NAD-dependent retinal dehydrogenase. Therefore, cellular retinol-binding protein (CRBP) apparently has at least 2 functions in retinoic acid synthesis: the apo form stimulates retinol mobilization from retinyl ester stores; the holo form delivers the retinol via direct transfer to dehydrogenase(s). Retinoic acid is converted into a mixture of at least 4 metabolites by testes microsomes which migrate closely on reverse-phase HPLC with 4-hydroxyretinoic acid, and may be mistaken for either 4-hydroxy or 4-oxo-retinoic acid. More rigorous analysis, however, shows that only one of them is 4-hydroxyretinoic acid, and another is 18-hydroxyretinoic acid. Two others remain unidentified. These metabolites are also formed in the presence of excess cellular retinoic acid-binding protein (CRABP), which increases the elimination half-life of retinoic acid, but does not prevent retinoic acid catabolism, suggesting that holo-CRABP may be a substrate for retinoic acid catabolism that modulates the steady-state concentrations of retinoic acid. Thus, both retinoid binding proteins, CRBP and CRABP, may each have direct roles as substrate in the biosynthesis and metabolism of retinoic acid, respectively.